Protease resistance of food proteins: a mixed picture for predicting allergenicity but a useful tool for assessing exposure

Jaap Akkerdaas, Muriel Totis, Brian Barnett, Erin Bell, Tom Davis, Thomas Edrington, Kevin Glenn, Gerson Graser, Rod Herman, Andre Knulst, Gregory Ladics, Scott McClain, Lars K. Poulsen, Rakesh Ranjan, Jean-Baptiste Rascle, Hector Serrano, Dave Speijer, Rong Wang, Lucilia Pereira Mouriès, Annabelle Capt, Ronald van Ree

Susceptibility to pepsin digestion of candidate transgene products is regarded an important parameter in the weight-of-evidence approach for allergenicity risk assessment of genetically modified crops. It has been argued that protocols used for this assessment should better reflect physiological conditions encountered in representative food consumption scenarios. To evaluate whether inclusion of more physiological conditions, such as sub-optimal and lower pepsin concentrations, in combination with pancreatin digestion, improved the performance of digestibility protocols used in characterization of protein stability. Four pairs of established allergens and their related non/weakly-allergenic counterparts (seed albumins, muscle tropomyosins, plant lipid transfer proteins [LTP] and collagens) plus fish parvalbumin, were subjected to nine combinations of pH (1.2-2.5-4.0) and pepsin-to-protein ratio (PPR: 10-1-0.1 U/µg) for pepsin digestion, followed by pancreatin digestion in the presence of bile salts. Digestion was monitored by SDS-PAGE in conjunction with Coomassie staining and immunoblotting using rabbit antisera and human IgE. At pH 4.0 and at PPR 0.1 most proteins, both allergen and non-allergen, were highly resistant to pepsin. Under conditions known to favor pepsin proteolysis, the established major allergens Ara h 2, Pru p 3 and Pen a 1 were highly resistant to proteolysis, while the allergen Cyp c 1 was not. However, this resistance to pepsin digestion only made Ara h 2 and to a lesser extent Pen a 1 and Pru p 3 stand out compared to their non-allergenic counterparts. Largely irrespective of preceding pepsin digestion conditions, pancreatin digestion was very effective for all tested proteins, allergens and non-allergens, except for Cyp c 1 and bovine collagen. Sub-optimal pH, low pepsin-to protein ratio, and sequential pepsin and pancreatin digestion protocols do not improve the predictive value in distinguish allergens from non-allergens. Digestion conditions facilitating such distinction differ per protein pair.