Home » Still In The RBF

Cyanthiwigin F   

9 July 2008 10,506 views 22 Comments

Stoltz and Enquist. Nature, 2008, 453, 1228. DOI: 10.1038/nature07046. Article PDF Supporting Information Group Website ResearchBlogging.org

We’re going high-concept again, here… but before it all gets complex, let’s examine the why before the how.  The cyanthiwigins are pretty diverse biological agents, with antimicrobial and antineoplastic activities, along with nerve growth factor stimulation; this particular beastie has some moderate cytoxicity versus human primary tumour cells. However, there’s another 29 with their own particular efficacies.  Indeed, as Stoltz mentions, it’s surprising that they’ve only been the product of two syntheses (Andrew Phillips and Jagadeeswar Reddy).

Key to Stoltz’s approach is a recognition of an element of centro-symmetry in the molecule, with the cyclohexanone moiety at it’s core.  Focus on the quaternary stereocentres, and the position of the double bond, and one can recognise that there’s something to work with there.  Stoltz went one-further, and realised that both stereocenters could be imparted using the same reagent-controlled asymmetric alkylation.  First, they had to make the substrate:

A bit of Dieckmann condensation action dimerised the 1,4-diallyl esters to give the required 1,4-cyclohexane-dione functionality very quickly – a nice use of an old-school reaction (and a nice reminder of Mini Dickmanns… mmm…).  This was then bis-methylated in the thermodynamic manner to give the substrate for the Pd-wizardry – in the form of a racemic mixture.  I must stress that at this point we have both (R,R), (S,S) and the meso compound as a statitisical mixture.

Now lets take this slowly.  I found the diagram in the original paper over-complicated, so I’ve simplified for the chemically aware (god-damned biologists…).  The first step is an unselective deallylative decaboxyaltion to lose one stereocenter.  This gives an enolate, and a remaining stereocenter, which may be either S or R, so we’ve got a racemic mixture of enantiomers.  Next up, the enolate center is alkylated using the enantiomerically enriched Pd catalyst to give only (mostly) the R configuration at the center. So now we’ve got a mixture of diastereoisomers, R,R and R,S.  However, the remaining allyl ester is then decarboxylated to give the respective enolate – but now we only have one stereocenter, and it is exclusively (mostly) R.  Lastly, the freshly installed enolate is alkylated, again attaining R configuration to give a predominance of the desired (R,R) product.  Phew…  Bloody awesome stuff, too…

Now, to complete the target.  A desymmeterisation established a triflate from one of the carbonyl groups, and allowed a Negishi coupling with an unsaturated partner.  RCM completed the medium ring, at the same time (quite impressively) hydroborating and oxidising the remaining terminal olefin to an aldehyde.  (I would have added this scheme too, but that would’ve been the whole paper…).  The aldehyde was then cyclised using a bit or radical chemistry, completing the final ring.

Lastly, Stoltz did a further triflation and a Pd-catalysed coupling reaction with an i-Pr-organocuprate.  I can see that this is very similar to Negishi chemistry, but they describe it as being ‘difficult’, and there’s no reference cited… interesting stuff – just wish I knew more about what was going on.

Fantastic synthesis – and I’ve not even mentioned the lack of protecting groups until now!

Enquist Jr, J.A., Stoltz, B.M. (2008). The total synthesis of (-)-cyanthiwigin F by means of double catalytic enantioselective alkylation. Nature, 453(7199), 1228-1231. DOI: 10.1038/nature07046

1 Star2 Stars3 Stars4 Stars5 Stars (2 votes, average: 4.50 out of 5)


  • InfMP says:

    i could have sworn the 51% was over 2 steps (methylation) in the first reaction??

  • Joe says:

    There’s an oxygen missing in the 1,4-diallyl ester.

  • Tok says:

    Great synthesis, I love to see tot. synth. in Nature.
    It’s interesting they chose to report the last step as 63% of a 1.8:1 mixture rather than saying 40% yield of the desired product with 23% reduction product. Maybe they couldn’t separate the two?

  • MCC says:

    I really like the double enantioselective allylation in this paper. Surly a tot. syn. worthy of Nature.

  • mr tin says:

    im wondering… can the dieckmann dimerization form the cyclopentane-dione also? could that be a side pdt?

  • petr says:

    I can not see any plane of symmetry in that meso compound :(

  • TheEdge says:

    Imagine an axis through the center of the ring, perpendicular to the plane of the molecule. If you rotate around it, the “R,S” become “S,R”.

  • Tot. Syn. says:

    @Joe – fixed, thanks!

  • tma says:

    the Pd-catalysed coupling of i-Pr cuprate is presumably difficult due to steric issues as they are kicked around by considerable reduction product formation (1.8:1). For an excellent coverage of organocuprates (and “everything” organometallic of synthetic worth) steal Manfred Schlosser’s Organometallics in Synthesis: A Manual 2002 from your library today. I gave it so much lovin’ that the library wrote it off as lost from catalogue. It covers organo-cuprates in excellent breadth and depth – including all the weaselly additives to improve yields. And it weights only 2.1kg !

  • cvengo says:

    I just wonder whether it is normal for Nature, that the paper has been submitted in November or so and it came out only now.

  • Jim says:

    @TheEdge – dosen’t that make it C2-symmetric rather than meso?

    Or am I stoopid?

  • HPCC says:

    tma: During my Ph. D., I signed out this book in June, 2003, and only returned it on my last day… November 2006!!!!!!!! I renewed it every four months. I swear by that bible only, when it comes to organometallics. I hope they’ll come up with an updated edition in 1-3 years, including more refinements, more references…

  • Potstirrer says:

    The meso compound is achiral not because it has a simple plane of symmetry. It has a center of inversion (i.e. it has Ci or S2 symmetry). Follow the Edge’s suggestion to rotate 180 degrees around an axis perpendicular to the ring but then reflect the resulting molecule through a plane perpendicular to the rotation axis. You will then end up with the same compound. Alternatively you can simply reflect every atom through a point at the middle of the ring. The R,R, and S,S molecules are chiral and have C2 symmetry.

  • TheEdge says:

    The R,R and the S,S are C2 symmetric around an axis perpendicular to the plane of the molecule. Rotating around that axis 180 deg gives you the same compound back.
    What I was trying to say was that the R,S and S,R compounds are identical, and can be shown to be so by rotating around an axis, so they must be meso. I think that’s faster than trying to find the plane of symmetry or point of inversion, but maybe that’s just me.
    Potstirrer’s analysis is the actual right way to go about these things.
    This is a really beautiful route.

  • milkshake says:

    Also, the reason they have 99% ee is that they got two asym steps in tandem, the enantioselectivity of each step is probably somewhere in low 90s, and all the wrong products end up as meso. Very nice.

  • TWYI says:

    Agreed milkshake.

    The traces contaminating the double asymmetric alkylation get removed as meso.

    Stoltz sells it as a double AA (and rightly so) but I think it is more than that.

  • Jim says:

    TheEdge + Poststirrer: Thanks for clearing that up. You see, this is why mere mortals such as myself could not design routes like this. Even when spelled out it is astounding.

    An incredible bit of work. Congratulations to all involved.

  • SiO2lungs says:

    Very nice work (and good simplification of the diagram too!)
    But I have to say I was surprised to know they used 1.5 equiv of AIBN for the radical cyclization :P… I guess the didn’t want to add Se or ann additional step to the menu.

  • whistling Kitty Chaser says:

    A beautiful synthesis indeed

  • Kennedy says:

    The compound is a meso, but it doesn’t have a plane of symmetry. It has inversion symmetry (also called a center of symmetry), which makes it optically inactive. It is the other (less common) qualifier for a meso compound.

  • Kennedy says:

    ha,sorry….didn’t see the other explanations already posted.

  • InfMP says:


    Stoltz full article.