Overman, Dounay, Humphreys and Wrobleski. JACS, 2008, ASAP. DOI: 10.1021/ja800163v.
A far smaller target than the last we examined, but that’s still a very interesting architecture. Although the biological activity of this beast isn’t mentioned, similar natural products are found in traditional medicines, so perhaps it’ll do something interesting if you eat it. Their retrosynthesis is fairly ambitious, with a plan to use palladium couplings to complete two of the rings, and to do these reactions as part of a pair of tandem reactions.
The interesting chemistry kicks-off pretty early, with their asymmetric Heck coupling of the aryl triflate onto the cyclohexadiene. The reaction performed reasonably under thermal conditions, but improved considerably in the microwave, allowing lowering of the catalyst loading and vastly truncated reaction times. After this success, they dumped some TFA into the mix, and got their desired iminium-ion cyclization, completing both five-member rings in excellent yield and control.
A substrate-controlled epoxidation, followed by PhSeNa opening/elimination gave them the allylic alcohol they wanted and appendage of a vinyl iodide left them with the substrate for a palladium-cascade. They hoped that the palladium would oxidatively insert into the vinyl iodide, add to the cyclohexene in a Heck manner to form a new carbon-palladium bond. This would then be trapped by carbon monoxide, completin the carbon skeleton of the natural product. However, the actual product from this reaction was quite surprising – a third five-member ring. As they write, it would be tempting to suggest that an alkenylpalladium species had inserted into the allylic C-H bond – but they postulate a far more reasonable mechanism in which the somewhat tetchy aminal moiety cleaves, rearranges the cyclohexene via an iminium ion and then performs a more traditional 5-exo-trig Heck cyclisation. Dissapointing for them, but interesting none-the-less…
A revised strategy allowed them to complete the molecule, but they didn’t quite get to have their cake and eat it – that tandem palladium chemistry wouldn’t go. Instead as stepwise approach prevailed, starting with a palladium-catalysed intramolecular enolate/vinyl iodide coupling. The cyclohexanone product was then converted to a triflate, and a bit more palladium allowed introduction of carbon-monoxide and trapping with methanol to give the unsaturated ester.
This is an awesome read – lots of detail for each transformation, and a logical discussion of the problems they encountered.