[2] PROTEIN CRYSTALLIZATION METHODS 13 [2] Overview of Protein Crystallization Methods By PATRICIA C. WEBER In~oducUon Crystallographic structure determination begins with growth of a suit- able crystal. With the availability of powerful X-ray sources, rapid data collection instruments, and faster computers, crystallization has increasingly become the rate-limiting step in macromolecular structure determinations. In this chapter, some of the physical principles that govern crystal growth are presented to assist the crystallographer in designing crystallization ex- periments and interpreting their results. The appearance of a macroscopic protein crystal containing roughly 1015 molecules begins with association of protein aggregates whose intermo- lecular contacts resemble those found in the final crystal. 1'2 These prenuclear aggregates eventually reach the critical nuclear size. Given stable nuclei, growth proceeds via addition of molecules to the crystalline lattice. Both crystal nucleation and growth occur in supersaturated solutions where the concentration of protein exceeds its equilibrium solubility value. The region of solution parameter space suitable for crystallization is generally repre- sented on the phase diagram by the solubility curve (Fig. 1). Supersaturation is a function of the concentration of the macromolecule and parameters that affect its solubility. It is achieved at high macromolecu- lar concentrations, and at increasing values of solution parameters that decrease macromolecular solubility. Many factors can influence protein solubility. Inclusion of additives such as alcohols, hydrophilic polymers, and detergents can decrease protein solubility. While these are commonly referred to as precipitants, they are solubility-influencing agents, with pro- tein precipitation being only one possible outcome of their addition to the solution. Protein solubility as a function of salt is usually an asymmetric bell-shaped curve with decreased solubility at both high and low salt concen- trations. Consequently, strategies involving both inclusion and exclusion of salt can induce protein crystallization. Solution parameters such as pH or temperature can also dramatically influence macromolecular solubility. 1 F. R. Salemme, L. Genieser, B. C. Finzel, R. M. Hilmer, and J. J. Wendoloski, J. Cryst. Growth 90, 273-282 (1988). 2 M. Jullien, M. P. Crosio, S. Baudet-Nessler, F. Merola, and J. C. Brochon, Acta Crystal DS0, 398-403 (1994). Copyright © 1997 by Academic Press, Inc. METHODS IN ENZYMOLOGY, VOL. 276 All rights of reproduction in any form reserved.14 CRYSTALS [21 E o. e~ a SupeTturatio n ~ ~ SupersoluNlity Solubility k ~ '~ % CUrVL~il e 2r. [Macromolecule] Fro. 1. Phase diagram. The solubility curve (solid) divides phase space into regions that support crystallization processes (supersaturated solutions) from those where crystals will dissolve (unsaturated solutions). The supersolubility curve (dashed) further divides the super- saturated region into higher supersaturation conditions where nucleation and growth compete (labile phase) and lower levels where only crystal growth will occur (metastable phase). As shown in the diagram, at higher macromolecular concentrations, supersaturation occurs at lower values of parameters that decrease protein solubility. For example, when crystals grow from salt solutions, less salt is required when higher protein concentrations are used. Con- versely, higher salt concentrations (greater values of this solubility-decreasing parameter) are needed at lower macromolecular concentrations. Ultimately, as diagrammed in Fig. 1, directional changes in solution parame- ters that result in decreased solubility aid the protein crystallization process. The supersaturation requirements for nucleation and crystal growth differ. This is shown on the phase diagram where the supersaturation region is further divided into regions of higher supersaturation (the labile region) where both growth and nucleation occur, and lower supersaturation (the metastable phase) where only growth is supported. The remainder of this article discusses how nucleation and growth conditions are achieved using common macromolecular crystallization methods. The aim is to provide a practical guide for manipulating crystallization conditions to obtain large and well-ordered crystals. Simplest Crystallization Method In batch crystallization methods, all components are combined into a single solution, which is then left undisturbed (Fig. 2). This simple method works well for hen egg white lysozyme, catalase, 3 and cytochrome C554 .4 3 W. Longley, J. Mol. Biol. 30, 323-327 (1967). 4 C. Nagata, N. Igarashi, H. Moriyama, T. Fugiwara, Y. Fukumori, and N. Tanaka, J. Biochem. 117, 931-932 (1995).['~,,] PROTEIN CRYSTALLIZATION METHODS 15 Batch Crystallization Initial Final E Conditions Conditions ~_ Microbatch Crystallization ~ a Initial Final Conditions Conditions Supersaturation Unsaturation [Macromolecule] FIG. 2. Schematic diagram of batch crystallization. Left: Batch and microbatch crystalliza- tion experiments. In batch experiments, vials containing supersaturated protein solutions are sealed and left undisturbed. In microbatch methods, a small (2-10/zl) droplet containing both protein and precipitant is immersed in an inert oil.which prevents droplet evaporation. Right: Circles on the phase diagrams shown in this and subsequent figures indicate the levels of supersaturation that occur in the various crystallization solutions. In batch methods, the initial solution is located within the labile region of the phase diagram (solid circle). Depending on the concentration of the solution following completion of crystallization, the equilibrium solution concentration of protein is likely to have decreased so that the solution is now within the metastable region (open circle). The technique can be miniaturized by immersing protein droplets as small as 1/.d into an inert oil. 5 Given the static nature of the batch crystallization experiment, success requires that supersaturation levels sufficient for nucleation be achieved on mixing (Fig. 2). Optimization then involves altering experimental condi- tions to control the number of crystals and the time required for crystals to reach the desired size. Crystal number reflects nucleation rate, which in general is strongly