Unveiling the potential of proteomics in addressing food and feed safety challenges.

The food and feed sector in Europe is rapidly evolving to address contemporary challenges, striving for fairer, safer, greener and more sustainable food systems. This includes the exploration of new protein sources for human consumption and animal feed such as protein derived from insects, algae or novel plant-derived proteins, and the re-evaluation of existing sources like processed animal protein (PAP). To generate reliable data on the diverse array of emerging protein sources for future food and feed safety assessments, a growing demand for the development and implementation of advanced analytical techniques exists. New approach methodologies (NAMs) including, mass spectrometry (MS)-based proteomics methods have been emerging as valuable techniques which potentially can be implemented in regulatory laboratory settings to complement conventional approaches in this realm. These MS-driven strategies have already proven their utility in diverse applications, including the detection of prohibited substances in feed, identification of allergens, differentiation of fish species in complex mixtures for fraud detection and the verification of novel foods and alternative protein sources. This EU-FORA programme was focused on three core objectives namely: (i) the training of the fellow in utilising MS-based proteomics for food and feed safety analyses, (ii) the involvement of the fellow in the development of standardised operating procedures (SOP) for targeted and non-targeted proteomic MS-based workflows for species and tissues specific PAP identification in a national reference laboratory (NRL) and (iii) the transfer and implementation of MS-based approaches and standardised protocols for PAP analysis at the fellow's home institution. Altogether, this programme facilitates the broadening and diversification of use of MS-based proteomic methodologies for reinforcing their significance within the domains of food and feed safety research and regulatory science applications.

Refined mackerel oil increases hepatic lipid accumulation and reduces choline and choline-containing metabolites in the liver tissue in mice fed a Western diet.

In this study, we aimed to evaluate the impact of consuming refined mackerel oil (MO) from rest raw material on hepatic fat accumulation, glucose tolerance, and metabolomic changes in the liver from male C57BL/6N mice. The mice were fed either a Western diet (WD) or a chow diet, with 30 g or 60 g MO per kg of diet (3% or 6%) for 13 weeks. Body weight, energy intake, and feed efficiency were monitored throughout the experiment. A glucose tolerance test was conducted after 11 weeks, and metabolomic analyses of the liver were performed at termination. Inclusion of MO in the WD, but not in the chow diet, led to increased liver weight, hepatic lipid accumulation, elevated fasting blood glucose, reduced glucose tolerance, and insulin sensitivity. Hepatic levels of eicosapentaenoic and docosahexaenoic acid increased, but no changes in levels of saturated and monounsaturated fatty acids were observed. The liver metabolomic profile was different between mice fed a WD with or without MO, with a reduction in choline ether lipids, phosphatidylcholines, and sphingomyelins in mice fed MO. This study demonstrates that supplementing the WD, but not the chow diet, with refined MO accelerates accumulation of hepatic fat droplets and negatively affects blood glucose regulation. The detrimental effects of supplementing a WD with MO were accompanied by increased fat digestibility and overall energy intake, and lower levels of choline and choline-containing metabolites in liver tissue.

Crude oil exposure of early life stages of Atlantic haddock suggests threshold levels for developmental toxicity as low as 0.1 µg total polyaromatic hydrocarbon (TPAH)/L.

Atlantic haddock (Melanogrammus aeglefinus) embryos bind dispersed crude oil droplets to the eggshell and are consequently highly susceptible to toxicity from spilled oil. We established thresholds for developmental toxicity and identified any potential long-term or latent adverse effects that could impair the growth and survival of individuals. Embryos were exposed to oil for eight days (10, 80 and 300 µg oil/L, equivalent to 0.1, 0.8 and 3.0 µg TPAH/L). Acute and delayed mortality were observed at embryonic, larval, and juvenile stages with IC50 = 2.2, 0.39, and 0.27 µg TPAH/L, respectively. Exposure to 0.1 µg TPAH/L had no negative effect on growth or survival. However, yolk sac larvae showed significant reduction in the outgrowth (ballooning) of the cardiac ventricle in the absence of other extracardiac morphological defects. Due to this propensity for latent sublethal developmental toxicity, we recommend an effect threshold of 0.1 µg TPAH/L for risk assessment models.

Seasonal variations in mercury, cadmium, lead and arsenic species in Norwegian blue mussels (Mytilus edulis L.) - Assessing the influence of biological and environmental factors.

BACKGROUND: Blue mussels (Mytilus edulis L.) can accumulate undesirable substances, including the potentially toxic elements (PTEs) cadmium (Cd), mercury, (Hg), lead (Pb), arsenic (As) and As species. In this study, the levels of PTEs and As species were determined in samples of blue mussels to assess the influence of environmental and biological factors, and evaluate the potential risk associated with blue mussels in terms of food and feed safety. METHODOLOGY: Blue mussels were collected monthly from one location in Western Norway from February 2018 to December 2018, and from April 2019 to April 2020. Samples were analyzed for PTEs using inductively coupled plasma mass spectrometry (ICP-MS), and high-performance liquid chromatography (HPLC) coupled to ICP-MS. Temperature, salinity and fluorescence (chlorophyll a) were monitored in the seawater column by STD/CTD, to assess the potential influence of these environmental factors on the PTE levels in the mussels. RESULTS: The results showed seasonal variations in the PTEs, with somewhat higher concentrations in spring and winter months. Unusually high levels of total As (101.2 mg kg-1 dw) and inorganic As (53.6 mg kg-1 dw) were observed for some of the time points. The organic As species arsenobetaine was generally the major As species (17-82% of total As) in the mussels, but also simple methylated As species and arsenosugars were detected. Principal components analysis (PCA) did not show a consistent relationship between the environmental factors and the PTE concentrations, showing contrary results for some elements for the periods studied. The condition index (CI) could explain variations in element concentration with significant correlations for Cd (r = -0.67, p = 0.009) and Pb (r = -0.62, p = 0.02 in 2019/20 and r = -0.52, p = 0.02 in 2018), whereas the correlation between As and CI was not significant (r = 0.12 in 2018, and r = -0.06 in 2019/20). Higher concentrations of iAs and arsenosugars coincided with increased signals of chlorophyll a, suggesting that phytoplankton blooms could be a source of As in the blue mussels. CONCLUSION: To our knowledge, this is the first study of As species in blue mussels collected over a time period of two years, providing an insight into the natural variations of these chemical forms in mussels. In terms of mussel as food and future feed material, concentrations of Cd, Hg and Pb were below the maximum levels (MLs) established in the EU food and feed legislation. However, levels of As and iAs in mussels at some time points exceeded the MLs for As in the feed legislation, and the margin of exposure (MOE) was low if these mussels were for human consumption, highlighting the importance of determining the chemical forms of As in feed and food.

Co-exposure to UV radiation and crude oil increases acute embryotoxicity and sublethal malformations in the early life stages of Atlantic haddock (Melanogrammus aeglefinus).

Crude oil causes severe abnormalities in developing fish. Photomodification of constituents in crude oil increases its toxicity several fold. We report on the effect of crude oil, in combination with ultraviolet (UV) radiation, on Atlantic haddock (Melanogrammus aeglefinus) embryos. Accumulation of crude oil on the eggshell makes haddock embryos particularly susceptible to exposure. At high latitudes, they can be exposed to UV radiation many hours a day. Haddock embryos were exposed to crude oil (5-300 µg oil/L nominal loading concentrations) for three days in the presence and absence of UV radiation (290-400 nm). UV radiation partly degraded the eggs' outer membrane resulting in less accumulation of oil droplets in the treatment with highest oil concentration (300 µg oil/L). The co-exposure treatments resulted in acute toxicity, manifested by massive tissue necrosis and subsequent mortality, reducing LC50 at hatching stage by 60 % to 0.24 µg totPAH/L compared to 0.62 µg totPAH/L in crude oil only. In the treatment with nominal low oil concentrations (5-30 µg oil/L), only co-exposure to UV led to sublethal morphological heart defects. Including phototoxicity as a parameter in risk assessments of accidental oil spills is recommended.

compareMS2 2.0: An Improved Software for Comparing Tandem Mass Spectrometry Datasets.

It has long been known that biological species can be identified from mass spectrometry data alone. Ten years ago, we described a method and software tool, compareMS2, for calculating a distance between sets of tandem mass spectra, as routinely collected in proteomics. This method has seen use in species identification and mixture characterization in food and feed products, as well as other applications. Here, we present the first major update of this software, including a new metric, a graphical user interface and additional functionality. The data have been deposited to ProteomeXchange with dataset identifier PXD034932.

Developing a framework for open and FAIR data management practices for next generation risk- and benefit assessment of fish and seafood.

Risk and risk-benefit assessments of food are complex exercises, in which access to and use of several disconnected individual stand-alone databases is required to obtain hazard and exposure information. Data obtained from such databases ideally should be in line with the FAIR principles, i.e. the data must be Findable, Accessible, Interoperable and Reusable. However, often cases are encountered when one or more of these principles are not followed. In this project, we set out to assess if existing commonly used databases in risk assessment are in line with the FAIR principles. We also investigated how access, interoperability and reusability of data could be improved. We used the OpenFoodTox and the Seafood database as examples and showed how commonly used freely available open-source tools and repositories can be implemented in the data extraction process of risk assessments to increase data reusability and crosstalk across different databases.

Dietary Selenomethionine Reduce Mercury Tissue Levels and Modulate Methylmercury Induced Proteomic and Transcriptomic Alterations in Hippocampi of Adolescent BALB/c Mice.

Methylmercury (MeHg) is a well-known environmental contaminant, particularly harmful to the developing brain. The main human dietary exposure to MeHg occurs through seafood consumption. However, seafood also contains several nutrients, including selenium, which has been shown to interact with MeHg and potentially ameliorate its toxicity. The aim of this study was to investigate the combined effects of selenium (as selenomethionine; SeMet) and MeHg on mercury accumulation in tissues and the effects concomitant dietary exposure of these compounds exert on the hippocampal proteome and transcriptome in mice. Adolescent male BALB/c mice were exposed to SeMet and two different doses of MeHg through their diet for 11 weeks. Organs, including the brain, were sampled for mercury analyses. Hippocampi were collected and analyzed using proteomics and transcriptomics followed by multi-omics bioinformatics data analysis. The dietary presence of SeMet reduced the amount of mercury in several organs, including the brain. Proteomic and RNA-seq analyses showed that both protein and RNA expression patterns were inversely regulated in mice receiving SeMet together with MeHg compared to MeHg alone. Several pathways, proteins and RNA transcripts involved in conditions such as immune responses and inflammation, oxidative stress, cell plasticity and Alzheimer's disease were affected inversely by SeMet and MeHg, indicating that SeMet can ameliorate several toxic effects of MeHg in mice.

Proteomic analysis of hepatic effects of phenobarbital in mice with humanized liver.

Activation of the constitutive androstane receptor (CAR) may induce adaptive but also adverse effects in rodent liver, including the induction of drug-metabolizing enzymes, transient hepatocellular proliferation, and promotion of liver tumor growth. Human relevance of CAR-related adverse hepatic effects is controversially debated. Here, we used the chimeric FRG-KO mouse model with livers largely repopulated by human hepatocytes, in order to study human hepatocytes and their response to treatment with the model CAR activator phenobarbital (PB) in vivo. Mice received an intraperitoneal injection with 50 mg/kg body weight PB or saline, and were sacrificed after 72-144 h. Non-repopulated FRG-KO mice were used as additional control. Comprehensive proteomics datasets were generated by merging data obtained by targeted as well as non-targeted proteomics approaches. For the first time, a novel proteomics workflow was established to comparatively analyze the effects of PB on human and murine proteins within one sample. Analysis of merged proteome data sets and bioinformatics data mining revealed comparable responses in murine and human hepatocytes with respect to nuclear receptor activation and induction of xenobiotic metabolism. By contrast, activation of MYC, a key regulator of proliferation, was predicted only for mouse but not human hepatocytes. Analyses of 5-bromo-2'-deoxyuridine incorporation confirmed this finding. In summary, this study for the first time presents a comprehensive proteomic analysis of CAR-dependent effects in human and mouse hepatocytes from humanized FRG-KO mice. The data support the hypothesis that PB does induce adaptive metabolic responses, but not hepatocellular proliferation in human hepatocytes in vivo.

Pyrrolizidine alkaloid-induced transcriptomic changes in rat lungs in a 28-day subacute feeding study.

Pyrrolizidine alkaloids (PAs) are secondary plant metabolites synthesized by a wide range of plants as protection against herbivores. These toxins are found worldwide and pose a threat to human health. PAs induce acute effects like hepatic sinusoidal obstruction syndrome and pulmonary arterial hypertension. Moreover, chronic exposure to low doses can induce cancer and liver cirrhosis in laboratory animals. The mechanisms causing hepatotoxicity have been investigated previously. However, toxic effects in the lung are less well understood, and especially data on the correlation effects with individual chemical structures of different PAs are lacking. The present study focuses on the identification of gene expression changes in vivo in rat lungs after exposure to six structurally different PAs (echimidine, heliotrine, lasiocarpine, senecionine, senkirkine, and platyphylline). Rats were treated by gavage with daily doses of 3.3 mg PA/kg bodyweight for 28 days and transcriptional changes in the lung and kidney were investigated by whole-genome microarray analysis. The results were compared with recently published data on gene regulation in the liver. Using bioinformatics data mining, we identified inflammatory responses as a predominant feature in rat lungs. By comparison, in liver, early molecular consequences to PAs were characterized by alterations in cell-cycle regulation and DNA damage response. Our results provide, for the first time, information about early molecular effects in lung tissue after subacute exposure to PAs, and demonstrates tissue-specificity of PA-induced molecular effects.

RNA sequencing and proteomic profiling reveal different alterations by dietary methylmercury in the hippocampal transcriptome and proteome in BALB/c mice.

Methylmercury (MeHg) is a highly neurotoxic form of mercury (Hg) present in seafood. Here, we recorded and compared proteomic and transcriptomic changes in hippocampus of male BALB/c mice exposed to two doses of MeHg. Mice were fed diets spiked with 0.28 mg MeHg kg-1, 5 mg MeHg kg-1, or an unspiked control diet for 77 days. Total mercury content was significantly (P < 0.05) increased in brain tissue of both MeHg-exposed groups (18 ± 2 mg Hg kg-1 and 0.56 ± 0.06 mg Hg kg-1). Hippocampal protein and ribonucleic acid (RNA) expression levels were significantly altered both in tissues from mice receiving a low dose MeHg (20 proteins/294 RNA transcripts) and a high dose MeHg (61 proteins/876 RNA transcripts). The majority but not all the differentially expressed features in hippocampus were dose dependent. The combined use of transcriptomic and proteomic profiling data provided insight on the influence of MeHg on neurotoxicity, energy metabolism, and oxidative stress through several regulated features and pathways, including RXR function and superoxide radical degradation.

Effects of seafood consumption on mercury exposure in Norwegian pregnant women: A randomized controlled trial.

BACKGROUND: Seafood provides nutrients that are important for optimal development of the unborn child. However, seafood is also a source of contaminants including mercury (Hg) and methylmercury (MeHg) that may have adverse effects on neurodevelopment of the fetus. Humans are predominantly exposed to MeHg through seafood consumption, however, levels of MeHg vary considerably between species. OBJECTIVES: Investigate, in a randomized controlled trial (RCT) during pregnancy, if an increased intake of Atlantic cod (Gadus morhua), a fish species with relatively low levels of MeHg contamination, influences total hair Hg (THHg) concentrations in humans. METHODS: Pregnant women (n = 137) were enrolled in the RCT "Mommy's Food" (2016-2017), which was designed to increase iodine status. Participants were randomly assigned to intervention (400 g of cod fillets per week) or control (continued habitual diet) groups for 16 weeks (gestational week 20-36). THHg concentrations were measured at baseline and post-intervention using thermal decomposition, amalgamation, and atomic absorption spectrophotometry (US EPA method 7473). The trial is registered in ClinicalTrials.gov, NCT02610959. RESULTS: Post-intervention, the intervention group had median (inter-quartile range) THHg concentrations of 554 (392-805) µg/kg, and the control group 485 (341-740) µg/kg (p = 0.186). When adjusting for baseline THHg concentrations, there was a significant difference between the groups in those participants with baseline THHg concentrations below 534 µg/kg. Post-intervention, 8% of the study population exceeded the US EPA reference dose in hair (1,000 µg/kg) (intervention group: n = 6, control group: n = 4). CONCLUSION: THHg concentrations were generally low in both study groups of pregnant women, despite the relatively high seafood intake. While the intervention with 400 g of cod per week slightly increased THHg concentrations, it did not lead to an increase in number of subjects exceeding the US EPA reference dose; a dose level at which no adverse effects are expected to occur over a period of lifetime exposure.

Hepatotoxic pyrrolizidine alkaloids induce DNA damage response in rat liver in a 28-day feeding study.

Pyrrolizidine alkaloids (PA) are secondary plant metabolites that occur as food and feed contaminants. Acute and subacute PA poisoning can lead to severe liver damage in humans and animals, comprising liver pain, hepatomegaly and the development of ascites due to occlusion of the hepatic sinusoids (veno-occlusive disease). Chronic exposure to low levels of PA can induce liver cirrhosis and liver cancer. However, it is not well understood which transcriptional changes are induced by PA and whether all hepatotoxic PA, regardless of their structure, induce similar responses. Therefore, a 28-day subacute rat feeding study was performed with six structurally different PA heliotrine, echimidine, lasiocarpine, senecionine, senkirkine, and platyphylline, administered at not acutely toxic doses from 0.1 to 3.3 mg/kg body weight. This dose range is relevant for humans, since consumption of contaminated tea may result in doses of ~ 8 µg/kg in adults and cases of PA ingestion by contaminated food was reported for infants with doses up to 3 mg/kg body weight. ALT and AST were not increased in all treatment groups. Whole-genome microarray analyses revealed pronounced effects on gene expression in the high-dose treatment groups resulting in a set of 36 commonly regulated genes. However, platyphylline, the only 1,2-saturated and, therefore, presumably non-hepatotoxic PA, did not induce significant expression changes. Biological functions identified to be affected by high-dose treatments (3.3 mg/kg body weight) comprise cell-cycle regulation associated with DNA damage response. These functions were found to be affected by all analyzed 1,2-unsaturated PA.In conclusion, 1,2-unsaturated hepatotoxic PA induced cell cycle regulation processes associated with DNA damage response. Similar effects were observed for all hepatotoxic PA. Effects were observed in a dose range inducing no histopathological alterations and no increase in liver enzymes. Therefore, transcriptomics studies identified changes in expression of genes known to be involved in response to genotoxic compounds at PA doses relevant to humans under worst case exposure scenarios.

Non-target and suspected-target screening for potentially hazardous chemicals in food contact materials: investigation of paper straws.

A non-target screening strategy was developed for the safety evaluation of potentially hazardous chemicals in paper food contact materials (FCMs). A tentative list of suspect analytes was generated using publicly available FCM substance inventories, the presence of contaminants in paper straws was confirmed by high-resolution Orbitrap mass spectrometry. Data-independent and data-dependent MS and MS/MS results for candidate compounds were processed using a workflow including peak detection by deconvolution, blank subtraction, retention time alignment, formula assignment and fragmentation spectra search against spectral libraries followed by in silico generated spectra annotation. This workflow allowed for the identification of 74 suspect compounds, of which 40 were assigned a high confidence level of detection. A tentative in silico toxicity evaluation for mutagenic and carcinogenic activities was carried out. Using quantitative structure-activity relationship (QSAR) models it was found that two of the detected compounds tested positive for mutagenicity and three for carcinogenicity.

Cadmium in brown crab Cancer pagurus. Effects of location, season, cooking and multiple physiological factors and consequences for food safety.

Brown crab Cancer pagurus is appreciated as seafood in several European countries. However, cadmium levels in crabs can be elevated and their consumption may pose a hazard for human health. To assess if cadmium poses a threat to food safety in Norway, crabs were sampled at two different locations along the Norwegian coast: one in the South of Norway and one in the North of Norway. Cadmium levels were determined in different tissues (claw meat, hepatopancreas and inner meat). To highlight specific risk factors for cadmium, the concentration of cadmium was related to different exogenous (location, cooking and season) and physiological (size, sex, moulting stage, gonad maturation stage, condition) factors. The results confirmed previous findings of much higher cadmium levels in brown crab sampled in the North of Norway compared to the South. Cooking of crabs further led to higher concentrations in claw meat. The effect of season on cadmium levels was different in the North and South and no clear patterns could be identified, probably due to a high inter-individual variation in cadmium levels. Size showed a correlation with the total amount of cadmium for crabs in the North indicating an accumulation of cadmium over time; together with a slower growth, this may lead to the higher cadmium levels, observed in the crabs from Northern Norway. The risk connected to cadmium exposure when consuming brown crab mainly depends on the consumption pattern, the parts of the crab consumed and the origin of the crab. Regardless of origin, the consumption of claw meat does not display a consumer health risk. However, the consumption of meals consisting of inner meat only and inner meat of brown crab from Northern Norway may pose a health risk.

Species-Specific Discrimination of Insect Meals for Aquafeeds by Direct Comparison of Tandem Mass Spectra.

Insect protein has the potential to become a sustainable feed ingredient for the rapidly growing aquaculture industry. In the European Union, insect derived protein is placed under the same legislation as processed animal proteins (PAP). It is therefore of interest to develop methods for regulatory use, which unambiguously identify the species origin of insect-based ingredients. We performed (i) total protein quantification of insect samples using the traditional nitrogen-to-protein conversion factor of 6.25 and the sum of anhydrous amino acids, (ii) quantitative amino acid profiling and (iii) high-throughput tandem mass spectrometry to describe and differentiate 18 different commercial-grade insect meal samples derived from Hermetia illucens (8), Tenebrio molitor (5), Alphitobius diaperinus (3) and Acheta domesticus (2). In addition, we investigated and compared different protein extraction and digestion protocols for proteomic analysis. We found that irrespective of sample preparation, shotgun proteomics in combination with direct spectral comparison were able to differentiate insect meal according to their taxonomic classification. The insect specific spectral libraries created in the present work can in future be used to develop more sensitive targeted methods of insect PAP identification and quantification in commercial feed mixtures.

Tolerance and dose-response assessment of subchronic dietary ethoxyquin exposure in Atlantic salmon (Salmo salar L.).

Ethoxyquin (EQ; 6-Ethoxy-2,2,4-trimethyl-1,2-dihydroquinoline) has been used as an antioxidant in feed components for pets, livestock and aquaculture. However, possible risks of EQ used in aquafeed for fish health have not yet been characterized. The present study investigated the toxicity and dose-response of subchronic dietary EQ exposure at doses ranging from 41 to 9666 mg EQ/kg feed in Atlantic salmon (Salmo salar L.). Feed at concentrations higher than 1173 mg EQ/kg were rejected by the fish, resulting in reduced feed intake and growth performance. No mortality was observed in fish exposed to any of the doses. A multi-omic screening of metabolome and proteome in salmon liver indicated an effect of dietary EQ on bioenergetics pathways and hepatic redox homeostasis in fish fed concentrations above 119 mg EQ/kg feed. Increased energy expenditure associated with an upregulation of hepatic fatty acid ß-oxidation and induction and carbohydrate catabolic pathways resulted in a dose-dependent depletion of intracytoplasmic lipid vacuoles in liver histological sections, decreasing whole body lipid levels and altered purine/pyrimidine metabolism. Increased GSH and TBARS in the liver indicated a state of oxidative stress, which was associated with activation of the NRF2-mediated oxidative stress response and glutathione-mediated detoxification processes. However, no oxidative DNA damage was observed. As manifestation of altered energy metabolism, the depletion of liver intracytoplasmic lipid vacuoles was considered the critical endpoint for benchmark dose assessment, and a BMDL10 of 243 mg EQ/kg feed was derived as a safe upper limit of EQ exposure in Atlantic salmon.

Factors influencing risk assessments of brominated flame-retardants; evidence based on seafood from the North East Atlantic Ocean.

Brominated flame-retardants (BFRs) such as polybrominated diphenyl ethers (PBDE) and hexabromocyclododecane (HBCD) are considered hazardous to human health. Due to their persistence, they are still present in the environment and in biota and seafood is major contributor of BFRs to human exposure. Here, we used data from >9700 samples of wild and farmed fish, fish feed and fish feed ingredients collected from the North Atlantic between 2006 and 2016 aiming to investigate factors influencing the risk assessments of BFRs. Due to most representative number of analyses, PBDEs were the main focus of investigation. Mean ∑PBDE in fillet samples ranged from below quantification in Atlantic cod fillet to 2.0 µg kg-1 in Atlantic halibut. The main congener contributing to the ∑PBDE in all species was BDE 47. Factors affecting the level of BFR in seafood were multifaceted, and the levels were within species mainly determined by fish age, geographical origin and time of sampling. BDE 47, 99, 153 and HBCD were selected for margin of exposure (MOE) evaluation. When other sources of BFR than seafood were excluded, our risk assessment showed low risk at the current dietary intake of seafood. However, the dietary intake of BDE 99 may be of concern for toddlers when all sources are considered. The choice of fish species, dietary studies, choice of statistics, as well as exposure from other sources than seafood, were all factors that influenced the final MOE of BFRs. We propose the use of regression on order statistics as a tool for risk assessment, to illustrate means and spreads in large surveillance datasets to avoid the issue of measurements below the limit of quantification. A harmonized, updated evaluation of the risk associated with exposure to BFRs from diet, air and dust is warranted, where the fish species most commonly consumed also is taken into consideration.

Subchronic dietary exposure to ethoxyquin dimer induces microvesicular steatosis in male BALB/c mice.

The use of the synthetic antioxidant ethoxyquin (6-ethoxy-2,2,4-trimethyl-1,2-dihydroquinoline; EQ) in animal feed results in the presence of EQ residues and metabolites, including the EQ dimer (1,8'-bi(6-ethoxy-2,2,4-trimethyl-1,2-dihydroquinoline); EQDM) in animal food products. To investigate the toxicity and dose-response of dietary exposure to EQDM, male BALB/c mice were exposed to one of six dietary doses of EQDM, ranging from 0.015 to 518 mg/kg body weight/day for 90 days. Doses above 10 mg/kg body weight/day affected whole body lipid metabolism resulting in increased liver weights and decreased adipose tissue mass. Metabolomic screening of livers revealed alterations indicating incomplete fatty acid ß-oxidation and hepatic oxidative stress. Histopathological evaluation and biochemical analyses of the liver confirmed the development of microvesicular steatosis and activation of the glutathione system. Hepatic protein profiling and pathway analyses suggested that EQDM-induced responses are mediated through activation of CAR/PXR nuclear receptors and induction of a NRF2-mediated oxidative stress response. Based on the development of microvesicular steatosis as the critical endpoint, a Reference Point for dietary EQDM exposure was established at 1.1 mg/kg body weight/day (BMDL10) from benchmark dose modelling. Applying an uncertainty factor of 200, an Acceptable Daily Intake of 0.006 mg EQDM/kg body weight was proposed.

Identification and evaluation of potentially mutagenic and carcinogenic food contaminants.

Heat processing of food gives rise to a plethora of chemical compounds whose toxicological effects are largely unknown. Due to a general lack of experimental toxicological data, assessing the risks associated with the consumption of these substances remains a challenge. Computer models that allow for an in silico prediction of physicochemical and toxicological characteristics, may be able to fill current data gaps and facilitate the risk assessment of toxicologically uncharacterised chemicals, their transformation products and their biological metabolites. The overall aims of the present project were for the fellow: (i) to get acquainted with the application of computational toxicological analyses tools in risk assessment based on results and experiences from previous research performed at the German Federal Institute for Risk Assessment (BfR); and (ii) to apply the newly gained skills on historic and novel data using updated and additional in silico tools. The project contributed to the continuous further education of the fellow in the use of computational toxicology tools, corroborated findings related to the safety of heat-induced contaminants and laid the foundations for future collaborations between the fellow's home institution, the Institute of Marine Research (IMR) in Norway, and the BfR in Germany.