Investigation of (suspected) genotoxic flavouring substances in cakes and sweet biscuits

Details

Supervisors

Collin, Sonia ; Dusart, Alexandre

Faculty

Faculté des bioingénieurs

Degree label

Master [120] : bioingénieur en sciences agronomiques, à finalité spécialisée

Abstract

In Europe, Member States are responsible for the monitoring of the consumption and use of flavourings, although no common methodology has been adopted to collect this information, mainly due to the lack of analytical method. In Belgium, to address this issue, a collaboration project called Flavouran was set up between Sciensano (the Belgian institute for health) and the UCLouvain (Unit of Brewery and Food Sciences). This master thesis focused on the solid matrices part of the Flavouran project sampling, namely cakes and sweet biscuits. Compounds of interest were a selection of flavouring substances for which the genotoxicity evaluation was still pending (suspected genotoxic flavouring substances) in addition to confirmed genotoxic flavouring substances. First, a routine analytical method allowing the quantitative analysis of several flavouring substances at the same time (multi-analyte method) and usable in a broad range of cakes and sweet biscuits with a high accuracy was developed. Isolation of flavouring substances was carried out using solvent-assisted flavour evaporation (SAFE) coupled to gas chromatography-mass spectrometry (GC-MS). To guarantee accurate results, quantification of flavouring substances was performed using standard addition method. For this purpose, samples were clustered based on their relative proportion of carbohydrates, fat, and proteins using the Association of Official Analytical Collaboration (AOAC) triangle. Thus, since standard addition is a time-consuming experiment, it was performed on samples chosen as a reference of samples of similar composition, namely crepe (Delhaize), petit beurre (Lu) and chocolate waffle (Lotus). Then, the method was validated in-house in terms of selectivity, limits of detection (LOD) and quantification (LOQ), linearity, matrix effects, recovery and intra- and interday repeatability. No guidance document had been established to set out the validation procedure to be followed as well as common performance criteria specific to the analysis of flavouring substances. Moreover, studies about flavouring substances generally did not mention any validation of the method used. Therefore, a validation procedure as well as performance criteria were chosen in the case of this master thesis based on Part 3 of Annex of Decision 2002/657/EC as it provides general recommendations in the field. In total, 27, 15 and 21 flavouring substances were validated in crepe (Delhaize), petit beurre (Lu) and chocolate waffle (Lotus), respectively. To finish, the method was used to determine the occurrence and concentration of 38 selected (suspected) genotoxic substances in a serie of 26 cakes and sweet biscuits samples and results were interpreted. An important result was the presence at high concentration of furan-2(5H)-one (up to 4320.3 ± 950.5 μg.kg-1) and 2,4-pentanedione (not validated in any sample) in the samples analysed whereas these compounds were both classified as genotoxic by the European Food Safety Authority (EFSA). As these compounds are mainly formed by heat treatment, further analysis on the impact of the temperature, cooking time and water content of the sample should be carried out. Also, high levels of 2-phenylcrotonaldehyde and 5-methyl-2-phenyl-2-hexenal were found in chocolate (290.2 ± 55.1 μg.kg-1 and 640.7 ± 185.8 μg.kg-1, respectively) and therefore in chocolate containing biscuits (up to 33.3 ± 11.3 μg.kg-1 and 29.6 ± 8.6 μg.kg-1, respectively) whereas the evaluation of these compounds was still to be completed by the Authority. Further analysis of different types of chocolate, both commercial and for food preparations, would be of great interest. Finally, p-mentha-1,8-dien-7-al was found in all citrus flavoured samples (up to 4.1 ± 1.2 μg.kg-1) whereas this compound was classified as genotoxic by EFSA. Therefore, more analysis of samples containing citrus should be carried out, including the analysis of citrus pulp and peel.