Paterson, Findlay and Noti. Chem. Asian J., 2009, ASAP. DOI: 10.1002/asia.200800445.
It’s been a rather slow week in the world of total synthesis – I normally have a check on the ASAPs and EarlyViews in JACS and Angewandte every day in Google Reader (BTW, tune directly into my brainwaves under the ‘What I’m Reading’ section on the right…), and a bit less often, I’ll read Org. Lett., Chem. Comm. and OBC. If nothing turns up, then it’s on to Chem. Eur. J, JOC, and maybe even Tetrahedron… but today I took a far eastern excursion and found this distinctly Scottish paper (66% at least…). The thoughts going through my mind at this point were firstly, 1. I recognise that structure, 2. Ah, it’s Ian… 3. …and Alison – a former housemate from when I was living in Cambridge. It wasn’t just a house we shared, though – we also shared group meetings, and I remember this strucuture coming up, as Alison had just finished Dolastatin (that’s some old-school TotSyn right there…). I actually found the pentaene moiety the more interesting chunk, but of couse Ian Paterson’s all about the aldol.
I can’t actually remember if I (or the team I was working with in our synthesis-problems group) recognised the repeating stereotetrad fragment, but I’m sure it came up in the discussion. Pulling this out of the retrosynthesis after dismembering the spiroketal leaves a far simpler problem, which I could write aldol all-over if this was a full retro. The key, then, is building the pieces quickly and efficiently, and getting the coupling in the right order. The key repeated unit was build very quickly, starting with an enatiomerically pure ketone, and doing a 1,4-syn-aldol using dicyclohexylboron chloride and methacrolein. This is a methodology that Paterson has been using for decades, and it shows, with a cracking yield and control of stereochemistry. All it needed was a diastereoselective reduction of the ketone to complete the unit, and they had two choices – Evans-Saksena (paper with Erick Carreira) or Evans-Tischenko (with Amir Hoveyda). The former was one step quicker, as the latter required saponfication of the resulting ester, but the two-stepper was not only more efficient, but apparently easier to scale up.
A few transformations allowed development of this fragement into the two components that were to be coupled. One required a further stereodefined hydroxyl, brought about by a Myers’ alkylation, whilst the other used a cuperate addition to append the unsaturated sidechain. Forming the boron enolate of the ketone first, and then addition of the aldehyde allowed the coupling to complete, generating a reasonable yield and diastereomeric excess. Various attempts at improving this yield by altering the protecting groups on the partners were performed, but nothing was gained – but that’s not to say that this wasn’t a tough reaction, and a good result.
The mixture of products was then methylated at that troublesome C-23 postion, setting up the molecule for spiro-cyclisation, brought on by removal of the acetonide protecting groups. The result, as expected, is the doubly anomerically-stabilised spiro-ketal, with only the thorny C-23 position providing issues. Happily, at this point the unwanted diastereomer could be removed, leaving the group ready to progress.
Treatment of this product with 9-BBN to effect a selective hydroboration of the less-substituted olfin, providing a synthetic handle for elaboration. Oxidation and Stork-Wittig olefination gave them a vinyl-iodide, ready to build the pentaene system. As shown in the retro, they were able to make a symmetrical bis-vinyl-stannane, and simply had to choose which way to couple the smaller and larger chunks. Ready for instability, they used ‘strict exclusion of light and employing base-washed amberised glassware‘, and gave both alternatives a go, with the favoured route being the small then large Stille’s if you get what I mean. Nice work, folks, and congrats to Alison for getting so much done in such a short timeframe.