Mulzer and Tiefenbacher. ACIEE, 2007, EarlyView. DOI: 10.1002/anie.200702852.
It’s back! I’ve not blogged any total syntheses of platensimycin for ohh… at least three months, but this is some smart work by Johann Mulzer at the University of Vienna. Rather than complete a full synthesis, they’ve done the same as a few other groups and taken an intermediate from Nicolaou’s work and completed a synthesis of that structure. This is the same intermediate present in Snider’s synthesis, so I guess the consensus is that this is a good disconnection. High praise indeed!
The synthesis kicks-off with the literature three-step synthesis of the acid shown in the scheme below. This compound is then converted to the acid chloride, and what I think is an Arndt-Eistert homologation. This is a cool piece of classical rearrangement chemistry which goes via formation of an alpha-diazo ketone, which then perfoms a Wolff rearrangement to give the ketene. This is then trapped by any nucleophile present – in this case the methoxy-phenol. This forms the five member ring, and a lot of the complexity is introduced in this step. However, the synthesis is racemic, which is quite a drawback. Mulzer states that if one performed this chemistry on an enantiomerically enriched acid SM, the reaction should be controlled by that centre. This asymmetry could be introduced in the final step in the creation of the acid by doing an asymmetric hydrodgenation or a resolution. So my question is why didn’t they do it??
Anyway, the racemic chemistry is cool, so it’s time to complete the formal synthesis with a further hydrodge. However, they have a selectivity issue in that there are two similar olefins (but one is clearly less accessible). They didn’t get the selectivity required, so simply reduced both moieties, and then reoxidised. I hadn’t seen this reaction before, but the restore the required enone using iodic acidâ€“dimethyl sulfoxide complex. Nice! Hypervalent iodine is awesome
ChemSpider is currently down, so no InChiKeys for now…