Organic Chemistry Portal
Reactions >> Total Syntheses

Totally Synthetic by Paul H. Docherty, 25 April 2007

Total Synthesis of Platensimycin

Nicolaou

K. C. Nicolaou, D. J. Edmonds, A. Li, G. S. Tria, Angew. Chem. Int. Ed. 2007, 3942-3945.

DOI: 10.1002/anie.200700586

As much as I appreciate all the work done on this molecule, I’m sure I’m not the only one who would like to see it have a little rest for a while. However, we can’t do that without one last flourish, again from KCN’s labs. Those of you who read his original racemic publication of last year (described here) will have noticed that had the cycloisomerisation step been enantioselective, the entire synthesis might follow in that manner.

However, as Nicolaou pointed out in the discussion in this paper, the Trost chemistry used in that synthesis wasn’t amenable for enantioselectivity. However, a publication by Zhang et al showed that a similar species of substrate could be cyclised in this manner using a rhodium catalyst. So they tried it, and with a little modification of the substrate, an excellent result could be achieved:

This product was a little different to that achieved in the prior paper, so they used a different approach for the cyclisation, firstly forming the Barton xanthate ester, looking for a radical decarboxylation. However, the reaction went a little further, isomerising via a 1,3-H shift to the more stable internal olefin. No matter - it still did the Sm mediated cyclisation as before, returning a common intermediate, except in enantiomerically enriched form.

Feeling done with this target yet? KCN wasn’t! They have a second route to the molecule in this paper, this time using a oxidative dearomatising cyclisation as the key step. A lot less needs to be said about this synthesis, which builds the cyclisation precursor through a more routine (though no less reasonable) fashion. The cyclisation with hypervalent iodine (PIDA) then went in appreciable yield to return an intermediate recognisable from his other work.

I prefer the former route to the latter (although I love hypervalent iodine as a reagent), but both are smart ideas. This may seem to be too much for one small molecule, but as I pointed out in my first post, it’s a novel antibiotic, and therefore worthy of this attention.