First X-Ray Diffraction Image of a Single Virus

KentuckyFC writes "X-ray crystallography has been a workhorse for chemists since the 1940s and 50s, revealing the 3D structure of complex biological molecules such as haemoglobin, DNA and insulin. But the technique has a severe limitation: it only works with molecules that form into crystals and that turns out to be a tiny fraction of the proteins that make up living things. But today, a team of US researchers say they have created the first image of a single uncrystallized virus using x-ray diffraction. The trick is to take a diffraction pattern of the virus and then subtract the diffraction pattern of its surroundings (abstract). The breakthrough paves the way for scientists to start teasing apart the 3D structures of the many proteins that have eluded biologists to date."

Comments

overreaching /. summary

[jschen]

This work is really cool, and it's interesting to muse about what else might be imaged this way. But while 22 nanometer resolution may give insight into the structure of a virus, that would be awfully lousy resolution for a macromolecule (say, a protein) or even a macromolecular complex.

Re:overreaching /. summary

[damn_registrars]

while 22 nanometer resolution may give insight into the structure of a virus, that would be awfully lousy resolution for a macromolecule (say, a protein)
You hit the nail on the head, there. Indeed, protein structures generally need to be solved at a resolution of 8 angstroms or less to be taken seriously. And of course there are 10 Angstroms to a nanometer, so 22 nanometer resolution is equal to 220 Angstroms.

Useful for a virus superstructure, but with a protein you wouldn't be able to distinguish one end from another.

First step towards X-ray microscopy

[vossman77]

This is very exciting, I remember during my Biophysics training that "blah, blah, blah you cannot focus X-rays like you can visible light with lenses, so we'll never have an X-ray microscope." Well, this looks promi

Re:First step towards X-ray microscopy

[Cevets]

Sorry, but people have been focusing x-rays for decades. As for x-ray microscopes, search up scanning transmission x-ray microscopy (STXM) or X-ray Photoelectron emission microscopy (X-PEEM), you should find them an interesting read.

Re:First step towards X-ray microscopy

[blueg3]

Yes, neither x-ray microscopy nor x-ray diffraction are new. This appears to be essentially small-sample x-ray diffraction, as you describe.

TEM of biological materials (like a virus or protein) is tricky because the TEM can destroy the structure of the protein and can't adequately discern its internal structure. I'm less familiar with SEM, so I couldn't say, but diffraction is usually interesting not for obtaining an image of the sample, but determining its structure.

And yeah, you use zone plates. Or, for some applications, optical fiber (which is what we use to do confocal X-ray fluorescence).

Re:First step towards X-ray microscopy

[Steve525]

SEM imaging is not going to offer much improvement over TEM. The main advantage of a SEM is that you (usually) look at the electrons scattered from the surface instead of having to transmit the electrons through the sample. This does have some important advantages - you don't need to thin samples, and you can use lower electron energies. Lower electron energies might lead to less damage, but all that energy is deposited right at the surface. Plus, you only end up imaging the surface, which is why it's useful for images, but not for looking at internal structure. The resolution of a SEM tops (bottoms?) out at 1-2 nm.

This is interesting work, but as others have pointed out it'll need to achieve better resolution to really be useful. (22 nm is similar to what can be achieved with an x-ray microscope). I'm not sure what's limiting the resolution right now, but I'm guessing it's a signal to noise problem. One exciting thing this work has to potential to do is 3-D reconstructions. Think of this technique as crystallography without needing the ordered crystal to produce the data. That's (perhaps) the potential of the technique. The reality is, however, that radiation damage might be too much of a problem no matter how many advances they make.

Re:Makes sense

[Anonymous Coward]

I think you are confusing terms. Fatty acids don't refer to amino acids. While it is true that there are certain amino acids that are hydrophobic and hydrophilic (among other properties) like many fatty acids, this doesn't mean they are fatty acids themselves.

As far as the rest of your comment, it is confusing. I assume you are talking about the primary structure of the protein when you talk about 'fatty' and 'wet' amino acids (which I'm guessing you are trying to say that the side chains of the primary structure are hydrophobic and hydrophilic). But then you jump to the tertiary structure by talking about 'fatty' amino acids being surrounded by 'wet' amino acids. I really don't follow what you are trying to say. Many proteins have surfaces designed for certain conditions (like transmembrane proteins that have both hydrophilic and hydrophobic surfaces to anchor them in a phospholipid bilayer), and indeed, their surfaces in the tertiary structure do cover over parts that may not be desirable on the surface. Are you trying to argue that this is wrong?

Re:Makes sense

[clonan]

That is exactly what I am arguing is wrong.

The internal consistency of a protein more closley matches a very tight crystal than a hydrophobic amalgum.

It is absolutly true that hydrophilic amino acids are often on the surface except for specific circumstances like you mentioned. However these amino acids tend to adopt the most energetically efficient form possible which is usually crystaline.

What I was refering to is that for most of modern biology there was an assumption that since the consistency of the individual hydrophobic amino acids was very much like a oil that therefore the internal consistency of the protein should be similar. A more detailed analysis of densities, x-ray diffraction, fNMR and computer modeling proved that the internal structure of a protein is actually a very hard crystal with very little mobility of the individual atoms.

The older assumption is still often taught up through undergraduate level. When I was in High school proteins were described as a drop of oil surronded by hydrophilic amino acids.

Re:Makes sense

[rnaiguy]

While the interior of a protein resembles a crystal in terms of atomic packing density, it is certainly not crystalline, as it lacks the crucial requirement of being arranged in a regular pattern. You need a regular crystal lattice in order to amplify x-ray diffraction signals to detect them. That's why the resolution on this is so bad compared to x-ray crystallography studies. However, since you don't need crystals, this can be very useful for larger complexes and stuff, though probably not single proteins (unless the resolutions comes up 2 orders of magnitude).

Re:Makes sense

[rnaiguy]

What you say is correct to a point, but i don't understand the point you're trying to make. As you correctly alluded, the local environment of atoms in the protein interior has a packing density resembling that of a salt crystal, BUT they lack the regular lattice structure that allows for constructive interference to occur between X-ray photons to produce detectable diffraction patterns. That repetitiveness is the essential component of a crystal! There are crystals with low packing densities (solid water is less densely packed than liquid, and protein crystals can have huge holes in between protein molecules), but what they all have in common is a regular repeating lattice. So, in short, the protein interior is densely packed, but NOT crystalline.

oh fuck off

[afxgrin]

You are so deluded with this belief that it's a behavioral issue.

I think there's a bunch of Libyan kids who'd like to crack you across the head with a crowbar.