Totally Synthetic by Paul H. Docherty, 24 June 2006

Total Synthesis of Amphidinolide X & Y

Fürstner

A. Fürstner, E. Kattnig, O. Lepage, J. Am. Chem. Soc. 2006, 128, 9194-9204.

DOI: 10.1021/ja061918e

Another pair of amphidinolides in the bag, Fürstner et al. have completed the synthesis of X (the only member of the series with an even-numbered macrocycle) and Y using a powerful iron catalysed process. Both products (as with most of the family) are cytotoxic, and contain the heavily functionalised THF moiety. This allowed the group to create an intermediate common to both campaigns, starting from a simple epoxide produced from an Sharpless epoxidation.

Treatment of this with n-propyl grignard and catalytic quantities of the iron catalyst generated the allene in a 8:1 dr (this system has been used by the group in other work; see: DOI: 10.1246/cl.2005.624, DOI: 10.1021/ja027190t, DOI: 10.1002/anie.200460504, plus further examples cited in the paper). The allene was then cyclised with silver nitrate and calcium carbonate, returning the DHP, which was augmented to the desired THF via bromoesterification.

This portion of the natural product was coupled using an alkyl Suzuki reaction to the rest of the molecule in both cases, along with macrolactonisation to furnish the major ring system. In amphidinolide Y, a boron-mediated aldol reaction was used to create the 1,4 anti relationship between a pair of hydroxyls in the C1 - C12 fragment, in a 4:1 dr. Inseparable at this point, they carried the mixture through to a diastereoselective methyl grignard addition.

The desired aldol product reacted diastereoselectively with the grignard following the 1,2-anti chelate-cram model, whereas the undesired aldol product reacted with far less control. This section of the synthesis is quite intriguing, and is discussed in far more detail in the paper, which is a truly excellent read.