Week 8 Solution Mega Chemist Challenge #MegaCC

Sorry it is late………..

Well done this week to @AzaPrins who sent in the only correct answer.

Mega Chemist 8 was Ed Anderson who currently holds the position of lecturer at Jesus College, Oxford (UK).  Just to elaborate on the clues a little; on Anderson’s return to the UK from Sorenson’s group in 2003 he took up a Junior Research Fellowship at Homerton College, Cambridge (UK) working in the lab of Ian Paterson – beat that for a C.V. (don’t forget his DPhil with Prof. Andrew Holmes). In 2007 he moved to Oxford on an Advanced Research Fellowship, and became a lecturer at Jesus College in 2009.

Anderson made the Mega Chemist Challenge because I saw him speak a couple of years ago, and I loved his talk.  He was young, enthusiastic and inspiring. He also took a roasting from Professer Jim Thomas which he handled with great aplomb – this I regard as a great achievement because Prof. Thomas turns me into a gibbering wreck as soon as he looks at me. Although I have known this to be true for a long time, my second year viva really consolidated this for me!

Back to Anderson – His talk was amazing, great total synthesis (and methodology being developed out of it all over the place), so I was really excited about going back and devouring his latest conquest. Unfortunately though, there was nothing forthcoming, Anderson’s recent publications have all been methodolgy:

One-Step Preparation of Functionalized (E)-Vinylsilanes from Aldehydes
Org. Lett. 201113 (18), 4806.

Palladium-Catalyzed Asymmetric Synthesis of 2-Alkynyl Oxacycles
Angew. Chem. Int. Ed. 201150, 11506.

Palladium-Catalyzed Cascade Cyclization of Ynamides to Azabicycles
Chem. Eur. J. 201117, 14366-14370.

That is of course if you do not count the 2nd generation approach to Spirastrellolide A Methyl Ester, just out of the Paterson lab. Anderson’s name on this paper, I am sure is a result of his phenomenal lead work on the first generation synthesis from his time in Paterson’s lab rather than a recent contribution, so I am afraid in my mind this doesn’t count (also macrolides just don’t do it for me). Here is Totally Synthetic’s article on this work.

Instead I have gone digging into the past a little and pulled out the first (and still only) total synthesis of the furanosteroidal antibiotic viridin that Anderson completed in the USA. I had a browse through a few papers of Anderson’s, but this had not only an elegant alkyne cyclotrimerisation, but a “tandem conrotatory electrocyclic ring-opening 6pi-disrotatory electrocyclisation” – and you can not really turn one of those down!

Viridin is a potent anti-fungal that covalently inhibits protein function. Inhibition is reportedly through protein amine residues forming covalent bonds with viridin through reaction at the doubly activated carbon of the unusual electron deficient furan ring fused in the pentacyclic core. Anderson’s/Sorenson’s approach (scheme 1) took into account the sensitivity of this functionality, and as such intended to undertake the oxidative functionalisation of the A ring at a late stage. They intended to complete the pentacyclic core from napthelenofuran 2 which in turn would be generated through the key tandem electrocyclic ring opening-ring closing of a derivative of 3. The furan group was to be introduced through furanyl lithium addition to 4 which in turn would be generated from poly-alkyne in a rhodium catalysed cyclotrimerisation.

Skipping an unexciting beginning we arrive directly at cyclotrimerisation substrate 6 as an inconsequential mixture of diastereoisomers. Treatment of the triyne with [RhCl(PPh)3] (3 mol%) in ethanol gave 10 in an excellent 88% yield and in only 20 minutes (presumably excluding the column). Having generated the aromatic C ring, the cyclopentane moiety D and in the cyclobutanol a handle for generating the B ring in an impressive single step, Anderson et al introduced the furan ring through nucleophillic addition of 9 with complete anti-addition with respect to the vicinal methyl group, giving 8, and thus setting up the tandem electrocyclic ring opening-ring closing (scheme 2).

Taking us through the key reaction stepwise (scheme 3), we start with the conrotatory electrocyclic ring opening of 8 to give intermediate 10. The silyl group reportedly confers a high degree or torquoselectivity, ensuring that the furan rotates inwards setting up the 6pi electrocyclisation. Cyclisation proceeds to give 11 and subsequent in situ DDQ oxidation gives aromatised tetracycle 12 in impressive overall yield.

Think about how impressive this is – we have gone from the acyclic triyne to tetracycle 12 in only 5 steps in a 51% overall yield. Brilliant.

The final ring of viridin was installed in only three more steps. A one pot selective desilylation-allylation gave 13 which on heating in mesitylene installed the quaternary center of 14, and set up the ring closing metathesis needed to finish the core of viridin, giving 15 (scheme 4).

From 15 the end game is not that thrilling, though it does not detract from how much I like and am impressed by the synthesis up to here. This incredibly rapid construction of this pentacyclic steroidal core in only eight steps form 5 is wonderful.

To finish up from 15 (scheme 5), a SeO2 mediated allylic oxidation followed by a DMP oxidation-NaBH4 reduction installed the allylic hydroxyl with the correct stereochemistry, giving 16. This allowed for a ‘Donohoe’ hydroxy-directed dihydroxylation giving 17, with all the stereocenters installed.  Protecting group manipulaion and oxidation took it the rest of the way through to viridin in an overall yield of 5.0% from pent-4-yn-1-ol. Fantastic.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s