Magnus, Cheung, Goldberg, Russell, Turnbull and Lynch. JACS, 2007, ASAP. DOI: 10.1021/ja0744448.
Another “big target”, another JACS… And this is another molecule that has been in the making for several years. Phil Magnus isn’t the first in this case – Nicolaou and Harran have both been there and done that. However, a worthy point made by Magnus is that both of the previous syntheses have involved rather inefficient H ring formation. That C-10 stereocenter is a pig! However, let’s look at a bit of a retro first:
I say a bit of a retro because this only takes us back to a key intermediate. But that’s enough for now. The fragments are made very quickly and with apparent ease… I particularly liked the alkyl-lithium alkylation – three deprotonations on the oxazoline! A peptide coupling put in the other fragment, leaving us set for more exciting transformations.
The macrocycle was first formed by simply deprotecting the aryl silylether, promoting displacement of the freshly installed chloride. (I’m actually interested in the stereochemistry of the chlorination, as thionyl chloride displacements of secondary alcohols are supposed to go SNi. However, this is apparently very substrate dependant. No way to tell in this case, however, as the both the product and SM are 1:1 mixtures of diastereoisomers…)
I must say that the aryl ether formation goes in particularly good yield. It might be intramolecular, but that bond formation looks fairly congested. I wonder if the adjacent carbonyl group assists…
Either way, the ether bond formation was great, and gave them an easier entry to the C-arylation. This rearrangement was performed both on the separated isomers of the aryl ether, and also on the 1:1 mixture, with the ratios above quoted for 100% conversion. The group performed these reactions with a broad range of solvents, varying the polarity to probe the reaction mechanism. They found from this that when using very polar solvents that stabilise separate ion formation, the product ratio is close to 1:1. However, using chloroform, a predominance of the desired product could be achieved. I’m not going into the full justification – this is a really nice piece of chemistry that is best explained in the paper. So read it.
The approach reminds me somewhat of Steve Ley’s chemistry; his synthese of both Rapamycin and Azadirachtin included formation of difficult bonds by “rearrangement of a tether”. Nice work!
InChIKey=YKBUODYYSZSEIY-FCHPBXHZBC InChIKey=SXDQQMTZSPEYFP-POEUFYPIBW InChIKey=MZJVSXCNDDXUKX-KCPSNSNRBT InChIKey=MZJVSXCNDDXUKX-VEIFWNHXBW InChIKey=ASSNSFRKKJMRAI-WNZYYCCLBQ InChIKey=FTSOZXRKWNWOSU-XKRKMPLYBQ