Cryogenic cooling of macromolecular crystals.

Macromolecular crystals can be used to determined detailed three-dimensional structures of proteins, DNA and RNA via X-ray diffraction, which in turn can aid in understand basic mechanisms in biology.

The crystals are sensitive to X-rays, so they are usually cryogenically cooled to ~100K prior to X-ray exposure. This slows the X-ray damage processes.

BUT, the cooling process itself can damage the crystal. We are trying to understand this cooling damage with the aim of improving cooling techniques. We are interested in the following related questions:

1. What is the nature of cooling damage to crystals?
2. How do cryoprotectants work to limit cooling damage?
3. Can cooling damage be reversed in a predictable way based on the nature of the damage?
4. How much information about a particular macromolecular crystal is required to choose a good cryoprotectant?

Our work and others' have suggested the following mechanism for cryoprotectant action. It appears that cryoprotectants work in part by adjusting the thermal contraction of the bulk solvent to match the contraction of the interstitial spaces (see Figure 1). 

Figure 1. Schematic comparing a salt crystal (left) with a macromolecular crystal (right). The main difference is one of scale.  This means that in the macromolecular case the interstices are large and filled with disordered bulk solvent. Also, in the macromolecular case only a small fraction of the atoms are involved in lattice contacts, which make the crystal relatively fragile.  When the crystal is cooled, the macromolecules get closer together, shrinking the interstitial space. The bulk solvent must therefore contract the right amount to compensate for the contraction of the interstitial space. The amount of solvent contraction can be adjusted by adding cryoprotectants, which are small molecules such as dimethyl sulfoxide, glucose and glycerol.

 

These results led to additional experiments to investigate why cooling induced lattice damage can sometimes be recovered by warming the crystal to room temperature and then recooling. This appears to work, at least in part, via the transport of water into or out of the crystal, altering the thermal contraction properties of the bulk solvent to better match the contraction of the interstitial space.

We are continuing this work by carrying out X-ray diffraction experiments with our new Oxford Diffraction Xcalibur Nova instrument on various protein crystals in concert with bulk measurements on the thermal contraction of solutions of water and various cryoprotective agents.