Home » Still In The RBF

Octalactin A   

20 April 2008 9,042 views 29 Comments

Toste, Radosevich, Chan and Shih, ACIEE, 2008, EarlyView. DOI: 10.1002/anie.200800554. Article PDF Supporting Information Group Website ResearchBlogging.org

Another paper from the Toste group that doesn’t use recycled Rolex watches in any of the steps. Indeed, the metal of choice today is vandadium – read-up on the methodology in this tasty JACS from 2005. Anyway, the target is octalactin A, a medium ring lactone with a bit of murine melanoma and colon tumor busting activity. Previous syntheses by Buszek, Clardy, Shiina and my boss with Andy Holmes, so it’s certainly seen a bit of action…

I’ve not bothered with a full-retrosynthesis, cause there’s a lot going on in a tight space, but much of the action revolves around the synthesis of the olefin shown below. Toste noticed that this bad-boy contains two units of similar relative stereochemistry and connectivity, and that they might be produced together as the two products formed in a kinetic resolution of a racemic mixture.

Said racemic compound was produced quickly using a rather neat Darzens condensation, rearrangement and oxidative hydroboration protocol. Then, treatment with a bit of vanadium catalyst along with some ligand (no actual figure of the ligand in the paper, so I’m going on ChemDraw’s ‘Name To Structure’ function) resulting in an excellent mass recovery of two products, both in good enantiomeric excess.

Transformation of these products into the desired partners for the Horner-Wadsworth-Emmons reaction was efficient and quick – with one particularly neat reaction. The α-keto ester required decarboxylation which they performed using an intreguing oxidative process, treating the analogous α-keto carboxylic acid with iodosobenzene. The procedure was developed by one Robert Moriarty (good name!), and follows this mechanism:

Transformation of the product into a mixed anhydride, and addition of a lithio-phosphonate gave the desired ketophosphonate, and allowed a high-yielding coupling of the two fragments.

Completion of the medium ring invoked an under-used protocol for formation of lactones; a Wolff rearrangement. Of course, there are two main methods for the expulsion of nitrogen – silver salts and light. In this case, they tried both, and found that the photo-Wolff was the more efficient process. Worth noting (and absent from the prose) was the high dilution required for the reaction. Presumably this is to prevent the competing intermolecular reaction.

Synthesis of the sidechain used the neat Kocienski procedure for the opening of DHPs via a 1,2- metalate rearrangement (discussed more fully in this previous post). Coupling of the two fragments using an NHK coupling completed the carbon skeleton, leaving them with only a few more steps to finish the molecule. Smart stuff…

Radosevich, A.T., Chan, V.S., Shih, H., Toste, F.D. (2008). Synthesis of (−)-Octalactin A by a Strategic Vanadium-Catalyzed Oxidative Kinetic Resolution. Angewandte Chemie International Edition DOI: 10.1002/anie.200800554

InChiKey Dump:

1 Star2 Stars3 Stars4 Stars5 Stars (No Ratings Yet)


  • Howard says:

    Anyone else find it weird that Toste numbered the ligand in the paper, but never drew it? I honestly thought I was just missing it until I read this post and saw it wasn’t just me.

    I also have to say that my first instinct upon reading this was that it doesn’t really teach the world much about chemistry. Sure, it is a nice application of a nice methodology, but how many targets happen to have stereomotifs that can be accessed from both products of a kinetic resolution, this KR specifically? It’s a neat idea, and I’m sure more chemists would use it if they could, but this is such a specific, “academic” or “small-market” application, I’m not sure it’s all that exciting or important. Also, I think whatever elegance Toste was going for in not having to throw away half his material early was kind of detracted from by having to throw away half of it late – the NHK coupling in the end-game gives a dr of 1.05:1, and they column away the undesired diastereomer.

    Sorry for being so negative, good work as always on the post, at least. Also, am I missing something with/can somebody explain the recycled Rolexes comment?

  • Hap says:

    Toste has tended to do lots of gold-based catalysis – hence the “Rolex watches” comment.

  • ZZZZZ says:


    …I liked the 8-membered ring formation. I was asleep before and after that though…


  • ZAL says:

    From the paper it’s not completely clear, but if you look in the supporting info it seems that he used the same ligand already reported by Bolm and Ellman for sulfide oxidations: that is, the Schiff base obtained by condensation of 3,5-di-tertbutylsalicylaldehyde and tert-leucinol. So I think the alkyl chain (and the stereogenic center) should be alfa to the nitrogen atom. Strange that he did not show the structure, though.

  • SMILES says:

    For the formation of 8-membered lactone ring I prefer Holmes Methodology ;-). I am sure some people here will agree with me.

  • anon says:

    “…I also have to say that my first instinct upon reading this was that it doesn’t really teach the world much about chemistry…”

    To me this synthesis is about showcasing a particular methodology – the resolution. In this case, what to me is impressive is that they haven’t needed to “force fit” their methodology.

    This raises an interesting point about the nature of total synthesis. There seems to be several aims to undertaking a synthesis. 1) to make the molecule; 2) prove it’s structure; 3) make it better than before either in less steps, higher yields or larger scale; and 4) to prove the usefulness and predictability of a new methodology. I think that all these goals are worthwhile but there is an undertone that the last is not as respected as the first three… any thoughts?

  • stork naked says:

    it’s a nice synthesis, but nothing spectacular. good to see an approach not relying on esoteric substrates for bond reorganizations

  • TWYI says:

    Congrats on the ChemComm Paul, nice work!

  • milkshake says:

    Not that I wish to imply the vanadium kinetic resolution was used to crack a ficticious synthetic challenge – and no-one is more deserving of the prize that prof Toste himself:

    The enantiomerically pure (R)-methylester of 2-methyl-3-hydroxypropionic acid costs $90/10mL at TCI-US. The (S)-enantiomer is cheaper stil…

  • milkshake says:

    also “…If it is iodine it must be professor Moriarty”

  • Howard says:

    Anon –

    I’m in agreement with and in support of your ideas about total synthesis – I just don’t think this necessarily fits them.

    I guess what I’m saying is that this methodology doesn’t need a total synthesis. If you were to tell me that I could make enantioenriched alpha-hydroxy esters catalytically and asymmetrically using oxygen as a reagent, I’m very interested in that result, and it is a good one. I just don’t think the world is taught anything by showing that these products are versatile building blocks – we already knew it. It is a clever disconnection, but it’s not going to help me in my total synthesis in any way – I already knew to try to think of clever disconnections that don’t waste material.

    Overall, I don’t really have a problem with the synthesis (besides the waste of NHK-product that I mentioned before), I just didn’t get a lot out of it.

  • J says:

    That hypervalent iodine-mediated alpha-keto carboxylic degradation rocked! I couldn’t be called a Moriarty reaction. I’d love to know the name of that reaction. Any takers?

  • Big_C says:

    Theres a lot of Toste hating going on in these comments… why does Baran get all the fanboys?

    I think this is a short, clever synthesis… esp. in comparison to the others.

  • Thanathos says:

    There is a review about hypervalent iodine, written by Moriarty himself, and he explains with a lot of humour that he loves iodine because of Holmes’ stories ;-)

  • pmgb says:

    mechanism: arrow should point to Iodine in stead of Ph

  • milkshake says:

    There are no haters here. A right method in the wrong place can make all the difference. So wake up.

  • milkshaken says:

    How exactly were you planning on taking that propionic acid from TCI and making it into either of those kinetic resolution products?

  • TheEdge says:

    In the context of this synthesis, you don’t necessarily need the kinetic resolution products, you need access to the aldehyde and the beta-keto phophonate that Toste makes from the kinetic resolution products. I think milkshake’s point is that you can buy a compound that will take an equivalent amount of functional group mainpulation to get to the desired coupling partners. The arguement over whether it’s better to buy or to make your first asymmetric center doesn’t have a good answer, so I think the whole thing is a push. It’s a nice synthesis, though.

  • Tot. Syn. says:

    I think that is an exceptionally important point, and one that I’ve laboured over to an extent – at what point does ‘buying-in’ asymmetry become inelegant? Many syntheses of macrolides involve Roche ester (I’ve seen a photograph of Ian Paterson with several drums of it!), which seems par-for-the-course. Sugar-bashing – even extract only one or two stereocentres – is fairly common too. But technically, a synthesis derived from an existing natural product is a semi-synthesis (but this is stretching the definition to breaking-point). Thoughts?

  • Eraser says:

    In my opinion, the whole art of total synthesis is in the assembly of the molecule and how you deal with the inherent challenges of working in an environment of multiple functional groups, stereogenic centers and/or rings. The source of the starting material whether it be chiral pool or asymmetric synthesis i largely irrelevant. After all going back to the source, all chirality is chiral pool derived. Even BINOL.

  • Eraser says:

    Unless you want to get out your tweezers.

  • milkshake says:

    To answer your question, protecting the OH of HOCH2CHMeCO2Me and adding LiCH2PO(OR)2 to carbonyl gives the right piece fro HWE. Infact Toste used this method, except that he made the precursor from a sideproduct in his kinetic resolution in two steps (not counting the in situ activation)

    The aldehyde-ester left piece for HWE would be best made from glyoxylate and propionaldehyde equivalent, either by organocatalysis or by Evans-MacMillan anti-aldol ( tin(II) with a Pybox ligand)…but I suppose you could get it by elaborating the commercial propionate ester piece also. Either way, the scheme it is going to be lot faster, with fewer steps. And you wont have to develop any new methodology.

    I am all for methodology driven papers but as as has been mentioned before, this particular methodology demonstration is somewhat forced.

  • InfMP says:

    I think that buying some stereocenters is fine as long the entire synthesis is impressive enough that the reviewers decide it is worthy (and don’t just look to see if it is a big name).
    But it has got to be an affordable starting material. I have strong feelings against people who say “two steps from readily available starting materials!!!” and then I look and see it is 200$/g.
    That is as “cheap” as titling a paper “a concise approach to x” only to find it is not a total synthesis, but a “blueprint”.

  • TheEdge says:

    There are a very small number of labs that focus on both interesting ways to make one or two stereocenters and on interesting ways of putting together larger fragments, which kind of explains why you see so many people starting from sugar or tartrate or roche ester derived starting materials. Why fiddle aroud trying to get somebody else’s methodology to work if the step you care about is 12 or 20 steps in? Just bash the stereocenters you need out of a sugar using known methodology and move on!

    I’ve actually got no beef with people purchasing stereocenters. In the end, like Eraser said, the asymmetry of everything we use is derived from something found in nature. My problem is when people buy a chiral compound and then strip all of the functionality off of it in order to install what they actually need, or start from something that forces them into a ridiculous protecting group scheme. The installation of fuctionality in the right desired state doesn’t just apply to aldehydes and amines, it applies to stereogenic centers as well.

    I think the problem is that it’s always easier to bash through from a natural source than to build an elaborate catalyst from equally (or more) expensive starting materials and optimize “someone else’s” reaction to work on your substrate. This is why there are dozens of papers published every week on making compounds with one or two stererocenters, yet very few of them are ever part of a total synthesis.

  • milkshake says:

    yeah, there is lots of salesmanship in high-profile synthetic chemistry – it always bothered me when people advertise the gee-wiz features of their work while burrying the inconvenient detail deep in the supplementary info. This kind of exaggeration has been going on for decades now and some famous people in the past even would not publish full experimental details of their total synthetic work. I think JJ LaClair hexacyclinol stunt was the Sokal hoax equivalent in our field.

  • cvengo says:

    this a bit off topic but still related o the application of the methodology in the total synthesis, I would like to ask, what is the simplest way to search for a natural product of a certain structure. Imagine you have developed a synthetic approach towards a certain structural motif and you’d like to find a natural product which is similar.

  • Potstirrer says:

    Use Beilstein Crossfire. Draw your substructure and then restrict your results using the “INP” (Isolation from Natural Product) data search function. Works great!

  • cvengo says:

    thanks Potstirrer

  • Merlin says:

    Forced? Forced can mean many things; a method to show its “practicality” or a final step giving a 2% yield for an overall 0.000001% yield for an entire synthesis of a famous molecule. Sometimes people tend to forget that the science we do, at least in the “beginning” of our naive careers, is to learn. Some people do hate, but for the wrong reasons. Groups run reactions, especially methods groups, to understand their mechanisms, not only to get another JACS (http://totallysynthetic.com/blog/?p=322). Dr. Radosevich did that in his first paper concerning oxidative kinetic resolutions. This, it seems to me, was just icing on the cake, a way for him to experience total synthesis in a non-total synthesis group. Dr. Radosevich has nothing to apologize for. No, I’m not a current Toste group member.