The European Food Safety Authority (EFSA) asked the Panel on Food Additives and Nutrient Sources Added to Food (ANS) to deliver a scientific opinion addressing the new scientific information that has become available since the adoption of the opinion on the re-evaluation of Allura Red AC (E 129) when used as a food colouring agents in 2009 and assess whether this would alter its conclusions.
This request originates from an opinion adopted by the FEEDAP Panel in April 2012 on the reevaluation of the substance for use in feedingstuffs for cats and dogs. In its opinion, the FEEDAP Panel had concluded that the genotoxic potential of Allura Red AC could not be excluded and that the data available were insufficient to demonstrate the safety of the substance for its proposed use.
In the context of the re-evaluation of Allura Red AC for use as a food additive, a first positive in vivo comet assay in mice by Tsuda et al. (2001) had already been considered by the ANS Panel. At the time, the Panel concluded that the positive findings observed in isolated nuclei in the mouse colon and in the stomach were not expected to result in carcinogenicity in the light of the negative results from the long-term toxicity studies in rats and mice.
Subsequent to the ANS Panel opinion in 2009, a new study was published by the same research group (Shimada et al., 2010) confirming observation of DNA damage in the mouse colon, three hours after administration of a single dose of 10 mg/kg bw by gavage. No such effects were observed in rats at doses up to 1 000 mg/kg bw.
The positive findings from the two in vivo comet assays in mice were considered by the FEEDAP Panel at the time of their re-evaluation of the safety of Allura Red AC for use in feedingstuffs. In its 2012 opinion the FEEDAP Panel concluded that, on the basis of these experimental results, it was not possible to exclude a genotoxic potential of Allura Red AC.
The ANS Panel critically reviewed the study by Shimada et al. (2010) and noted that these results were observed by a single group of researchers using an in-house developed protocol that uses isolated nuclei from homogenised tissues. Although this protocol was considered to meet the minimum criteria for acceptance of the in vivo comet assay (EFSA, 2012), the Panel noted that such protocol had not been included in the international validation exercise. Under the same experimental conditions, no effect could be observed in the rat. The Panel noted that the same findings observed for Allura Red AC were also reported by Shimada et al. (2010) for the two other authorised food dyes tested in this study, Amaranth and Ponceau 4R.
The Panel acknowledged the role of the comet assay as an indicator test for the detection of DNA lesions and/or DNA repair activity and concluded that the findings reported cannot be considered conclusive evidence of mutagenicity, i.e. of induced permanent genetic changes. The Panel also noted that the in vivo comet assay is typically performed as part of genotoxicity testing strategies for the follow-up of in vitro positives and, as such, its results should be interpreted in conjunction with the findings obtained from other genotoxicity tests.
In this specific case, however the Panel was faced with a limited dataset on the mutagenicity of Allura Red AC. A read-across exercise was therefore conducted in order to gather additional information from the relevant data available for a number of other structurally related sulphonated mono azo dyes, currently authorised as food and feed additives, namely Amaranth (E 123), Ponceau 4R (E 124), Sunset Yellow FCF (E 110), Tartrazine (E 102), and Azorubine/Carmoisine (E 122). The review performed has covered the data on genotoxicity and carcinogenicity previously included by the ANS Panel in the scientific opinions on the re-evaluations of these food colours and published studies which have become available in the literature since then. A similar exercise has been done for the data on metabolism, absorption, distribution and excretion, with a particular focus on the possible differences between metabolism in mouse and rat that could explain the different responses observed in the two species in the in vivo comet assays. Consideration has also been given to structure relationship activities and possible modes of actions have been explored, such as the generation of reactive oxygen species, which could provide explanations of the different findings of the in vivo comet assay observed in mice and in rats.
When considering the structurally related sulphonated mono azo dyes as a chemically-related group, a number of genotoxicity studies have been retrieved, however their quality was inconsistent in terms of the reporting and conduct, some of the endpoints investigated were of only limited relevance to the investigation of the safety of the dyes under consideration and could not provide insight into any possible mode of action involved that could explain the species differences in the findings observed in the in vivo comet assay. On the other hand, from the data considered in this review, there appears to be a pattern of effects shared by the structurally related sulphonated mono azo dyes that would warrant further investigation.
The Panel concluded that the new data by themselves were insufficient at this time to change the overall weight of evidence in its 2009 opinion on the re-evaluation of Allura Red AC (E 129) as a food additive and that there is no reason to revise the ADI at this time.
After considering the available information on the metabolism of sulphonated and non-sulphonated mono azo dyes and species differences between rat and mouse, the Panel acknowledged that until the uncertainties regarding the in vivo comet assay were investigated it was not appropriate to undertake a full evaluation of the other points listed under the Terms of Reference at this stage.
The Panel therefore recommended a repetition of the in vivo comet assay in mice, to be performed in compliance with the more recent and internationally validated experimental protocol, using whole cells instead of isolated nuclei from the range of tissues included in the previous assays (glandular stomach, colon, urinary bladder, lung, liver, brain, kidney and bone marrow), for all the sulphonated mono azo dyes included in this review.