Movassaghi, Tjandra, Qi. JACS, 2009, ASAP. DOI: 10.1021/ja903790y.
Okay, I missed my ‘I’ll blog it by Saturday‘ deadline by a day, but it’s taken quite some time to get my head round these structures. And beer. However, I think I’ve got there – so ladies and gents – on with one of the best syntheses this year.
First up scheme-wise is a rather nice Suzuki coupling. Taking a standard boronic acid, mixing in a vinyl-gem-dibromide, they achieved a selective E,E- product in cracking yield. However, this came at a price, as they had to use thalium carbonate in the mix; not a pleasent compound. Quoting directly from the SI, ‘Tetrakis(triphenylphosphine)palladium (1.54 g, 1.30 mmol, 8.00 mol%) and thallium carbonate (15.7 g, 33.0 mmol, 2.00 equiv) were added sequentially…’ – urgh! I’m sure Movassaghi tried several sets of conditions, but I’d have used quant. tetrakis before I touched the thalium bottle! However, the student in question get permission from me to use the Bas-Ass Pot-Stirrer badge in this post.
The free vinyl bromide (the less-reactive of the pair) was then avaliable to a bit of Buchwald chemistry, doing a copper-mediated coupling with azetidinone. Normally these amides are crap with BH amination chemistry, but I reckon the strained-ring improves the avaliability of the N lone-pair, and gives the reaction a bit of a poke. A bit of enolate shennanigans, and they were set to use their tetra-ene in the Diels-Alder fashion. Four stereocenters were set in excellent yield and selectivity, but that wasn’t enough – a bit of tickle-four allowed a Mukaiyama aldol to complete a third ring, and provide a nice synthetic handle.
Time to put in two more rings – but this time they bought one in (no point in over-doing it…). Lithiation of the imine and Michael addition into the enone provided the expected imine intermediate. This tautomerised to the enamine, which added back into the ketone, providing a cyclohexanol in cracking yield and efficiency (I don’t have the actual figures for this). This follows a hypothesised biosynthetic route, but it’s still pretty inventive!
As can be seen from the tightly caged structure, they had a few more ring junctions to form. Even though they’ve only got one protecting group in place, a remarkable selectivity was had. A bit of the Vilsmeir reagent allowed an enolate-type addition into the chloro-imine to form a highly-activated imine intermediate (+ 1 carbon). The free (but highly hindered) tertiary alcohol was then able to add-into the imine, spitting-out dimethyl amine to give a dihydrofuran.
This DHF didn’t hang-around long – addition of DDQ opened the ring to form an alcohol and an aldehyde, which was immediately oxidised to the free-acid and methylated. (No pussy TMS-diazamethane for the Movassaghi group – they’re using free diazamethane for this! Gotta look-after that non-ground-glass kit!). Removal of the sole protecting group then had them ready for yet another impressive cyclisation. Treatment of the cyclohexenone with NCS promoted addition of the now-free piperidine into the ring, and completion of the carbon skeleton.
I’m not going to get into the mechanism for that transformation Movassaghi did quite a bit of work on it, and if I did I’d have transribed almost all the paper. Read it. Really, one of the best paper so far this year, cementingMovassaghi as a top-rank total-synthesiser.