The CY Subunit of the Human IgE Receptor (FcERI) Is Sufficient for High Affinity IgE Binding*

The alpha subunit of the FcERI binds IgE with high affinity. Previous studies have demonstrated that alpha subunit expression requires the presence of beta and/or gamma subunits, and it is not known how these two subunits contribute to the ability of the alpha subunit to bind IgE. In this report, we describe the expression and characterization of a human chimeric alpha subunit. The data demonstrate that high affinity IgE binding does not require the presence of the beta and/or gamma subunits and that this activity is localized to the extracellular domain (residues 26-201) of the human alpha subunit. Permanent cell lines expressing the chimeric receptor were used to characterize the binding parameters of the alpha subunit. These cell lines provide a means of identifying therapeutic agents which may be effective in the treatment/management of allergic diseases.

The high affinity receptor (FcERI) for immunoglobulin E (IgE) that is found on mast cells and basophils plays a central role in the allergic response (1). In the rat this receptor consists of three different subunits: an IgE binding (Y chain, one /3 chain, and two y chains (l-5). Previous rat and mouse studies have demonstrated that high affinity cell surface IgE binding is dependent on the coexpression of the three subunits (5,6). In the human system at least the cy and y subunits must be present for high affinity cell surface binding (7,8). The functional contribution of the /3 and/or y subunits at present is unknown. In this report we describe the construction and functional expression of a chimeric human LY subunit.

IgE Binding
to FcERI a Subunit of ""I-IgE was assessed as previously described (20) and ranged between 50 and 75%. Binding assays were performed as described previously (21), and for competition assays, samples at various dilutions were preincubated with the cell suspensions (4 X 10' cells/ml) for 30 min prior to the addition of 250 rig/ml ""I-IgE. in the absence of the y subunit. In an effort to obtain cell surface expression of the a subunit to study in detail its binding properties, a chimeric receptor was engineered by substituting the transmembrane and cytoplasmic domains of the a subunit with those coding for the ~55 IL-2 receptor. The ~55 IL-2 receptor subunit is expressed efficiently on the cell surface independent of other subunits. Transfection of COS cells with the wild type FcERI a, pLJ1101, did not yield any cell surface binding of IgE or antipeptide antisera (Fig. 1, B and D). Analysis of permeabilized cells using the antipeptide antisera clearly demonstrated that the a subunit was indeed synthesized (results not shown). However, transfection of COS cells with a chimerit cy subunit, pLJ1275, resulted in efficient cell surface expression of an a subunit which was able to bind IgE and was recognized by antipeptide antisera (Fig. 1, A and C). The binding of biotinylated IgE was completely inhibited by a 50fold excess of unlabeled IgE (results not shown, see below).
These results suggest that the a subunit transmembrane and/or cytoplasmic domains are involved in preventing the cell surface expression of the a subunit. The block exerted by these domains is obviously overcome by the presence of the y subunit (5). It should now be possible to identify the regions which are exerting this block as well as the regions which are interacting with the y subunit. Binding Analysis-Permanent cell lines expressing the chimerit receptor were established in CHO cells using gene-  To clearly demonstrate that the extracellular domain of the a subunit is sufficient for high affinity IgE binding, the equilibrium association constant was established by independent determinations of the association and dissociation rate constants as previously described for the human and rodent receptors. The receptor number on the CHO cells was determined by incubating the cells with 10 pg/ml of ""I-1gE for 2 h. Receptor density ranged between 4 and 10 X lo" molecules/cell.
The kinetics of association and dissociation between l"I-IgE and the chimeric IgE cell surface receptor were determined at 25 "C. A representative experiment is shown in Fig.  2 where the cells expressed 6.03 x lo5 surface receptors, and the V, was determined during the first 100 s. The k, calculated from these results is 1.4 x lo5 M-' s-l. The dissociation rate of bound lZ51-IgE over a 5-6-h period was calculated from the slope of plot shown in Fig. 3, and the km1 is 1.5 x 10m5 s-l. From three independent experiments the kl and km1 calculated from our data are 1.1 f 0.26 X lo5 M-l s-' and 1.3 f 0.18 X 10m5 s-', respectively. The average KA from our data is 0.85 X 10" M-'. These data are similar to those described for the high affinity FcERI receptors on cultured cord blood basophils and COS cells transfected with the (Y and y subunits (7,24 Specificity of receptor binding was evaluated in radioreceptor competition experiments (Table I). Dose-dependent inhibition of lz51-IgE (PS) binding was observed with both human and rat IgE but not with human and rat IgG. Inhibition of 50% binding (IC50) with two human myeloma IgE proteins and a chimeric recombinant human IgE was achieved with approximately 2-3 pg/ml of protein.
Polyclonal human IgE (from hyper-IgE plasma) was 6-fold less effective in the competition assay. It is unknown whether this reflects a difference in the sample's intrinsic binding activity or partial inactivation during storage and purification. Rat IgE was at least lo-fold less active in the competition assay, yet the same rat IgE inhibited 50% binding in a rat radioreceptor binding assay at 2-3 pg/ml (data not shown). By Scatchard analysis, the affinity of rat lZ51-IgE to the human FcERI was approximately 7-fold lower than that of human IgE (data not shown). Human IgE (PS) did not bind to the rat IgE receptor on RBL cells. These data confirm the crossspecies specificity of IgE binding on rat and human FcERI. Early studies using human basophils indicated that approximately 15-fold more rat IgE was required for 50% inhibition of lz51-IgE (PS) binding than unlabeled IgE (25). Subsequent studies using human cord blood basophils demonstrated no significant difference in equilibrium constants for rat and human IgE (24). More detailed binding analysis of rat IgE on the recombinant FcERI is required to characterize this species specificity.
It is clear from our data that the binding of human IgE to FcERI is localized to the extracytoplasmic region of the receptor and that the B and y subunits of the receptor complex do not play a critical role in IgE binding. The chimeric receptor overexpressed on CHO cells provides us with the opportunity to study in detail the interaction of IgE and its receptor and to develop high flux receptor binding assays to identify antagonists of IgE/FcERI binding.