All of these compounds were proposed as oxidation products from the β-carotene ozonolysis in solution during the present study, based on their tentative

identification AZD0530 through LC-MS. Secocarotenoids, such as 4,9,13,17,17-pentamethyl-16,21-dioxo-docos-2,4,6,8,10,12,14-heptaenal and 3,7,11,11-tetramethyl-10,15-dioxo-hexadec-2,4,6,8-tetra-enal, have not been assessed in the literature to date, since oxidation products originating from the breakdown of the ring’s double bond, producing a keto function, are not very common (Britton, 1995). The 5,6-seco-β-carotene-5,6-dione is a possible exception, although it has been identified as product of β-carotene oxidation in permanganate solutions (Chou and Labuza, 1984), thus in a different condition of this work. Other compounds observed, including β-cyclocitral, 15-apo-β-carotenal, 14´-apo-β-carotenal, 12´-apo-β-carotenal, 5,6-epoxy-12´-apo-β-carotenal and 5,6-epoxy-10´-apo-β-carotenal, had been identified previously by other researchers, although using different model systems, as, for instance,

exposure to UV light (Chou & Labuza, 1984), in combination with photo-sensitizers (Ojima et al., 1993 and Stratton et al., 1993), through autooxidation at 20 and 80 °C (Ojima, Sakamoto, Ishiguro &Terao, 1993) and in the presence of permanganate (Rodriguez et al., 2007), amongst other methods. It is generally accepted that the initial compounds Cyclic nucleotide phosphodiesterase Fulvestrant cell line formed, during the oxidation of β-carotene, are epoxides and apocarotenals. β-cyclocitral is frequently mentioned as a product of the reaction of the double bond between the C7–C8 carbons of β-carotene (Glória et al., 1993 and Sommerburg et al., 2003), since this bond has a high mobility index which favours its break-down and results in the formation

of this carbonyl compound. β-Ionone (9-apo-β-carotenone) has been mentioned in several studies (Glória et al., 1993 and Waché et al., 2002) as an oxidation product of β-carotene. However, this compound was not detected in our experiments. Since β-ionone still has double bonds in its structure which can react with ozone, this study proposes that β-ionone could have been completely oxidised during the experiments, giving rise to secondary oxidation products. As predicted, in our experiments the ozonolysis of β-ionone gave rise to three carbonilic compounds which had been also tentatively identified as products of β-carotene ozonolysis, namely methyglyoxal, β-cyclocitral and 6,6-dimethyl-undec-3-en-2,5,10-trione. It is worth to mention that methylglyoxal and β-cyclocitral had also been found previously in the gas-phase reactions between β-ionone and ozone in Teflon chambers (Forester, Ham & Wells, 2007). The oxidation of β-carotene, under different ozone concentrations, was found to follow a zero order kinetic model relative to β-carotene in the main region of the curves.