Synthesis of a Complete Series of O-Methyl Analogues of Naringenin and Apigenin

Abstract

Flavonoids have a variety of biological effects in numerous mammalian systems in vitro as well as in vivo. Most of the in vitro studies are performed on flavonoid itself and not on its metabolites as these latter are not readily available from commercial sources. However, no systematic preparation of flavonoid metabolites has been reported presumably due to the difficulty in controlling the regioselectivity. Particularly, regioselective synthesis of the 5-Omethylflavones or 5-O-methylflavanones has rarely been accomplished. It is well-known that the hydroxyl function on the 5 position of flavonoids resists to alkylation under basic conditions due to the acid-weakening effect of an intramolecular H-bond between the 5-hydroxyl group and the 4-keto group. The aim of this study was thus to prepare a complete series of O-methyl naringenins (flavanones) and apigenins (flavones) with emphasis on the syntheses of 5-Omethyl analogues. For the purpose of preparing every possible O-methyl analogues of naringenin and apigenin, various regioselective methylation and protection was attempted by using narinenin as a starting material (Scheme 1). The hydroxyl function on the 7 position of naringenin is the most acidic site among its three hydroxyl groups (7-OH, 5-OH and 4'OH). Thus, treatment of naringenin with equimolar amount of the alkylating reagents such as Me2SO4 and BzCl under basic conditions gave the corresponding 7-O-alkylated products 1 and 11, from which a variety of O-methyl naringenins and apigenins were prepared. Oxidation of 1 with DDQ provided 7-O-methylapigenin 2. On the other hand, treatment of 1 with Me2SO4 and TBDMSCl provided 7,4'di-O-methyl naringenin 3 and 7-O-methyl-4'-O-tert-butyldimethylsilyl naringenin 4, respectively. It is worth to note that, compared with dimethylation of naringenin by use of excess amount of Me2SO4, stepwise methylation of naringenin via 7-O-methyl analogue 1 always provided the 7,4'O-dimethylated naringenin 3 in higher yield. With these monoand dimethylated naringenin analogues 3 and 4 in our hands, we set out divergent syntheses of 5-O-methyl apigenins 9 and 10: methylation followed by oxidation (upper routes: 3 → 5 → 9 and 4 → 7 → 10; Scheme 2) and the other way around (lower routes: 3 → 6 → 9 and 4 → 8 → 6,10; Scheme 2). First, treatment of 3 and 4 with excess amount of Me2SO4 provided the corresponding 5-O-methyl naringenin analogues 5 and 7, respectively, which were then treated with DDQ in refluxing 1,4-dioxane to give clean conversions to the corresponding 5-O-methyl apigenin analogues 9 and 10. 5,7,4'-Trimethyl apigenin 9 was also prepared by methylation of 6 which was obtained by DDQoxidation of 3. The H NMR spectrum of 9 was found identical to the literature data. The structure of 9 was further characterized by 2DNOESY experiment in which irradiations of 5-O-Me, 7-OMe and 4'-O-Me resulted in increases of H-6, H-6/H-8, and H-3' resonance frequencies, respectively. Throughout this study, these characteristic nuclear Overhauser effects between adjacent protons were used to unequivocally identify the positions of the O-methyl groups. On the other hand, methylation of 8 gave an inseparable mixture of apigenin analogues which were purified after removal of protecting groups by TBAF in THF to give two dimethylated apigenin analogues 6 and 10. Presumably, the compound 6 was formed by partial loss of the 4'-O-TBS group followed by methylation. Thus, we tried to install other protecting groups such as benzyl or benzoyl groups at the 4' position but, in these cases, DDQ-oxidation did not proceed at all (data not shown), which suggests the 4'-OTBS group as the orthogonal protecting group of choice to the 7-O-Me of naringenin.