Analysis of the IgE epitope profile of soybean allergen Gly m 4

Background

Individuals with birch pollinosis may show allergic reactions after consumption of soybean-containing food. This is caused by cross-reaction of IgE directed against the major birch pollen allergen Bet v 1 with its homologue Gly m 4 from soybean. Birch pollen allergic subjects without soybean allergy often present IgE binding to Gly m 4.

Objective

To identify IgE epitopes of Gly m 4 in birch pollen-allergic subjects with and without allergy to soybean.

Methods

Sera from 33 birch pollen-allergic patients with and without clinical reactivity to soy as determined by oral food challenge or convincing history of soy allergy were included in the study. Specific IgE against Bet v 1 and Gly m 4 was determined by ImmunoCAP. IgE binding of the sera to rGly m 4 and soy extract was tested in western blot analysis. To predict putative IgE epitopes, 20 IgE-binding phage-displayed peptide mimics and heptamers representing the total Gly m 4 amino acid sequence were bioinformatically mapped onto the molecular surface of Gly m 4. To verify predicted epitopes, Gly m 4 variants with substituted epitope residues were expressed in and purified from E. coli. Secondary and tertiary structures were assessed by CD and NMR spectroscopy. IgE binding of variants was analyzed by competitive immunoblotting and inhibition ELISA.

Results

Gly m 4- and Bet v 1-specific IgE levels ranged from 0 to >100 kU_A/L in sera of both patient groups. CAP values were above 3.5 kU_A/L in 18 sera (86%) of patients with clinical soy allergy and in 7 sera (58%) of patients sensitized, but without clinical reactivity to soy, respectively. Intensity of IgE binding in western blot corresponded to ImmunoCAP analysis. Bioinformatic mapping resulted in four distinct epitopes on Gly m 4 surface. Several amino acid residues were chosen for generation of substitutional variants. Purified rGly m 4 variants showed a typical Bet v 1-like structure according to CD spectra and 1D-¹H-NMR spectroscopy. IgE binding to several substitutional variants was reduced, indicating a multiple number of epitopes on Gly m 4 surface.

Conclusion

No significant differences in Gly m 4-specific IgE binding could be observed between birch pollen-allergic patients with and without clinical soy allergy in ImmunoCAP and immunoblot analysis. However residual IgE reactivity persists in substitutional rGly m 4 variants, indicating the presence of further IgE epitopes not yet identified.