A food allergy is an adverse reaction of the immune system to allergenic food proteins. Food allergies are becoming increasingly common, posing a growing public health challenge. The phenomenon of cross-reactivity represents an additional challenge, in which an individual's immune system recognizes and reacts to proteins that are similar to the primary allergen. This makes it difficult to make an accurate diagnosis and thus to determine safe foods for individuals with allergies.
In our thesis, we investigated whether the similarity of the three-dimensional (3D) structures of proteins is a better predictor of cross-reactivity than the similarity of their primary structures. As a case study, we used the allergenic peanut protein Ara h 1.
Through a review of the scientific literature, we identified instances of documented cross- reactivity between the investigated protein pairs. In the first part of the thesis, we obtained the primary protein structures of Ara h 1 and related allergens using COMPARE allergen database. Using COMPASS bioinformatics tool, we identified proteins with similar primary structures and then assessed the similarity of the primary and 3D structures of the protein pairs using BLASTp and PDBeFold tools.
We confirmed a high level of 3D similarity among cross-reactive proteins, even in cases where primary structure similarity was low. The similarity of 3D structures is therefore a better predictive factor for the potential of cross-reactivity than the similarity of primary structures. The lack of experimentally determined structures limits the scope of such analyses. This limitation could be overcome in future research by including computationally predicted 3D structures, which would allow for the comparison of a larger number of proteins.
We also investigated potential structural similarities between Ara h 1 (and related proteins) and bacterial proteins from the human gut microbiota. We identified several instances of structural similarity but found a low likelihood of cross-reactivity. Most of the proteins were derived from probiotic bacteria commonly used in food supplements.
We used DALI tool alongside PDBeFold to assess the similarity of 3D structures between Ara h 1 (and related proteins) and bacterial proteins from the human gut microbiota. The results of DALI and PDBeFold showed a high degree of consistency between the two methods, confirming the reliability of the findings on 3D structural similarity.
|
