Malondialdehyde trapping by food phenolics.

The reactions between malondialdehyde and 2,5-dimethylresorcinol, orcinol, olivetol, and alkylresocinols were studied in an attempt to investigate both if this lipid oxidation product is trapped by phenolics analogously to other reactive carbonyls and to elucidate the chemical structures of the produced adducts. After being formed, malondialdehyde is both partially fractionated to acetaldehyde and oligomerized into dimers and trimers. All these compounds react with phenolics producing three main kinds of derivatives: 5(or 7)-alkyl-7(or 5)-hydroxy-4-methyl-4H-chromene-3-carbaldehydes, 7-alkyl-9-hydroxy-6H-2,6-methanobenzo[d][1,3]dioxocine-5-carbaldehydes, and 4-(3-formylphenyl)-7-hydroxy-4H-chromene-3-carbaldehydes. A total of twenty-four adducts were isolated by semipreparative high-performance liquid chromatography (HPLC) and characterized by mono- and bi-dimensional nuclear magnetic resonance (NMR) spectroscopy and mass spectrometry (MS). Reaction pathways to explain the formation of all these compounds are proposed. Obtained results show that phenolics can trap malondialdehyde producing stable derivatives. The function(s) that such derivatives can play in foods remain(s) to be elucidated.

Reactive carbonyl-scavenging ability of 2-aminoimidazoles: 2-amino-1-methylbenzimidazole and 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP).

The carbonyl-scavenging ability of 2-amino-1-methylbenzimidazole (AMBI) and the heterocyclic aromatic amine 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) was investigated in an attempt to identify new routes that can modify the carbonyl content of foods. The reaction of both AMBI and PhIP with 2-alkenals, 2,4-alkadienals, 4-oxo-2-alkenals, 4,5-epoxy-2-alkenals, and 4-hydroxy-2-nonenal produced fluorescent adducts, whose structure was determined for the adduct produced between AMBI and 2-pentenal. This adduct was isolated and identified by one- and two-dimensional nuclear magnetic resonance and high-resolution mass spectrometry as 2,10-dihydro-2-ethyl-10-methylpyrimido[1,2-a]benzimidazole. The formation of these adducts was parallel to the elimination of AMBI and PhIP. The Ea of the reaction between PhIP and 4-oxo-2-nonenal was 27.4 kJ/mol. All these results suggest that 2-aminoimidazoles can be employed to modify the carbonyl content of foods. At the same time and because the reaction produces the disappearance of the amino compound, lipid-derived carbonyl compounds can be employed to eliminate 2-aminoimidazoles, which suggests a new strategy for the elimination of heterocyclic aromatic amines in foods.

Ammonia and formaldehyde participate in the formation of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) in addition to creati(ni)ne and phenylacetaldehyde.

The formation of formaldehyde from phenylacetaldehyde and phenylalanine, and the contribution of both formaldehyde and ammonia to the production of PhIP from phenylacetaldehyde and creatinine were studied in an attempt to clarify the reaction pathways that produce PhIP. Formaldehyde was produced by thermal degradation of phenylacetaldehyde and, to a lesser extent, also by degradation of phenylalanine, phenylethylamine, styrene, and creatinine. When formaldehyde was added to a mixture of phenylacetaldehyde and creatinine, PhIP yield was multiplied by nineteen. When formaldehyde and ammonia were simultaneously present, PhIP yield was multiplied by fifty and the Ea of the reaction decreased by 61%. All these results point to formaldehyde and ammonia as the two additional reactants required for PhIP formation from both phenylacetaldehyde/creati(ni)ne and phenylalanine/creati(ni)ne mixtures. A general pathway for PhIP formation is proposed. This pathway is suggested to be the main route for PhIP formation in foods.

Effect of amino acids on the formation of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) in creatinine/phenylalanine and creatinine/phenylalanine/4-oxo-2-nonenal reaction mixtures.

2-Amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) formation in mixtures of creatinine, phenylalanine, amino acids and 4-oxo-2-nonenal was studied, to analyse the role of amino acids on the generation of this heterocyclic aromatic amine. When oxidised lipid was absent, cysteine, serine, aspartic acid, threonine, asparagine, tryptophan, tyrosine, proline, and methionine increased significantly (p < 0.05) the amount of PhIP formed in comparison to the control. When lipid was present, only the addition of methionine, glycine, and serine increased significantly (p < 0.05) the amount of PhIP produced, while histidine, cysteine, lysine, tryptophan, tyrosine, and alanine reduced significantly (p < 0.05) PhIP. These results may be a consequence of the different competitive reactions that occur. Thus, in the absence of lipids, thermal decomposition of the amino acids produced reactive carbonyls that converted phenylalanine into phenylacetaldehyde as a key step in the formation of PhIP. When oxidised lipid was present, amino acids competed with phenylalanine for the lipid, and amino acid degradation products were formed, among which alpha-keto acids seemed to play a role in these reactions. These results suggest that PhIP can be produced by several alternative reaction pathways from all major food components, including amino acids and lipids, in addition to carbohydrates.

Strecker-type degradation of phenylalanine initiated by 4-oxo-2-alkenals in comparison to that initiated by 2,4-alkadienals, 4,5-epoxy-2-alkenals, or 4-hydroxy-2-nonenal.

The conversion of phenylalanine to phenylacetaldehyde as a consequence of its reaction with 4-oxo-2-alkenals was studied both to characterize the reaction pathway and to compare the reactivities and kinetic constants of oxoalkenals with those of other lipid oxidation products: 2,4-alkadienals, 4,5-epoxy-2-alkenals, and 4-hydroxy-2-nonenal. Oxoalkenals produced the Strecker aldehyde through imine formation, which was then decarboxylated and hydrolyzed. In the course of the reaction the lipid was converted into an unsaturated hydroxylamine that eventually cycled to 2-alkylpyrrole. The Ea of phenylacetaldehyde formation in the presence of oxoalkenals was 55-64 kJ/mol. This Ea was similar to the Ea determined for the other tertiary lipid oxidation products assayed (58-67 kJ/mol), but higher than the Ea determined for alkadienals (28-38 kJ/mol). However, this difference in Ea only correlated with the amount of phenylacetaldehyde produced at 37 °C. At higher temperatures, 4-oxo-2-nonenal was the lipid-derived carbonyl compound that produced the highest amount of the Strecker aldehyde, therefore pointing to this oxoalkenal as the most efficient Strecker aldehyde forming compound derived from lipids. For this reason, oxoalkenals should be expected to play a significant role in reactions in which Strecker aldehydes are recognized intermediates, as occurs in the formation of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP).

Comparative formation of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) in creatinine/phenylalanine and creatinine/phenylalanine/4-oxo-2-nonenal reaction mixtures.

The comparative formation of the heterocyclic aromatic amine 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) in both creatinine/phenylalanine (CRN/Phe) and creatinine/phenylalanine/4-oxo-2-nonenal (CRN/Phe/ON) systems was studied to analyse the ability of lipid-derived reactive carbonyls to promote PhIP formation. Although PhIP was produced to some extent in the CRN/Phe system, the presence of the oxidized lipid increased considerably the amount of PhIP produced. This increase seemed to be a consequence of the decrease in the E(a) of the reaction when the lipid was present, which diminished from 112.9 to 80.9 kJ/mol. On the other hand, the addition of the lipid did not seem to produce PhIP by an alternative mechanism because PhIP was formed analogously in both CRN/Phe and CRN/Phe/ON systems as a function of pH, creatinine concentration, phenylalanine concentration, time, temperature, oxygen concentration in the reaction atmosphere, and the addition of different amounts of ammonia. All these results suggest that the ability of lipid oxidation products to produce PhIP is related to their capacity to induce the Strecker degradation of phenylalanine to phenylacetaldehyde. Therefore, any other reactive carbonyl compound that can produce the Strecker degradation of phenylalanine should also be considered as a potential inducer of PhIP formation under appropriate conditions.

Effect of lipid oxidation products on the formation of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) in model systems.

Ternary mixtures of creatinine, phenylalanine and lipids (or carbohydrates) were heated at 200°C for 1h to determine the potential contribution of lipids to the formation of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP). Although unoxidised lipids did not contribute to PhIP formation, their oxidation produced many compounds that significantly increased the formation of PhIP. Among the different lipid oxidation products (LOPs) studied, which included ω-6 and ω-3 derived lipid hydroperoxides, 2,4-alkadienals, 2-alkenals, 4,5-epoxy-2-alkenals, 4-oxo-2-alkenals, and 4-hydroxy-2-nonenal, 4-oxo-2-nonenal was the most reactive compound. It produced 32.48 pmol of PhIP/µmol of creatinine in comparison with the 7.92 pmol of PhIP/µmol of creatinine produced by the control phenylalanine/creatinine reaction mixture. 4-Oxo-2-nonenal reactivity was similar to that of most carbohydrates; although ribose and arabinose produced more PhIP (44-46 pmol of PhIP/µmol of creatinine). In addition to single LOP, the addition of oxidised soybean oil for 24-144 h at 60°C also increased PhIP formation. All these results pointed out to a potential contribution of LOP to the formation of PhIP in food products. This contribution will depend on the lipid content of the food product and its easiness to be oxidised.