Specific Antagonism of Type I IL-4 Receptor with a Mutated Form of Murine IL-4¹

Interleukin-4 is a multifunctional cytokine produced by different cells of hemopoietic origin, such as T cells, mast cells, and basophils (reviewed in 1 . IL-4 stimulates the proliferation of B cells, enhances the expression of cell surface molecules such as MHC class II and the low affinity Fc receptor for IgE (CD23), and is, at least in mice, indispensable for the Ig heavy chain class switching to IgE (reviewed in 2 . Ag presentation as well as the microbicidal effects of cells of the monocytic lineage are enhanced by IL-4, while the production of proinflammatory cytokines is down-regulated in the presence of IL-4 (reviewed in 3 . IL-4 is absolutely essential for the development of Th2 cells as demonstrated by several experimental approaches in a variety of infectious disease models (4, 5, 6, 7, 8). As shown for the first time in the model of murine cutaneous leishmaniasis, IL-4-mediated induction of Th2 cells in susceptible strains of mice is responsible for severe disease due to the induction of cytokines leading to an insufficient activation of the leishmanicidal functions of macrophages (reviewed in 9 . The fact that transgenic mice overexpressing IL-4 develop allergy-like diseases and autoimmunity is another example of a pathophysiologic role of IL-4 (10). In line with these experimental findings, in allergic diseases of humans there is a strong correlation of the frequencies of IL-4-producing Th cells and serum IgE concentrations (reviewed in 11 .

The pleiotropic activities of IL-4 are mediated by high affinity receptor(s) (dissociation constant (K_d) = 20–300 pM) that are expressed in relatively low numbers (100–5000) on a variety of cell types (12). A majority of cells express the 140 kDa IL-4R α-chain in association with the common γ chain (γc chain)⁴ (13), which is necessary, at least in T cells, for IL-4 signal transduction (14). The trimolecular complex initiates the signal transduction by activation of Janus kinase 1 (JAK1) and STAT6 via the IL-4R α-chain and JAK3 via the γc chain (summarized in 15 , respectively. More recently, IL-4 was demonstrated to induce functions on cells that express the IL-4R α-chain but lack the γc chain (16, 17). These observations indicate that there are at least two classes of IL-4R: type I IL-4R, containing the IL-4R α-chain and the γc chain; and type II IL-4R, consisting of the IL-4R α-chain and IL-13-binding protein recently identified and cloned by two research groups independently and which associate with the IL-4R α-chain (18, 19). Several experimental findings support the hypothesis that these IL-13R molecules are essential components of the type II IL-4R complex in the sense of cytokine binding as well as signaling. First, a point-mutated human IL-4 (IL-4.Y124D) displayed cross-competitive activity for both human IL-4 and IL-13 (20). Second, blocking Abs against the human IL-4R α-chain also inhibited the effects of IL-13 (21). Third, IL-13 as well as IL-4 was able to activate STAT6 in transfected CHO cells expressing both IL-4R α- and IL-13R α-chains, while no activation was observed in cells expressing one of these receptor molecules alone (22).

Only two reports have been published so far concerning murine IL-4 mutants. Morrison and Leder (23) analyzed the receptor-binding domain of mouse IL-4 using a solid-phase binding assay and in vitro mutagenesis. They reported that the 16 amino-terminal residues and the 20 carboxyl-terminal residues are required for species-specific receptor binding. Not surprisingly, IL-4 mutants no longer able to bind to the receptor were also unable to induce T cell proliferation. More recently, Grunewald et al. (24) expressed and characterized a double mutant (IL-4.Q116D.Y119D) that completely antagonized all IL-4 functions tested, including the proliferation of T and B cells as well as the up-regulation of CD23 on splenic B cells. This mutant, therefore, resembles complete antagonists of human IL-4 previously reported independently by two groups (20, 25). The generation of a human IL-4 variant, by replacing tyrosine 124 with aspartic acid, led to an antagonist that inhibited T and B cell proliferation (26) as well as IL-4- or IL-13-induced IgG4 and IgE synthesis (27) and CD23 expression (26). Biochemical and structural evidence from computer modeling of human IL-4R have shown that the site on human IL-4 defined by Y124 and R121 is required for the interaction with the human γc chain. In this study, we investigated the effect of a mutated form of murine IL-4 with regard to wild-type IL-4 functions that have not yet been studied, e.g., the expression of soluble IL-4R (sIL-4R) and the enhancement of parasite killing by infected macrophages. Additionally, we tested whether a differential effect on distinct IL-4 functions can be achieved by introducing a single point mutation into mouse IL-4. The observations strongly suggest that the murine IL-4.Y119D mutant is a type I IL-4R-specific antagonist.

Female mice of the inbred strain BALB/c were obtained from Charles River Breeding Laboratories (Sulzfeld, Germany). IL-4−/− mice (Sv129 × C57BL/6) (8) as well as sex- and age-matched control animals were a generous gift from Dr. M. Kopf, Basel Institute for Immunology (Basel, Switzerland). TNFR55−/− mice (28) as well as sex- and age-matched control animals were kindly provided by Dr. K. Pfeffer (Munich, Germany).

Leishmania major promastigotes of the strain MHOM/IL/81/FEBNI were grown in vitro in blood agar cultures as previously described (29). Stationary phase promastigotes were washed in PBS and were added to the macrophage cultures, as indicated below.

Recombinant murine IL-4 (LPS content <100 pg/μg protein) was purchased from ICChemikalien (Ismaning, Germany) and recombinant murine IFN-γ (LPS content <100 pg/μg protein) was from R&D Systems (Wiesbaden, Germany). IL-2 was obtained from supernatants of transfected X63Ag8-653 cells (30). Anti-murine IL-4 mAbs, BVD4-1D11, and biotinylated BVD6-24G2 for the measurement of IL-4 in cell culture supernatant by capture ELISA were obtained from PharMingen (Hamburg, Germany), used as previously described (6) and as recommended by the manufacturer. Purified rat anti-mouse CD23-biotin mAb and rat anti-mouse I-A^d-FITC mAb were purchased from PharMingen. Purified rat anti-mouse IL-4 mAb 11B11 was obtained from Dianova (Hamburg, Germany), and Polymyxin B was from Sigma (Deisenhofen, Germany). Purified recombinant murine and human sIL-4R were kindly provided by Dr. Seiler (Behringwerke, Marburg, Germany). Streptavidin-FITC was purchased from Boehringer Mannheim (Mannheim, Germany).

The cDNA encoding murine IL-4.Y119D was generated by mismatch-PCR using murine IL-4 cDNA as template to introduce a point mutation leading to the tyrosine 119 to aspartic acid replacement. For expression in Escherichia coli, the sense primer 5′-CCGAATTCCATATCCACGGATGCGACAAAAAT-3′ and the antisense primer 5′-GGCTCAGTACTACGAGTCATCCAT-3′ were used. The resulting PCR fragment was purified by gel electrophoresis, digested with RsaI and SmaI, and cloned into His tag expression plasmid pQE30 (Qiagen, Hilden, Germany), which had been linearized with RsaI and SmaI. The resulting plasmid, pQE30-IL-4.Y119D, was transformed into the E. coli strain C600 (Stratagene, Heidelberg, Germany). To analyze the expression of IL-4.Y119D protein, transformed bacterial cells were induced with 2 mM isopropyl-β-d-thiogalactoside for 1 to 5 h, and total bacterial proteins were analyzed on 15% SDS-PAGE. Optimized conditions of isopropyl-β-d-thiogalactoside induction (3 h) were used for preparation of large amounts of rIL-4.Y119D. The bacterial cell pellets were mixed thoroughly in 2 volumes of homogenization buffer (50 mM Tris-HCl (pH 7.4)/10 mM MgCl₂/0.2 M KCl/5% glycerol) and passed through a French press (SLM Aminco, Rochester, NY). The cell homogenates were centrifuged at 20.000 × g for 30 min at 4°C. The pellets were washed in 50 mM Tris-HCl (pH 7.4)/10 mM MgCl₂ and treated with 6 M guanidine-HCl/0.1 M NaH₂PO₄/0.01 M Tris (pH 8.0) for 4 h at room temperature. The extracted IL-4.Y119D was affinity purified over Ni-NTA-resin (Qiagen) and eluted by reducing the pH, as recommended by the manufacturer. The denatured protein was refolded by slowly adding 9 volumes of 50 mM Tris-HCl (pH 7.4)/50 mM NaCl/0.005% Tween 20/2 mM reduced glutathione and 0.2 mM oxidized glutathione and incubated at room temperature for 4 h. The refolded protein was bound to a Hi-Trap affinity column (Pharmacia Biotech, Freiburg, Germany) coupled with anti-murine IL-4 mAb, 11B11, eluted with 100 mM glycine (pH 2), and dialyzed against PBS (pH 7.4).

For expression in a eukaryotic system, the following primers were used: sense primer, 5′-TCTCAACCCCCAGCTAGTTGTCAT-3′; and antisense primer, 5′-CCGAATTCCATATCCACGGATGCGACAAAAAT-3′. The PCR fragment was digested with BamHI and cloned into the BamHI linearized eukaryotic expression plasmid pCEP4 (Invitrogen, Leek, The Netherlands). The episomal pCEP4-IL-4.Y119D expression plasmid was transfected into the human embryonic kidney expression cell 293 EBNA (Invitrogen) by electroporation. Transfected cells selected by adding 300 μg/ml hygromycin B constitutively produced IL-4.Y119D. All cloned cDNAs were sequenced using the Dye Terminator Cycle Sequencing Ready Reaction Kit (Perkin-Elmer/Applied Biosystems, Warrington, U.K.) as recommended by the manufacturer.

Purified human or murine rIL-4R was immobilized onto microtiter plates at a concentration of 5 μg/ml. After blocking nonspecific binding sites with FCS, serially diluted IL-4 or IL-4.Y119D was added to the wells, and the receptor-bound cytokine molecules were detected as described for the IL-4 ELISA applying the biotinylated mAb BVD6-24G2 (PharMingen).

Recombinant murine IL-4 or IL-4.Y119D was radiolabeled by the Iodogen (Pierce, Oud Bijerlad, The Netherlands) method as described by the manufacturer and by Lowenthal et al. (12). Briefly, 5 μg of purified cytokine was injected into glass reaction vessels coated with 0.5 μg of dried Iodogen before 2.5 μCi of ¹²⁵I (Amersham, Braunschweig, Germany) was added. After 15 min at room temperature, an equal volume of phosphate buffer was added. The iodinated cytokines were purified by affinity chromatography as described above. BSA (0.1%) was added as a carrier protein to enhance ¹²⁵I-IL-4 and ¹²⁵I-IL-4.Y119D stability and to reduce loss caused by nonspecific adsorption to tubes.

The determination of the binding affinities of ¹²⁵I-IL-4 and ¹²⁵I-IL-4.Y119D to the IL-4R was performed as described previously (12) using TF-1 cells stably transfected with the murine IL-4R and expressing ∼20,000 IL-4R molecules per cell (31). Briefly, 3-fold serial dilutions of ¹²⁵I-IL-4 and ¹²⁵I-IL-4.Y119D were prepared in 100 μl (final volume) of complete Clicks RPMI medium in microfuge tubes at 4°C. Then, 2 × 10⁶ cells in 100 μl were added to each microfuge tube and incubated at 4°C with gentle shaking until binding equilibrium was reached (90 min). Cell-bound ligand was separated from nonbound (free) by centrifugation of the cells through an oil gradient. Nonspecific binding was measured in the presence of a 200-fold excess of unlabeled IL-4. Saturation binding curves and Scatchard plots were calculated using GraphPad Prism software (San Diego, CA).

D10.4G.1 (32), CTLL-2 (33), L1/1 (34), HT-2 (35), and B9 (36) were used for cell proliferation studies. The cells were propagated in complete medium (Clicks/RPMI medium (Life Technologies, Eppenstein, Germany) supplemented with 10% FCS (Biochrom, Berlin, Germany), 2 mM l-glutamine, 10 mM HEPES, 100 μg/ml penicillin, 60 ng/ml streptomycin, 13 mM NaHCO₃, and 5 × 10⁻⁵ M 2-ME) and stimulated in the presence or absence of various concentrations of IL-2, IL-4, and IL-4.Y119D, respectively, in 96-well flat-bottom microtiter plates (Nunc, Wiesbaden, Germany). D10.4G.1, CTLL-2, B9, and HT-2 cells were pulsed after 48 h of culture with [³H]thymidine (18.5 kBq/well; Amersham, Braunschweig, Germany) for 16 h and processed for beta counting. B9 and L1/1 cells were incubated with 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; 50 μg/well) for 4 h after 72 h of stimulation, stopped with 0.01N HCl/10% SDS, and colorimetrically analyzed at 550 nm.

cDNA prepared from B9 cells was used as template for PCR with primers for the complete IL-13R α-chain (sense primer, 5′-CCGGATCCGCGAGGGCCTGCATGGCGCGGCCA-3′; antisense primer, 5′-CCGAATTCCACTTCTCCCCATCAAGGAGCTGC-3′). The resulting cDNA fragment was cloned into the BamHI/EcoRI-linearized eukaryotic expression vector pM5neo, which is based on the Moloney murine leukemia virus and contains a neomycin resistance gene (kindly provided by W. Ostertag, Hamburg, Germany). After confirmation of the nucleotide sequence, 15 μg of purified receptor plasmid DNA was electroporated into 1 × 10⁷ CTLL-2 cells in 0.8 ml of medium at 900 μF and 260 V in an Easyject electroporation unit (Eurogentec, Seraing, Belgium). Stably transfected cells were selected by cultivation in the presence of 400 μg/ml G418 for 2 wk, then subjected to single-cell cloning. IL-13Rα cell surface expression was analyzed by flow cytometry, applying affinity-purified rabbit IgG raised against the extracellular domain of the murine IL-13R α-chain expressed in E. coli.

B cells were purified from naive murine spleen cells by magnetic separation with a MACS column (Miltenyi Biotech, Bergisch Gladbach, Germany) according to the manufacturer’s instructions. The purity of the cells negatively selected with anti-CD4, -CD8, and -CD11b mAb-coupled microbeads (Miltenyi Biotech) was analyzed by flow cytometry leading to a 95 to 97% purity of B cells. Purified splenic B cells (1 × 10⁶) were cultured in the presence or absence of different amounts of recombinant murine IL-4 and IL-4.Y119D for 16 h, respectively. The expression of CD23 and MHC class II was measured by flow cytometry.

Spleen cells from IL-4−/− mouse were prepared as single-cell suspensions and cultured in vitro for 72 h at a density of 2 × 10⁶/ml in complete medium in the presence or absence of various concentrations of IL-4 and IL-4.Y119D, respectively. After 72 h of culture, the release of the sIL-4R into the supernatant was determined by a specific sandwich ELISA as described previously (6).

After RNA extraction from murine lymphocytes with acidic guanidinium thiocyanate (37), cDNA was synthesized for each time point in 20-μl reactions containing 1 μg of total RNA, 50 mM Tris-HCl (pH 8.3), 5 mM MgCl₂, 1 mM of each dNTP, 2.5 mM oligo(dT), 32 U RNAguard, and 17 U AMV-RT (all from Pharmacia Biotech) at 42°C for 90 min. The cDNA was amplified in a 40-μl reaction volume containing 50 mM Tris-HCl (pH 8.3), 2.5 mM MgCl₂, 10 mM of each dNTP, 1 U Taq polymerase (Pharmacia Biotech), and 100 nM primers during 35 cycles (1 min denaturation at 94°C, 1 min annealing at 58–63°C, and 1 min extension at 72°C). Samples were analyzed on 1.5% agarose gels containing 0.2 μg/ml ethidium bromide. Primers used were as follows: spliced IL-4R sense primer, 5′-GCCCCAGTGGTAATGTGAAGCCCCT-3′, and spliced IL-4R antisense primer, 5′-CTCACCACCGCAGCCCCCAAGGTCA-3′ (amplified fragment of 373 bp); IL-13R sense primer, 5′-ACAGAAGTTCAGCCACCTGTGACG-3′, and IL-13R antisense primer, 5′-CTAGGAGTTTTGCTCCTTACCTATACT-3′ (amplified fragment of 941 bp); β-actin sense primer, 5′-CACCCGCCACCAGTTCGCCA-3′, and β-actin antisense primer, 5′-CAGGTCCCGGCCAGCCAGGT-3′ (amplified fragment of 574 bp); γc chain sense primer, 5′-CCCAGAGAAAGAAGAGCAAGCACC-3′, and γc chain antisense primer, 5′-GGGGTCCTGGAGCTGGACAACAAA-3′ (amplified fragment of 429 bp).

For quantification of the mRNA molecules coding for sIL-4R, a competitive PCR was used as described previously (38). Briefly, each cDNA was first amplified in the presence of diluted β-actin control fragment (1:10 in a first approach followed by 1:2 dilution intervals for exact monitoring) of known concentration to determine the titer of control fragments needed to compete successfully with this cDNA. Then, relative concentrations of sIL-4R cDNA were determined in β-actin-normalized cDNA samples using the same two-step protocol with a sIL-4R-specific control fragment. All reactions were repeated three times independently. The relative amounts of cDNA molecules were calculated and expressed on the basis of equivalent numbers of β-actin cDNA molecules.

Macrophage monolayers were prepared and tested for purity as described elsewhere (39). Briefly, thioglycolate-elicited peritoneal exudate cells (PEC) were seeded into Labtek tissue culture chamber slides (Nunc), and nonadherent cells were removed, after 4 h of incubation, by three intensive washings. Before infection, PEC were incubated for 4 h either in complete medium or in complete medium supplemented with cytokines as indicated in the text. The macrophages were then infected with L. major promastigotes (parasite:cell ratio, 8:1) in the respective media for 4 h. Thereafter, nonphagocytosed parasites were removed by two intensive washings, and the cultures were further incubated in the respective media for 72 to 96 h. Intracellular amastigotes were assessed by fluorescence microscopy, after staining with ethidium bromide (50 μg/ml) and acridine orange (5 μg/ml) in PBS, as described in detail elsewhere (39). The percentages of infected macrophages (infection rate) were determined. Differences between treated and control cultures were tested for statistical significance by Student’s t test for unpaired samples.

A mutant murine IL-4 protein with an amino acid replacement of tyrosine to aspartic acid near the carboxyl terminus at position 119 (Y119D) was cloned as described in Materials and Methods. This protein was expressed in E. coli bacteria, as well as in eukaryotic 293 EBNA cells, to analyze the influence of lack of glycosylation, the degree of correct folding, and the influence of the added His tag and to avoid LPS contamination of E. coli-synthesized IL-4.Y119D. The two-step affinity purification of the mutated IL-4 expressed in bacteria yielded an apparently homogenous protein as judged by silver-stained SDS-PAGE, which could be detected in Western blots with Abs specific for murine IL-4 as well as for the His tag (Fig. 1). The purified IL-4.Y119D was folded correctly, since the neutralizing 11B11 mAb used for the second affinity chromatography binds only to IL-4 possessing a correct tertiary structure (data not shown). In addition, no differences with regard to either the biologic effects of the IL-4 mutant or the EC₅₀ values measured were detected when IL-4.Y119D expressed in 293 EBNA cells was used, suggesting that the glycosylation of this protein is not critical for its biologic functions (data not shown). In line with this finding, no difference between wild-type IL-4 produced in our laboratory (as glycosylated protein in 293 EBNA cells) and the commercially obtained E. coli-expressed mouse IL-4 was detected in the different assays performed.

First, the receptor-binding properties of the mutated IL-4 were compared with those of the wild-type cytokine. In a solid phase binding assay, applying either purified human or murine sIL-4R immobilized on plastic surfaces, no differences of the resulting binding curves were observed between IL-4 and IL-4.Y119D. Both proteins displayed specific binding only to murine and not to human IL-4R, showing the species specificity of the murine IL-4 forms (Fig. 2). To analyze ligand interactions with IL-4R complexes, IL-4 and IL-4.Y119D were iodinated and tested for binding to TF-1 cells expressing high numbers of IL-4R. Both IL-4 and IL-4.Y119D displayed similar binding curves and K_d values of 525 pM and 447 pM, respectively (Fig. 3, A and B), demonstrating that the affinity of IL-4 to its receptor is also not influenced by the carboxyl-terminal amino acid substitution when the IL-4R α-chain is expressed on the surface of cells.

Although binding of IL-4.Y119D to the murine IL-4R α-chain was not affected as a result of this mutation, the IL-4 mutant was unable to induce proliferation of several T cell lines, such as CTLL-2, L1/1, HT-2 (Fig. 4), and D10.4G.1 cells (data not shown); in contrast, these cell lines were highly responsive to wild-type IL-4 (Fig. 4). To investigate whether IL-4.Y119D is able to antagonize IL-4-induced proliferation, the T cell lines were cultured with IL-4 or IL-2 in the presence of an excess of the IL-4 mutant. Indeed, IL-4.Y119D was a potent antagonist of wild-type IL-4, leading to a complete inhibition of IL-4-induced T cell proliferation in the presence of a sufficient (40-fold) molar excess (Fig. 5). The IL-2-induced proliferation of the T cells was unaffected by IL-4.Y119D, thus excluding nonspecific or toxic effects. To evaluate whether the presence of a type II IL-4R complex leads to agonistic effects of IL-4.Y119D on T cells, CTLL-2 cells were stably transfected with an expression vector encoding the murine IL-13R α-chain. Several isolated transfected clones expressing the IL-13Rα not only displayed a proliferative response to IL-13 but, more importantly, also to IL-4.Y119D. As shown in Table I, IL-13- and IL-4.Y119D-induced proliferation reached comparable levels on IL-13R-expressing CTLL-2 cells, while no IL-13- and IL-4.Y119D-induced proliferation could be measured on wild-type or control-transfected (pM5neo vector) CTLL-2 cells. These experiments show that the reconstitution of a type II IL-4R complex on these T cells confers responsiveness to the IL-4 mutant.

Resting B cells were enriched from spleen cells of BALB/c mice using MACS for the depletion of CD8⁺, CD4⁺, and CD11b⁺ cells, yielding 95 to 97% pure B cell populations. After 16 h of culture in the presence or absence of IL-4 or IL-4.Y119D, respectively, the expression of CD23 and MHC class II Ags was analyzed by flow cytometry. As depicted in Figure 6, A and B, both surface molecules were up-regulated by wild-type IL-4 as well as IL-4.Y119D in a dose-dependent manner, although the calculated EC₅₀ values indicated that IL-4.Y119D is ∼50-fold less efficient than wild-type IL-4. Experiments with B cells from IL-4-deficient mice showed similar results, excluding a significant influence of endogenously produced IL-4 (data not shown).

The murine plasmacytoma cell line B9 has been shown previously to proliferate in response to IL-4 and IL-13 (40). Like wild-type IL-4, the mutated form of IL-4 induced proliferation of B9 cells in a dose-dependent manner (Fig. 7). IL-4.Y119D-induced proliferation was independent of LPS contamination, since it was also observed with IL-4.Y119D expressed in 293 EBNA cells, which did not contain detectable amounts of endotoxin (LPS content < 100 pg/μg protein). Furthermore, the addition of Polymyxin B into the cultures of B9 cells stimulated with the E. coli-derived protein did not influence IL-4.Y119D-induced proliferation (data not shown). The calculated EC₅₀ were 8.5-fold higher for the IL-4.Y119D (EC₅₀ = 170.6 ng/ml) as compared with wild-type IL-4 (EC₅₀ = 19.7 ng/ml), irrespective of whether MTT or [³H]thymidine incorporation was measured. As expected, due to the agonistic function of the IL-4 mutant, no inhibitory effect on IL-13-induced proliferation of B9 cells was detected (data not shown).

The IL-13R α-chain has been shown to participate in the formation of functional type II IL-4R complexes, while the γc chain appeared to be indispensable for IL-4 effects mediated through the type I IL-4R (14). Among the cell lines analyzed, only B9 cells expressed mRNA for the IL-13R α-chain, as detected with an IL-13R α-chain-specific primer pair, described in detail above, while mRNA for the γc chain was also detectable in the T cell lines (Fig. 8). The correlation between IL-4.Y119D-induced cell proliferation and IL-13R α-chain expression, as well as the fact that IL-13R α-chain transfected CTLL-2 cells proliferated in response to IL-4.Y119D (see above), strongly argues for a type II IL-4R-mediated effect of the IL-4 mutant and a type I IL-4R-specific antagonism.

An additional function of IL-4, which has been previously demonstrated to be independent of the γc chain, is the induction of sIL-4R (38). IL-4 and IL-4.Y119D were added to spleen cell cultures obtained from IL-4-deficient mice to avoid the complicating and ill-defined effects of endogenously produced IL-4. As depicted in Figure 9,A, both forms of IL-4 stimulated the production of sIL-4R with similar kinetics. As already known for IL-4 (38), the sIL-4R release induced by IL-4.Y119D was due to the induction of the alternatively spliced form of IL-4R mRNA (Fig. 9 A, inset).

The evaluation of the dose-response relationship for the induction of the sIL-4R release revealed that IL-4.Y119D is similar to wild-type IL-4 in its efficiency. The EC₅₀ values calculated from the experiments differed between the two forms of IL-4 only by a factor of 2 (Fig. 9 B).

As reported previously (41), IL-4 (50 ng/ml) in combination with IFN-γ (20 ng/ml) increases the killing activity of macrophages to eliminate intracellular L. major amastigotes as long as both cytokines are added simultaneously 4 h before infection (Fig. 10). This effect was not observed when PEC from TNF R55−/− mice were used, confirming previous studies (42) showing that the macrophage-stimulatory effect of IL-4 is mediated by enhanced production of endogenous TNF-α. Supernatants of transfected 293 EBNA cells were applied to analyze the effect of IL-4.Y119D on L. major-infected macrophages (since the LPS-content of the IL-4.Y119D produced in E. coli precludes experiments with macrophages). These experiments clearly demonstrate that IL-4.Y119D, like wild-type IL-4, is able to enhance the killing of intracellular parasites in the presence of IFN-γ (Fig. 10). The percentage of infected macrophages was reduced to a similar extent as with wild-type IL-4, and this effect was not observed with macrophages obtained from TNF R55−/− mice. The specificity was confirmed by the use of supernatants of control-transfected 293 EBNA cells, which were inactive, (Fig. 10) and by the inhibition of the IL-4.Y119D-mediated leishmanicidal activity after adding 10 μg/ml neutralizing anti-IL-4 mAb 11B11 (data not shown). The analysis of the expression of the IL-13R α-chain mRNA by RT-PCR showed that this molecule is present in the adherent PEC (data not shown). Thus, it is tempting to speculate that the effect of IL-4.Y119D on macrophages might also be mediated independently of the γc chain by type II IL-4R.

In this study, we expressed and functionally characterized a mutant of mouse IL-4 with a single amino acid substitution near the carboxyl terminus (IL-4.Y119D). This IL-4 mutant displayed an unexpected split response, since it was able to antagonize IL-4-induced proliferation of T cells, while in contrast, it acts as an agonist for B cell proliferation, induction of MHC class II and CD23 molecules, and the release of sIL-4R, as well as the enhancement of parasite killing by macrophages.

The receptor-binding behavior of IL-4.Y119D was indistinguishable from that of wild-type IL-4 in two types of assays. In addition to solid phase binding assays, we iodinated both mutated and wild-type IL-4 to directly measure the receptor binding on cells expressing high numbers of functional IL-4R complexes. The observed K_d values were not significantly different between IL-4 and its mutant, which argues for an identical or very similar binding of IL-4.Y119D to the heteromolecular IL-4R complexes, at least on the surface of these cells. According to currently available models of the tertiary structure of human IL-4, the critical tyrosine at position 119 of mouse IL-4 is located on helix D of this 4-helix-bundle protein (43). Cross-linking experiments using the human IL-4 double mutant IL-4.R121D.S125D support the hypothesis that helix D of human IL-4 directly interacts with the γc chain within the type I IL-4R (44). Analogous to previous experiments with mutants of human IL-4 (20, 25, 26), we observed that the substitution of the most carboxyl-terminal tyrosine abolished the T cell-stimulatory activity of mouse IL-4, resulting in a competitive IL-4 antagonist on T cells. The indispensability of the murine γc chain for IL-4-induced growth of mouse T cells has been defined previously by using mAbs to block this receptor molecule (14). The observation that thymocytes from γc chain-deficient mice do not proliferate in response to IL-4 (45) confirmed this conclusion. Together with our findings this suggests that IL-4.Y119D antagonizes γc chain-dependent functions mediated via the type I IL-4R. In line with this hypothesis, IL-4 antagonism of the IL-4 mutant is converted to agonistic functions on CTLL-2 cells expressing a functional type II IL-4R after transfection with the IL-13R α-chain.

Interestingly, and in contrast to the results with T cells, IL-4.Y119D induced the expression of CD23 and MHC class II on B cells as well as the proliferation of B9 plasmocytoma cells. Three experimental findings published by other groups indicate that the γc chain is not essentially involved in IL-4 effects on B cells. First, the induction of MHC class II or CD23 molecules on murine B cells by IL-4 was virtually unaffected by mAbs against the γc chain (14). Second, B cells deficient for the γc chain derived from patients with X-SCID were able to proliferate in response to IL-4 and IL-13 (16, 17). Third, IL-4 was able to induce Cε germline transcripts as well as IgE isotype switching and, in addition, the dose-response curves to IL-4 for X-SCID and normal B cells were indistinguishable (46). Interestingly, the human IL-4.Y124D mutant inhibited responses of X-SCID B cells to IL-4 and IL-13 (17), clearly demonstrating that this human IL-4 mutant is an IL-4 antagonist for functions mediated via IL-4Rs that lack the γc chain. Furthermore, indirect evidence suggests that the up-regulation of CD23 on the surface of B cells is also a γc chain-independent function of IL-4. Oakes et al. (47) reported that on human B cell lines from patients lacking JAK3, IL-4 induced CD23 expression, albeit less efficiently than on normal B cells. Since it has been shown previously that JAK3 is activated via the γc chain (reviewed in 15 , these findings suggest that CD23 expression might also occur in the absence of γc chain signaling. In purified splenic B cells, however, no signal for the IL-13Rα could be detected by RT-PCR (data not shown). We thus considered three possibilities by which IL-4.Y119D exerts its effects on B cells in the absence of the γc chain. First, the expression of the IL-13R α-chain, although of functional significance, might be too low to be detected by RT-PCR, but this appears to be unlikely due to the high sensitivity of the method. Second, in B cells as opposed to T cells, IL-4 signaling might be initiated by homodimerization of IL-4R α-chains in the absence of additional receptor molecules. Third, B cells might express an alternative IL-13R chain (IL-13Rα2), which has recently been cloned from human cells. Since no sequence data on the murine homologue have been published or deposited in gene data banks, we are currently unable to distinguish experimentally between the two latter possibilities.

Another cell type in which the γc chain appears not to be essential for the effects of IL-4 and IL-13 is the macrophage. Anderson et al. (48) reported recently that IL-4 up-regulated MHC class II molecules and inhibited nitric oxide production by macrophages from γc chain-deficient mice in response to LPS and IFN-γ to the same extent as in wild-type macrophages. The γc chain-deficient macrophages were likewise fully responsive to IL-13, consistent with the model of a type II IL-4R complex, which functions in the absence of the γc chain. Of special interest to the overall context of our study was the finding that the mouse IL-4 double mutant IL-4.Q116D.Y119D completely antagonized IL-4 and IL-13 responses on macrophages (48). Thus, together with the three IL-4 functions previously reported to be completely inhibited, e.g., the proliferation of T and B cells as well as the up-regulation of CD23 on splenic B cells, the IL-4.Q116D.Y119D double mutant is undoubtedly a complete type I and II IL-4R antagonist. In contrast to this scenario, we demonstrate in this study that IL-4.Y119D antagonizes IL-4 functions mediated through the γc chain-containing IL-4R type I complex but maintains γc chain-independent functions of IL-4, i.e., the stimulation of B cells and macrophages, mediated via the type II IL-4R. However, the precise molecular interactions of IL-4.Y119D with the different molecules of the IL-4R complexes are still poorly defined, and especially the analysis of the interaction of the mutant with the IL-13 α-chain awaits the development of reagents such as purified IL-13R α-chain or Abs directed against this molecule, which are not yet available.

So far, we can only speculate on the reasons for the different dose-response relationships observed for the different effects of IL-4 and its mutant. The several IL-4 receptor chains and signal-transducing molecules such as STAT6, JAK1, JAK3, IRS-1, and IRS-2 might be either differentially expressed, involved in different cell types, and/or be of dissimilar importance for varied IL-4 effects such as up-regulation of cell surface molecules or the soluble IL-4R and cell proliferation.

To our knowledge, the only other example of a cytokine mutant distinguishing between two functional responses is human IL-5.E12K (49). This IL-5 mutant is a specific and potent antagonist for both IL-5-mediated proliferation of TF-1 cells and the adhesion of eosinophils. In contrast, IL-5.E12K is a full agonist in a human eosinophil survival assay, although with reduced potency compared with the wild-type protein. However, the exact molecular mechanisms leading to this split antagonism/agonism are not yet defined.

We think that the newly described murine IL-4.Y119D represents a valuable tool for the dissection of different IL-4 functions in vitro as well as in vivo. For example, there currently appears to be no other way of distinguishing γc chain-dependent from γc chain-independent IL-4 functions, since the complete inhibition of the γc chain by Abs or gene knock-out also definitively affects the function of other cytokines that are dependent on this molecule, such as IL-2, IL-7, IL-9, and IL-15. Thus, our future goal is to analyze the possible therapeutic effects of IL-4.Y119D in the model of murine cutaneous leishmaniasis, where certain IL-4 effects, such as the promotion of Th2 development, are disadvantageous for the host, while others, such as the enhanced killing of parasites by IL-4-activated macrophages, are desirable.

We thank Mrs. Christine Kugler for excellent technical assistance and Dr. Seiler for supplying sIL-4R and Drs. Pfeffer and Kopf for supplying gene-deficient mice, respectively. We are grateful to Dr. Klaus Schröppel for stimulating discussion and for critical reading of the manuscript.