Chackalamannil, Shah, Ganguly, Chelliah, Kolotuchin, Buevich and McPhail, JACS, 2006, ASAP. DOI: 10.1021/ja065198n
An interesting synthesis by a group working at the Schering-Plough Research Institute. Certainly, this family of targets has received a load of attention by the pharma industry. Their synthesis starts with an aldehyde produced by Evans in his synthesis of (+)-A83543A, shown below:
This advanced fragment was elaborated with some nice, but fairly well-known chemistry to the cyclisation precursor. Treatment of this compound with base prompted little reaction, but heating in a microwave with HCl for one hour (quite a long time, really) allowed the aza-Michael addition to the now open unsaturated acid.
The intermediate then decarboxylated, and retro-Michael addition occured, returning to the original skeleton, sans-lactone. This is interesting, as the intermediates produced resemble the natural product very closely, but do not lead there directly. The group then resubmitted the product to an aza-Michael, this time using Sc(OTf)3 in acidic chloroform. This preformed the desired cyclisation, allowing them to complete the synthesis by reduction of the ketone. However, sodium borohydride delivered the hydride to the wrong face, resulting in epi-himgaline. They overcame this problem by using sodium (triacetoxy)borohydride, which formed a chelate to the neighbouring hydroxyl. This allowed delivery of the hydride form the correct face, and completion of the target.