Short Communication Bioavailable Flavonoids: Cytochrome P450-Mediated Metabolism of Methoxyflavones

Methoxylated flavones were recently shown to be promising cancer chemopreventive agents. Their high metabolic stability compared with the hydroxylated analogs was shown in our laboratory using the human hepatic S9 fraction with cofactors for glucuronidation, sulfation, and oxidation. In the present study, the resistance of methoxylated flavones toward oxidative metabolism was investigated with human liver microsomes and recombinant cytochrome P450 (P450) isoforms. Among 15 methoxylated flavones investigated, the two partially methylated compounds, tectochrysin and kaempferide, were among the most susceptible to microsomal oxidation (Clint 283 and 82 ml/min/kg). Of the fully methylated compounds, 5,7-dimethoxyflavone and 5-methoxyflavone were the most stable (Clint 13 and 18 ml/min/kg, respectively), whereas 4 -methoxyflavone, 3 -methoxyflavone, 5,4 -dimethoxyflavone, and 7,3 -dimethoxyflavone were the least stable (Clint 161, 140, 119, and 92 ml/min/kg, respectively), emphasizing the importance of the positions of the methoxy substituents in the flavone ring system. Among the five P450 isoforms tested, CYP1A1 showed the highest rate of metabolism of fully methylated compounds, followed by CYP1A2 and CYP3A4. CYP2C9 and CYP2D6 gave minimal disappearance of the parent compound. Finally, in incubations with hepatic S9 fraction with cofactors for oxidation and both conjugation reactions, partially methylated flavones, as expected, were much less metabolically stable than fully methylated flavones, confirming that oxidative demethylation is the ratelimiting metabolic reaction for fully methylated flavones only. In summary, the rate of oxidative metabolism of methoxylated flavones, mainly involving CYP1A1 and CYP1A2, varied widely, even between compounds with very similar structures. Promising biological effects of dietary flavonoids and other polyphenols as chemopreventive agents against cancer, cardiovascular disease, and other diseases have been shown in cell culture studies (Middleton et al., 2000; Williams et al., 2004; Walle, 2007). However, in vivo, in particular in humans, using moderate, clinically relevant doses, these effects have not been replicated. This is mainly because of very low oral bioavailability, as has been shown in clinical studies of, for example, chrysin (5,7-dihydroxyflavone) (Walle et al., 2001a) and quercetin (3,5,7,3 ,4 -pentahydroxyflavone) (Walle et al., 2001b; Manach and Donovan, 2004; Williamson and Manach, 2005). This poor bioavailability of dietary polyphenols is highly dependent on their free hydroxyl groups, making them susceptible in particular to glucuronidation and sulfation and much less so to oxidation (Otake et al., 2002), which commonly, but not always, removes the biological activity of these compounds. In contrast, the methoxylated flavones are much more resistant to hepatic metabolism and have higher intestinal absorption through Caco-2 cell monolayers compared with their unmethylated analogs (Wen and Walle, 2006a,b). Among the few in vivo disposition studies of the methoxylated flavones, nobiletin (5,6,7,8,3 ,4 -hexamethoxyflavone) reached significant tissue levels after p.o. administration to rats (Murakami et al., 2002), whereas the coadministered unmethylated flavone luteolin (5,7,3 ,4 -tetrahydroxyflavone) gave much lower tissue levels. Similarly, in the rat, 5,7-dimethoxyflavone (5,7-DMF) reached high tissue levels (Walle et al., 2007), whereas coadministered chrysin was undetectable. Tangeretin (5,6,7,8,4 -pentamethoxyflavone) administered in the feed to hamsters was absorbed but extensively metabolized (Kurowska and Manthey, 2004). Whereas the antioxidative properties of flavonoids with free phenolic groups have been considered necessary for effects (Rice-Evans, 2001), recent studies have shown that some of the methoxylated flavones, including the citrus flavonoids nobiletin and tangeretin, have potent antiproliferative activities (Pan et al., 2002; Morley et al., 2007; Walle et al., 2007), inhibit the bioactivation of benzo[a]pyrene by CYP1A1 and 1B1 and thereby cancer initiation in various cell lines (Wen and Walle, 2005; Wen et al., 2005; Tsuji and Walle, 2006; Walle and Walle, 2007), and are inhibitors of aromatase (Ta and Walle, 2007), an important target in hormone-sensitive cancers. The present study focused on the oxidative metabolism of a number of methoxyflavones (Fig. 1) by human liver microsomes and pure P450 isoforms to try to elucidate whether there are particular structural features that confer resistance or susceptibility to metabolism. This should be important when selecting methoxyflavones for testing as potential chemopreventive agents in human disease. Materials and Methods Materials. 5-MF, 7-MF, 3 -MF, 4 -MF, 5,7-DMF, 5,3 -DMF, 5,4 -DMF, 7,3 -DMF, 7,4 -DMF, 3 ,4 -DMF, 5,7,4 -trimethoxyflavone (5,7,4 -TMF), sinensetin (5,6,7,3 ,4 -pentamethoxyflavone), tectochrysin (5-hydroxy-7-MF), This study was supported by National Institutes of Health Grant GM55561. Article, publication date, and citation information can be found at http://dmd.aspetjournals.org. doi:10.1124/dmd.107.016782. ABBREVIATIONS: DMF, dimethoxyflavone; MF, methoxyflavone; TMF, trimethoxyflavone; UDPGA, UDP-glucuronic acid; PAPS, 3 -phosphoadenosine-5 -phosphosulfate; HPLC, high-performance liquid chromatography. 0090-9556/07/3511-1985–1989$20.00 DRUG METABOLISM AND DISPOSITION Vol. 35, No. 11 Copyright © 2007 by The American Society for Pharmacology and Experimental Therapeutics 16782/3267220 DMD 35:1985–1989, 2007 Printed in U.S.A. 1985 at A PE T Jornals on Jne 5, 2017 dm d.aspurnals.org D ow nladed from