Amphidinolide H and G
FÃ¼rstner, Bouchez, Funel, Liepins, PorÃ©e, Gilmour, Beaufils, Laurich and Tamiya. ACIEE, 2007, EarlyView. DOI: 10.1002/anie.200704024.
Apologies for the lack of blogging – turns out this writing-the-thesis thing is quite time consuming. Feel free to talk amongst yourselves… Anyway, normal service has been resumed for today, with another Amphidinolide in the bag (actually a pair – Amphidinolide H & G). As with a lot of macrolide syntheses, there’s plenty of leg-work to get the fragments together, and a degree of familiarity due to the type of chemistry used to make ‘em, but there’s still some pretty smart stuff here…
Looking at the retrosynthesis, there’s that degree of familiarity I mentioned. They used metathesis to close the macrolactone, after using the 1,4-anti aldol to unite the two largest fragments. The ester was formed using using DCC/DMAP, rather than the more complex systems I’m used to reading, and a Stille-type coupling was used to make the diene. This is where things get interesting for me, as the “Stille” wasn’t mediated by palladium, and wasn’t catalytic either. Nope, they used copper chemistry in the form of copper thiophene-2-carboxylate, or CuTC to you and me. This reagent is often employed in situations when the normal Stille conditions fail, and worked really quite well for them (82%).
The fragment synthesis also interested me, as they needed to hydrostannylate to form the internal stannane. Tricky reaction, but they overcame this using the commercial Bu3SnSiMe3, and a bit of palladium catalyst. The silylstannation went well, and following removal of the TMS (and TBS) groups, left the desired internal stannane. Now, the interesting bit for me is the regioselectivity of the silylstannation – why did it go that way? It’s probably in the references – I’ll read them and annotate this later.
Another intersting reaction in the synthesis of the fragments was the use of a Reetz hydrodgenation. Although this is quite well known, the result is sweet and worth highlighting. The use of such a small amount of phosphite ligand and metal precatalyst is awesome!
I haven’t highlighted the trouble they had – the final deprotections screwed the advanced material over, which is such appalling luck, but switching protecting groups in the fragments allowed completion of the synthesis. Nice work…