Scientific Opinion on the risks for public health related to the presence of zearalenone in food

Zearalenone is a mycotoxin produced by several Fusarium species. It is commonly found in maize but can be found also in other crops such as wheat, barley, sorghum and rye. The European Commission asked the European Food Safety Authority to review the safety of zearalenone and the risk to consumers of a possible increase of the maximum level (ML) for zearalenone in breakfast cereals. A total of 13,075 analytical results obtained on food samples and 9,877 results on unprocessed grains sampled by 19 European countries in 2005 2010 were used in the evaluation. The highest concentrations of zearalenone were reported for wheat bran, corn and products thereof (e.g. corn flour, cornflakes). Grains and grain-based foods, in particular grains and grain milling products, bread and fine bakery wares, made the largest contribution to the estimated zearalenone exposures. Vegetable oils also made an important contribution to the zearalenone exposure. The critical effects of zearalenone result from its oestrogenic activity. Based on recent data in the most sensitive animal species, the pig, and taking into account comparisons between pigs and humans, the Panel on Contaminants in the Food Chain established a tolerable daily intake (TDI) for zearalenone of 0.25 μg/kg b.w. Estimates of chronic dietary exposure to zearalenone based on the available occurrence data are below or in the region of the TDI for all age groups and not a health concern. A potential increase in the ML for zearalenone in breakfast cereals from 50 µg/kg to 75, 100, 125 or 150 µg/kg is unlikely to result in a chronic dietary exposure exceeding the TDI. In a worst case scenario it is possible that an individual could consume the same batch of breakfast cereal containing zearalenone at the ML every day for 2 to 4 weeks, in which case exposures may exceed the TDI.

© European Food Safety Authority, 2011

Zearalenone is a phenolic resorcyclic acid lactone mycotoxin produced by several Fusarium species, particularly F. graminearum (formerly called F. roseum) and also F. culmorum, F.equiseti and F. verticillioides. It is commonly found in maize but can be found also in other crops such as wheat, barley, sorghum and rye throughout various countries of the world. Generally, the Fusarium species grow and invade crops in moist cool field conditions. F graminearum also produces trichothecenes, such as deoxynivalenol, 15-acetyldeoxynivalenol, 3-acetyldeoxynivalenol, nivalenol, 4 acetylnivalenol and fusarenon-X. Whilst zearalenone is primarily a field contaminant, toxin production may also occur under poor storage conditions.

Zearalenone has previously been evaluated by the Joint Food and Agriculture Organization of the United Nations (FAO)/World Health Organization (WHO) Expert Committee on Food Additives (JECFA) which established a provisional maximum tolerable daily intake (PMTDI) of 0.5 µg/kg bodyweight (b.w.) in 2000, based on the oestrogenic activity of zearalenone and its metabolites, in the most sensitive animal species, the pig. Also, in 2000, the Scientific Committee on Food (SCF) established a temporary TDI (t-TDI) of 0.2 µg/kg b.w. This TDI was designated as temporary and included an additional uncertainty factor because of some ’deficiencies in the data base (e.g. the question of a higher sensitivity of prepubertal vs adult pigs raised by new information from the study of Bauer et al., 1987)’. The SCF recommended that additional studies were needed to determine the no-hormonal-effect level in pre-pubertal pigs, on the potential genotoxicity of zearalenone, on species differences in metabolism, and on blood levels of zearalenone in humans in order to help clarify the toxicokinetic behaviour.

The European Commission (EC), in considering if changes are needed to the current legal provisions for the presence of zearalenone in bran and breakfast cereals, asked the European Food Safety Authority (EFSA) to provide a scientific opinion on the effects on consumer health risk of a possible increase of the maximum level (ML) for zearalenone in breakfast cereals. The request also specified a review of the opinion of the SCF in the light of results of more recent toxicological studies.

Methods for analysis of zearalenone in food and feed are well established and are also suitable for analysis of biological samples with appropriate clean-up procedures, such as use of immunoaffinity columns. Analysis mostly uses high performance liquid chromatography coupled to fluorescence detection or triple quadrupole mass spectrometers. Quantification can be achieved via matrix calibration or by using stable isotope labelled standards.

Following a call for data issued in July 2010, a total of 13,075 analytical results obtained on food samples and 9,877 results on unprocessed grains sampled by 19 European countries in 2005-2010 were used in the evaluation. Zearalenone was reported at quantifiable levels in 15 % of the samples. The highest concentrations of zearalenone were reported for wheat bran, corn and products thereof (e.g. corn flour, cornflakes). Notably high levels have been found in corn germ oil and wheat germ oil. There were indications that soy can be contaminated with zearalenone but there were insufficient data to draw conclusions. The EFSA Panel on Contaminants in the Food Chain (CONTAM Panel) used a lower bound-upper bound (LB-UB) approach in its assessment of the occurrence data. The lower bound assigns a value of zero to non-detects; the upper bound assigns the value of the limit of detection (LOD) or limit of quantification (LOQ) to results below the LOD and LOQ, respectively.

Levels of zearalenone in the group ‘Unprocessed grains’ were considerably higher than in the group ‘Grains for human consumption’. This suggests that cleaning and selection steps applied to grains after harvesting result in lower zearalenone concentrations in grains intended for human consumption. In general, zearalenone is redistributed between the milling fractions. The by-products from cleaning the raw cereal grains (dust, hulls and others) were characterised by 3- to 30-fold higher zearalenone concentrations than the cleaned cereal grains while bran contained up to 2-fold higher concentrations. Zearalenone is generally stable during cooking, except under alkaline conditions or during extrusion cooking (heating under a high degree of pressure).

The CONTAM Panel estimated total chronic dietary exposures to zearalenone across 19 European countries, using LB and UB mean concentrations of zearalenone in foods, and consumption data for different age groups. For adults the minimum LB to maximum UB was 2.4 to 29 ng/kg body weight (b.w.) per day for average consumers (average consumption in total population), and 4.7 to 54 ng/kg b.w. for high consumers (95^th percentile consumption in total population).The highest exposure estimates are for toddlers (aged ≥ 12 months to < 36 months), at 9.3 to 100 ng/kg b.w. per day for average consumers, and 23 to 277 ng/kg b.w. for high consumers.

Grains and grain-based foods, in particular grains and grain milling products, bread and fine bakery wares, made the largest contribution to the zearalenone exposure in all age classes. Vegetable oils, especially corn germ oil and wheat germ oil, make an important contribution to the zearalenone exposure.

From the average values across the European countries, breakfast cereals provide a contribution of 0.4-17 % to total dietary exposure of zearalenone in adults. Increasing the ML from 50 µg/kg up to 150 µg/kg has the potential to increase chronic total dietary exposure to zearalenone by up to 35 %. From the average values across the European countries, breakfast cereals provide a contribution of 0.1-5.1 % to total dietary exposure of zearalenone in toddlers (age ≥ 12 months to < 36 months). Increasing the ML from 50 µg/kg up to 150 µg/kg has the potential to increase chronic total dietary exposure to zearalenone by up to 16 %. Short-term mean dietary exposure in consumers of breakfast cereals could increase to up to 357 ng/kg b.w. per day and the 95^th percentile exposure up to 1029 ng/kg b.w. per day if the ML for zearalenone were to be increased from 50 µg/kg up to 150 µg/kg.

Limited data indicate that dietary exposure of vegetarians to zearalenone could be up to 2-fold higher than for the general population.

Zearalenone is extensively absorbed and metabolised by three major routes. Reduction results in formation of α-zearalenol, which is more oestrogenic, and β-zearalenol, which is less oestrogenic than zearalenone. Efficient glucuronidation of zearalenone in the small intestine and liver significantly reduces the amounts of unconjugated (i.e. receptor-active) parent compound that reaches the circulation. It is possible that fetuses and neonates could be more susceptible than adults to the oestrogenic effects of zearalenone, based on higher internal exposures due to metabolic and physiological immaturity. Cytochrome P450-mediated oxidation produces catechol metabolites that are subject to redox cycling to reactive quinones.

Of the laboratory and domestic animals studied, pigs are the most sensitive species for oestrogenic effects of zearalenone, with females being more sensitive than males. The greater formation of α zearalenol in the pig, relative to other animal species studied, may contribute to this sensitivity.

In well-conducted carcinogenicity bioassays, no increase in tumors was observed in two rat studies, while in mice, significant increases in pituitary and liver adenomas, but not carcinomas, were observed in one study, providing limited evidence of carcinogenicity. Zearalenone does not cause gene mutations in bacterial test systems but is clastogenic and aneugenic in vitro and has been confimed as an in vivo clastogen in the mouse. A plausible mechanism for the clastogenic effects has been proposed, namely formation of catechols that can be oxidised to quinones that undergo redox-cycling.

The oestrogenic effects of zearalenone in pigs are observed at doses around three orders of magnitude lower than doses reported to cause clastogenicity and increases in adenomas in mice. The CONTAM Panel therefore decided to establish a TDI for zearalenone based on its oestrogenic effects. Adverse effects of zearalenone and its metabolites on testosterone synthesis, sexual behaviour, sex organ weights, testicular histology and spermatogenesis have been observed in male animals. In females, adverse effects of zearaleneone on the reproductive tract, fertility and embryo survival have been reported. The female pig is the most sensitive, with immature pigs possibly more sensitive than mature pigs. Effects include disturbance of the oestrous cycle, ovulation, conception and implantation, embryonic death, reduced fetal weight, reduced litter size and impaired neonatal survival. In female pigs, the tissues that are most sensitive to the oestrogenic effect of zearalenone and its metabolites are the ovary, uterus and vulva. Lowest-observed-effect-levels (LOELs) for these tissues in mature and immature gilts range from 17 to 200 μg/kg b.w. per day, with an overall no-observed-effect-level (NOEL) of 10 μg/kg b.w. per day.

Toxicodynamic information indicates that it is likely that the human female would not be more sensitive to zearalenone and its metabolites than the female pig. For derivation of a TDI, it is therefore not necessary to include an uncertainty factor of 2.5 for toxicodynamic differences between pigs and humans. Using the NOEL of 10 μg/kg b.w. per day and an uncertainty factor of 40 (4 for interspecies differences in toxicokinetics and 10 for interhuman variability), a TDI of 0.25 μg/kg b.w. can be derived. As a number of relevant studies, including in the pig, have become available since the previous t-TDI was established by the SCF in 2000, the CONTAM Panel concluded that a full TDI of 0.25 μg/kg b.w. can now be established.

Estimates of chronic dietary exposure to zearalenone based on the available occurrence data are below or in the region of the TDI for all age groups and not a health concern. A potential increase in the ML for zearalenone in breakfast cereals from 50 µg/kg to 75, 100, 125 or 150 µg/kg is unlikely to result in a chronic dietary exposure exceeding the TDI.

In a worst case scenario it is possible that an individual could consume the same batch of breakfast cereal containing zearalenone at the ML every day for 2 to 4 weeks. The highest estimated short-term exposure is for children aged ≥ 3 to < 10 years old and, at the current ML, is 60 % above the TDI. Increasing the ML from 50 µg/kg up to 150 µg/kg has the potential to increase the short-term exposure of children aged ≥ 3 to < 10 years old to up to approximately 1 µg/kg b.w. per day. High fibre breakfast cereals, which are the most likely to be contaminated with zearalenone due to the high content of wheat bran, are more likely to be consumed by adults than by children. For adults the highest estimates of short-term exposure are below the TDI for ML scenarios up to a concentration of 100 µg/kg zearalenone in breakfast cereals.

The mycotoxins that usually co-occur with zearalenone do not have oestrogenic effects. Combined effects of zearalenone and other mycotoxins are not expected to arise in humans at dietary exposures below the respective health-based guidance values of the individual toxins. The possible impact of combined exposure to zearalenone with other oestrogenic substances in food (such as phytoestrogens in soya) or the environment could be additive or antagonistic.

Mycotoxins, Fusarium, zearalenone, food, wheat bran, breakfast cereals, analysis, occurrence, dietary exposure, risk assessment, toxicity, tolerable daily intake (TDI)