J. Am. Chem. Soc. 134, 893–896 (2012)

Halichondrin C (pictured) is a polyether macrolide whose isolation from a marine sponge was first described in the mid-1980s. Thanks to the work of Yoshito Kishi and co-workers from Harvard University, it has now succumbed to total synthesis. The total synthesis of another family member — halichondrin B, which lacks the hydroxyl group highlighted in green — was reported by Kishi and co-workers almost 20 years ago. Biological testing showed that the right-hand side of the molecule was most important for its anticancer properties — knowledge that eventually led to the development of the drug Halaven (Erbulin) which is used to treat metastatic breast cancer.

The key to the synthesis of these molecules was always the construction of the C8–C14 polycycle, highlighted in blue. Kishi and co-workers planned to use a method analogous to that used in their halichondrin B synthesis. Specifically, they had previously used an oxy-Michael addition to an enone, and now they wished to develop a similar addition to an ynone — a strategy that, if successful, might be suited to the synthesis of halichondrin A as well. The ynone Michael addition worked but was followed by a second, unplanned, oxy-Michael addition to the resulting enone. The researchers were, however, able to convert this to the polycycle found in halichondrin C using a Lewis-acid-mediated isomerization.

Kishi and co-workers used some of the building blocks developed in their previous work (C14–C38 and C39–C54), but improved the synthesis along the way. They were able to use a chromium/nickel-mediated coupling to form no fewer than seven C–C bonds of the target (disconnections shown in red).