B.R.S.M. All this happened, more or less.


Sunday Evening Maths 1: ovatoxin-a

Among other things in JACS last week (such as more exciting welwitindolinone stuff from Rawal and Garg), I noticed this enormous (and enormously toxic) palytoxin-type natural product (DOI: 10.1021/ja210784u). Although I'm not particularly excited as a synthetic chemist, and I don't understand the biology, I do like compounds like this because they're so silly.

Structural elucidation was a bit tricky as the group only got 700 μg of compound. They say that as a result they 'could not afford any reliable 13C spectrum', by which they presumably mean they couldn't afford the 6 months of NMR time it'd probably take to get a good one, or that the one they got resembled the black forest prior to the industrial revolution. That's 0.26 μmol for you.

Regarding stereochemistry, there are some 4,611,686,018,427,387,904 possible diastereomers for this behemoth, which sounds like a quite a lot. Thankfully, comparison with palytoxin allowed assignment of 28 of the 62 stereocentres, meaning that actually only 17,179,869,184 diastereomers are left. That's not bad, as they've straight away ruled out 99.999999996275% of the possible structures, without setting foot in the lab. Depending how precise you're being, that's all of them. The fairly hefty molecular weight of 2648.13 means that if someone made 1 mmol of each that'd be 45,441 tonnes of material. To put this is perspective, may I remind you that this is approximately the same weight of tomatoes exported by New Zealand every year (to 2 significant figures). The authors conclude:

"Eventually, the availability of pure OVTX-a, which is by far the main toxin produced by O. cf. ovata, heralds the potential of characterizing the risks it poses to humans as well as of establishing the tolerable daily intake for seafood consumers."

However, as  the 'mouse lethality assessment' found that:

"one to 3 min after injection, mice started wriggling and all died within 30 min with paralyzed limbs. These preliminary studies on OVTX-a’s toxicity led us to conclude that the toxin mouse lethality lies below 7.0 μg/kg."

I think I might play it safe and have none.


Comments (9) Trackbacks (0)
  1. Woah. Some crazy maths there.

  2. I had to use WolframAlpha to actually figure out how many stereocenters you counted.

  3. Just for perspective, by my calculations those 7 μg equal 15,918,772,869,911,975 molecules. So a 20 g mouse will be killed by 318,375,457,398,239 molecules of this stuff. Which, to the nearest microgram, is 0 μg.

  4. I’d like to see the biosynthetic genes for this.

  5. nice… a new target for nicolaou is born 😉 until he finishes maitotoxin…
    maybe somewhere in this molecule some C2 symmetry can be found so the synthesis can be halved…
    thx for this perspective

    • I agree. I was just about to say this looks like a target for Nicolaou. However, I don’t see any C2 symmetry…..Looks like a collaboration project with 2-3 groups (Kishi, Nicolaou, …?)

      • … and A.B. Smith III ? or Kocienski? I would really like to see a total synthesis in a full account of this monster…

        • Would you, though? This molecule is really similar to palytoxin, and as epic as Kishi’s synthesis of that was, I didn’t learn all that much from it, and I’m not sure the rest of the world did either. I guess it might be interesting to see how we’ve moved on in the last couple of decades (I suspect quite a bit), but I would not like to write a comparison of the two routes!

Leave a comment

No trackbacks yet.