Studies of highly functionalized bicyclic ring systems directed towards the total synthesis of [(positive or negative symbol)]-vernolepin

Abstract

Highly functionalized derivatives of the bicyclo[4.3.1]-dec-3-ene (1) ring system were examined as part of an effort to synthesize the sesquiterpene (+)-vernolepin (2). Although vernolepin has been synthesized in other laboratories, each of these syntheses (with one exception) has had one major drawback. That drawback has been the production of a structural isomer, vernomenin (3), as a by-product. The studies done in this laboratory were designed to take advantage of the stereochemistry inherent in the bicyclo[4.3.1]dec-3-ene ring system. By doing so it was expected that vernolepin could be produced without concomitant production of vernomenin. The bicyclic ring system was constructed in the form of dione 122. After a double bond was introduced at C-7 and the C-10 carbonyl was protected, the final two carbons needed to complete the synthesis were added via a Michael addition. This produced vinyl adduct 129. A series of steps involving selective reduction of carbonyl groups, deprotection of a carbonyl, and protection of hydroxyl groups led from 129 to the key intermediate 140. Completion of the synthesis required that the disubstituted double bond of the ring system be selectively epoxidized. However, as model studies using enone 127 showed, the desired selectivity was not achieved. Several attempts were made to surmount this problem using other bicyclo[4.3.1]dec-3-ene ring systems, but these attempts were unsuccessful. Among the attempts was an effort to synthesize epoxide 164. However, low yields were obtained for several steps in this sequence, and their cumulative effect prevented the attainment of 164. Finally an attempt was made to construct an intermediate from Greico's first synthesis (151). Again the cumulative effect of poor yields prevented construction of 151. Thus, vernolepin has not yet been synthesized using this methodology.