Njarðarson , McGrath, Bartlett, Sittihan. ACIEE, 2009, EarlyView. DOI: 10.1002/anie.200903347.
It’s back! Twice in as many weeks, too, as KC has decided to drop the details on his many syntheses of platensimycin and platensin into a full JACS paper. However, I’ve spent enough time looking at his work in this area, so it’s a pleasure to examine this synthesis by Jon Njarðarson of Cornell. I should point out that I don’t feel that this is necessarily the best route, but it’s got some damn smart chemistry in it, so what more do you want?
The first bit of chemistry to catch my eye was his use of Molander’s ‘BeeEffThreeKay salts’ (or at least that’s what I call ‘em). I’d be remiss not to mention them for two reasons – 1) my lab-mate Keith worked on these for his post-doc, and 2) it’s pretty smart stuff, and a lot nicer to work with than boronic acids. Sure, sometimes one can use boronic esters directly, or even use disiamyl-boranes (like I did in my PhD), but these salts are a more stable way to go. There is some concern that they can let-go of a bit of HF (and are often delivered in plastic pots), but you’ve really gotta maltreat them to provoke that kind of reaction.
This is one of the more impressive implementations of this chemistry I’ve seen, as sp3-aryl couplings aren’t the easiest chemistry even with ‘normal’ Suzuki conditions. However, Njarðarson has achieved a decent yield here with pretty respectable catalysis. </salespitch>
Sweet as that was, it wasn’t Njarðarson’s chemistry – this is. What we’ve got here is a neat, selective epoxidation of an alkene and rearrangement of the allylic system to give a DHP. More on this methodology here. The use of the trityl hydroperoxide was key to the regioselectivity of the nucleophilic epoxidation – more about that in this JACS by John Porco.
With that tricyclic system in place, a bit of reduction installed a further stereocenter, and left the group ready to perform a radical reaction to close the final ring in the core. However, in the earlier work the group did, no product could be found. They tried a few simple structural modifications to bring the prospective partners closer together, but to no avail. This is an unfortunate problem with radical chemistry – if the reaction decides not to work, switching reagents rarely works. Often the only solution is to change the substrate, but in target-oriented synthesis, this is not only a synthetic bastard, but a bit more ‘diversity’ oriented.
Luckily, the group had a further idea on how to do this – switch to an ionic reaction, and try that way. First they had to prepare the desired functionality, requiring silyl protection on the phenol. Using a little triethyl silane and a more exotic borane (which looks pretty pokey!), they achieved both removal of the methyl ether and reprotection. Cool chemistry – more here. Now set to go, a high-temperature deprotection produced a species with plenty nucleophilcity to spare, closing on the primary tosylate to give the desired ring, converging with literature synthesis to finish a formal synthesis.
Now, who hasn’t learnt something neat (put you hand down, Milkshake…)