Regulation of Interleukin-13 Receptor Constituents on Mature Human B Lymphocytes*

Human B cells stimulated through both their immunoglobulin and CD40 receptors up-regulate interleukin (IL)-13 ligand binding sites with an affinity of 0.91 6 0.08 n M within 24 h. IL-13 binds primarily to the IL-13R a 1 with subsequent sequestration of the IL-4R a into the complex. IL-13R a 1 may also be found in those receptors capable of binding IL-4. g chain ( g c ) partici-pates in receptors capable of binding IL-4 but is not found in association with bound IL-13. Dimeric receptors composed of the IL-4R a complexed with either the IL-13R a 1 or g c occur simultaneously within defined B cell populations. mRNAs for all receptor constituents are increased subsequent to immunoglobulin stimulation alone, while maximal expression of IL-13R a 1 is more dependent upon co-stimulation of immunoglobulin and CD40 receptors. mRNA levels for IL-13R a 1 vary over a wider range subsequent to surface stimulation than other receptor components. Although g c is not bound to IL-13 in B cells under the conditions evaluated, it may influence IL-13 binding by competing with IL-13R a 1 for association/sequestration with the IL-4R a chain. IL-13R a 2 does not participate in the IL-13 receptor that is up-regulated upon activation of quiescent tonsillar

Receptor structure for IL-4 1 and IL-13 (1)(2)(3)(4)(5) has been studied in diverse cell types including those of lymphoid, epithelial, and mesenchymal derivation (6 -13). Several configurations for the IL-4 receptor have been postulated based on variations in the smaller molecular weight component of a 140-kDa/70-kDa multimeric complex. In T cells, the 70-kDa entity has definitively been shown to be ␥ c , a protein that is shared among several receptor types, including those for IL-2, IL-7, IL-9, and IL-15 (8, 14 -19). A 70-kDa protein, distinct from ␥ c , has been proposed as part of a second variant of the IL-4 receptor. This moiety has been reported to also function as part of the IL-13 receptor (9, 11-13, 20, 21). Several reports of the cloning of distinct 70-kDa IL-13 binding chains strengthen the proposed role of this component(s) in a shared IL-4/IL-13 receptor configuration (22)(23)(24)(25)(26). The 70-kDa human protein encoded by the cDNA isolated by Aman et al. (23), by Miloux et al. (25), and by Gauchat et al. (26), originally based on the murine cDNA isolated by Hilton et al. (22), is herein referred to as IL-13R␣1; the human protein encoded by the cDNA isolated by Caput et al. (24) is referred to as IL-13R␣2.
The receptor capable of binding IL-13 appears to vary structurally and functionally depending upon the cell of origin (6,(11)(12)(13)27). Human cell populations that bind IL-13 with high affinity include several renal cell carcinoma lines, the glioma line U251, and the ovarian cancer cell line IGROV-1 (13,28). It has recently been hypothesized that in these cell lines isodimers of the IL-13 binding proteins may result in the reported high affinity IL-13 binding (28). Multiple cell types including the ovarian cancer cell line PA1, an epidermal carcinoma line A431, a primate fibroblast line COS-7, and normal human endothelial cells bind both IL-4 and IL-13, although some cell types preferentially bind IL-4. It has been suggested from a combination of functional and structural studies that these populations possess IL-13 receptors composed of an IL-13-specific binding component in conjunction with IL-4R␣ (13). Evaluation of the erythroleukemia cell line TF-1 suggests that the ␥ c protein may participate as part of a functional IL-13 receptor. A potential role for ␥ c has also been suggested from studies evaluating phorbol dibutyrate and ionomycin-stimulated B cells exposed to high IL-13 concentrations (29). Whether ␥ c acts directly as a binding element or as part of the signal transduction mechanism in these cells remains to be definitively determined.
Evaluation of the IL-13 receptor on human B cells has mainly focused upon Epstein-Barr virus-transformed cell populations as well as functional studies on B cells derived from X-SCID patients (21,28,30). Examination of IL-13 receptorrelated proliferation has recently been extended to cells derived from patients with non-Hodgkin's lymphoma as well as partially purified B cells from reactive lymph nodes (31). In the present series of experiments, normal, nonvirally transformed, purified mature human B lymphocytes have been directly evaluated both structurally and functionally. Prior examination of receptor binding on Epstein-Barr virus-transformed or freshly isolated B cells has been hindered by relatively low binding of radiolabeled native IL-13 in part secondary to low receptor numbers (10,11,13). Herein, these difficulties have been par-* This work was supported in part by National Institutes of Health Grants CA45148 (to A. L. M.) and R29-DK49649 (to J. M.) and by funds from the Department of Pathology Teaching and Research Foundation. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. tially overcome by utilization of a mutant IL-13 moiety that can be labeled to high specific activity in association with specific external activation conditions that substantially up-regulate ligand binding components (12). In addition, we report the initial utilization of antibodies specific for the IL-13R␣1 binding proteins with direct identification of those elements involved in cytokine binding and receptor coaggregation to permit a detailed, direct assessment of receptor structure on normal lymphoid cells.
Measurement of Proliferation-Cell proliferation, as measured by thymidine incorporation, was performed as described previously (33). The S.E. values of triplicates within an individual experiment were invariably less than 10%. The percentage of cells in S phase was determined by propidium iodide staining and flow cytometric analysis as described by Look et al. (35).
Radiolabeling and Binding of [Phe 43 ]IL-13-GYGY-Biologically functional 125 I-labeled [Phe 43 ]IL-13-GYGY ( 125 I-IL-13, 60 -100 Ci/g) was prepared by methods essentially as described previously (12). Modifications to the procedure included a stepwise addition of four 2-l aliquots of 30 nM chloramine T at 1-min intervals and fractionation of the labeled product via Sephadex G-25 chromatography. Ligand affinity was evaluated utilizing a two-step paradigm. Initially, the binding of the radiolabeled ligand was determined to evaluate effects introduced by the labeling procedure. The dissociation constant (K d ) of the labeled ligand was then used to calculate the affinity of the native ligand according to the procedures of Munson et al. using Ligand software (36). The receptor number and binding affinity were determined from the Scatchard and displacement analyses of 3-5 replicate experiments. The receptor competitions for specific cytokines were performed in RPMI medium containing 0.25% bovine serum albumin with 2 nM 125 I-labeled IL-4 and 30 nM 125 I-labeled [Phe 43 ]IL-13-GYGY as indicated in the figure legends. The total binding reaction volume was 35 l or 0.3 ml as indicated in figure legends.
Affinity Cross-linking of Receptor Complexes-Activated B lymphocytes were removed from culture; washed twice with RPMI 1640, 0.25% bovine serum albumin; and subsequently incubated with radioiodinated IL-4 or IL-13-GYGY for 30 min at 37°C or for 2-3 h at 4°C. The cells were incubated in the absence or presence of excess unlabeled ligand. Reactions were stopped by the addition of cold phosphate-buffered saline containing 0.1% azide and 100 M Na 3 VO 4 . Receptor crosslinking was performed by incubation of washed cells in 1 mM disuccinimidyl suberate or BS 3 , at 14 or 4°C, respectively, for 20 -30 min. The washed cells were resuspended in lysis buffer (50 mM HEPES, pH 7.5, 0.5% Brij 96, 50 mM NaCl, 50 mM NaF, 5 mM EDTA, 1 mM Na 3 VO 4 , 2 mM PMSF, 10 mM aprotinin, 10 g/ml leupeptin) incubated for 15-30 min on ice, and centrifuged for 15 min at 14,000 ϫ g at 4°C. Supernatants were either boiled in Laemmli buffer and analyzed by 7-7.5% SDS-PAGE or were further immunoprecipitated with appropriate antibodies.
Immunoprecipitation of Cross-linked Receptor Complexes-Solubilized samples to be immunoprecipitated were exposed overnight, at 4°C, with either 10 g of purified, polyclonal rabbit anti-human receptor component antibody (anti-IL-3R␣1, anti-IL-4R␣) or 3 l of rat monoclonal anti-human ␥ c ascites (Tugh4 kindly provided by Dr. Sagamura). The anti-IL-4R␣ utilized was a rabbit polyclonal antiserum against a C-terminal peptide corresponding to amino acids 801-820 of the mature protein (Santa Cruz Biotechnology, Inc., Santa Cruz, CA). The anti-IL-13R␣1 antibody was prepared as described above. Immunoprecipitated complexes, isolated with protein G-Sepharose beads, were resolved by 7% SDS-PAGE.
Evaluation of mRNA Expression-IL-13R␣1, IL-13R␣2, IL-4R␣, and ␥ c mRNA levels were assessed using RT-PCR with primers derived from published sequences (see below) (19,23,24,37,38). B lymphocytes (10 7 ) were harvested, and total RNA was prepared using Trizol® (Life Technologies, Inc.) reagent or RNeasy spin columns (Qiagen). 2-5 g of total RNA were used as a template for oligo(dT)-primed cDNA synthesis using a cDNA cycle kit (InVitrogen, Carlsbad, CA or Life Technologies, Inc.). cDNA quality and quantity were standardized using RT-PCR for glyceraldehyde-3-phosphate dehydrogenase and/or ␤-actin (CLONTECH). A hot start technique was employed in 50-l reactions with GeneAmp® buffer (Perkin-Elmer), 1.5 mM MgCl 2 , 200 M Gene-Amp deoxynucleotide triphosphates (Perkin-Elmer), 5-10 Ci of [␣-32 P]dCTP (Amersham Pharmacia Biotech), and 1.5 units of ampli-Taq DNA polymerase (Perkin-Elmer). For semiquantitative reactions, 30 cycles (94°C for 1 min; 59°C for 1 min; 72°C for 1.5 min) of amplification were performed, and a minimum of three dilutions were evaluated for each sample. Nested reactions were performed using 5 l of the primary reaction as template for the secondary reaction. PCR products were separated by electrophoresis on 2% agarose or 10% acrylamide gels, stained with ethidium bromide, photographed, dried, and autoradiographed. Band intensity was quantitated and presented in relative arbitrary units.
For quantitative competitive PCR, cloned competitive substrates were prepared from cDNA as described using the indicated primers (39). A constant known amount of competitive substrate was added to each PCR reaction, and the amount of cDNA was evaluated in reference to the intensity of the competitive band. Amplicon number was calculated based on the concentration of the competitive substrate. Sequences of the primer pairs utilized are as shown in Table I.

RESULTS
Comparative Efficacy of Cytokines in Stimulating the Proliferation of Quiescent B Cells-Mature human B cells were evaluated for their proliferative response to IL-13. Without exogenous stimulation, Ն98% of these cells were in G 0 /G 1 . B lymphocytes were stimulated with either insolubilized antiimmunoglobulin M (anti-) and/or soluble anti-CD40 antibodies in the presence or absence of individual cytokines including IL-2, IL-4, IL-6, IL-13, IL-15, and tumor necrosis factor-␣. Stimulation of B cells by ligation of both the antigen and CD40 receptors in combination with a cytokine provided the strongest proliferative response. The relative mitogenic efficacy, dependent upon dual surface receptor ligation, was IL-4 Ն IL-13. These cytokines showed a consistent proliferative effect when IL-13 Receptor on Human B Cells utilized at Յ10 Ϫ9 M. IL-2 and IL-15 produced significant effects only at concentrations Ն10 Ϫ8 M. Similar rank orders of proliferative efficacy were seen when cells were solely stimulated by either ligation of the immunoglobulin receptor or CD40 receptor alone followed by cytokine exposure, with soluble anti-CD40 being the least effective stimulus. Fig. 1, A-F, presents binding curves and Scatchard analyses for radiolabeled IL-4 and IL-13. The affinity constants given in the legend to Fig. 1 were used to calculate native ligand binding values. Table II shows the affinity and binding site determinations for native IL-4 and IL-13 as a function of surface stimulation. Anti-stimulation alone resulted in the appearance of 1498 Ϯ 224 IL-4 sites/cell of high affinity with a K d of 41 Ϯ 10 pM. After cross-linking both surface receptors, surface immunoglobulin and CD40, 1398 Ϯ 151 IL-4 receptor sites/cell were found to have a K d of 37 Ϯ 7 pM. Examining B cells stimulated with anti-alone, the K d and site number for IL-13 could not be reliably calculated due to the low level of specific binding, which resulted in high coefficients of variation. In the presence of both anti-immunoglobulin and anti-CD40 antibodies, 745 Ϯ 51 IL-13 binding sites/cell were found with an intermediate affinity of 0.91 Ϯ 0.08 nM.

Expression of IL-4 and IL-13 Receptors as a Function of Surface Stimulation-
Differential Receptor Competition as a Function of Cytokine Exposure- Fig. 2A demonstrates that 125 I-IL-4 binding was completely inhibited by unlabeled IL-4 and that unlabeled IL-13 competed for approximately 50% (control lanes, no preincubating cytokines) of such binding (n ϭ 3). Conversely, as seen in Fig. 2B, unlabeled IL-13 completely blocked radiolabeled IL-13 binding, and unlabeled IL-4 displaced 87% of the ligand (n ϭ 4) (control lanes, no preincubating cytokines). Neither radiolabeled IL-4 nor IL-13 were displaced by excess unlabeled IL-2 (Fig. 2, A and B, control lanes, respectively). The actual binding assays depicted in Fig. 2, A and B, were conducted with an incubation period of 1-2 h at 4°C, conditions associated with minimal receptor internalization/down-regulation. However, incubation of cells with ligand at 37°C is associated with such internalization/down-regulation (40,41). To determine the efficacy of each cytokine to occupy and/or internalize IL-4 and IL-13 receptors, high density cells were stimulated in culture with anti-and anti-CD40 antibodies in the presence or absence of unlabeled ligands. After the cultures were maintained at 37°C for 24 -36 h, the cells were tested for radiolabeled ligand binding and for competition of that binding by an unlabeled growth factor. As seen in Fig. 2A, preincubation at 37°C with IL-2 failed to modulate subsequent 125 I-IL-4 binding, while unlabeled IL-4 resulted in Ն95% reduction in specific 125 I-IL-4 binding. Exposing cells to IL-13 during the 37°C incubation phase resulted in a 60% decrease in 125 I-IL-4 binding. In experiments that measured 125 I-IL-13 sites (Fig.  2B), preincubation at 37°C with either the homologous ligand or IL-4 resulted in near elimination of subsequent 125 I-IL-13 binding, while preincubation with IL-2 induced no measurable change in binding.
The observed reduction in radiolabeled IL-4 binding subsequent to IL-13 preincubation at 37°C could be due to IL-13 occupancy of shared receptor elements, IL-13 down-regulation/ internalization of shared receptor components, or steric hindrance of radiolabeled IL-4 binding by IL-13 bound to its specific receptor. To rule out the blocking effect of residually bound cytokine to either a shared or contiguous receptor, cells were washed after the 24 -36-h incubation and exposed for 2 min to RPMI 1640 medium (10% fetal calf serum) adjusted to pH 3.0. Such acidic incubation releases surface-bound ligand without loss of cell viability (40). These treated cells were tested for 125 I-IL-4 binding in comparison with cells that had been washed and not exposed to acidic conditions. As seen in Fig. 2A (IL-13 preincubation, pH 3.0-treated columns), such a comparison of 125 I-IL-4 binding both before and after residual cytokine (IL-13) release revealed that 10 -15% of the reduction in binding seen after the 24 -36-h incubation with IL-13 could be accounted for by receptor competition/blockade by IL-13, while the remainder of the IL-13 effect on IL-4 binding was due to actual receptor down-regulation. This latter result suggests that most of the IL-13 competitive effect can be accounted for by a sharing/utilization of common receptor elements.
Experiments were next directed at specifically quantitating the variations in site number due to homologous or heterologous ligand pre-exposure. Quiescent B cells were stimulated with either anti-immunoglobulin and/or anti-CD40 antibody in the presence of IL-13. Under these conditions, IL-13 binding should be quantitatively reduced as previously demonstrated. Measurement of IL-4 binding sites on those cells incubated overnight with IL-13 after dual surface ligation with antiimmunoglobulin and anti-CD40 indicated that the number of measurable IL-4 binding sites was reduced by approximately 50% to 702 Ϯ 2 sites (Table III) per cell, compared with those cells not treated with IL-13 during the culture period (approximately 1400 IL-4 sites, refer to Table II). Furthermore, the sites exhibited a high affinity with a K d of 22 Ϯ 14.6 pM, similar to those cells not treated with IL-13 where the K d for IL-4 was 37 pM (refer to Table II). In those cells stimulated with antiimmunoglobulin antibody alone, which do not possess detectable high affinity IL-13 sites yet do possess high affinity IL-4  (Fig. 3A), while at 37°C both ligand binding and receptor protein coaggregation occur (Fig. 3B). The electrophoretic profile of the whole cell lysates for IL-4 specif-ically bound at 4°C demonstrated a predominant band at 150 -160 kDa, consistent with a 140-kDa binding protein bound with labeled ligand. A less intense 85-kDa band, consistent with a 70-kDa protein bound with labeled ligand, is also seen (n ϭ 5). Fig. 3A additionally demonstrates that 125 I-IL-13 bound predominantly to a band of 80 kDa, consistent with a binding protein of 70 kDa complexed with labeled ligand. There was faint binding detected to a higher molecular weight complex of 150 kDa indicative of a 140-kDa protein associated with the labeled ligand. Fig. 3B reveals that at 37°C the ligand binding/ coaggregation band pattern for 125 I-IL-4 replicates that seen at 4°C. The pattern seen with IL-13 at 37°C differs from that seen at 4°C in that a band corresponding to a protein of 140 kDa coaggregates with the 70-kDa protein, the predominant moiety seen at the lower temperature.
The identity of the binding proteins was investigated using Western analysis and immunoprecipitation. Experiments examined the contribution of the common ␥ c protein to receptors reactive with either IL-4 or IL-13. Eighteen-hour activated B cells were exposed to either native IL-4 or IL-13 for 30 min and subsequently immunoprecipitated with either anti-IL-4 or anti-IL-13 antibody, respectively. Following resolution of the immunoprecipitate on 7.5% SDS-PAGE and transfer to membranes, the samples were blotted with an affinity-purified polyclonal antibody prepared against the ␥ c C terminus (Santa Cruz Biotechnology). Immunological recognition of this protein as a 70-kDa band, as seen in Fig. 4, was demonstrated in those

IL-13 Receptor on Human B Cells activated cells incubated with IL-4 and immunoprecipitated
with anti-IL-4. The ␥ c protein was not identifiable in those immunoprecipitates derived from activated cells exposed to IL-13 and immunoprecipitated with a polyclonal antibody to IL-13.
Experiments next utilized immunoprecipitation to evaluate the presence of IL-4R␣, IL-13R␣1, or ␥ c within the ligandbinding complexes formed with either IL-4 or IL-13. Activated B cells were exposed to either radiolabeled IL-4 or IL-13-GYGY at 37°C for 30 min, cross-linked, and immunoprecipitated, and proteins were separated by 7.5% SDS-PAGE. As seen in Fig.  5A, the anti-IL4R␣ antibody immunoprecipitated a two-band complex of 140 and 70 kDa in activated cells incubated with radiolabeled IL-4 at 37°C. The relative band intensity was similar to that seen previously in that the predominant band was 140 kDa, with 70 kDa being of lesser intensity. The anti-IL-4R␣ immunoprecipitate of activated cells exposed to radiolabeled IL-13 at 37°C also revealed a two-band profile but with a predominance of the 70-kDa moiety. This is in contradistinction to the cross-linked whole cell lysate at 4°C (refer to Fig.  3A), which reveals a predominant component of 70 kDa without substantial presence of a 140-kDa protein. The anti-IL-13R␣1 polyclonal antiserum immunoprecipitated a two-band profile from cells exposed to either radiolabeled IL-4 or IL-13 (refer to Fig. 5B). Therefore, exposure of cells to either IL-4 or IL-13 at 37°C for 30 min resulted in the immunoprecipitation of complexes that are composed of both the IL-4R␣ and the IL-13R␣1. Activated cells were also immunoprecipitated with anti-␥ c after a 37°C incubation period with radiolabeled cytokine. For IL-4-exposed cells, this procedure resulted in a radiolabeled im- munoprecipitate of 140 and 70 kDa for incubations with cytokine ranging from 5 to 30 minutes. In contrast, anti-␥ c immunoprecipitation of IL-13-exposed cells did not yield any radiolabeled immunoprecipitate at any incubation time point tested (data not shown).
mRNA Expression for Receptor Constituents-Experiments next evaluated message levels for the protein elements of the IL-4 and IL-13 receptors. Fig. 6 shows semiquantitative RT-PCR with dilutions of cDNA. These experiments permit eval-uation of changes of a specific message that occur as a function of stimulation but do not permit evaluation of relative changes between receptor components. IL-13R␣1 message was substantially up-regulated as a function of ligation of both the immunoglobulin and CD40 receptors. The IL-4R␣ and ␥ c were upregulated as a function of activation through the immunoglobulin receptor. Semiquantitative evaluation seen in Fig. 6, A-C, reveals that dual surface ligation does not lead to further significant increases in these message levels. The time course for message up-regulation reveals that within 6 -12 h subsequent to cell activation, message for all constituents had been increased (data not shown).
More precise quantitations of mRNA levels were next undertaken using competitive PCR. Examination of Table IV reveals that IL-13R␣1 shows a consistent increase in message amount with maximal values following dual surface ligation. Interestingly, both ␥ c and the 140-kDa moieties show significant change subsequent to anti-immunoglobulin stimulation alone. ␥ c shows an additional small increase in message with dual surface ligation. The mRNA abundance of these different species indicates that the 140-kDa component is apparently in limiting amounts.
In addition to the IL-13R␣1, a second IL-13 binding protein has been identified and termed IL-13R␣2 (24). To evaluate its presence within the lymphoid populations isolated from the tonsil, cells were fractionated into high density quiescent cells and two levels of lower density cells corresponding to populations inclusive of centrocytes and centroblasts (42,43). Cells were either unstimulated or stimulated by anti-immunoglobulin alone or with anti-CD40. After 24 h, cells were collected, and RNA was isolated and evaluated by standard RT-PCR for the presence of message for IL-13R␣1 or IL-13R␣2. IL-13R␣1 was present in all fractions studied and varied as a function of cell activation (data not shown). Utilization of a standard RT-PCR revealed the presence of message for IL-13R␣2 only in that cellular fraction corresponding to the lowest density cells, which contains the centroblastic population. The high density (i.e. those quiescent lymphocytes utilized in the present receptor studies) and moderate density fractions (centrocytic cells) failed to demonstrate the presence of mRNA for IL-13R␣2 by either standard or nested RT-PCR (data not shown).

DISCUSSION
Depending on the specific mode of surface activation, the proliferation of mature human B cells induced by IL-13 is either equal to or less in magnitude than that induced by IL-4. The differential effectiveness of IL-4 and IL-13 noted in the present studies was most pronounced in cells on which only the immunoglobulin receptor had been ligated. Surface stimulation with anti-immunoglobulin alone results in up-regulation of a high affinity IL-4 receptor, while IL-13 binding under these conditions is minimal. Dual surface ligation by anti-immunoglobulin and anti-CD40 up-regulates a second receptor type that binds and transmits signals from both IL-4 and IL-13. It therefore seems plausible that the quiescent human B cell may be induced to express two forms of IL-4 receptor that differ in their ability to bind and/or respond to IL-13. Results from both thymidine incorporation and Scatchard analyses favor a model in which both receptor types coexist on an individual cell. Coexistence on the same cell would require a receptor configuration in which both the ␥ c chain of the IL-4 receptor and the specific IL-13 binding chain simultaneously compete for the shared 140-kDa binding chain. This model is compatible with (a) the binding of the cytokines in the presence or absence of the specific IL-13 binding component, (b) the results of immunoprecipitation and Western analyses, and (c) the observation that quiescent density-fractionated cells stimulated by both FIG. 4. Identification of ␥ c as part of the IL-4 receptor and its absence in the IL-13 receptor. Immunochemical identification of receptor binding chains was undertaken following immunoprecipitation of the receptor components as they are complexed with either native IL-4 or IL-13. Cells stimulated for 18 h with either antialone or anti-with anti-CD40 were exposed to cytokine for 30 min at 37°C and then immunoprecipitated with either anti-IL-4 or anti-IL-13. Following resolution of the immunoprecipitated proteins on 7.5% PAGE and subsequent transfer to membranes, the samples were blotted with anti-␥ c antibodies. Immunological recognition of ␥ c occurred with activated cells incubated with IL-4 and immunoprecipitated with IL-4. The ␥ c protein was not identifiable in immunoprecipitates derived from cells stimulated with IL-13 and immunoprecipitated with anti-IL-13. anti-immunoglobulin and anti-CD40 antibodies respond similarly to either cytokine. In addition, differential sequestration of IL-13 receptor constituents has been suggested by studies examining cells lacking the cytoplasmic domain of the IL-13R␣1 (44). It is interesting to note that studies on cells derived from human X-SCID patients that lack functional ␥ c demonstrate proliferative sensitivity to either IL-4 or IL-13 (21,30). These data and others suggest that the ␥ c protein is a sufficient although not a necessary component for B cell proliferation in response to IL-4 and that an IL-4 receptor composed of the 140-kDa binding unit and the IL-13-specific binding chain is sufficient for B cell proliferation to either IL-4 or IL-13.
IL-4 binds to its receptor with approximately 50 pM affinity and associates with both 140-and 70-kDa receptor components at 4 and at 37°C, with more radiolabeled IL-4 complexed with the 140-kDa IL-4R␣. In contrast, IL-13 binds to its receptor with at least 10-fold lower affinity and is bound principally to the 70-kDa IL-13R␣1 at 4°C. IL-13 binding to the 140-kDa protein is minimal unless the activated cells are exposed to ligand at 37°C. Even at 37°C, IL-13 shows more association with the 70-kDa chain than the 140-kDa component. These data are consistent with a model wherein IL-4 binds predominantly to the 140-kDa IL-4R␣, which subsequently coaggregates with a 70-kDa component that can be either ␥ c or IL-13R␣1. Conversely, IL-13 predominantly binds to the IL-13R␣1, which subsequently coaggregates with the 140-kDa IL-4R␣. Whether this differential utilization of receptor components underlies the variation in affinity noted above requires further investigation (45).
IL-13 does not appear to bind to and/or coaggregate with the ␥ c component in human B cells stimulated as described herein. The absence of ␥ c in the human IL-13 receptor has been suggested for several human cell types including cell lines derived from renal cell carcinoma, glioma, and epidermoid carcinoma (13). ␥ c has been implicated in an IL-13 receptor described for the TF-1 cell line derived from a human erythroleukemia (46). The role of ␥ c in Epstein-Barr virus-transformed B cells is less clear, but a direct or indirect interaction of this protein with the IL-13 receptor complex has been suggested (10,13). Functional studies on endothelial cells and human B cells derived from X-SCID patients have demonstrated that these cells are able to functionally respond to IL-13 in the absence of detectable ␥ c expression (6,21,30,47). In contradistinction, human B cells activated with phorbol dibutyrate and ionomycin demonstrate rapid coaggregation of the 140-kDa IL-4R␣ and the ␥ c protein after exposure to high concentrations of IL-13. This latter result has been used to argue that IL-13 utilizes ␥ c in its receptor complex (29). These experiments with ionomycin and phorbol dibutyrate-stimulated B cells do not distinguish between initial binding events and downstream effects subsequent to IL-13 binding. In the present work, the absence of the ␥ c protein in the B cell IL-13 receptor was demonstrated by its specific absence from the IL-13 binding complex. This was assessed both by immunoprecipitation of receptor ligand complexes with anti-IL-13 with subsequent blotting for ␥ c and by evaluation of radiolabeled IL-13 in anti-␥ c immunoprecipitates. As mentioned above, it appears from experiments presented herein and elsewhere that some form of competition exists between the IL-13R␣1 and ␥ c for aggregation with the IL-4R␣ (44,45). Whether this is simply a function of which ligand interacts FIG. 6. Effects of co-stimulation on receptor mRNA expression. High density B lymphocytes were unstimulated or activated by anti-/ or by anti-/ and anti-CD40 for 24 h. Equal quantities of isolated RNA were used as a template for single-stranded cDNA synthesis. Samples were subsequently utilized undiluted (lanes 1, 4, and 7), diluted 1:10 (lanes 2, 5, and 8) and 1:100 (lanes 3, 6, and 9) for IL-4R␣ and ␥ c . For IL-13R␣1, samples were utilized undiluted (lanes 1, 4, and 7), diluted 1:7 (lanes 2, 5, and 8), and diluted 1:50 (lanes 3, 6, and 9). All samples were subjected to a 30-cycle amplification protocol. Films were scanned and quantified into relative, arbitrary units for each message component separately. The sample with the greatest absorbance was given a value of 1000 units, and all other dilutions of the specific message component were compared with that value.

TABLE IV
Competitive PCR quantitation of mRNA for receptor-specific proteins The number of amplicons was determined using a Life Technologies, Inc. Superscript™ cDNA synthesis kit. The efficiency of cDNA synthesis using this kit is at least 1 log greater than with a similar product from Invitrogen (Carlsbad, CA). This was particularly evident in terms of the IL-4R␣, where the calculated differences approached 2 logs. These results are in agreement with data derived from RNase protection assays (Riboquant™; Pharmingen, San Diego, CA) (data not shown).

Cell stimulation a
Amplicons/g total RNA (ϫ10 Ϫ4 ) b a RNA was purified either directly after isolation of high density cells (unstimulated) or 24 h subsequent to stimulation by anti-immunoglobulin or anti-immunoglobulin plus anti-CD40.
b Results presented are those derived from replicate determinations of mRNA amounts from a single experiment with multiple (n ϭ 4) experiments indicating the same findings. The results presented are the mean Ϯ S.D. ϫ 10 Ϫ4 . with the cell surface remains to be determined. Furthermore, ␥ c may indirectly affect IL-13 binding, since IL-13 receptor affinity is altered in those transfected cell populations overexpressing ␥ c (24,45).
In the present series of experiments, the role of IL-13R␣1 in the IL-13 receptor has been confirmed utilizing antisera specific for its C terminus. These are the first such studies to demonstrate the actual presence of this protein in a functional receptor apart from the assembled receptors in transfected cell populations. The observation confirming the presence of IL-13R␣1 in mature B cells and the relative paucity/absence of IL-13R␣2 in mature lymphoid cells is of substantial interest. Future studies establishing the identity of that cellular phenotype in the centroblast fraction of the tonsillar B lymphocytes that contains both forms of the IL-13 receptor binding proteins should provide further essential information about the functional significance of each receptor element.