Xie, Du, Li, Li. ACIEE, 2009, ASAP. DOI: 10.1002/anie.200902908.
A second showing for this complex little target, Zhixiang Xie’s route uses the same end-game as that developed by Nicolaou two years ago. However, getting to that point is a very different matter, using some quite powerful chemistry. The RHS of the molecule is a distinct natural product – hyperolactone C, with the LHS also natural – zingiberene. The latter was produced as in Nicolaou’s paper, but hyperolactone – the more challenging portion – was produced in a rather different manner.
First up is a synthesis of the actual lactone. The SM for this chemistry is a rather sensitive-looking tricarbonyl compound, with a suspicious looking diazo-group. Selective diazotisation of the beta-keto-ester postion was achieved in a simple manner – by working with the benzyl alcohol and oxidising the product with DMP. Of course, a little rhodium catalyst did the usual trick with diazo groups, extruding nitrogen to form a carbonyl ylide, which cyclised to give a furanone.
It was then time for the center-piece – a Pd-AAA. For those out-of-the-know, this is an abbreviation for a palladium-catalyzed asymmetric allylic alkylation, pioneered by Barry Trost. In fact, I’ve written a couple of posts about it – Tamiflu, Terpestacin. Perhaps a better place to read about this chemistry is in this review. However, one case left untouched by Trost is the creation of two adjacent stereocenters in this reaction – a case Xie intended to tackle. In this case, he was using a racemic isoprene monoepoxide along with a bit of Pd catalyst and a that beta-keto-ester to do just that, in a manner reminiscent of a further Trost paper. However, in that paper, Trost’s beta-keto-ester is unsubstituted, so doesn’t create such a complex system. Using those conditions, Xie has acheived a respectable yield in awesome enantiomeric excess, with a pretty handy d.r. too. What more do you want?! Well, if I’m fair, a bit mroe product would be nice, but the problem is the reaction electronics. Putting a para-chloro substituent on the phenyl ring changes things quite a bit, bringing the yield up to 68%, and improving the d.r. However, getting rid of such substituents is a pain, so I hope they have future reagent-based success – perhaps experimenting with Lewis acid / base additives could help…
As is obvious, completion of hyperolactone was now no stretch – a little PTSA was enough to snap that ring shut. And we’re basically done and dusted, but I will reiterate the synthesis of biyouyanagin A, which as I said uses Nicolaou’s chemistry. What we need here is a [2+2] cycloaddition, so it’ no surprise that the reaction is done in a tanning-bed, but the interesting thing is that this reaction is biomimetic – I wasn’t aware of [2+2]s in nature. Neatly done; seven steps in all.