Skip to main content

Chemical News

Macrocycles For the Making

I meant to write about this paper at the time, but there’s no harm in highlighting it now. A group at the University of Toronto reports a neat way to make some unusual macrocycles, by closing down an amine and a carboxylic acid into an oxadiazole with the known isonitrile phosphorane reagent shown. You bring in one more carbon from an added aldehyde (propionaldehyde in most of their examples, so you turn an X-atom-long chain into an X+2-sized macrocycle.

The paper has a whole list of example where oligopeptides are closed to peptidic oxadizole macrocycles, giving 15- to 24-membered rings. Yields are often down in the 30s and 40s, sometimes below, but sometimes up into the 60% range as well. You may well be wondering why that works even to that extent, since closing macrocycles is often a capricious process. The phosphorane reagent, though, takes the mechanism through a stage where one end of the chain has a positive charge and the other end is a negatively charged carboxylate, so you have electrostatics bringing things together.

The resulting macrocycles seem to partake of some of the interesting features of their type, specifically enhanced cell permeability. (Here’s a fairly recent book on the topic). The authors took a list of their macrocycles through a PAMPA assay (which for those who haven’t run into it is an in vitro test with an artificial membrane), and found that in every case the macrocycles were significantly more permeable than the open-chain precursors. We certainly don’t understand what’s going on with such compounds, once you get past some bulk physical properties, but they do have a lot of potentially useful features.

My first thought when I saw this paper was “Hey, I’ll bet that works on things that aren’t peptides”, and I’m sure that it does. There are surely a huge number of interesting structures that you can put together relatively quickly that have an open primary or secondary amine on one end and a carboxylic acid on the other, and which could be zipped up into a library of most unusual macrocycles. I note that the folks behind this are the founders of Encycle Therapeutics, so similar thoughts have no doubt occurred to them as well!

25 comments on “Macrocycles For the Making”

  1. Hap says:

    Baran has an imine peptide macrocyclization in JACS:

    1. Derek Lowe says:

      I have to say, that if I were forced to pick, I’d rather have an oxadiazole in my final product than an imine. Baran’s reaction is neat, though.

      1. Hap says:

        If you could reduce the imine moiety chemoselectively, it could be useful, but the stability of imines (particularly with all of the dynamic combinatorial chemistry work) would not make me comfortable.

    2. anon says:

      So performing a reductive alkylation with standard conditions gets you a JACS…

      1. anon says:

        Depends on who publishes it. And you have to have schemes that look like infographics too. I like plain old Chemdraws.

        1. Derek Lowe says:

          Me, too. If I had plain old Chemdraw, that’s what I’d be using (!)

    3. Pepe Peptide says:

      Except for the overwhelming scope, the underlying chemistry is pretty meh.

      1. Phsh says:

        Cmon broe the substraight scoap is epic

      2. Activation mode says:

        0/10 not enough photoredox

  2. Anon says:

    The link to the paper seems to be broken (I tried on multiple browsers)

    1. Derek Lowe says:

      OK, that should be fixed – thanks!

  3. M. Welinder says:

    I wonder how big those cycles need to be before you can start making chain-mail-like structures.

    1. zero says:

      Not very, but how do you propose threading the chains through properly before closing?

      1. Lassotides says:

        This also recent peptide macrocycle paper from Bode seems like a good start:

        I wonder if these would also have improved cell permeability?

  4. anonymous says:

    I wonder what will happen if we take a linear cell penetrating peptide (CPP) or functionally active peptide and transform it into oxadiazole incorporating cyclic peptides and expect that to retain its activity? I reckon not! But then the cyclic variants will lead to molecules that are enzymatically more stable!

  5. Curious Wavefunction says:

    There are interesting differences in physicochemical properties between 1,2,4 and 1,3,4 oxadiazoles: a 2012 J. Med. Chem. publication by Boström et al. (you can Google “Oxadiazoles in Medicines Chemistry”) found that the log D for the latter can be an order of magnitude lower. I wonder if this method can make both of them (don’t have access to it currently).

    1. Derek Lowe says:

      Nope, this is the 1,3,4 only (which is apparently a good thing!)

    2. Peter Kenny says:

      The differences between 1,2,4-oxadiazoles and 1,3,4-oxadiazoles stem from from differences in hydrogen bond basicity. Interactions between adjacent aromatic nitrogen lone pairs can be seen as a manifestation of the alpha effect which is also invoked as an explanation for the surprising (in view of pKa) nucleophilicity of hydrazine and hydroperoxide. I have linked an article on hydrogen bond basicity as the URL for this comment.

  6. Jonathan says:

    One other interesting aspect of Yudin’s work is the potential to mitigate the high dilutions typically required for macrocyclizations with rings of this size, which can sometimes pose a substantial barrier to scale-up. It’s not at all uncommon to use concentrations of 1-5 millimolar for macrocyclizations (RCM, macrolactonization, etc.). The whole zwitterionic control mechanism is really interesting, and I wish Yudin had discussed it more in the paper. I would like to know if these reactions could be done at even higher concentrations than the 25-100 mM reported here.

    1. milkshaken says:

      Dr. Yudin seems delighted to discuss any interesting synthetic/medicinal chemistry, and even more so the research coming from his own group. Here is his blog

  7. Things I won't work with says:

    Off topic:

    Just sayin’.

    1. milkshaken says:

      I really like the method mentioned in the Wiki article – you just have to make some tetrafluoroammonium perchlorate NF4(+) ClO4(-) and pyrolyze it

  8. dvizard says:

    “We certainly don’t understand what’s going on with such compounds”

    Well, just that they don’t stick with the common assumptions made for linear stuff, surprisingly.

Comments are closed.