A non-enzymatic synthesis of (S)-(-)-rosmarinic acid and a study of a biomimetic route to (+)-rabdosiin

Article abstract of DOI:10.1139/v97-612

The synthesis of (S)-(-)-rosmarinic acid (30) in 9% overall yield is described. The synthesis was achieved by a convergent route in which 3-(3′,4′-dihydroxyphenyl)-(S)-lactic acid (23) and caffeic acid (25), both appropriately protected, were coupled to produce a pentaallyl precursor 29, which was then deprotected to give (S)-(-)-rosmarinic acid (30). A triallyl derivative 35 was similarly prepared and converted to (+)-rabdosiin (41) and its (1R,2S) isomer (42) via a biomimetic oxidative free radical coupling-cyclization followed by deallylation. The coupling-cyclization gave a ratio of rabdosiin diastereomers unlike that found in nature. A preliminary study showed that methyl (R)-mandelyl sinapate (15) could be dimerized diastereoselectively to give a 1,2-trans thomasidioate diester (16). The synthesis of (S)-(-)-rosmarinic aid (30) in 9% overall yield is described. The synthesis was achieved by a convergent route in which 3- (3',4'-dihydroxyphenyl)-(S)-lactic acid (23) and caffeic acid (25), both appropriately protected, were coupled to produce a pentaallyl precursor 29, which was then deprotected to give (S)-(-)-rosmarinic acid (30). A triallyl derivative 35 was similarly prepared and converted to (+)-rabdosiin (41) and its (1R,2S) isomer (42) via a biomimetic oxidative free radical coupling- cyclization followed by deallylation. The coupling-cyclization gave a ratio of rabdosiin diastereomers unlike that found in nature. A preliminary study showed that methyl (R)-mandelyl sinapate (15) could be dimerized diastereoselectively to give a 1,2-trans thomasidioate diester (16).