Photodecomposition of retinyl palmitate in ethanol by UVA light-formation of photodecomposition products, reactive oxygen species, and lipid peroxides.

Photodecomposition of retinyl palmitate (RP), an ester and the storage form of vitamin A (retinol), in ethanol under UVA light irradiation was studied. The resulting photodecomposition products were separated by reversed-phase HPLC and identified by spectral analysis and comparison with the chromatographic and spectral properties of synthetically prepared standards. The identified products include 5,6-epoxy-RP, 4-keto-RP, 11-ethoxy-12-hydroxy-RP, 13-ethoxy-14-hydroxy-RP, anhydroretinol (AR), palmitic acid, ethyl palmitate, and four tentatively assigned cis and trans isomeric 15-ethoxy-ARs. AR was formed as a mixture of all-trans-AR, 6Z-cis-AR, 8Z-cis-AR, and 12Z-cis-AR with all-trans-AR predominating. 5,6-Epoxy-RP, 4-keto-RP, 11-ethoxy-12-hydroxy-RP, and 13-ethoxy-14-hydroxy-RP were also formed from reaction of RP with alkylperoxy radicals generated by thermal decomposition of 2,2'-azobis(2,4-dimethylvaleronitrile). Formation of these photodecomposition products was inhibited in the presence of sodium azide (NaN3), a free radical inhibitor. These results suggest that formation of 5,6-epoxy-RP, 4-keto-RP, 11-ethoxy-12-hydroxy-RP, and 13-ethoxy-14-hydroxy-RP from photoirradiation of RP is mediated by a light-initiated free radical chain reaction. AR and the isomeric 11-ethoxy-ARs were not formed from reaction of RP with alkylperoxy radicals generated from 2,2'-azobis(2,4-dimethylvaleronitrile), and their formation was not inhibited when NaN3 was present during the photoirradiation of RP. We propose that these products were formed through an ionic photodissociation mechanism, which is similar to the reported formation of AR through ionic photodissociation of retinyl acetate. RP and all its identified photodecomposition products described above (i) were not mutagenic in Salmonella typhimurium tester strains TA98, TA100, TA102, and TA104 in the presence and absence of S9 activation enzymes, (ii) were not photomutagenic in Salmonella typhimurium TA102 upon UVA irradiation, and (iii) did not bind with calf thymus DNA in the presence of microsomal metabolizing enzymes. These results suggest that RP and its decomposition products are not genotoxic; however, photoirradiation of RP, 5,6-epoxy-RP, and AR with UVA light in the presence of methyl linoleate resulted in lipid peroxide (methyl linoleate hydroperoxides) formation. The lipid peroxide formation was inhibited by dithiothreitol (DTT) (free radical scavenger), NaN3 (singlet oxygen and free radical scavenger), and superoxide dismutase (SOD) (superoxide scavenger) but was enhanced by the presence of deuterium oxide (D2O) (enhancement of singlet oxygen lifetime). These results suggest that photoirradiation of RP, 5,6-epoxy-RP, and AR by UVA light generated reactive oxygen species resulting in lipid (methyl linoleate) peroxidation.

Synthesis and characterization of N-demethylated metabolites of malachite green and leucomalachite green.

Malachite green (MG), a triphenylmethane dye used to treat fungal and protozoan infections in fish, undergoes sequential oxidation to produce various N-demethylated derivatives (monodes-, dides(sym)-, dides(unsym)-, trides-, and tetrades-) both before and after reduction to leucomalachite green (LMG). The close structure resemblance of the metabolites with aromatic amine carcinogens implicates a potential genotoxicity from exposure to MG. The availability of the synthetic standards is important for metabolic and DNA adduct studies of MG. This paper describes a simple and versatile method for the synthesis of MG, LMG, and their N-demethylated metabolites. The synthesis involves a coupling of 4-(dimethylamino)benzophenone or 4-nitrobenzophenone with the aryllithium reagents derived from appropriately substituted 4-bromoaniline derivatives, followed by treatment with HCl in methanol. The resulting cationic MG and their leuco analogues showed systematic UV/vis spectral and tandem mass fragmentation patterns consistent with sequential N-demethylation. The extensive (1)H and (13)C spectral assignments of the metabolites were aided by the availability of (13)C(7)-labeled MG and LMG. The results indicate the existence of a resonance structure with the cationic charge located in the central methane carbon (C(7)). The synthetic procedure is general in scope so that it can be extended to the preparation of N-demethylated metabolites of other structurally related N-methylated triphenylmethane dyes.

The effect of deuterium and fluorine substitution upon the mutagenicity of N-hydroxy-2,6-dimethylaniline.

2,6-Dimethylaniline (2,6-DMA) is an intermediate in the manufacture of several products, including pesticides, dyestuffs, and synthetic resins. It is also present in nanogram amounts in tobacco smoke, and is a major metabolite of the potent anesthetic and antiarrhythmic drug lidocaine, as well as a nasal carcinogen in rats. As with other aromatic amines, 2,6-DMA can undergo metabolic activation through cytochrome p450-mediated N-hydroxylation, followed by O-esterification to a reactive derivative capable of forming DNA adducts. We have recently characterized four DNA adducts resulting from this metabolic pathway. Three of the adducts arose from reaction of the exocyclic heteroatoms of deoxyadenosine and deoxyguanosine with the carbon para to the arylamine nitrogen. The fourth adduct resulted from reaction of the 2,6-DMA nitrogen with the C8 atom of deoxyguanosine. In order to investigate the relative contribution of the exocyclic heteroatom adducts as compared to the C8-deoxyguanosine adduct to the toxicities elicited by 2,6-DMA, we synthesized and compared the mutagenicity of N-hydroxy-2,6-DMA, N-hydroxy-4-deutero-2,6-DMA, 2,6-dimethylnitrosobenzene, 4-deutero-2,6-dimethylnitrosobenzene, and N-hydroxy-4-fluoro-2,6-DMA. In Salmonella typhimurium TA100, the two deuterated compounds and their non-deuterated analogues gave similar mutagenic responses ( approximately 25 revertants/nmol). Likewise in S. typhimurium TA98, a similar mutant frequency ( approximately 0.7 revertants/nmol) was obtained with the four compounds. With N-hydroxy-4-fluoro-2,6-DMA, the mutant frequency was reduced by approximately 90% in S. typhimurium TA100 and approximately 50% in S. typhimurium TA98. The results suggest that multiple adducts contribute to base substitution mutations detected by S. typhimurium TA100 while the C8-deoxyguanosine adduct is primarily responsible for the frameshift mutations detected by S. typhimurium TA98.

Characterization of DNA adducts derived from syn-benzo[ghi]fluoranthene-3,4-dihydrodiol-5,5a-epoxide and comparative DNA binding studies with structurally-related anti-diolepoxides of benzo[ghi]fluoranthene and benzo[c]phenanthrene.

This paper reports structural characterization of the adducts and tetraols formed from syn-benzo[ghi]fluoranthene-3,4-dihydrodiol-5,5a-epoxide (syn-B[ghi]FDE, 3) and comparative DNA-binding and mutagenicity studies involving 3, anti-B[ghi]FDE (2), and anti-benzo[c]phenanthrene-11,12-dihydrodiol-13,14-epoxide (anti-BcPDE, 5). The structures of nine DNA adducts and two racemic tetraols derived from 3 have been determined spectroscopically. Similar characterization of adducts obtained from the anti-isomer 2 was described in the preceding paper in this issue [Chang et al. (2002) Chem. Res. Toxicol. 15, 187-197]. The majority of DNA adducts with 3 are those from the trans- or cis-opening of the epoxide at C5a by the exocyclic amino groups of dG, dA, and dC. The diolepoxides 2 and 3 are rigid structure analogues of anti- and syn-BcPDE (5 and 6), respectively, thus serving as models for probing molecular deformity and diol conformation in diolepoxide-DNA interaction. Comparative DNA binding experiments indicate that 57% of 2 and 33% of 3 were converted into DNA adducts, whereas a 71% conversion was observed for 5. In general, lower percentages were observed with denatured calf-thymus DNA. As for base selectivity, 2 showed a greater affinity for dA relative to dG (dA/dG ratio, 0.79) than 3 (0.56) when reacted with native calf-thymus DNA. A much higher dA/dG ratio (1.41) was obtained for 5. The overall dA/dG ratios were lower with denatured DNA, indicating the importance of the secondary structure of DNA for both adduct formation and chemical selectivity. The T-shape pseudo-diaxial diols of 3 appears to have favorable electrostatic interactions with the nearby phosphate backbone in the minor groove of DNA, thereby yielding greater amounts of dG adducts than the pseudo-diequatorial 2. The anti-isomer 2 was found to be seven times more mutagenic than 3, but they are significantly less mutagenic than the nonplanar analogue 5 when tested in Salmonella typhimurium TA 100.