Williams and Smith, ACIEE, 2007, EarlyView. DOI: 10.1002/anie.200705421.
It’s fair to say (in some respects at least) that this isn’t the freshest total synthesis report ever, but it’s certainly one of the most interesting, as a chapter of synthetic chemistry that has caused serious controversy has now been closed. This whole affair revolves around Woodward’s formal total synthesis of quinine, first published way back in 1944. His work with Doering led to an advanced intermediate, d-quinotoxine, which had featured in a synthesis of quinine performed by Rabe and Kindler way way back in 1918. However, some concern was voiced as the original 1918 paper was somewhat scant on detail. Indeed, there wasn’t a proper experiment (even by 1918 standards) – only a reference to the use of similar chemistry in analogous work.
Though this might seem to stitch these works together, one Gilbert Stork was still rather disgruntled with this work. (A full and interesting discussion of the history, politics and science of this affair can be read in Jeffrey Seeman’s awesome Angewandte) Somewhat interestingly, Stork originally voiced his opinion in 1944 by way of an unanswered letter to Woodward asking
“…Would you also tell me whether Rabe’s conversion of quinotoxine into quinine has been repeated by you in your recent work”
With no answer recieved, I guess Storks concern grew, along with his interest in the molecule. His own work on Quinine was complete in 2001 with a publication titled “The First Stereoselective Total Synthesis of Quinine” detailing Storks impressive stereoselective synthesis of our molecule. However, in that paper he once again suggests that the Rabe synthesis is flawed, and that there is no evidence that the chemistry works. With the Rabe work in question, the Woodward synthesis did indeed seem to be in doubt.
Recently, though, vindication was achieved for both Woodward and Rabe, as Williams has recently completed the Rabe chemistry “deploying the experimental protocols originally described by Rabe and his co-workers”. So, we finally get to some chemistry!
The Rabe chemistry begins with d-quinotoxine, available from an acidic degradation of quinine itself (thus turning this work into a sort-of-relay-synthesis). The SM was oxidised to the N-bromide, and the bromine displaced with the enolate of the ketone using a little base. This resulting (initially) in exclusive formation of quinidinon, which was found to isomerise to an equilibrium mixture of the desired quininone and quinidinone. Then to the crux of the problem – the reduction of the ketone to deliver quinine. Rabe states that they used “powdered aluminium” to perform this, so Williams did what any sensible 21st century chemist would do and cracked out the Aldrich pot. However, no product was isolated. To cut a long (and interesting) story short, they found that what they needed was Edwardian Al-powder, made from ground-up Sopwith Camels. With some aged Al-powder – quinine! Okay, the yield was rubbish, and the product was a mixture of diastereoisomers, but they did it. (I think the recrystalisations were somewhat traumatic, though – the group need to grow more facial hair if they want to improve their technique).
With the Woodward chemistry validated, I thought we’d look back at that 1944 chemistry and marvel at the skills involved. Please knock your synthetic brain into 1940′s mode though – no NMR, flash chromatography or TLC. Hard-core.
Woodward took a few steps about making our SM, starting with an elimination to give an exo-cyclic methylene, which was reduced. Then, time for a bit more hydrogen, taking that pyridine down entirely. The amine was protected, and then the remaining aryl group reduced again! Now we’ve reduced-away all the aromaticity, the alcohol was oxidised to form the corresponding cyclohexanone as a mixture of cis/trans-decalins.
Then, a bit of Woodward genius: a nitrite ester ring clevage (when was the last time you did that?). The mechanism is in Classics in Total Synthesis II. The oxime was then reduced and eliminated, and after a bit of protecting group juggling, the key Dieckmann condensation used to append the quinoline moiety, completing the formal synthesis. For sure!
A bit of ‘spect goes out to Jeffrey Seeman for the paper I mentioned above, and an amazing lecture he gave on this subject last year in Oxford. Also, I want to reaffirm my admiration for Gilbert Stork – good scientists are critical of the literature.