Totally Synthetic by Paul H. Docherty, 17 November 2009

Total Synthesis of Fostriecin

McDonald

O. Robles, F. E. McDonald, Org. Lett. 2009, 11, 5498-5501.

This target is new to me - but not to the chemical community at large. Indeed, the list of prior syntheses is quite unnerving for the prospective student-synthesizer, including Kobayashi, Boger, Cossy, Shibasaki, Reddy… This brings a certain amount of knowledge, perhaps about unviable intermediates or recalcitrant couplings - but also demands that something new be brought to the field.

With a triene in the target, there’s no surprise that palladium mediated couplings are the order of the day. Rather neatly, McDonald managed two in quick succession, using a gem-dibromo olefin. The first Negishi coupling goes for the more reactive trans-bromine, generating a good yield of the diene. Then, a little more palladium and ligand, along with dimethyl zinc allowed methylation of the remaining bromine.

I wonder if there was any hope of doing both couplings in the same pot… but the ligand is a bit different, so perhaps the two are incompatible.

With that done, it was time to oxidise that somewhat reactive looking diene. As can be seen in the target, what they needed was a syn diol, so Sharpless seems to be the method of choice. However, there’s a bit of a selectivity issue here, with only a methyl group differentiating the two halves of the diene. It appears that is enough, though, with a pretty decent yield of the desired regioisomer (13% went the wrong way).

There’s quite a bit left to do, but it was done quickly; revealing of the acetylene and formation of a vinyl iodide by iodation and diimide reduction brought them to an intermediate in Jacobsen’s synthesis. However, rather than go to the pub, they went off-piste, and coupled the RHS using Suzuki rather than Stille conditions. However, to get the reaction to go at RT (the triene doesn’t like heat, unsurprisingly…), they needed to use thalium carbonate.