Deciphering the origin of total estrogenic activity of complex mixtures.

Introduction: Identifying compounds with endocrine properties in food is getting increasingly important. Current chemical analysis methodology is mainly focused on the identification of known substances without bringing insight for biological activity. Recently, the application of bioassays has been promoted for their potential to detect unknown bioactive substances and to provide information on possible interactions between molecules. From the toxicological perspective, measuring endocrine activity cannot inform on endocrine disruption and/or health risks without sufficient knowledge on the nature of the responsible factors. Methods: The present study addresses a promising approach using High Performance Thin-Layer Chromatography (HPTLC) coupled to bioassays were analyzed using the Liquid Chromatography Mass-Spectrometry (LC-MS). The estrogen receptor activation was assessed using the transcription activation Estrogen Receptor Alpha Chemical Activated LUciferase gene eXpression assay (ERα- CALUX) and the HPTLC coupled to the Estrogen Screen Yeast assay (p-YES). Results: Seven isoflavones were identified in the soy isolates. Estrogen receptor activation was assessed for both, the identified isoflavones and the soy isolates with ERα-CALUX test. Correlation between the soy isolates extracts and the identified isoflavones was shown. Moreover, p-YES revealed the presence of an estrogenic bioactive zone. Analysis of the bioactive zone through LCHRMS highlighted signals corresponding to several isoflavones already detected in the isolates as well as two additional ones. For all detected isoflavones, an estrogenic activity dose-response was established in both bioassays. Conclusion: Finally, genistein, daidzein, and naringenin were found as the most active substances. A concordance analysis integrating the analytical and bioassay data indicated that genistein and daidzein were the drivers of the estrogenic activity of these soy protein isolates. Altogether, these data suggest that the integration of HPTLC-bioassay together with chemical analysis is a powerful approach to characterize the endocrine activity of complex mixtures.

Limitations of currently available in vitro oestrogenicity bioassays for effect-based testing of whole foods as the basis for decision making.

The idea that previously unknown hazards can be readily revealed in complex mixtures such as foods is a seductive one, giving rise to the hope that data from effect-based assays of food products collected in market surveys is of suitable quality to be the basis for data-driven decision-making. To study this, we undertook a comparative study of the oestrogenicity of blinded cereal samples, both in a number of external testing laboratories and in our own facility. The results clearly showed little variance in the activities of 9 samples when using a single method, but great differences between the activities from each method. Further exploration of these findings suggest that the oestrogenic activity is likely an inherent part of the natural food matrix which the varying sample preparation methods are able to release and extract to differing degrees. These issues indicate the current poor suitability of these types of datasets to be used as the basis for consumer advice or food decision-making. Data quality must be improved before such testing is used in practice.

High resolution mass spectrometry workflow for the analysis of food contaminants: Application to plant toxins, mycotoxins and phytoestrogens in plant-based ingredients.

An analytical workflow including mass spectral library, generic sample preparation, chromatographic separation, and analysis by high-resolution mass spectrometry (HRMS) was developed to gain insight into the occurrence of plant toxins, mycotoxins and phytoestrogens in plant-based food. This workflow was applied to 156 compounds including 90 plant toxins (pyrrolizidine alkaloids, tropane alkaloids, glycoalkaloids, isoquinoline alkaloids and aristolochic acids), 54 mycotoxins (including ergot alkaloids and Alternaria toxins) and 12 phytoestrogens (including isoflavones, lignans and coumestan) in plant-based protein ingredients, cereal and pseudo-cereal products. A mass spectral library was built based on fragmentation spectra collected at 10 different collision energies in both positive and negative ionisation modes for each toxin. Emphasis was put on a generic QuEChERS-like sample preparation followed by ultra-high-pressure liquid chromatography using alkaline mobile phase allowing the separation of more than 50 toxic pyrrolizidine alkaloids. HRMS acquisition comprised a full-scan event for toxins detection followed by data-dependent MS2 for toxin identification against mass spectrum. Method performance was evaluated using fortified samples in terms of sensitivity, repeatability, reproducibility and recovery. All toxins were positively identified at levels ranging from 1 µg kg-1 to 100 µg kg-1. Quantitative results obtained by a standard addition approach met SANTE/12682/2019 criteria for 132 out of 156 toxins. Such a workflow using generic, sensitive and selective multi-residue method allows a better insight into the occurrence of regulated and non-regulated toxins in plant-based foods and to conduct safety evaluation and risk assessments when needed.

Fate of acrylamide during coffee roasting and in vitro digestion assessed with carbon 14- and carbon 13-labeled materials.

Acrylamide (AA) formation during coffee roasting happens rapidly, reaching a peak value within the first minutes of roasting followed by a fast decrease to reach an asymptote at approximately 200 µg/kg. Today, the mechanisms by which AA is reduced during roasting remain unclear. In this research, the fate of AA during roasting followed by drip brewed-like extraction was studied using 14C-radiolabeled (14C-AA) and 13C-labeled (13C3-AA) materials. Results showed that 28% of the spiked 14C-AA was lost during the roasting process, presumably by degradation to volatile compounds and 25% was non-extractable; therefore, appeared bound to the matrix. About 50% of initial AA went into the water extract, either unchanged or transformed by conjugation/binding. The release of bound acrylamide was further evidenced by increasing levels of 13C3-AA over prolonged roasting times. In addition, the absence of 14C activity in the hexane extracts suggested acrylamide not to bind to any lipophilic material.

Thermal degradation of 2-furoic acid and furfuryl alcohol as pathways in the formation of furan and 2-methylfuran in food.

This study reports the heat-induced formation of furan by decarboxylation of 2-furoic acid, and 2-methylfuran by dehydration of furfuryl alcohol under dry conditions. Model systems were incubated at temperatures up to 190 °C, followed by quantitative determination of furan and 2-methylfuran performed by isotope dilution headspace gas chromatography-mass spectrometry. Results show that 2-furoic acid decarboxylation and furfuryl alcohol dehydration are activated as from about 140-160 °C. Furfuryl alcohol and 2-furoic acids were measured in a selection of roasted coffee products by isotope dilution liquid chromatography-high resolution mass spectrometry, and the data evidenced a strong correlation between the two compounds, suggesting an intimate mechanistic relationship between them. The possible oxidation of furfuryl alcohol to furfural and 2-furoic acid in heated food is raised with particular emphasis on coffee roasting. These findings are relevant for better understanding the formation of furan and alkylfurans in food, and ultimately opening avenues for mitigation.

Quantification of folpet and phthalimide in tea and herbal infusions by LC- high-resolution MS and GC-MS/MS.

Two methods based on a modified QuEChERS sample preparation and either LC coupled to atmospheric pressure ionisation and high-resolution MS or GC coupled to electron ionisation and tripled quadrupole MS have been assessed for the quantification of folpet and phthalimide in tea and other dry herbal infusions. Both methods have been fully validated in green tea and further checked in black tea, verbena and rooibos, and they performed according to the SANTE/11813/2017 criteria at the target LOQ concentration level (50 µg/kg). These methods allow the accurate quantification of folpet in the selected matrices according to the new EU residue definition, which includes phthalimide. Phthalimide is the main metabolite and degradation product of folpet, although according to recent studies, it could be generated from different sources than folpet breakdown, such as food processing or analysis by GC.

Quantification of folpet and phthalimide in food by gas chromatography and mass spectrometry: Overcoming potential analytical artefacts.

Accurate quantification of folpet is problematic because it degrades into phthalimide during sample preparation and analysis by gas chromatography (GC). Thus, EU regulation was recently modified to include phthalimide in the folpet residue definition. However, recent studies have shown that phthalimide could also be generated from different sources, which could lead to an overestimation of the phthalimide content and therefore to false positives. GC coupled with either negative chemical ionisation and single quadrupole mass spectrometry, or electron ionisation with triple quadrupole mass spectrometry (GC-EI-MS/MS), were evaluated for the determination of folpet and phthalimide in food. Both methods were validated in 4 different matrices namely apple puree, rice flour, raspberry puree and infant formula. Better selectivity and precision were obtained with GC-EI-MS/MS. Negligible amounts of phthalimide was found in blank matrices, and validation results met the SANTE/11813/2017 criteria in all matrices at the LOQ concentration levels by using GC-EI-MS/MS.