Starting from protoanemonin (7), synthetic investigation of dl-coriamyrtin (1) was undertaken. The 1,6 adduct (6) from (7) and (8) was subjected to react with isopropenylmagnesium bromide-CuI to afford the compound (9) and (10). All trials to convert these compounds to the (4) type compounds possessing the correct stereo-chemistries for the present synthesis were unfruitful. On the other hand, a series of reactions of (27) from (6), acidic methanolysis, isopropenylmagnesium bromide-CuI and acidic hydrolysis, gave stereoselectively the compound (4) in four steps. The compound (4) was derived to the lactone (3) by the mixed-anhydride method and treatment of (3) with NBS gave the bromo-ether (31). After several unfruitful trials for introducing the C_1 unit to the carbonyl group at C_9 of (31), (32) and (33), the prolactone (34) was treated with Ph_3P=CH_2 and Me_2S=CH_2 to give (37) and (36), respectively. The subsequent conversions of (36) and (37) to the natural product, however, were unsuccessful. The Wittig reaction of the saturated ketone (34) under the reaction condition reported by Conia gave the exomethylene compound (47), which was transformed to the bromo-ether (49). Oxidation of (49) by the Sharpless methods gave a single product (51), in good yield. Dehydration of (51), followed by epoxidation afforded a diastereomeric mixture of diepoxides (53). Separation and definite identification of the synthetic product with bromo-coriamyrtin are in progress.
