1.0 Introduction

1.0.1 Background

All bio-macromolecular crystals suffer from time-dependent radiation-induced loss of diffracted intensity to some degree. In many of these cases the radiation damage at room temperature makes it impossible to collect a complete dataset from a single crystal. This problem can often be overcome by flash-cooling the crystal to very low temperature. This chapter contains a step-by-step guide to that method, which will reduce and, in most cases, eliminate the problem of radiation damage and may also offer a way to assist in trapping a short-lived intermediate for crystallographic studies.

One of the obstacles to structure determination of macromolecules by X-ray crystallography is radiation-induced crystal decay. Crystal decay appears to be a result of X-ray radiation and thermal damage to the macromolecule and, subsequently, to the lattice. This damage hinders uninterrupted data collection and introduces experimental errors to the measurements. In the past the only way to overcome this obstacle was to collect data from several crystals, switching to the next when the previous one decayed beyond usable levels. This strategy suffers from problems of scaling datasets from crystal to crystal. Radiation and thermally induced crystal decay can be slowed and even stopped entirely by collecting data from cooled crystals.

1.0.2 Crystal Decay

In most cases, crystals will degrade when heated, thus crystals will exhibit thermally induced decay by the X-ray beam. Thermally induce decay will be noted mostly at high-flux synchrotron X-ray sources and will be negligible on a home-lab X-ray source. Yet the more profound cause for decay is radiochemistry. The main source of radiation damage (Hendrickson, 1976; Talmon, 1987; Henderson, 1990) is believed to be free radicals photochemically produced by X-rays (Coggle, 1973; Davies, 1987). Radicals react with the protein in random ways, thus breaking down the order of the lattice (Mayer, 1985; Bachman and Mayer, 1987). Disruption of order in the crystal manifests itself as radiation-dependent decay of the intensity of individual Bragg reflections.

Figure 1.1: Typical decay of ASA-DH crystal at 4oC
Unlike the photochemical production of free radicals, the rate at which free radicals are produced in non-photochemical processes (diffusion-dependent propagation) and react with the macromolecule is temperature-dependent. At cryo-temperatures (in this chapter cryo refers to temperatures in the range of -180 to -150 oC), diffusion of free radicals and reaction rates within the crystals are greatly reduced. In many cases, use of cryo-temperature may essentially eliminate radiation damage (Hope, 1988; Hope, 1990). The photochemical production of free radicals is not inhibited at cryo-temperatures; this can only be achieved at much lowed temperature closer to the absolute zero.

1.0.3 Flash Cooling

The way a protein crystal is taken to cryo-temperatures is crucial for the success of any cryo-technique. Gradually lowering the temperature to cryo-levels causes formation of ice crystals within or around the protein crystal lattice. Ice formation degrades the diffraction from the protein lattice in two ways: by contributing to the diffraction pattern and by shearing the crystal lattice. The primary objective of flash-cooling (Haas and Rossmann, 1970; Parak et al., 1981; Hartmann et al., 1982) is to overcome the problem of radiation damage while avoiding the problem of ice formation that occurs in slower cooling methods. Other aspects of low temperature methods have been described in Alber et al., 1976; Hope, 1988; Hope, 1990; Earnest et al., 1991; Watenpaugh, 1991; Tilton et al., 1992.

To flash-cool, in this context, means to take a specimen from non-cryo-conditions to cryo-conditions rapidly. Plunging a protein crystal into cryo-temperature causes the aqueous solution, in and around the crystal, to freeze amorphously, like glass, in a process known as vitrification (Mayer, 1985; Dubochet and Schultz, 1988). In principle, vitrification should not change the protein crystal in terms of lattice integrity and the solute structure. In practice, flash-cooling of a crystal can affect the lattice (Low et al., 1966; Singh et al., 1980; Kellenberger, 1987). As a consequence the crystal may exhibit a change in the degree of order of the lattice (the mosaic spread). In most cases, the mosaicity will increase upon flash-cooling, but there are cases where the opposite effect is observed (Zalonga and Sarma, 1974; Young et al., 1990). This behavior is greatly influenced by the solute composition. To increase the rate of success of flash-cooling, an additive can be introduced to the mother liquor. The term "cryo-protectant" refers to the chemical or chemicals that permits the cooling of a protein crystal with little or no increase in its mosaicity (Haas and Rossmann, 1970; Petsko, 1975; Casico et al., 1984; Dewan and Tilton, 1987; Dubochet and Schultz, 1988; Hope et al., 1989). Those chemicals, which have so far been members of the alcohol, sugar or sulfoxide families, serve to slow the nucleation of ice, raise the viscosity of the solution as it cools (and thus raise the glass transition temperature) and break the propagation of ice formation in the mother liquor. Those additives are added in amounts ranging from as low as 2% to as high as 30% by volume.

Other parameters that affect the freezing process are the characteristics of the crystal itself such as size and shape, mechanical stability and density. When subjected to flash-cooling, the crystal is suspended on a support surrounded by the aqueous solvent. The nature of this support and the amount of solvent surrounding the crystal is another important parameter in the freezing process. In the design of the support the mechanical strength of the crystal should be considered, as well as minimization of the amount of mother liquor surrounding it.

Once a crystal has been frozen, it can no longer be taken out of cryo-conditions without almost total loss of diffraction. As the solvent warms up, slowly or rapidly, it will go through several phase changes and ice will form within and around the crystal, causing the breakdown of the lattice. Thus, storage of flash-frozen crystals at cryo-temperatures is imperative.


Press here to go to the next section.