Home » Still In The RBF

Gonyautoxin 3   

31 August 2008 19,044 views 49 Comments

Du Bois and Mulcahy. JACS, 2008, ASAP. DOI: 10.1021/ja805651g. Article PDF Supporting Information Group Website

One quick inspection, and you know we’re looking at a dinoflagellate neurotoxin, with more guanidine groups and derivatives than you can shake a stick at.  Unlike other large families of natural products with common functionalities, there doesn’t seem to be a universal strategy that (whilst ugly) could synthesise the majority of the family.  This means we get to see a whole variety of strategies towards these targets (and ultimately, the daddy, palau’amine).

Du Bois’ approach starts with the use of L-serine as the starting material, and source of asymmetry.  Smart choice, as in only four steps they had dihydropyrrolo[1,2]pyrimidin-1(2H)-one product shown below.  It was the cyclisation step (as ever) that caught my eye, treating the aldehyde with allylamine to get the imine, and then Lewis acid mediated cyclisation.  Great result; the yield is reasonable, but the d.r. is excellent.  I thought I’d read a bit about the procedure, but I was disappointed to see that the SI is extremely sparse – there’s plenty of data, but no experimental procedures.  Grrr…!

A few steps further into the synthesis comes another sweet cyclisation; this time using a rhodium catalyst to effect formation of the final 5-membered ring.  Again, the yield was cracking, and the d.r. apparently complete.  Mechanistic insights are somewhat less clear (as this is still a topic of some debate).  Du Bois suggests initial formation of an aziridine, and then opening of the allylic aziridine with acetic acid to give the product.  He then reduced out the O,N-acetal (which was causing them serious problems) using an interesting reagent combination: Et3SiH and BF3·OEt2.

With the ring-system complete, they attempted a ketohydroxylation, but as in their synthesis of saxitoxin, this proved problematic.  Instead, they did a standard Upjohn Oxidation, protected the less hindered hydroxyl and then oxidised the remaining.  The completion of the synthesis was quite sweet; deprotection of all protecting groups with a bit of hydrogen, acetic acid and then some ammonia resulted in a strange beast – the α hydroxy ketone hydrate.  Then, addition of DMF·SO3 resulted in selective sulfation of the C11 alcohol, and the target.  Nice!

1 Star2 Stars3 Stars4 Stars5 Stars (5 votes, average: 3.60 out of 5)


  • optional says:

    Nice work again by Dubios. I wonder if PA Evans was working on this target??

  • GYA says:

    dubois has done some impressive work in the guanidine alkaloid area

  • shin says:

    The key step is a another extension of carbene-type chemistry deveoped by Dubois: nitrenoid add to the double bond of pyrrole to form aziridine and ring-fragmentation (quite reasonable). I wonder what would be the result of simple oxidation using mCPBA or similar reagent such as halonium species???

  • PotStirrer says:

    Can anyone give me a good reason why a JACS total synthesis can come without full experimental details these days? I expect more from DuBois, whose work I highly respect, and our field in general.

  • anniechem says:

    PotStirrer….wow, that’s pretty shocking! I just took a look, there’s only characterization data in the SI.

  • antiaromatic says:

    I was pretty surprised by this myself. I was under the impression that characterization wasn’t sufficient. Maybe they’re hoping to put out a full paper with the full experimentals?

  • % says:

    nice blog, but the title structure is incorrect (no Tces in (+)gtx 3)

  • br says:

    very nice the first cyclisation, the second in quite blurred for me. about the blog, good job! I really like the choice of the syntheses

  • Tot. Syn. says:

    Wooo… That’s a piss-poor mistake even by my shoddy standards. I’ll rectify it asap, but I’m off to the BBC tonight to see some TV getting recorded… Just a little excited!

  • PotStirrer says:

    Putting out a full paper a year or two afterwards and only then including full experimentals should be a thing of the past. Full papers are great as they often give the authors a chance to describe their journey in more detail, and give readers a chance to really learn how syntheses evolve; however, there’s just no excuse anymore for putting off the experimental disclosure. I would say most of us often delve quite deeply into those details to discern how effective the synthesis and upfront the reporting of yields are. Knowing that my peers will hopefully :) be analyzing my work with that level of scrutiny is always in my mind when I am working in the lab and this is without a doubt a positive development for our field.

  • optional says:

    I agree. The guidelines for publication does indeed state that any-body should be able to reproduce the work. How can one look at the characterization data and set up the experiment in the “hood”

  • optional says:

    One needs the stoichiometry, conc’s, time, temp, rate and order of addition of reaction components, the monitoring of the reaction, etc. Nothing against this beautiful synthesis, but no one should be given favors in scientific world.

  • milkshake says:

    BF3.Et2O is a very nice with Et3SIH, especially if you add some TFA, because the BF3 complex of TFA is much more powerful acid than TFA alone and the sideproduct is TES-F which is sufficiently volatile to be pulled off on rotavap and migrates in fron on silica. (TES-OH could be pain to remove completely, especially if the Rf is similar to the product.)

  • Potstirrer says:

    Milkshake: I don’t understand how you would make TES-F with TFA as an additive. I assume the fluoride is supposed to come from the TFA. So, what is the fate of the TFA? Surely not HOOCCF2OH. BTW, one time my only chance to reduce a very similar looking O,O acetal was with DIBAL. Everything else I tried including BF3.OEt2/Et3SiH resulted only in elimination and subsequent aromatization.

  • milkshake says:

    The F(-) comes from BF3.Et2O, of course. As an example of Et3SiH+TFA+BF3.Et2O, you can look up Org Prep Daily November 15, 2006.

  • Potstirrer says:

    Allright, I thought you were saying that you get TES-F only when you add TFA.

  • pi* says:

    These guys know how to cyclize

    “I was pretty surprised by this myself. I was under the impression that characterization wasn’t sufficient. Maybe they’re hoping to put out a full paper with the full experimentals?”

    antiaromatic and potstirrer, this is the standard procedure… Right or wrong, when competing for scholarships, grants tenure etc, you’re getting compared (usually by #papers/impact factor) to fields far less work intensive than tot syn…usually by people who dont know a thing about chemistry

  • milkshake says:

    On reading the supporting info, I have two small objections: There are spectra but no experimentals in supporting info, the yields are in the scheme but one cannot see if these are fairly typical results or “best yields”. Other great (Nobel-awarded) chemists got pretty notorious for doing the same thing but it would be nice to provide full procedure for the key steps even in a communication.
    Also, I noticed they carry unprotected carbamoyl through a number of steps, which is inconvenient as CH2OCONH2 group is pretty polar so they had to purify their compounds in straight EtOAc, etc. They put carbamoyl on using CCl3CONCO, then solvolysed trichloroacetyl with methanol. In this case I would recommend using trityl or methoxytrityl isocyanate (it is made by refluxing trityl chloride and NaOCN in anh acetonitrile) – the trityl protected carbamates and ureas are nice crystalline solids that behave well on silica and trityl could be easily cleaved off by acid or hydrogenated away on Pd-C(they have hydrogenation deprotection later on)

  • Avi says:

    Interesting work, but does anyone know where Du Bois is going?

  • ... says:

    The strange thing about the Supporting Information for me is that what got left out is the easy stuff. Mr. Mulcahy has completed a beast of a molecule. Looking at the spectral data in the SI, which is pristine, it’s clear that this guy has a high bar for quality. I reckon that Mr. Mulcahy knows how it feels to be in the lab well after midnight and I also reckon that translating the procedures in his lab notebook to a Word document could be achieved on one pot of coffee. So, pi*, I don’t imagine the submission of this paper was hastened by leaving out procedural data – writing that up is child’s play compared to what is presented in the paper. It’s a shame that the information didn’t get presented and I can only hope that the field doesn’t follow this example.

  • Tot. Syn. says:

    Typo fixed. Thanks for that!

  • Nagu says:

    Indeed nice piece of chemistry

  • ... says:

    The lack of experimental details appears to be a du Bois lab policy on total synthesis communications. If you look at his JACS communications for the initial disclosure of tetrodotoxin, manzacidin and saxitoxin; there’s enough spectral data to prove that the molecules were made but no recipes. I guess they are working hard to keep a leading edge over synthetic competition. This might make sense on a 6 month time scale but it’s going to be pretty frustrating for the grad student who could benefit from this information in the year 2038.

  • PJ says:

    The grad student in 2038 will be able to look at the full paper that gets put in 12-18 months time.

    Don’t see the problem here – the spectra are 100% proof. No need for the experimental, unless you want to repeat recently published work? If so, a quick e-mail is often enough to get the experimental for the step you are interested in

  • antiaromatic says:


    See, I think that’s a fundamental problem with the process. Even though the “understanding” is that a communication should be followed up with a full paper, that most certainly does not always (and in my experience, usually) happen, especially with established groups. It’s those cases that make it a real headache 50 years from now when the professor/students are no longer available to answer those questions.

  • bad wolf says:

    PJ: You don’t see the problem here? I see nothing but problems with this approach.

    Really, why even do it if you’ve got nothing to share with the world.

  • PJ says:

    Bad wolf, this paper has gone through JACS referees.

    He shared his spectra that proved beyond doubt he made the NP.

  • Tot. Syn. says:

    PJ, you’re missing the point or sidestepping the issue here. The fact that he made the natural product is beyond debate – it’s clear and obvious. But what’s the point in doing peer-reviewed, non-commercial research if you don’t explain what it was you did? How can the field of organic chemistry progress if researchers don’t share experimental details? I’m sure Du Bois will publish details in a full paper to come, but there’s no reason why that data shouldn’t be in the SI, especially in JACS.

    Consider the all-to-frequent SciFinder search that turns up exactly what you want. The yield looks great, and the substrate is a doppelgänger for the stuff in your RBF. You click through to get a prep and then the train comes off the tracks ’cause it’s a sodding 1980s Tet Lett, and therefore has no bastard experimental. Possibly the most annoying moment in any literature search (including turning-up a 200 page patent with no text-search facility).

    Progress in science comes from communication, sharing data, and mixtures of stimulants and depressants.

  • anniechem says:

    “Progress in science comes from communication, sharing data, and mixtures of stimulants and depressants.”

    Tot. Syn., you crack me up! So true :)

  • optional says:

    Tot. Syn… I agree with you 100%. PJ, not so much…probably, not at all.
    Lets say, If you (PJ) had to use a simlar method to make guanidines, would you care for the NP’s spectra or the experimental???

  • MAB1 says:

    You can get the experimental by e-mail easily. It takes ten seconds.

    Du Bois has no case to answer here.

  • Herr Lipp says:

    OK, MAB1, prove how easy it is.

    Send the e-mail, get some experimental procedures and post them. If you don’t get the experimental, post the email exchange.

  • milkshake says:

    stimulants and depressants are best used separatedly – I had a terminal meth junkie for a neighbour once, and he was also drinking heavily to take the “edge” off – the jitterness and clenched jaw I suppose. Unforgettable experience – he was very loud, violent and angry most of time, and paranoid too. He believed that somehow I was responsible for his trouble with police.

    Smokes and drinks pair up nicely though…

  • antiaromatic says:


    No one is questioning that Du Bois is willing to share experimental. The point is you shouldn’t have to bug the author for experimentals when the work has, you know, already been published. And I agree with “…”: it doesn’t take a month to write up an experimental.

  • InfMP says:

    people rarely bother the prof for his experimental. and years later, the email is no longer valid. There are a million times people say “research is ongoing” and never say another word. and full papers only seldom happen too/.

    Last time i saw this was Evans’ azaspiracid (spectral pictures ONLY). I thought he got away with it because it’s Angew, but clearly you can pull the same moves in JACS now.

  • Jose says:

    I have emailed PIs for experimental details on probably 10 occasions. Once, and only once, did I get a response. The number of communications that are followed up with a full paper is also in the 10% range. It is unacceptable to publish a JACS paper like that; writing up experimentals is a pain, but come on!

  • ZZZZZ says:


    …I agree with Jose…


  • Nothing says:

    Its nice work and wondering how come referees accepted this paper with incomplete supporting information that to in total synthesis.

  • ,,, says:

    To MAB1
    “It takes ten seconds” – I doubt that… Do you know how many emails PIs get every day? And you think yours about supp will be answered shortly?.. What about a couple of years later?

  • Avi says:

    Anyone know if DuBois is leaving Stanford?

  • % says:

    Word on the street is that DuBois is on the top of Harvard’s recruiting list.

    And Harvard chemistry is hurting pretty badly at the moment, so I wouldn’t be surprised…

  • thezorro says:


    Give some time to Ritter, man, and you will see a new Kentucky colonel on his prime

  • Avi says:

    % you gotta agree that Harvard Chem is hurting though

  • cunningsynthesist says:

    On what basis do you say that the Harvard Chemistry department is hurting? (Jacobsen, Myers, Evans, Shair, Corey, Kishi, are all alive and well)

  • % says:

    Well, they certainly have some highly influential chemists on board now, but their 10-year plan is what I was referring to in-particular (evans, kishi and corey likely will have the sashes closed by then, possibly sooner). Jacobsen is still tearing it up like usual, but Myers and Shair have slowed down a good bit in the last few years.

    But above all, I know they’re having a tough time recruiting the “elite” students. They need someone like DuBois (yes, Ritter is very good for their department) if they are to return to their synthesis powerhouse status of the 90’s…

  • snack says:

    the obvious choice for harvard is baran.

  • cunningsynthesist says:

    Yeah%, that makes sense in the context of a 10 year plan. I think the Harvard chemistry department has lost its direction somewhat, with an increasing emphasis on chemical biology despite its synthetic roots. I can’t imagine what would attract Baran to Harvard at the moment. However, it seems that the Shair lab is picking up its pace as far as number of group members, and recent publications.

  • European Chemist says:

    Shair is stepping it up, but Myers is still slowing down. And Ritter, my friends, has been there since 2006 now, has a group of over 10 co-workers and only puts out 2 papers a year? On stoichiometric fluorinations of aryl stannanes? Is he going to be the new Barry Trost? I’d be very surprised.