Totally Synthetic by See-Arr-Oh, 16 June 2011
Total Synthesis of Cyanolide A Aglycon
M. R. Gesinski, S. D. Rychnovsky, J. Am. Chem. Soc. 2011, 9727-9729.
When I first saw the dimeric gamma-substituted pyran motif in this molecule, I said “This has to be Rychnovsky” (or maybe Wender). Throughout his career at UC-Irvine, Scott Rychnovsky has been the master of the TMS-promoted diastereoselective Prins cyclization, which he has applied to the synthesis of at least 10 macrolides. Until now, however, none have been attempted without protecting groups!
Cyanolide A was isolated in 2010, and it’s already been the target of four total syntheses. Perhaps this is due to its intriguing biological profile; I usually assume new natural products will be antibacterials, antifungals, or might cure cancer. Not cyanolide A: it’s a molluscide. Rychnovsky hopes it could one day treat schistosomiasis, a disease caused by a parasitic worm carried by snails living in contaminated water.
The authors point out that most of the syntheses to this point have been based on either Yamaguchi or Shiina macrolactonization of two identical pyrans to form the C2-symmetric compound. Our team wants to use the double Prins here, but realize the required gem-dimethyl synthon will divert the reaction from the expected 2,3-pyran to form a THF ring instead. So they forge ahead with a Sakurai dimerization:
The authors believe that their monomeric unit might be unstable under Lewis acidic conditions, so they opt for a dimethyl acetal instead…wait, isn’t that a protecting group? (we’ll let it slide for now)
Troubles were afoot at the first step, when the authors tried a Ce(III)-mediated addition to a hindered ethyl ester, only to find out that they couldn’t install their desired allylsilane. Luckily, they devised a workaround: 2-step enol triflate formation, Kumada coupling to install the allylsilane, then pTSA deprotection led to an aldehyde, where a simple aldol reaction set them up for synthon 2. Right?
Wrong! All attempts to use a thiazolidinethione auxiliary with standard tin(II) or titanium(IV) Lewis acids just swapped the silane for a proton. Using modified Sammakia conditions (sparteine, PhBCl2), the authors were able to promote their aldol reaction without suffering subsequent protodesilylation. Simple LiOH hydrolysis of the auxiliary provided 2.
Now, how do we append the other five carbons of the lactone? Nucleophilic addition of a thioacetal to an epoxide, followed by oxidative methanolysis forms their desired pentyl alcohol.
Yamaguchi conditions unite 2 with 3, although they had to test quite a few other lactonizations first, since 2 tended to form the ?-lactone. From here, Prins Lewis acid conditions conditions (TMSOTf in DCM) are enough to form the dimeric, symmetric backbone, and we're only an Upjohn dihydroxylation, diol cleavage, and reduction away from the target:
A pretty clean 10-step LLS, with an overall yield at 18%, and (mostly) protecting-group free.