Temperature effect on IgE binding to CD23 versus Fc epsilon RI.

A chimeric soluble CD23, consisting of the extracellular domain of mouse CD23 and a modified leucine zipper (lz-CD23), has been shown to inhibit IgE binding to the FcepsilonRI. A similar human CD23 construct was also shown to inhibit binding of human IgE to human FcepsilonRI. In both systems, the inhibition was found to be temperature dependent; a 10-fold molar excess of lz-CD23 gave 90-98% inhibition at 4 degrees C, dropping to 20-30% inhibition at 37 degrees C. Surface plasmon resonance analysis of lz-CD23 binding to an IgE-coated sensor chip suggested that the effective concentration of lz-CD23 was lower at the higher temperatures. Analysis of (125)I-IgE binding to CD23(+)-Chinese hamster ovary cells also indicated that increased temperature resulted in a lower percentage of IgE capable of interacting with CD23. In contrast, IgE interacts more effectively with FcepsilonRI(+)-rat basophilic leukemia cells at 37 degrees C compared with 4 degrees C. The results support the concept that the open and closed IgE structures found by crystallography interact differently with the two IgE receptors and suggest that temperature influences the relative percentage of IgE in the respective structural forms. Changes in CD23 oligomerization also plays a role in the decreased binding seen at physiological temperatures.

Necessity of the stalk region for immunoglobulin E interaction with CD23.

Previously, a soluble mouse CD23 chimera, composed of an N-terminal trimeric isoleucine zipper motif (lz) followed by the entire extracellular region (amino acids 48-331) of CD23 (lz-CD2348-331), was prepared and exhibited strong binding to rodent immunoglobulin E (IgE). In the current study, we report the construction of a similar human chimeric protein (lz-huCD2345-321), as well as a series of murine chimeric lz-CD23 mutants with incremental portions of stalk deleted, to further investigate the role of the stalk region in mediating the CD23-IgE interaction. All chimeric proteins were designed such that the predicted heptad structure of the stalk was retained. IgE binding, as determined by the capacity to inhibit 125I-IgE from binding to FcepsilonRI-bearing RBL-2H3 cells, and by surface plasmon-resonance analysis using an IgE-coated sensor chip, was unchanged from the original lz chimera and the binding parameters were similar to those of cell-surface CD23. The minimal murine chimera that retained IgE-binding activity was lz-CD23139-331, which still contains 35 amino acids of the stalk region. When the lz motif was linked to CD23 amino acid 157 (or higher), significant IgE-binding capacity was lost. With human lz-CD23, as with mouse, deletion of the stalk greatly reduced IgE-binding ability. In summary, the data support the concept that at least a portion of the stalk region of CD23 plays a crucial role in maintaining high-affinity/avidity interaction with IgE. The lz-CD23 constructs represent a possible alternative for both blocking the IgE/FcepsilonRI interaction and inhibiting IgE production by B lymphocytes.