Come and see!

Gazing at the very first model of a protein structure determined by X-ray crystallography Max Perutz, one of the founding fathers of the technique, was disappointed. It may have gone unspoken at the time but the initial coiled, worm-shaped creation resembled nothing so much as a giant turd. “Could the search for ultimate truth really have revealed so hideous and visceral-looking an object?” lamented Perutz.

Polished, but still ugly? (From London’s Science Museum)

But that early dismay was soon forgotten. The first model of myoglobin, an oxygen-storage molecule from the muscles of sperm whales, was based on relatively low resolution data and revealed only the overall fold of the protein polypeptide. As crystallographers worked their crystals, getting them to scatter X-rays at higher and higher angles, the atomic details of the molecular structure were soon revealed. The consternation at the irregularity, the lack of symmetry in proteins—where was the slender elegance to mirror DNA’s double helix?—soon evaporated before the glittering galaxy of atoms and bonds.

Which brings me to a wet weekend in Hamburg. Last weekend in fact when I found myself at the synchrotron, along with two members of my group, peering into a microscope at three O’Clock in the morning and trying to fish out tiny protein crystals with a miniscule loop of nylon attached to the end of a metal pin. Well, they weren’t that miniscule by modern standards; in fact the crystals were rather chunky—about 0.3 mm on each side.

Once caught, the crystal was placed in the direct line of fire of an intense X-ray beam. We retreated from the lead-lined experimental hutch and instructed the computer to open the shutter.

And when we did so, our mouths dropped open too: the crystal diffracted to 1.4 Å! In the lab it had only gone to 2.7. We were hoping to do better than that at the synchrotron but 1.4 took our expectations and booted them out of the stadium. What can I say? We were very pleased!

And why exactly? What do these numbers mean? It tells you how much detail you can expect to see in your electron density maps, the three-dimensional chicken-wire that we compute from the data to reveal the shape of our molecule.


And I have just this afternoon gazed upon the electron density map. And now you can see it too. It still looks like worms but this time they are knurled and knobbly; you can just about see that the molecule is constructed of tiny spheres. Atoms. In places, even the hydrogen atoms, each with just a single electron, can be seen protruding shyly from the density. In all my time as a crystallographer I have never seen a molecule in such glorious detail, and neither has anyone else in the lab. So today everyone is wandering around with smiles on their faces. And nobody is mentioning poo.

Coda: This morning, although I (gently!) mocked Richard’s heroic attempt to generate a buzz around science by encouraging people to post in something like real time as they did their experiments, I guess I have endeavoured to capture something of the immediacy of a real result in this post. So I’m a hypocrite. So sue me.

This entry was posted in Uncategorized. Bookmark the permalink.

10 Responses to Come and see!

  1. Richard P. Grant says:

    Never mind the resolution, show me your R factors.

  2. Stephen Curry says:

    Rfree is 28% after the first round of simulated annealing. We’ve not even built in the bound waters yet so the map can only get even better…

  3. Richard P. Grant says:

    Oooh. Watch it drop when you add water, eh?

  4. Heather Etchevers says:

    As a non-specialist, it sounds like fun…
    A friend of mine sent me a list of gene candidates for a horrible and thankfully rare pathology in a chromosomal region this morning from which to pick out my favorites – and I have only a couple, and they’re humdingers! – I’ll get back to y’all after the sequencing has happened. That is real time – and it’s uninteresting to all but those concerned.
    What we all really like are pre-preprints like this, Stephen. With pretty pictures.

  5. Stephen Curry says:

    As a non-specialist, it sounds like fun…
    It was fun once the data started coming in. But there’s a whole back-story about getting samples ready for the trip, re-booking tickets the day before we left because a member of the team was forced to pull out at the last minute, dragging yourself to Hamburg tired before you started yet knowing that we had over 48 hours of beamtime to use. All that wasn’t so much fun but it was certainly a necessary precursor…
    Look forward to hearing your humdinger story (and seeing the pretty pictures!)

  6. steffi suhr says:

    And next time you’re in Hamburg, Stephen, you’ll let Martin and me know – we’ll take you out for a beer after you’re done ‘beaming’ and before you collapse… deal?

  7. Stephen Curry says:

    @steffi – we’ll take you out for a beer…
    Definitely! Usually it’s all a bit of a panic and we’re stuck on the beamline for hours on end. But the data collection software is getting more sophisticated so you can line up a few samples at a time and leave it running for a few hours – just enough time for a pleasant detour to the pub. But last Saturday/Sunday, we worked almost all night, crashed at 4 am, got a taxi to the airport at 5.15 am and were back in London at 8. But not really in a serviceable condition…

  8. steffi suhr says:

    Worth it, though. I think the image above should go on a postcard…

  9. steffi suhr says:

    Yours, that is, not the turd.

  10. Stephen Curry says:

    I think there might be a market for both…! But your comment reminds me that I should add an acknowledgement to the Science Museum, where the photo of the myoglobin model comes from.