Total synthesis of (+)-spongistatin 1 and studies towards the gram -scale synthesis of (+)-spongistatin 1

This dissertation describes the total synthesis of the potent anticancer agent (+)-spongistatin 1 (1). Chapter one summarizes the biological activity, structural elucidation and efforts of several research groups toward the total synthesis of spongistatin 1 and 2 to date. Chapter two details the Smith group\u27s synthetic venture with the focus on the synthesis of the EF hemisphere. An improved synthesis of (−)-215, employed in the first generation of spongistatin 2 (2), has been achieved in 8 steps from 1,5-pentanediol (30% overall yield). Also noteworthy, 300 g (−)- 215 was prepared using this sequence. The synthesis of F ring (+)- 216 was then completed in 12 steps from D-isoascorbic acid in 9% overall yield. This sequence, however, is not suitable for large scale reaction, because the sequence contained two problematic steps.* After careful analysis of EF fragment, we reasoned that dihydropyran F 261 would provide more flexible for the side chain addition. In addition to the obvious alkylation of the kinetic enolate derived from ketone 213 with allyl iodide 214, we could also introduce the side chain via the epoxide opening method.* The synthesis of F dihydropyran aldehyde (+)-261 was first achieved in 12 steps from propionyl chloride in 8% overall yield. An improved route for (+)-261 was then completed in 5 steps from the known propiolaldehyde 295 in five steps and in 42% overall yield. Unfortunately, removal of the dithiane functionality and the acetonide group in the presence of the enol ether moiety in coupled product (+)- 300 proved to be problematic. To circumvent the problem of removal of the acetonide and dithiane, we decided to use a hydroxyl group to mask the enol ether prior to removing the acetonide and dithiane groups. Therefore, tetrahydropyran 303 was chosen as a suitable F ring subtarget. The synthesis of the fully functionalized EF fragment was thus achieved employing 24 steps in the longest linear sequence with an overall yield of 1.8%.* In conclusion, we have achieved an effective stereocontrolled total synthesis of (+)-spongistatin 1 (1), highlighted in turn by a highly selective dithiane addition to F-pyran aldehyde (+)-303, the construction of allylstannane (+)-260, and the efficient union of key fragments. The synthesis proceeded in 29 steps for the longest linear sequence (based on the EF subunit), with an overall yield of 0.5%. The total number of steps required to prepare (+)-spongistatin 1 from commercially available materials was 112. Importantly, the linear sequence for the synthesis is 11 steps shorter than our previous reported synthesis of the closely related spongistatin 2 (2), and is the shortest synthesis of a spongistatin reported to date (currently the longest linear sequence of the published routes are ≥32 steps). (Abstract shortened by UMI.) *Please refer to dissertation for diagrams