research papers
Macromolecular crystals are usually cooled to ∼100 K for X-ray diffraction experiments in order to diminish lattice damage arising from the ionizing radiation. Such cooling often produces lattice disorder, but this disorder can sometimes be substantially reduced by cycling the crystal between low and higher temperatures (called annealing). Here, two related aspects of cryocooling and annealing are investigated using crystals of β-galactosidase and thermolysin. Firstly, as has been reported with other systems, there is an optimal cryoprotectant concentration above and below which diffraction is poor, with high mosaicity, diffuse scatter and low signal to noise. Measurements of the bulk density of the respective cryosolvents are consistent with the idea that at the optimal cryoprotectant concentration the contraction of the bulk solvent on cooling largely compensates for the contraction of the macromolecular lattice. Secondly, by controlling the relative humidity of the gas that contacts the crystal during the high (room) temperature phase, it is found that water is either imported into or exported out of the crystals during the melting phase of annealing. This water transport appears to change the concentration of the cryoprotectant solution and in so doing alters its thermal contraction. Thus, annealing appears to be involved, at least in part, in the tuning of the thermal contraction of the bulk solvent to best compensate for lattice contraction. Furthermore, it is found that if the cryoprotectant concentration is initially too high then annealing is more successful than if the concentration is initially too low. This result suggests that the search for optimal cryoprotectant conditions may be facilitated by equilibration of the crystal to relatively high cryoprotectant concentration followed by annealing.