Discussion
We found that barley-OFC results were associated with wheat intake at the time of OFC. Based on these results, OIT for wheat was conducted for patients with a wheat allergy who had positive barley-OFC results, and it was found that barley allergy was improved after wheat OIT. Although the cross-reactive protein between wheat and barley was not identified from the results of immunoblotting inhibition, we have shown for the first time, both clinically and immunologically, that barley allergy associated with wheat allergy is caused by cross-reactivity from wheat.
The positive rate of barley-OFC performed at our hospital (36.3%) was lower than previous studies (47.6%, 48.1%)4,18, which reported that the sIgE titers to wheat and omega-5 gliadin18 and sIgE titers to barley4were useful predictors of the result of a barley-OFC. In addition to these titers, we found that the results of barley-OFC were related to wheat intake at the time of barley-OFC. Therefore, the relatively lower positive rate of barley-OFC is due to the high proportion of cases with high wheat intake at the time of OFC. Moreover, when the results of barley-OFC were examined in patients with low wheat intake (wheat protein, ≤260 mg) at the time of barley-OFC, the sIgE titers to wheat and omega-5 gliadin were higher in patients with a positive barley OFC result (p =0.03 and 0.01, respectively, data not shown).
Based on the barley-OFC results, the change of barley-OFC results in the earlier stage of wheat OIT and after OIT were evaluated. The second barley-OFC, performed after reaching a desensitized state of at least 2,000 mg of wheat protein by wheat OIT, was negative in all patients. These results suggested that barley allergy may be clinically caused by cross-reactivity with a wheat allergy. In addition, it was proven by ELISA inhibition that barley allergy complicated by patients with a wheat allergy was caused by cross-antigenicity of both antigens, and wheat was the source of common antigenic sensitization.
Several components of barley allergy have been reported in the past. The water-soluble fractions of barley, α-amylase/trypsin inhibitor, α-amylase, and β-amylase, are known to be allergenic components of baker’s asthma.19-21 Heat-stable LTP and protein Z4 are the components of beer that can cause allergic symptoms.22-26 Adult patients with a barley allergy whose allergic symptoms are triggered by beer consumption are rarely complicated by wheat allergy, and the pattern of barley allergy development in children and adults may be different. Using serum from patients with WDEIA, it was reported that γ-3 hordein has cross-antigenicity with omega-5 gliadin.7
This study detected various spots in immunoblotting with barley extract using the serum of patients with a barley allergy complicated with the wheat immediate-type allergy. In particular, the spots of 30–35, 40–45, 60–80 kDa were detected in all cases and may be the main component. Furthermore, according to the WHO/IUIS database, the molecular weight of γ-3 hordein is 34 kDa, which may correspond to the 30–35 kDa spot detected in this study. In the future, we plan to identify the barley allergens for the spots detected in this study.
If the amino acid sequences of the epitope sites are highly homologous even for different allergen components, the IgE antibody binds to both, causing cross-antigenicity. Omega-5 gliadin has a repeating structure containing many glutamine residues in the molecule and has many continuous epitopes, QQX1PX2QQ (X1=L, F, S, I; X2=Q, E, G).15,27 On the other hand, γ-3 hordein is also a protein-rich in glutamine residues and has an amino acid sequence that is similar but not identical to the epitopes of omega-5 gliadin.28 Although the amino acid sequence similarity between omega-5 gliadin and γ-3 hordein was not high (23.9%) according to a Basic Local Alignment Search Tool search, the high amino acid homology between the epitopes of both components suggests that cross-antigenicity is likely to occur between wheat and barley allergens. Furthermore, we showed by ELISA inhibition that the response to barley extracts was more inhibited in wheat than in barley at low inhibitor concentrations. This may be because the key amino acid sequences of the epitopes for omega-5 gliadin and γ-3 hordein binding to IgE antibodies are identical, but the epitope sequences are not completely identical, resulting in a difference in affinity, or because the number of similar epitopes is higher in omega-5 gliadin than in γ-3 hordein. However, in the immunoblotting inhibition, most of the spots of barley disappeared due to inhibition by wheat extracts, and it is likely that other barley proteins are also involved in the cross-antigenicity with wheat proteins. In the future, it is necessary to identify barley allergens and conduct epitope analysis to elucidate the mechanism of cross-reactivity.
There are several limitations to this study. First, no patient underwent a second barley-OFC without OIT for wheat. Hence it cannot be denied that the barley allergy was not improved by the wheat OIT but was the acquisition of tolerance during the natural course. Therefore, in the future, patients with a positive barley OFC result without OIT for wheat will also be evaluated for barley allergy status over time. Second, the protocol for barley-OFC was not consistent in all cases. However, patients with negative OFC results were instructed to consume sufficient amounts of cooked barley at home to confirm the negative result once again.
In conclusion, we showed that wheat OIT increased the symptom-inducing threshold of barley allergy. In addition, barley allergy complicated by patients with an immediate-type wheat allergy was caused by cross-antigenicity of both antigens, and wheat was found to be the source of sensitization. In the future, it is expected that epitopes to which IgE antibodies react in cases of immediate-type wheat and barley allergy will be identified, and the mechanism of cross-antigenicity will be clarified.