Following a request from the European Commission to the European Food Safety Authority (EFSA), the Scientific Panel on Food Additives and Nutrient Sources added to Food (ANS Panel) was asked to provide a scientific opinion re-evaluating the safety of tocopherol-rich extract of natural origin (E 306), synthetic α-tocopherol (all-rac-α-tocopherol; dl-α-tocopherol; E 307), synthetic γ-tocopherol (dl-γ-tocopherol; E 308) and synthetic δ-tocopherol (E 309), which are used as antioxidants in foods to inhibit the peroxidation of fats and lipids.
The Panel was not provided with a newly submitted dossier and based its evaluation on previous evaluations, additional literature that has become available since then and data provided following a public call for data. The Panel noted that not all of the original studies on which the previous evaluations were based were available for re-evaluation by the Panel.
Tocopherols belong to the group of substances named vitamin E. Vitamin E is the collective term for a family of structurally related substances, namely tocopherol- and tocotrienol-derivatives, that exhibit, qualitatively, the biological activity of the naturally occurring d-α-tocopherol. Vitamin E is an essential vitamin and is naturally present in plant-derived foods, particularly fruit and vegetables.
All tocopherols evaluated in this opinion are used as antioxidants in food, either individually or in combination, and are authorised under Annex II of Regulation (EC) No 1333/2008 on food additives.
The Scientific Committee on Food (SCF) has not set an Acceptable Daily Intake (ADI) for tocopherols, but considered the use of tocopherols as antioxidants in food acceptable. The SCF, in its evaluation of vitamins and minerals, established a Tolerable Upper Intake Level (UL) of 300 mg/day for vitamin E (SCF, 2003). Effects on blood clotting were used as the basis for deriving this UL of 300 mg/day for vitamin E. This UL also applies to pregnant and lactating women. The UL was scaled for children in the age ranges 1–3, 4–6, 7–10, 11–14 and 15–17 years to give ULs of 100, 120, 160, 220 and 260 mg/day, respectively.
The current ADI for α-tocopherol, established by the Joint FAO/WHO Expert Committee on Food Additives (JECFA), is 0.15–2 mg/kg body weight (bw)/day. The JECFA ADI is based on clinical experience in humans and takes into account the fact that α-tocopherol is an essential nutrient.
Specifications for tocopherol-rich extract of natural origin (E 306) and synthetic α-tocopherol (E 307) have been defined in Commission Regulation (EU) No 231/2012 and by JECFA. Specifications for synthetic γ-tocopherol (E 308) and synthetic δ-tocopherol (E 309) have been defined in only Commission Regulation (EU) No 231/2012.
Tocopherols are relatively stable in foods, but oxidation may occur when exposed to air, heat, acids, alkalis or metal ions. During storage, when peroxyl radicals are formed in oils or in the presence of unsaturated fatty acids, α-tocopherol reacts with these radicals, leading to the formation of tocopheroxyl radicals that further react with other peroxyl radicals to form non-radical products (Burton and Traber, 1990). The non-radical oxidation products of vitamin E have been identified as α-tocopheryl quinone, epoxy-alpha-tocopherylquinones and 8a-(lipid-dioxy)-alpha-tocopherones (Yamauchi et al., 2002). A decrease in tocopherol content occurs during food processing as a result of oxidation or thermal degradation depending on the processing procedures, storage time, conditions and type of food. In sunflower oil heated to 180 °C, α-tocopherol reacts partly to produce oxidation and degradation products, such as α-tocopheryl quinone and α-tocopheryl fatty acids (Kreps et al., 2015).
Based on toxicokinetics data for α-, γ- and δ-tocopherol and tocopherol-rich extract, α-tocopherol is the most biologically active and is, therefore, the form that is most often used in toxicity testing. For the purpose of this opinion, it is assumed that α-tocopherol is representative of the other tocopherols as a worst case, based on the fact that this form of vitamin E is the only form that the liver can re-secrete into the plasma, hence maintaining plasma concentrations and prolonging time in the plasma before elimination. In testing α-tocopherol, it is likely that the most conservative No Observed Adverse Effect Level (NOAEL) is determined. Therefore, results from toxicity studies on α-tocopherol are assumed to also apply to γ-tocopherol, δ-tocopherol and tocopherol-rich extract.
Absorption of the tocopherols varies with dose, and the efficiency of absorption decreases with increasing intake. Absorption across the gastrointestinal tract occurs along with dietary lipids and is dependent on biliary secretion of bile acids and salts, which aid the emulsification process, to form micelles with the hydrolysed fat. The micelles containing the vitamin E are then absorbed by passive diffusion at the brush border enterocytes of the small intestine. Once absorbed, the tocopherols enter the blood and lymphatics in chylomicrons, and are transported to the tissues. The liver appears to select α-tocopherol using α-tocopherol transport protein, whereas most ingested γ- and δ-tocopherol are eliminated by the liver into the bile and excreted in the faeces.
The acute oral toxicity of α-tocopherol is very low, with LD50 values for α-tocopherol reported to be greater than 2 000 mg/kg bw/day for rats.
The critical effects on haematology (blood clotting) and clinical chemistry (total cholesterol, total lipids and phospholipids) are not expected to vary qualitatively among the various forms of tocopherols, i.e. prolonged coagulation times are common to all tocopherols and are thought to be due to reduced absorption of vitamin K from the gastrointestinal tract, as supplementation with vitamin K prevents the effects of the tocopherols on blood clotting. The NOAEL for this critical effect observed in a 13-week oral (gavage) toxicity study in rats was 125 mg/kg bw/day.
Studies in humans give conflicting results with respect to the potential of vitamin E to affect the incidence of cardiovascular disease and subsequent mortality. It appears that, as was observed in animal studies, the principal adverse effect of the tocopherols is on prothrombin times and factors related to blood clotting, via an interaction with vitamin K or vitamin K-dependent proteins. However, it does appear that in humans this effect is only observed at high doses, which are not relevant to the use of tocopherols as food additives. Susceptible subgroups appear to be individuals with already compromised blood clotting capacity, and those with low vitamin K status. Although not all studies are consistent, it appears that while cardiovascular disease is not affected by vitamin E in healthy individuals, it has the potential to reduce primary cardiovascular outcomes, such as stroke, in patients with existing cardiovascular disease.
There is no evidence to suggest that α-tocopherol is genotoxic. Well-conducted genotoxicity studies (three Ames tests, an in vitro chromosomal aberration test and an in vivo mutagenicity study), available through the public call for data, were all negative. There are no in vitro mammalian cell gene mutation studies. However, there are in vitro and in vivo studies that support the antioxidant properties of vitamin E being protective against genetic damage.
In a chronic toxicity study in rats (duration of up to 16 months), a NOAEL of 125 mg/kg bw/day was determined, based on reduced body weight and increased heart and spleen weight.
One limited carcinogenicity study has been performed in rats. A NOAEL for general systemic toxicity of α-tocopherol could not be established in this study due to the effects on blood clotting and the liver. The NOAEL for carcinogenicity was > 2 000 mg/kg bw/day, the highest dose tested. The Panel concluded that there is no concern that α-tocopherol is carcinogenic.
There are insufficient studies to address the reproduction and developmental endpoints, as there are no multigeneration studies conducted in accordance with appropriate test guidelines.
Reported use levels from industry give information on the amount of the food additive added to food. The use of these data results in an estimate of the exposure to α-tocopherol (E 307) at the moment the food was produced. Considering that tocopherols are degraded during processing and storage, the loss of tocopherols in food is very likely to have an impact on the overall exposure estimates calculated using the reported use levels. Therefore, the Panel calculated additional exposure estimates for the additive itself, including potential loss factors, intended to more closely reflect the exposure to α-tocopherol (E 307) via foods as consumed.
The total exposure to α-tocopherol from all food sources (food additives, enzyme preparations, nutrient as vitamin and from natural sources) would reach up to 6.3 mg/kg bw/day in toddlers at the high level (with the exception of children in one survey from one country, achieving up to 9.7 mg/kg bw/day). From the use of α-tocopherol (E 307) as a food additive itself, the exposure (non-brand loyal scenario, considering loss factors) would range from 0.3 mg/kg bw/day in infants to 2.7 mg/kg bw/day in toddlers at the mean and from 0.9 mg/kg bw/day in adolescents to 5.9 mg/kg bw/day in toddlers at the high level.
The Panel estimated that, when comparing all sources (i.e. from the additive itself, from natural sources and from all food sources), the contribution of α-tocopherol (E 307) from its use as a food additive may represent, on average, approximately 71 % (range 59–80 %) of the overall exposure to α-tocopherol, and around a two- to five-fold higher intake than from natural sources, with the exception of infants whose intake from the use of α-tocopherol (E 307) as a food additive is likely to be 0.9-fold that of the intake from natural sources.
The Panel considered that the uncertainties identified would tend overestimate the actual exposure to α-tocopherol (E 307) as a food additive, particularly for the maximum level scenario, and to underestimate the actual exposure to α-tocopherol from all sources in European countries.
Taking into account that:
- vitamin E is widely consumed via human food;
- it is an essential nutrient;
- there is no indication of genotoxic or carcinogenic potential;
- animal and human studies available have not shown adverse effects, except effects on blood clotting at high levels;
- the exposure to α-tocopherol resulting from all food sources does not exceed the ULs for vitamin E (SCF, 2003) in any population group, except in children in one survey from only one country,
the Panel considered that α-tocopherol (E 307) at the reported uses and use levels as a food additive is not of safety concern.
The Panel noted that the exceedance of the UL observed in children in one survey from one country may be a result of different methodologies used among dietary surveys for reporting the amounts of food supplements consumed.
The Panel considered that the database on γ- and δ-tocopherol was too limited to be included in the safety assessment of tocopherols. However, the Panel is aware that much lower concentration levels and fewer uses for γ- and δ-tocopherols are reported in food than for α-tocopherol. The Panel considered that data on α-tocopherol can be read-across to the other tocopherols, based on the similarities in the chemical structure, and the fact that α-tocopherol represents a worst case, as it is the form which the body selectively retains. The Panel noted that it would be prudent to re-assess the appropriateness of this read-across as new data on γ- and δ-tocopherols become available. Therefore, the Panel considered that, overall, the use of tocopherols (E 306–E 309) as food additives would not be of safety concern at the levels used in food.
The Panel noted that in Annex II of Regulation (EC) No 1333/2008, use levels of tocopherols (E 306–E 309) in food for infants under the age of 12 weeks are included in categories 13.1.1, 184.108.40.206 and 220.127.116.11. The Panel considered that these uses would require a specific risk assessment in line with the recommendations given by JECFA (1978) and the SCF (1998) and endorsed by the Panel. Therefore, the current re-evaluation of tocopherols (E 306–E 309) as food additives is not considered to be applicable to infants under the age of 12 weeks. The Panel concluded that the current re-evaluation of tocopherols (E 306–E 309) as food additives is not applicable to infants under the age of 12 weeks.
The Panel recommended that the maximum limits for the impurities of toxic elements (arsenic, lead and mercury) in the EC specifications for tocopherols should be revised in order to ensure that tocopherols (E 306–E 309) as food additives will not be a significant source of exposure to these toxic elements in food.
The Panel recommended re-assessing the appropriateness of the read-across from α-tocopherol to the other tocopherols as new data on γ- and δ-tocopherols become available.