Phorbasin C

Micalizio and Macklin. JACS, 2009, ASAP. DOI: 10.1021/ja809491b. 

An interesting test-case for new methodology, this family of diterpenes has a pretty novel architecture. The unsaturated sidechain is the bit I find intreguing, as skipped-conjugation is tricky to keep that way. Oh, there’s some bio-stuff to provide a little rationale (including a bit of bacteria growth inhibition, and some cytotoxicty), but nothing startling. Micalizio is in it for the chemistry, and concern about the C-11 stereocenter, which was unassigned when they started their work…

As they didn’t know which way that methyl group was going, they designed a route where the sidechain could be appended late-stage, and thus easily varied. This left them with the cyclohexenone as the initial target, and they decided to start with the ring already in place.

The SM is cyclohexadiene bearing an asymmetric glycol and vinyl bromide; pretty handy for their route. However, it comes at quite a price – £174 (that’s $254 in colonial cash) for five grams – quite a pretty penny. The other problem is instability – I quote from theSigma Aldrich website:

To recover the pure product from the suspension:
(1)Thaw the frozen suspension and filter the solid. (2) Rinse bottle with a few milliliters of base-washed (aqueous Na₂CO₃) ethyl acetate. (3) Use the rinsing to wash the solid. (4) Collect solid. (5) To further collect more product from the filtrate: extract filtrate with equal volumes of base-washed (aqueous Na₂CO₃) ethyl acetate (repeat three times). (6) Dry the ethyl acetate extract over MgSO₄. (7) Evaporate the solvent – DO NOT HEAT! (8) Combine solids collected from the suspension and the filtrate. Pure crystals should be stored at -78° C. A suspension of the product in phosphate buffer is stable at 0° C.

Hmm… so quite a start then! But from a chemistry point of view, it leads straight (well, a few steps later) into their synthetic centrepiece – a titanium mediate coupling of teeemess propyne. This allows direct coupling (or rather metallo-[3,3] rearrangement) of an unfunctionalised alkene with the silyl acetylene with control over the stereochemistry (‘exquisite selectivity…‘). Damn neat coupling, with more on the goods here.

From a damned complex reaction to a damned simple; deprotect an acetate. Should be simple, but there’s two in this system, so they had to be careful. A bit of scandium triflate did the job, providing site-selective deprotection. Presumably the rationale is neighbouring group participation…

Unfortunately, things take a turn for the ugly in the last few steps – but it doesn’t appear to have been avoidable. To complete their work, a pair of Suzuki couplings with the enantiomeric sidechains was required; however, this reaction seems to have have given the group a bit of a head-ache, as they needed to use several (20) equivalents of thalium carbonate. I’m sure I don’t need to describe the nastiness of this reagent in any detail… But this only goes to emphasise how crap incomplete our understanding of such coupling is; in my PhD I did several Suzuki couplings using similar substrates, and got away with 5% Pd(OAc)₂, 10% PPh₃, THF / 2M LiOH = 90% yield. I was probably just very lucky (those were the first conditions I tried too…), but where’s the rationale? Regardless, the student in question has my permission to print out this image as a badge.

However, their approach was ultimately successful, as one of their products bore an exactly opposing optical rotation to the isolate, meaning this was the enantiomer. So their synthesis may have been of ent-phorbasin C, but they did all the hard work for the group that makes the natural enantiomer.