As other researchers attempted to expand on the early digestive evaluation using greater numbers of allergens and non-allergens, it became clear that the purported correlation between digestive stability and allergenicity did not hold (Fu et al., 2002; Herman et al., 2007). It appeared that the selection of allergens and non-allergens evaluated by Astwood and colleagues (1996) confounded the protein function with the allergenicity status of the proteins (Schnell and Herman, 2009). When additional proteins were selected based on considerations for protein function and family, no discernable correlation between digestive stability and the allergenic status of proteins remained (Akkerdaas et al., 2018). It has also been recently observed that oral intake of known allergens at an early age reduces allergy later in life confirming that exposure in the gut is important to tolerization against allergy (Logan et al., 2020).

The developing evidence that sensitization to food allergens can occur through dermal and inhalation exposure to food dust helps explain why gastro-intestinal digestive stability does not predict allergenic risk (Herman and Ladics, 2018). It is now hypothesized that initial sensitization to food allergens may occur via inhalation or dermal exposure to food dust (or through food handling or application of cosmetics containing crop by-products), followed by elicitation of allergy due to subsequent consumption of the food, and that initial gut exposure to the relevant protein in food may favor tolerization against allergy (dual allergen exposure hypothesis) (Kulis et al., 2021; Yakaboski et al., 2021).

The modern understanding that the state of human microbiome has a great impact on allergy development also helps explain why acid-suppressant medications increase allergy rates (Robinson and Camargo, 2018). It is now understood that the effect of acid-suppressant medications on the microbiome not only results in an increase in the frequency of food allergy, but also dermal and inhalation allergy (Robinson and Camargo, 2018). Clearly, the digestive fate of allergens in the gut is not relevant to eczema or rhinitis, and thus decreased digestion due to intake of acid-suppressant medications cannot explain why the frequency of allergy increases via all three exposure pathways. Similar to antibiotics, alteration of the human microbiome is the likely mechanism by which acid-suppressant medications increase allergy rates (Jordakieva et al., 2019; Herman, 2020; Herman et al., 2020).

An example has recently been published where modifying an allergenic protein to decrease its digestive stability might actually increase its allergenic risk. In this study, the digestive stability of a known food allergen (carp parvalbumin Cyp c 1) was reduced by modifying its amino acid sequence (Freidl et al., 2020). In a mouse model, it was found that oral exposure to the digestively stable native allergen tolerized mice against allergy, while the version modified to be less digestively stable did not. Mice initially exposed to the stable native allergen did not exhibit allergenic symptoms when later exposed to the proteins in food, while those mice exposed to the version modified to be more digestible were subsequently found to display allergenic symptoms. Thus, modifying a food protein with an allergenic risk to be less digestively stable could increase sensitization to that protein (Herman et al., 2021).

The preceding example highlights the potential risk of government regulation for genetically engineered crops based on the erroneous belief that gastro-intestinal digestibility decreases allergenic risk. Current government regulations for newly expressed proteins in genetically engineered crops universally consider gastro-intestinal stability to be a risk factor for allergenicity (Ladics, 2008). Thus, technology developers that discover proteins that would be useful if expressed in genetically engineered crops determine the digestive stability of these proteins in standardized in vitro digestion assays. If the protein is found to be stable in these assays, developers are motivated to modify the amino acid sequence to be more digestible while maintaining its beneficial activity (Parisi et al., 2020). While developers evaluate other components of the weight-of-evidence allergenicity assessment (e.g., bioinformatics, history of safe exposure, concentration in food, etc.) to minimize allergenic risk, the erroneous belief that digestibility is associated with reduced allergenic risk could, in some cases, increase the allergenic risk by reducing the tolerizing properties of a protein in the gut (Figure 1).