Glycidol exposure evaluation of humans who have ingested diacylglycerol oil containing glycidol fatty acid esters using hemoglobin adducts

Internal exposure to heat-induced food contaminants in omnivores, vegans and strict raw food eaters: biomarkers of exposure to 2- and 3-monochloropropanediol (urinary excretion) and glycidol (hemoglobin adduct N-2,3-dihydroxypropyl-Val)

Fatty acid esters of 2/3-monochloropropanediol (2/3-MCPD) and glycidol are formed mainly during heat processing (deodorization) of vegetable oils, and are hydrolyzed by lipases in the gastrointestinal tract leading to the absorption of 2/3-MCPD and glycidol. The International Agency for Research on Cancer (IARC) has classified 3-MCPD as possibly and glycidol as probably carcinogenic to humans. The aims of the current work were to clarify the exposure to 2/3-MCPD and glycidol associated with different dietary habits (omnivore, vegan, raw-food eating), and the exposure development between 2017 and 2021 in German study participants. The questions were addressed using the daily urinary excretion of 2/3-MCPD and the hemoglobin adduct N-(2,3-dihydroxypropyl)-Val (DHP-Val) formed from glycidol as biomarkers of exposure, which were determined in two dietary studies including 36 omnivores, 36 vegans and 16 strict raw food eaters (abstaining from any heated food for at least four months). The median urinary excretion of 2- and 3-MCPD in non-smoking omnivores and vegans was 0.87 and 1.35 µg/day (2-MCPD), respectively, and 0.79 and 1.03 µg/day (3-MCPD), respectively. The 2/3-MCPD concentrations in urine samples of raw food eaters were usually below the limit of detection. The median DHP-Val levels in non-smoking vegans and omnivores were 3.9 pmol/g Hb each, and 1.9 pmol/g Hb in raw food eaters. Between 2017 and 2021, the exposure to 3-MCPD and glycidol did not change, however, the median 2-MCPD excretion decreased (p = 0.02, omnivores and vegans combined). The correlation between daily excretions of 2/3-MCPD determined 4 years apart was weak, whereas a moderate correlation was observed for DHP-Val (rS = 0.66) in this timeframe. In conclusion, the exposure to glycidol in omnivores and vegans was alike, whereas the 2/3-MCPD exposure was somewhat (albeit not significantly) higher in vegans. While 2/3-MCPD were hardly detectable in urine samples of raw food eaters, the median DHP-Val level (about 50% of those in omnivores) indicates a glycidol source independent of the dietary exposure.

A comparison of the exposure system of glycidol-related chemicals on the formation of glycidol-hemoglobin adducts

Glycidol fatty acid esters that are present in foods are degraded in vivo to the animal carcinogen glycidol, which binds to the N-terminal valine of hemoglobin (Hb) to form N-(2,3-dihydroxypropyl)valine (diHOPrVal) adducts. The existence of other chemicals that are converted to glycidol is unknown. To determine the effect of different exposure conditions on the formation of diHOPrVal adducts, several glycidol-related chemicals (3-monochloropropane-1,2-diol; 3-MCPD, epichlorohydrin, glyceraldehyde, acrylic acid, and 1,2-propanediol) were evaluated using in vitro and in vivo (single/repeated dose) methods. In vitro, the reaction of 3-MCPD or epichlorohydrin with human Hb produced 17% and 0.7% of diHOPrVal, as compared to equimolar glycidol, respectively. Following a single administration of glycidol-related compounds to ICR mice, diHOPrVal formation was observed only in the epichlorohydrin-treated group after day 5 of exposure. After 14 days of repeated dosing, the amounts of diHOPrVal produced by epichlorohydrin and 3-MCPD in vivo were <1% of diHOPrVal produced by an equal molar concentration of glycidol. Furthermore, glyceraldehyde group produced 0.2% of diHOPrVal at the same molar concentration of glycidol equivalents, in which diHOPrVal formation could not be confirmed by the in vitro assay. The results indicate the usefulness of diHOPrVal as an exposure marker for glycidol; however, the contribution of its formation in vivo by exposure to various chemicals will be necessary to validate and interpret the results.

Potential Role of Lipase Activity on the Internal Exposure Assessment of Glycidol Released from Its Fatty Acid Esters

Glycidyl fatty acid esters (GEs) can be found in food, and they can be converted into genotoxic animal carcinogen glycidol in vivo by the action of lipase. This study examined whether human ingestion of charbroiled pork containing high levels of GEs (300 µg/day) increased glycidol-hemoglobin adduct (diHOPrVal), a marker of internal exposure to glycidol using LC-MS/MS. Contrary to expectation, the diHOPrVal value before ingesting charbroiled pork was 3.11 ± 1.10 pmol/g globin, which slightly decreased to 2.48 ± 0.47 pmol/g globin after 5 days of consumption. The decrease in lipase activity caused by the continuous consumption of lipid-rich foods such as meat in humans might decrease internal exposure to glycidol released from its esters. Thus, lipase activity was measured in C57/BL6J mice fed a high-fat diet (HFD) for 8 weeks, and diHOPrVal formation was measured after the administration of glycidyl oleate. Lipase activity was significantly lower in the HFD group than in the normal diet group. The amount of diHOPrVal was reduced in the HFD group. Therefore, the lipase activity was reduced by HFD, thereby decreasing the degradation of glycidol from glycidyl oleate. These results indicate that changes in lipase activity depending on the amount of lipids in the diet may affect the assessment of GEs exposure, and monitoring the lipase activity would provide a comprehensive understanding of exposure assessment.

Development of physiologically based toxicokinetic models for 3-monochloropropane-1,2-diol and glycidol

3-Monochloropropane-1,2-diol (3-MCPD), glycidol, together with their fatty acid esters are commonly presented in various food and have shown carcinogenicity in various laboratory animals. Public health risk assessment of 3-MPCD and glycidol exposure relies on quantitative tools that represent their in vivo toxicokinetics. In order to better understand the absorption, distribution, metabolism, and excretion profiles of 3-MCPD and glycidol in male rats, a physiologically based pharmacokinetic (PBTK) model was developed. The model's predictive power was evaluated by comparing in silico simulations to in vivo time course data obtained from experimental studies. Results indicate that our PBTK model successfully captured the toxicokinetics of both free chemicals in key organs, and their metabolites in accessible biological fluids. With the validated PBTK model, we then gave an animal-free example on how to extrapolate the toxicological knowledge acquired from a single gavage to a realistic dietary intake scenario. Three biomarkers, free compound in serum, urinary metabolite DHPMA, and glycidol-hemoglobin adduct (diHOPrVal) were selected for in silico simulation following constant dietary intakes, and their internal levels were correlated with proposed external daily exposure via reverse dosimetry approaches. Taken together, our model provides a computational approach for extrapolating animal toxicokinetic experiments to biomonitoring measurement and risk assessment.

Simultaneous quantification of eight hemoglobin adducts of genotoxic substances by isotope-dilution UHPLC-MS/MS

Various genotoxic carcinogens ubiquitously present in the human environment or respective reactive metabolites form adducts in DNA and proteins, which can be used as biomarkers of internal exposure. For example, the mass spectrometric determination of Val adducts at the N-termini of hemoglobin (Hb) peptide chains after cleavage by an Edman degradation has a long tradition in occupational medicine. We developed a novel isotope-dilution UHPLC-MS/MS method for the simultaneous quantification of Val adducts of eight genotoxic substances in Hb after cleavage with fluorescein-5-isothiocyanate (FIRE procedure™). The following adducts were included [sources in square brackets]: N-(2,3-dihydroxypropyl)-Val [glycidol], N-(2-carbamoylethyl)-Val [acrylamide], N-(2-carbamoyl-2-hydroxyethyl)-Val [glycidamide], N-((furan-2-yl)methyl)-Val [furfuryl alcohol], N-(trans-isoestragole-3′-yl)-Val [estragole/anethole], N-(3-ketopentyl)-Val [1-penten-3-one], N-(3-ketooctanyl)-Val [1-octene-3-one], and N-benzyl-Val [benzyl chloride], each of which was quantified with a specific isotope-labeled standard. The limits of quantification were between 0.014 and 3.6 pmol/g Hb (using 35 mg Hb per analysis); other validation parameters were satisfactory according to guidelines of the U.S. Food and Drug Administration. The quantification in erythrocyte samples of human adults (proof of principle) showed that the median levels of Hb adducts of acrylamide, glycidamide, and glycidol were found to be significantly lower in six non-smokers (25.9, 12.2, and 4.7 pmol/g Hb, respectively) compared to those of six smokers (69.0, 44.2, and 8.6 pmol/g Hb, respectively). In summary, the method surpasses former techniques of Hb adduct quantification due to its simplicity, sensitivity, and accuracy. It can be extended continuously with other Hb adducts and will be used in epidemiological studies on internal exposure to carcinogens.

Factors Influencing the Formation of Chemical-Hemoglobin Adducts

Hemoglobin (Hb) adducts have been used as biomarkers for the internal exposure to chemicals. Simultaneous exposure to chemicals that bond with the N-terminal valine of Hb to form adducts, such as glycidol, acrylamide, and glucose, may affect the formation of the individual Hb adducts. In this study, various factors influencing the formation of chemical–Hb adducts were analyzed using in vitro and in vivo systems. In the in vitro assays, the formation of glycidol– and acrylamide–Hb adducts was altered in the presence of glucose, serum albumin, and other chemicals. In contrast, in the in vivo experiments, glycidol– and acrylamide–Hb adduct formation was unchanged in mice exposed to glycidol and acrylamide. The interaction between glycidol and acrylamide with residues other than the N-terminal valine of Hb was analyzed using the protein thermal shift assay. Glycidol and acrylamide also interacted with amino acid residues other than the N-terminal valine of Hb. The presence of other blood components, such as amino acids, may affect the formation of chemical–Hb adducts. Further research is expected to elucidate the remaining unknown factors that affect the formation of chemical–Hb adducts.

Glycidol Fatty Acid Ester and 3-Monochloropropane-1,2-Diol Fatty Acid Ester in Commercially Prepared Foods

Glycidyl fatty acid esters (GEs), which are the main pollutant in processed oils, are potential mutagens or carcinogens. 3-Monochloropropane-1,2-diol fatty acid esters (3-MCPDEs) are also well-known food processing contaminants. 3-MCPDEs are believed to be a precursor to GEs in foodstuffs. In vivo, lipase breaks down the phosphate ester of GEs and 3-MCPDEs to produce glycidol and 3-MCPD, respectively, which are genotoxic carcinogens. Thus, it is important to determine human exposure to GEs and 3-MCPDEs through foodstuffs. There are only reports on the amount of GE and 3-MCPDE in cooking oils and cooked foods. The content in multiple types of foods that are actually on the market was not clarified. In this study, 48 commercially prepared foods were analyzed to identify other sources of exposure to GE and 3-MCPDE. All of them contained relatively high amounts of GEs and 3-MCPDEs. The correlation between GEs and 3-MCPDEs in individual foods was examined. There was a correlation between the amounts of GEs and 3-MCPDEs in the food products (r = 0.422, p < 0.005). This is the first report on the content in multiple types of commercially prepared foods that are actually on the market was clarified.

Detection of a Hemoglobin Adduct of the Food Contaminant Furfuryl Alcohol in Humans: Levels of N-((Furan-2-yl)methyl)-valine in Two Epidemiological Studies

SCOPE

Furfuryl alcohol is a heat-induced food contaminant, classified as possibly carcinogenic to humans. The proximal carcinogen 2-sulfoxymethylfuran leads to adduct formation in DNA and proteins (e.g., N-((furan-2-yl)methyl)-Val (FFA-Val) in hemoglobin).

METHODS AND RESULTS

This study analyzed human erythrocyte samples from two studies for the presence of FFA-Val: the Risks and Benefits of a Vegan Diet study (RBVD; 72 adults) and the ENVIRonmental influence ON early AGEing birth cohort study (ENVIRONAGE; 100 mother-newborn pairs). In the RBVD study, FFA-Val levels are lower in vegans compared to omnivores (median 13.0 vs 15.8 pmol g^-1 hemoglobin, p = 0.008), and lower in non-smokers compared to smokers (median 14.1 vs 17.0 pmol g^-1 hemoglobin, p = 0.003). In the birth cohort, FFA-Val levels are distinctly higher in maternal compared to newborn samples (median 15.2 vs 2.2 pmol g^-1 hemoglobin, p < 0.001).

CONCLUSIONS

FFA-Val, hitherto detected only in blood samples of mice, is quantifiable in all human samples, indicating a general exposure to furfuryl alcohol. The low adduct levels in blood samples from newborn children suggested that the placenta is a barrier to furfuryl alcohol. Dietary habits and tobacco smoking are two main influencing factors on the formation of FFA-Val, which may be of use as a biomarker of exposure to furfuryl alcohol.

Does External Exposure of Glycidol-Related Chemicals Influence the Forming of the Hemoglobin Adduct, N-(2,3-dihydroxypropyl)valine, as a Biomarker of Internal Exposure to Glycidol?

Glycidyl fatty acid esters (GE) are constituents of edible oils and fats, and are converted into glycidol, a genotoxic substance, in vivo. N-(2,3-dihydroxypropyl)valine (diHOPrVal), a hemoglobin adduct of glycidol, is used as a biomarker of glycidol and GE exposure. However, high background levels of diHOPrVal are not explained by daily dietary exposure to glycidol and GE. In the present study, several glycidol-related chemicals (glycidol, (±)-3-chloro-1,2-propanediol, glycidyl oleate, epichlorohydrin, propylene oxide, 1-bromopropane, allyl alcohol, fructose, and glyceraldehyde) that might be precursors of diHOPrVal, were administered to mice, and diHOPrVal formation from each substance was examined with LC-MS/MS. DiHOPrVal was detected in animals treated with glycidol and glycidyl oleate but not in mice treated with other chemicals (3-MCPD, epichlorohydrin, propylene oxide, 1-bromopropane, allyl alcohol, fructose, and glyceraldehyde). The amount of diHOPrVal per administered dose produced from other chemicals was negligible compared to the amounts associated with dietary glycidol and GE. The present study provides important knowledge for exploring other sources for internal exposure to glycidol.

Metabolomics-based biomarker analysis of dihydroxypropyl mercapturic acid isomers from 3-monochloropropane-1,2-diol and glycidol for evaluation of toxicokinetics in rats and daily internal exposure in humans

3-Monochloropropane-1,2-diol (3-MCPD), glycidol, and their esters are some major sources of risk factors during food processing. Here we showed the biomarker analysis of 2,3-dihydroxypropyl mercapturic acid (DHPMA) isomers which derived from the metabolism of 3-MCPD, glycidol, and their esters in urine of rats and humans. Iso-DHPMA, a novel urinary metabolite, was discovered and detected in urine of rats, which were orally administered with glycidol but not 3-MCPD. Using the quadrupole-orbitrap high-resolution mass spectrometry, we confirmed that iso-DHPMA appeared a specific biomarker which derived from glycidol. The limit of quantification (signal-to-noise ratio, 10:1) of the analytes in urine of rats and humans were 1.33 ng/mL and 1.56 ng/mL, respectively. Acceptable within-laboratory reproducibility (RSD<9.0%) and spiking recovery (94.7%-100.1%) substantially supported the use of current method for robust biomarker analysis, which was successfully applied to the toxicokinetic study of DHPMA in rats and short-term internal exposure to 3-MCPD and glycidol in humans.

Effect of chloride on the formation of 3-monochloro-1,2-propanediol fatty acid diesters and glycidol fatty acid esters in fish, meats and acylglycerols during heating

The effects of the presence of chloride on the formation of 3-monochloro-1,2-propanediol fatty acid esters (3-MCPDEs) and glycidol fatty acid esters (GEs) in saltwater fish, meats and acylglycerols (diacylglycerol and triacylglycerol) during heating were investigated in this study. Five saltwater fish species (salmon, saury, yellowtail, mackerel and Spanish mackerel) were grilled with a fish griller. 3-MCPDEs and GEs were detected in all of the grilled fish samples. The total amount of GEs was higher than 3-MCPDEs. Beef and pork patties with or without sodium chloride (1.5%) were cooked using gaseous fuel. The formation of 3-MCPDEs was significantly increased by the addition of sodium chloride to the meat patties, whereas the concentration of GEs in the cooked meat patties was not changed by the content of sodium chloride. Hexadecane solutions of diacylglycerol or triacylglycerol containing FeCl₃ were heated at 240°C. The formation of 3-MCPDEs was greatly increased by adding FeCl₃ to the solutions of triacylglycerol. The amounts of 3-MCPDEs decreased with the extension of the heating time. From these results, it is suggested that 3-MCPDEs and GEs are formed in saltwater fish and meats by cooking, and that the formation of 3-MCPDEs was affected by chloride in foodstuffs.

The hemoglobin adduct N-(2,3-dihydroxypropyl)-valine as biomarker of dietary exposure to glycidyl esters: a controlled exposure study in humans

Fatty acid esters of glycidol (glycidyl esters) are heat-induced food contaminants predominantly formed during industrial deodorization of vegetable oils and fats. After consumption, the esters are digested in the gastrointestinal tract, leading to a systemic exposure to the reactive epoxide glycidol. The compound is carcinogenic, genotoxic and teratogenic in rodents, and rated as probably carcinogenic to humans (IARC group 2A). Assessment of exposure from occurrence and consumption data is difficult, as lots of different foods containing refined oils and fats may contribute to human exposure. Therefore, assessment of the internal exposure using the hemoglobin adduct of glycidol, N-(2,3-dihydroxypropyl)-valine (2,3-diHOPr-Val), may be promising, but a proof-of-principle study is needed to interpret adduct levels with respect to the underlying external exposure. A controlled exposure study was conducted with 11 healthy participants consuming a daily portion of about 36 g commercially available palm fat with a relatively high content of ester-bound glycidol (8.7 mg glycidol/kg) over 4 weeks (total amount 1 kg fat, individual doses between 2.7 and 5.2 µg/kg body weight per day). Frequent blood sampling was performed to monitor the 2,3-diHOPr-Val adduct levels during formation and the following removal over 15 weeks, using a modified Edman degradation and ultrahigh performance liquid chromatography-tandem mass spectrometry (UHPLC-MS/MS). Results demonstrated for the first time that the relatively high exposure during the intervention period was reflected in corresponding distinct increases of 2,3-diHOPr-Val levels in all participants, following the expected slope for hemoglobin adduct formation and removal over time. The mean adduct level increased from 4.0 to 12.2 pmol 2,3-diHOPr-Val/g hemoglobin. By using a nonlinear mixed model, values for the adduct level/dose ratio (k, mean 0.082 pmol 2,3-diHOPr-Val/g hemoglobin per µg glycidol/kg body weight) and the adduct lifetime (τ, mean 104 days, likely the lifetime of the erythrocytes) were determined. Interindividual variability was generally low. 2,3-DiHOPr-Val was therefore proven to be a biomarker of the external dietary exposure to fatty acid esters of glycidol. From the background adduct levels observed in our study, a mean external glycidol exposure of 0.94 µg/kg body weight was estimated. This value is considerably higher than current estimates for adults using occurrence and consumption data of food. Possible reasons for this discrepancy are discussed (other oral or inhalational glycidol sources, endogenous formation, exposure to other chemicals also forming the adduct 2,3-diHOPr-Val). Further research is necessary to clarify the issue.

Exposure assessment of process-related contaminants in food by biomarker monitoring

Exposure assessment is a fundamental part of the risk assessment paradigm, but can often present a number of challenges and uncertainties. This is especially the case for process contaminants formed during the processing, e.g. heating of food, since they are in part highly reactive and/or volatile, thus making exposure assessment by analysing contents in food unreliable. New approaches are therefore required to accurately assess consumer exposure and thus better inform the risk assessment. Such novel approaches may include the use of biomarkers, physiologically based kinetic (PBK) modelling-facilitated reverse dosimetry, and/or duplicate diet studies. This review focuses on the state of the art with respect to the use of biomarkers of exposure for the process contaminants acrylamide, 3-MCPD esters, glycidyl esters, furan and acrolein. From the overview presented, it becomes clear that the field of assessing human exposure to process-related contaminants in food by biomarker monitoring is promising and strongly developing. The current state of the art as well as the existing data gaps and challenges for the future were defined. They include (1) using PBK modelling and duplicate diet studies to establish, preferably in humans, correlations between external exposure and biomarkers; (2) elucidation of the possible endogenous formation of the process-related contaminants and the resulting biomarker levels; (3) the influence of inter-individual variations and how to include that in the biomarker-based exposure predictions; (4) the correction for confounding factors; (5) the value of the different biomarkers in relation to exposure scenario's and risk assessment, and (6) the possibilities of novel methodologies. In spite of these challenges it can be concluded that biomarker-based exposure assessment provides a unique opportunity to more accurately assess consumer exposure to process-related contaminants in food and thus to better inform risk assessment.

Alpha-linolenic acid-enriched diacylglycerol oil does not promote tumor development in tongue and gastrointestinal tract tissues in a medium-term multi-organ carcinogenesis bioassay using male F344 rat

Alpha-linolenic acid (ALA)-enriched diacylglycerol (DAG) oil is an edible oil enriched with DAG (>80%) and ALA (>50%). The present study investigated whether ALA-DAG oil promotes tumorigenesis in the tongue and gastrointestinal tract, using a rat medium-term multi-organ carcinogenesis bioassay model. Rats were treated with five genotoxic carcinogens to induce multi-organ tumorigenesis until week 4, and from 1 week after withdrawal, fed a semi-synthetic diet (AIN-93G) containing ALA-DAG oil at concentrations of 0, 13,750, 27,500, and 55,000 ppm. Rats fed AIN-93G containing 55,000 ppm ALA-triacylglycerol or a standard basal diet served as reference and negative control groups, respectively. Animals were euthanized at week 30. ALA-DAG oil was shown to have no effects on survival, general condition, body weight, food consumption, or organ weight. More discolored spots were observed in the stomachs of the 13,750- and 55,000-ppm ALA-DAG groups than in those of the control groups; however, there were no differences in the frequency of histopathological findings across groups. There were no meaningful increases in the incidence of pre-neoplastic and neoplastic lesions in the tongue and gastrointestinal tract among the groups. We therefore conclude that ALA-DAG oil does not promote tumor development in the digestive system.

Genotoxicity evaluation of alpha-linolenic acid-diacylglycerol oil

•

We evaluated the genotoxicity of ALA-DAG oil using standard tests.

•

Bacterial reverse mutation and in vitro/in vivo micronucleus tests were conducted.

•

No genotoxicity was observed under the testing conditions.

The alpha-linolenic acid (ALA)-diacylglycerol (DAG) oil is an edible oil enriched with DAG (>80%) and ALA (>50%). Although DAG oil, which mainly consists of oleic and linoleic acids has no genotoxic concerns, the fatty acid composition could affect the chemical property of DAG. Therefore, the purpose of this study was to evaluate the genotoxicity of ALA-DAG oil using standard genotoxicity tests in accordance with the OECD guidelines. ALA-DAG oil showed negative results in the bacterial reverse mutation test (Ames test) and in vitro micronucleus test in cultured Chinese hamster lung cells with and without metabolic activation, and in the in vivo bone marrow micronucleus test in mice. Our results did not show any genotoxicity, suggesting that the fatty acid composition had no deleterious effects. We conclude that ALA-DAG oil had no genotoxicity concerns under the testing conditions.