Totally Synthetic by Paul H. Docherty, 21 April 2009
Total Synthesis of Agelastatin A
P. M. Wehn, J. Du Bois, Angew. Chem. Int. Ed. 2009, 48, 3802-3805.
Du Bois' total synthesis of Agelastatin A is the third approasch described here (see also: Trost, Tanaka, Yoshimitsu). As the biology has been explained already, let's get straight into the synthesis. Du Bois starts with an opening of an optically active, bicyclic lactam (2-Azabicyclo[2.2.1]hept-5-en-3-one), which is commercially available, but not cheap. Protection, reductive ring-opening and sulfamate formation give us the starting material for the first scheme: a catalytic aziridination, that offers high diastereomeric control. That’s no real suprise, as the intramolecular tether isn’t very long, so forming a cis-ring junction would be heavily disfavoured. However, the next reaction is from a selectivity standpoint more impressive, as azide attack occurs with an 9:1 preference for their desired position. More about that selectivity in this paper.
With all four stereocenters now inplace around the core, it was time to start building the rest of the tetracyclic ring system. The tether needed to be remove first by using the inherent reactivity of sulfamates. Sulfonates are highly reactive and good leaving groups, whereas sulfonamides are quite stable protecting groups. So it’s clear why displacement occurs at the right position to give the desired phenyl selenide.
Elimination of the selenide gave the exo-methylene group, whilst a Staudinger reduction gave them a carabamate in the end. Lastly, a Paal-Knorr pyrrole synthesis allowed construction of the lower ring. Of course, they’ve now destroyed a stereocenter, but that situation was quickly resolved by doing an oxidative cleavage of the alkene. The intermediate ketone was then attacked by intramolecular cyclisation, providing the N,O-acetal as a single stereoisomer. The final ring was then installed using a base followed by a bromination to complete the target. Nice work.