RESUMEN
Western society is facing an increase in the number of food-allergic individuals, with rising incidence in the past years. Therefore, allergen-free food and accurate and reliable analysis of allergen contamination are essential for the protection of consumers. Yet, there is limited understanding on the effect of food processing on allergenicity and on the ability of available methods to detect trace contamination in processed food. Available studies addressing this have relied on sample processing on a laboratory scale. In this study, industry-like processing under precisely defined conditions (ranging from 110 to 150°C roasting temperatures) was employed to better understand the limitations of state-of-the-art methods for detecting traces of hazelnut and almond in processed food. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis analysis indicated an overall reduction in extracted proteins from roasted nut samples, and with matrix-assisted laser desorption ionization time-of-flight Cor a 9 and Prunin, were identified as majorly expressed proteins for hazelnut and almond, respectively. A commercial ELISA kit detected nut traces only up to a 130°C roasting temperature. Untargeted MS (Orbitrap) analysis was able to detect traces of nuts roasted up to 150°C while also confirming Cor a 9 and Prunin as the major expressed proteins for hazelnut and almond, respectively. Preparing cookie dough spiked with roasted nut samples, a complex food matrix was simulated. Analysis by ELISA showed the same limitations encountered for pure nuts samples, hardly detecting traces of nuts roasted above 130°C. Targeted MS (linear ion trap) using multiple reaction monitoring methods for one proteotypic peptide for Cor a 9 and Prunin, respectively, enabled a detection of nut traces up to 150°C. The results indicated that a reduced extractability because of temperature-related effects (e.g., protein denaturation, cross-linking, poor solubility) caused the significant differences between the ELISA and MS analysis. Overall, the results of this study may form the basis to improve allergen detection after roasting through improved extraction methods and refined ELISA formats.