Home » Still In The RBF


17 August 2006 4,781 views 17 Comments


Stoltz, Tambar and Ebner, JACS, 2006, ASAP. DOI: 10.1021/ja0651815

Another short but sweet synthesis by Stoltz, this time of the isopavine alkaloid amurensinine (or at least the enantiomer of the natural product). The biological activity of this target is certainly appealing enough, as it seems to be interest with regard to Parkinson’s and Alzheimer’s disease. Their synthesis involves C-C and C-H bond insertions, the former with an aryne. Lets start with a bit of C-H activation:


Okay, I guess you’ve seen this style of thing in you undergrad course, but it’s nice to see a real-world example, and a hell of a yield! But there’s a thorny selectivity issue – why only one regioisomer, and why that one? No answer in the paper, even though the same question is posed. Perhaps it’s because the H is para to an OMe, or is it just steric effects… but I can’t actually believe that. On with the C-C activation:


Now that’s a funky transformation. Taking the product of the former reaction, and treating it with an aryne, they get addition and then rearrangement to get the to product in a none-too-shabby yield. With this substrate in hand, they did a diastereoselective reduction with L-Selectride, and an oxidative resolution to set up the asymmetry, requiring only a few more steps to get to the target.

1 Star2 Stars3 Stars4 Stars5 Stars (No Ratings Yet)
Loading ... Loading ...


  • Canadian Chromatographer says:

    To me, I would not jump too early to conclusions, the step that puzzles the authors of this blog probably has a mixture of so-called C-H activation and of Friedel-Crafts character. Thus, while the “C-H”‘s have the same theoretical proclivity to undergo C-H activation, or insertion, electron density in a Friedel-Crafts sense is still the same at both of them.
    In an F.C. acylation or arylation, you should get at least 8:1 selectivity in the worst cases, i.e. with AlCl3 at 80 ºC. Assuming this transformation goes at lower temperatures, my 2-cents is that this is merely a Friedel-Crafts on a metal carbenoid.

    Nice combination of two approaches, still. Kudos!

  • Canadian Chromatographer says:

    If I had proof-read myself, I would have added (!!) what I meant that according to a Friedel-Crafts process, i.e. the para / ortho ratio should be 8:1. So the C-H that got “activated” is the one with no ortho-methoxy steric interactions going on.

    But still, rationally speaking, you’d expect those methoxies to direct the Rh by complexation…

    One proxy, two proxies…
    One Stubby, two Stubbies (a brand of beer only available in parts of Canada!)

  • Hi says:

    Why can’t you “actually believe” sterics? It’s sterics.

  • Tot. Syn. says:

    I have a bit of difficulty with that, beacuse it’s not a massively hindered system. An ortho OMe isn’t a great deal of bulk, and whilst I could understand it’s influence were there a ratio of regioisomers, exclusive reaction on one site seems unlikely based solely on steric grounds.
    I think the Canuck has reasoned this rather well, and I’m happy with that argument.

  • aa says:

    The selectivity issue here can also be attributed to an electronic contribution (although I would say that sterics is the dominating determinant), if this does follow a Friedel-Crafts related mechanism. Relatively speaking, the para position of anisole is more electron-rich than the ortho because of the inductive electron-withdrawing effect of oxygen counteracting the electron-donation by resonance. Some evidence against a purely steric argument:

    In my own work on Friedel-Crafts acylations I have observed >99:1 selectivity for intramolecular acylations using 3,4-dimethoxyphenyl substrates. As well, the 3,4-methylenedioxyphenyl substrates, where the steric bulk of the substituents has been decreased substantially, gives the same selectivity. Food for thought…

  • Chemist of Sorts says:

    I don’t buy sterics as an argument. Methoxy is hardly a steric force to be reckoned with. This definitely looks like an electronic effect.

  • lucagh says:

    very nice molecule for synthesis. This is 1955, right?

    wonder when slotz will make diffcult molecule?

  • Ochemist says:

    This paper should have been published in Tet.Lett.

  • guigui le chimiste says:

    Stolz is the only one that made lemonomycin which is a really difficult target. A lot of different groups have worked on Lemonomycin for years and Stolz is the only one that made it.

  • secret milkshake says:

    it is better to do a clever synthesis of a small natural molecule that has a reasonable chance of being used in pharma, rather than taking on a monster that takes 4 grad students 4 postdoc and 4 years to finish. Also, people have to graduate, write up and get a job. It is easier to do it if you have your own molecule, as opposed to some “C-D ring portion” of some marine monster.

  • European Chemist says:

    Hi everyone

    Can I drop my two cents in it? It seems that rhodium carbenoids are actually rather sensitive to steric effects, and if I can remember it correctly, FC (or any other electrophilic aromatic substitution) reactions always run away from ortho substituents if given a chance. Specially in this case, where the electronic density of both positions by mere resonance considerations appears to be the same.

    But I really wanted to comment the 2nd step. It’s really reminding of the DeMayo reaction! But doing it with an aryne is quite nice.

    Personal appreciation of the chemistry: it’s a clever synthesis. Nice chemistry. But the kinetic resolution (it IS a standard kinetic resolution, half the material goes to scrap!!) and the relatively high number of steps needed to elaborate to the final product (because of that cursed racemisation problem) sort of tarnish the final result… and I think we’ve seen similar material make it to Org Lett only…

    Final thoughts: I think that doing short and controlled syntheses of “small” molecules can be as challenging as assembling a monster beast of 1500 Daltons. First of all, you never quite predict how things are going to evolve if you have compact and densely functionalised compounds; second, since the molecule is “small” you actually have to be as creative as possible ’cause people won’t forgive you going over the 10-15 step-barrier or lower than 50% yields for tricky steps. Just look at the case in consideration.

    Oh, BTW: Congratulations for setting up this great blog!

  • Canadian Chromatographer says:

    As an answer to European Chemist,

    You could not be more on the point as for “small” molecule total synthesis. Look at Nicolaou’s endless efforts on azadirachtin! Will they ever finish it? It seems that those 5-6-7 oxygens in line keep ping-ponging with each other.

    And I thought than the golden rule stated that one wants to conduct a kinetic resolution in the first 2-3 steps of a total synthesis! Not right in the middle…

    At least, that’s the Canadian talk on the streets! :)

  • John H says:

    I’d definitely cast a vote in the direction of electronic activation; that’s certainly what I’ve read in the past on the topic (not that I’m any expert!), and I’d question the steric argument. Not impossible, but I’d have thought that electronics were the main deciding factor here.

    If I recall correctly the rhodium carbenoid mechanism hasn’t been proved actually yet; the role of the metal hasn’t been confirmed. (LA loss of N2 is the other possibility argued… I think I prefer carbenoid myself, but who’s to say?)

    The cycloaddition step has been out there before; is quite similar to this, I believe:

    Caubere, JOC 1980, 45, 240 – 246, which remains one of my favourite reactions ever.

  • gaussling says:

    I’ve done some molecular modeling (CAChE) of rhodium carbenoid complexes in C-H insertion reactions. The transition states in insertion reactions are fairly congested, so it isn’t too much of a stretch to imagine why the reaction goes as it does. The C-H insertion begins while the carbenoid lone pair is still coordinated to the Rh. Obviously, the rxn product comes from a transition state with an ortho H and the unseen product would come from a transition state bearing an ortho methoxy. The rhodium acetate dimer is quite large. Even though the methoxy doesn’t seem too sterically challenging, the delta delta G between the respective transition states obviously drives the ortho H pathway. I have not modeled this system, so this is just hand-waving. When you imagine mechanisms of carbene insertions via rhodium acetate, you have to imagine the insertion happening up against the “acetate wall” of the catalyst. These things do not proceed as the free carbene.

  • Chinese chemist says:

    I would like to comment about the alkyne-ketoester reaction step. It’s obviously an application of Samuel Danishefsky work that published in the middle or late 1990s. Unfortunatley, no any citation of Samuel Danishefsky work is given.

  • Chinese chemist says:

    should be benzyne-ketoester step.

  • [...] using enantioselective oxidation (resolution).  It’s great, but I covered it back in 2006 (synthesis of amurensinin).  Then I went through Amos Smith’s lyconadin A and B, but we looked at that last year.  [...]