Registration Dossier

Administrative data

Link to relevant study record(s)

Description of key information

Key value for chemical safety assessment

Additional information

According to Belsito et al. (Food and Chemical Toxicology 45 (2007) S130–S167):

1. Absorption

There are neither dermal nor oral pharmacokinetic studies available from which the bioavailability of this class of compounds can be quantitatively determined. Based on metabolic studies on alpha-ionone (Prelog et al., 1951) and beta-ionone (Bielig and Hayasida, 1940; Ide and Toki, 1970) in which ionone-specific metabolites were recovered in the urine of treated rabbits, and in the urine of dogs treated orally with b-ionone (Prelog and Meier, 1950), oral absorption of these compounds does occur to some extent. These studies, however, were not designed as pharmacokinetic investigations suitable to determine oral absorption.

Although no direct data is available for beta-ionone, the structurally similar methyl-ionone (CAS 1335-46-2) can be used for read-across. Since also logPow of methyl-ionone (4.5) and beta-ionone (4) are very similar, the study is regarded as suitable.

"In an in vitro dermal penetration/permeability study, only 0.7% or undetectable amounts of methyl-ionone (mixture of isomers) were recovered in the fluid beneath the skin preparations of rats and pigs, respectively, 6 h after application of a 3000 µg dose(600 µg/cm² over 5 cm² of skin) (RIFM, 1984a). In this study, approximately 50% (rat) and 10% (pig) of methyl-ionone-14C penetrated into, but not through the epidermisand dermis, while another 30% was lost to evaporation."

2. Distribution and pharmacokinetics

Data available describing the distribution and pharmacokinetics of ionones/rose ketones following absorption are limited to a single study in mice reporting the presence of beta-ionone at trace levels (<0.1 ng/ml) in the blood 30–90 min following a 1-h inhalation exposure (Buchbauer et al., 1993).

3. Metabolism

All the compounds discussed in this group are simple molecular modifications of the basic ionone and damascene structures, which are in essence cyclohexene derivatives carrying a butanone side chain. Therefore ionone and damascone can be regarded as being archetypal for the group as a whole. Furthermore, it is anticipated that compounds in this group will show a high degree of metabolic homology, bearing in mind that, in general, the same functional groups will be involved in biotransformation reactions. The a- and b-ionones are structural positional isomers as are also the alpha- and beta-damascones. The only structural differences between the ionones and the rose ketones are the position of the allylic double bond and of the ketone in the butanone side chain. The ionones and rose ketones, because of their highly lipophilic nature would be expected to be extensively metabolized in vivo and eliminated as transformation products. This appears to be the case as in several studies involving the administration of alpha- or beta-ionone to rabbits and dogs little unchanged compound was recovered from the urine compared to the relatively large amounts of transformation products that could be isolated (Bielig and Hayasida, 1940; Prelog and Meier, 1950; Ide and Toki, 1970). Based upon the molecular structures of the ionones and rose ketones several metabolic options might be predicted:

1. hydroxylation/oxygenation of the cyclohexene ring;

2. reduction of the butenone group to a secondary alcohol;

3. oxidation of the angular methyl groups;

4. reduction of the double bond in the exocyclic alkenyl side chain to form dihydro derivatives;

5. conjugation of the hydroxylated metabolites with glucuronic acid;

6. conjugation with glutathione.

Finally there could be various combinations of these pathways to produce an array of metabolites. Overall, while the empirical metabolic data are limited to studies primarily on b-ionone, it should be noted that the ionones and rose ketones are close structural analogues, both having a cyclohexa(e)ne ring with an allylic side chain containing a ketone moiety. Differences in the structures are related to the presence of an additional ketone group [e.g., 1-(2,6,6-trimethyl-1-cyclohexen-1-yl)-butane-1,3-dione], unsaturation of the cyclohexene ring (e.g., dihydro-c-ionone), unsaturation of the allylic side chain, differences in the points of methylation of the cyclohexene ring, the position of the double bond in the allylic side chain (i.e., the ionones versus the rose ketones), and various combinations of the above. While these differences would be expected to lead to the production of compound specific metabolites without a common terminal metabolite, some generalizations can be made. As reported by JECFA (1999), alpha-ionone, dihydro-alpha-ionone, methyl-alpha-ionone, alpha-irone, alpha-iso-methylionone, and allyl-alpha-ionone would likely share a common metabolic pathway, with differences in rates of metabolism only. Likewise, beta-ionone, dihydro-beta-ionone, and methyl-beta-ionone, could be expected to be metabolized in a very similar manner. For the other compounds, while common pathways cannot be clearly established, similar metabolic processes would be expected to occur and could include various combinations of hydroxylation/oxygenation of the cyclohexene ring, reduction of the butenone group to a secondary alcohol, oxidation of the angular methyl groups, reduction of the double bond in the exocyclic alkenyl side chain to form dihydro derivatives, and conjugation of the hydroxylated metabolites with glucuronic acid. Other metabolic routes such as epoxidation may potentially be available to certain ionones and rose ketones, but no metabolites indicative of this pathway have been reported. It should be noted that the rose ketones, which are more likely to undergo epoxidation have not been subjected to metabolic study. For most ionones and rose ketones, the endocyclic unsaturated bond is structurally hindered by methyl substituents which likely impede epoxidation reactions at this site. Similarly, based on in vitro studies with two archetypal a,b-unsaturated ketones included in the chemicals under assessment, namely 4-(2,6,6-trimethylcyclohex-1-enyl)-2-buten-4-one and 1-(2,6,6-trimethylcyclohexa-1,3-dienyl)-2-buten-1-one, reactivity with glutathione, and hence the potential for electrophilic reactions with biological molecules, was concluded to be minimal (Portoghese et al., 1989). The authors concluded that these compounds exhibit low reactivity towards glutathione because the electrophilic centers are sterically hindered by directly attached substituents (methyl groups) and neighboring groups. Reactivity with other nucleophilic centers (e.g., guanine components of nucleotides) would be expected to be dramatically less than with glutathione. As a result, the metabolism of the a,b-unsaturated ketone in the side chain of the rose ketones is not expected to produce reactive intermediates of greater toxicity than similar metabolism of the more sterically hindered alpha,beta-unsaturated ketone side chain of the ‘‘ionone’’ series. Three rose ketones (trans,trans-d-damascones; d-damascone; damascone) as well as dehydrodihydroionone, have an additional and unhindered double bond in the cyclic ring structure that could provide a potential site for epoxidation to occur. Similarly, for methyl-d-ionone, the cyclohexene ring contains a point of unsaturation less hindered by the presence of methyl groups, possibly increasing the likelihood of epoxidation. Epoxidation of these specific chemicals could produce products with higher reactivities/ toxicities than other members of this class. In summary, empirical metabolic data on ionone isomers demonstrate the activity of various metabolic pathways leading to polar metabolites, both in free and conjugated forms. The primary differences in the chemical structure of members of this class of compounds that could affect metabolism, and potentially the toxicity of metabolites, are the position of the double bond in the allylic side chain (ionones versus rose ketones) and the potential for epoxidation depending upon the number and position of the double bonds in the cyclohexene ring. Since the allylic side chain of the rose ketones does not appear to have strong electrophilic activity, based on in vitro data (Portoghese et al., 1989), the damascone metabolites are unlikely to be of greater toxicity than those of the ionones. However, based on metabolic considerations, unique epoxide metabolites could be generated for each of trans,trans-d-damascones; d-damascone; 1-(2,6,6-trimethyl-3-cyclohexa-1-e-dienyl)-2-buten-1-one); dehydrodihydroionone, and methyl-dionone. Thus, these compounds may have greater toxic potential than other members of this class. The most complete in vivo metabolic data are from animal studies; there are no human data for these compounds. The most extensive data are for b-ionone with a limited amount of data for the a-isomer; the metabolic data available can be viewed as being representative for the class as a whole. Following administration of b-ionone to a male rabbit (oral gavage, 1 g/day for seven days), Ide and Toki (1970) isolated from the urine and characterized the following transformation products (numbered on the CAS system): 3-oxo-beta-ionone, 3-oxo-beta-ionol, dihydro-3-oxo-beta-ionol and 3-hydroxy-beta-ionol together with the glucuronides of 3-oxo-beta-ionol and dihydro-3-oxo-beta-ionol. Only a small amount of unchanged b-ionone (circa 1% of dose) was recovered from the urine of the dosed animal. In an earlier study, Bielig and Hayasida (1940) isolated b-ionol and dihydro-b-ionol as reduction products from the urine of dogs fed b-ionone; three additional hydroxylated metabolites were detected but not characterized. Prelog and Meier (1950) confirmed these findings and identified 3-oxo-b-ionol and 3-hydroxy-b-ionol or 3 -hydroxy-b-ionone. In the single metabolic study of a-ionone in mammals, Prelog et al. (1951) isolated a trans-formation product in urine of rabbits which appeared to be an oxidation product, tentatively identified as 4-oxo-tetrahydro-ionone. There is no available information on the metabolic fate of the ionones and rose ketones in humans, but one might reasonably presume that it would be similar to that seen in mammals such as the rabbit and dog, i.e., oxidative and reductive transformation followed by conjugation. Support for this view comes from the pattern of metabolism of other compounds containing the ionone structure. For example, the retinoids, such as 13-cis-retinoic acid (isotretinoid) contain the ionone ring structure. 13-cis-Retinoic acid undergoes extensive metabolism in humans by oxidation and conjugation, including oxidation of the ionone nucleus to give the 4-oxo-13-cis-retinoic acid metabolite (Vane et al., 1990; Kraft et al., 1991). This position of oxidation is analogous to the 3-oxo metabolites of b-ionone as numbered using the CAS system of nomenclature. In summary, the available evidence indicates that the ionones and rose ketones are extensively metabolized in vivo by pathways involving oxidation, reduction and conjugation. These metabolites do not raise issues of toxicological concern.