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) was asked to provide a scientific opinion on the safety of glycerol esters of tall oil rosin (GETOR) when used as a stabilising and emulsifying food additive in certain non-alcoholic flavoured cloudy drinks and certain cloudy spirit drinks to a maximum level of 100 mg/l. In view of the limited toxicity studies for GETOR, analytical data were submitted to demonstrate that GETOR are chemically equivalent to GEWR which have already been authorised as a food additive by Directive 95/2/EC.
GETOR have been evaluated by the Joint FAO/WHO Expert Committee on Food Additives (JECFA) in 2009. The Committee concluded, in principle, that the data from GEWR could be used in the evaluation of GETOR; however, the Committee did not have adequate information on the composition of GETOR, considering that the source material and production processes are different, which may result in different by-products. The Committee decided that it could not evaluate GETOR without additional information on its composition, in order to clarify the extent and significance of any differences, relative to other glycerol esters of rosins. Tentative specifications for GETOR were adopted, pending the submission by the end of 2010 of infrared spectra that correspond to the commercially available products, data on the resin acid composition obtained with updated chromatographic techniques, and additional information on methods, that enable the identification of the individual glycerol esters of rosins and their differentiation. The Committee also requested information on the identity of the sulphur compounds in the commercial product. No ADI was allocated for GETOR at the 71st meeting of JECFA in 2009.
The petitioner describes GETOR as a complex mixture of mono-, di- and tri-glycerol esters of resin acids from tall oil rosin (TOR). TOR is obtained by distillation of crude tall oil, which is a by-product of Kraft (paper) pulp processing and derives botanically from coniferous wood, especially pine (Pinus species). TOR is composed of approximately 90% resin acids and 10% “neutrals” (non-acidic substances). As outlined by the petitioner, GETOR contains 15-28% of “triglycerides”, 59-67% of “diglycerides”, 9-15% of “monoglycerides” and 3.4-4.5% free (unreacted) resin acids.
The Panel noted that mono-, di- and tri-glycerides are usually defined as esters from glycerol with one, two or three fatty acids, respectively, not comprising resin acids which presumably are referred to here. Upon request of the Panel the petitioner explained that the terms “mono-, di-, and triglycerides” in GETOR correspond to esters of resin acids and to esters of fatty acids.
The Panel also noted that the description of GETOR presented by the petitioner differs somewhat from the definition given in the JECFA monographs, in which GETOR are described as a complex mixture of tri- and di-glycerol esters of resin acids from TOR, with only a residual fraction of monoglycerol esters. Furthermore the Panel noted that glycerol esters of rosins usually contain the following three fractions:
- the fraction of total glycerol esters of resin acids (mono-, di- and tri-esters), hereafter called fraction (a),
- the fraction of free resin acids, hereafter called fraction (b), and
- the fraction of other substances, hereafter called fraction (c).
The percentage values on the composition of GETOR given by the petitioner do not take into account the presence of “other substances” (fraction (c)) which comprises other saponifiables (e.g. other esters) and unsaponifiables (e.g. diterpenes, sterols), including sulphur compounds.
Concerning the occurrence of sulphur-containing compounds in GETOR carried over from the Kraft process used in pulping wood, only the results of the analysis of GETOR for elemental sulphur were presented yielding 308-398 mg sulphur/kg. According to the petitioner the identity of the sulphur-containing compounds in GETOR is unknown, but some of the sulphur compounds present in crude tall oil have been identified. These compounds include hydrogen sulphide, methyl mercaptan, dimethyl sulphide, and dimethyl disulphide. According to the petitioner the most likely residual sulphur-containing compounds in GETOR would be sodium salts of sulphuric acid and possible trace amounts of dimethyl sulphide, and dimethyl disulphide.
The Panel noted that the detected sulphur may derive from various inorganic or organic sulphur compounds which may differ in their toxicity profiles. Information concerning qualitative and quantitative analyses of sulphur compounds in GETOR is missing.
The Panel noted that the petitioner did not provide data on the identity and quantity of individual components in fractions (a), (b) and (c) of GETOR.
Concerning the composition of the source material TOR, the total of identified resin acids amounted to 82.6–91.9%, the total of identified fatty acids amounted to 0.9–1.2%, and the total of unidentified substances (resin acids, fatty acids, neutrals) amounted to 6.9–16.3%. In the fraction of resin acids, aside from the main component abietic acid, dehydroabietic acid, palustric acid, neoabietic acid, isopimaric acid, pimaric acid, sandaracopimaric acid and communic acid have been identified. Among the unsaponifiables, β-sitosterol and diterpene alcohols such as pimarinol, sandaracopimarinol, isopimarinol, elliotinol, abietinol and neoabietinol, have been detected. According to the petitioner, the fraction of the so-called “neutrals” of TOR moreover comprises, among other substances, esters of resin acids, lignins and sulphur compounds in low levels carried over from the Kraft pulping process. These substances have neither been identified nor quantified individually.
The Panel noted that due to the severe manufacturing conditions of GETOR, including heating at high temperatures, isomerisation, oxidation and dehydrogenation/hydrogenation reactions of the resin acids (whether as individual free acids or bound in esters) may occur. Decarboxylation of resin acids may also take place. In consequence, the composition of resin acids in GETOR is expected to differ from that of the parent material TOR.
The Panel noted that according to the description of the petitioner, the TOR used as a source in the production of GETOR meets the definition of the wood rosin as source of GEWR laid down in the Commission Directive 2008/84/EC as far as the content of approximately 10% “neutrals” in TOR is concerned. This so-called “neutral fraction” is considered to be, at least partly, retained during GETOR production and is presumed to result in fraction (c) (other substances) of GETOR defined above.
Overall, the Panel noted that the data provided to chemically characterise GETOR are not sufficient, since qualitative and quantitative data on individual components of the three fractions are missing, and data obtained by analysing the source material TOR have to be regarded with reservations.
Based on studies with 14C-labelled GEWR in vitro and in vivo in rats JECFA had concluded that food-grade GEWR are metabolically stable in the gastrointestinal tract and that only a minor fraction, most likely the monoglycerol ester fraction, undergoes partial hydrolysis (JECFA, 1996b). Therefore the Panel noted that setting limits for glycerol monoesters of resin acids might be relevant for the specifications of GETOR. Furthermore the Panel noted that information concerning qualitative and quantitative analyses of sulphur compounds and of other components in fraction (c) might be relevant for the specifications.
The Panel noted that for GETOR, apart from two acute toxicity studies, no studies are available on: i) absorption, distribution, metabolism and excretion, ii) short-term and sub-chronic toxicity, iii) genotoxicity, iv) chronic toxicity and carcinogenicity, v) reproductive and developmental toxicity.
Considering free resin acids present as fraction (b) in GETOR, selected data on absorption, distribution, metabolism and excretion and on genotoxicity as reviewed in a previous JECFA monograph are available. The Panel concluded that the toxicokinetic data available for free resin acids are not sufficient to evaluate the absorption, distribution, metabolism and excretion of GETOR. The Panel also noted that there are no data from in vivo genotoxicity testing to overrule positive in vitro findings.
In view of the limited toxicity data available for GETOR, the petitioner submitted analytical data to demonstrate that GETOR are chemically equivalent to GEWR and GEGR, claiming that the toxicological data for GETOR, GEWR and GEGR are for all practical purposes interchangeable.
The Panel noted that even though GETOR and GEWR are similar with respect to IR and NMR spectra and some physical and chemical parameters, the available data show however that they are not identical in their composition due to differences in botanical source and manufacturing process. Thus, GETOR contains sulphur compounds in contrast to GEWR and GEGR and show further compositional differences in the fraction of other substances (fraction (c)), concerning the composition of diterpenes and the occurrence of steroids, as to be expected from the analyses of the source materials TOR, wood rosin and gum rosin. No data are available concerning the identification or quantification of individual components in fractions (a), (b) and (c) of GETOR, GEWR and GEGR. The distribution of the resin acids as part of the glycerol esters of fraction (a) in GETOR and GEWR is unknown and analytical data on the percentage of the fraction of other substances (fraction (c)) in GETOR and GEWR are not presented. For GEWR there is a tendency to a lower percentage of glycerol triesters and a higher percentage of glycerol diesters in comparison with GETOR.
Dermal contact sensitisation is seen with different types of rosins and modified rosins and their components as well as with GETOR and 1-glyceryl monoabietate. Since allergic responses have not been reported from ingestion of glycerol esters of rosins with beverages, the Panel did not consider it likely that the oral exposure via the intended use and use levels of GETOR in beverages would be associated with a relevant risk of adverse effects for individuals with a known contact hypersensitivity to rosin or modified rosin products.
The exposure to GETOR was calculated based on the information available from the EFSA Concise Database in Exposure Assessment. The average exposure in Europe would range from 0.07 to 0.74 mg/kg bw/day, and at the 97.5th percentile from 0.5 to 3.3 mg/kg bw/day. The Panel also calculated the exposure to GETOR based on the UK data for children, using the intake of non-alcoholic flavoured drinks, and the intended use level of 100 mg/l. The exposure was estimated to be 1.7 mg/kg bw/day on average and 5.8 mg/kg bw/day at the 97.5th percentile. The Panel noted that these estimates are conservative, since GETOR is not intended to be added to all non-alcoholic flavoured drinks but to citrus-fruit based drinks only.
The Panel concluded that the chemical and toxicological characterisation of GETOR is not adequate and that the limited toxicological data on GETOR prevents the evaluation of the safety of GETOR. Furthermore the Panel could not conclude that GETOR is chemically equivalent to GEWR.Therefore, the toxicological data obtained with GEWR could not be used for read across to GETOR.
The Panel concluded that the available data are too limited to conclude on the safety of GETOR as a food additive for the proposed uses and use levels.