Food processing, especially thermal treatment, may have implications on mycotoxins in products available for consumers. This research work aimed to study how mycotoxin levels may be influenced by modifying the technological parameters of both whole grain and cocoa biscuit-making processes. The study was mainly focused on the following mycotoxins: deoxynivalenol, deoxynivalenol-3-glucoside, and the minor metabolite culmorin. Special emphasis was given to the recipe formulation, and to the baking conditions, using an industrial-scale operation, starting from naturally contaminated raw materials. Exploiting the power of Design of Experiments (DoE) and a dedicated LC-MS/MS method, the complexity of the different processes was investigated. The models obtained within this study showed a high goodness-of-fit suggesting that the pH and the baking time play important roles for minimizing mycotoxins in the final products, while the recipe formulation has an impact on the mycotoxins extractability by affecting the biscuit microstructure.

In the scientific field, there is a progressive awareness about the potential implications of food processing on mycotoxins especially concerning thermal treatments. High temperatures may cause, in fact, transformation or degradation of these compounds. This work is aimed to study the fate of mycotoxins during bakery processing, focusing on deoxynivalenol (DON) and deoxynivalenol-3-glucoside (DON3Glc), along the chain of industrial rusk production. Starting from naturally contaminated bran, we studied how concentrations of DON and DON3Glc are influenced by modifying ingredients and operative conditions. The experiments were performed using statistical Design of Experiment (DoE) schemes to synergistically explore the relationship between mycotoxin reduction and the indicated processing transformation parameters. All samples collected during pilot plant experiments were analyzed with an LC-MS/MS multimycotoxin method. The obtained model shows a good fitting, giving back relevant information in terms of optimization of the industrial production process, in particular suggesting that time and temperature in baking and toasting steps are highly relevant for minimizing mycotoxin level in rusks. A reduction up to 30% for DON and DON3Glc content in the finished product was observed within an acceptable technological range.

Deoxynivalenol (DON) is the most prevalent trichothecene in Europe and its occurrence is associated with infections of Fusarium graminearum and F. culmorum, causal agents of Fusarium head blight (FHB) on wheat. Resistance to FHB is a complex character and high variability occurs in the relationship between DON content and FHB incidence. DON conjugation to glucose (DON-3-glucoside, D3G) is the primary plant mechanism for resistance towards DON accumulation. Although this mechanism has been already described in bread wheat and barley, no data are reported so far about durum wheat, a key cereal in the pasta production chain. To address this issue, the ability of durum wheat to detoxify and convert deoxynivalenol into D3G was studied under greenhouse controlled conditions. Four durum wheat varieties (Svevo, Claudio, Kofa and Neodur) were assessed for DON-D3G conversion; Sumai 3, a bread wheat variety carrying a major QTL for FHB resistance (QFhs.ndsu-3B), was used as a positive control. Data reported hereby clearly demonstrate the ability of durum wheat to convert deoxynivalenol into its conjugated form, D3G.