It is well known that the crystallization of proteins is strongly dependent on the crystallization conditions, which are sometimes very sensitive to environmental disturbances. Parameters such as the concentration of precipitants or protein, pH, temperature and many others are known to affect the probability of crystallization, and the task of crystallizing a new protein often involves a trial-and-error test using numerous combinations of crystallization conditions. These crystallization parameters, such as the concentration of either the protein or the precipitant, are important because they directly affect the driving force of crystallization: the supersaturation of the solution. Although it is common sense that the concentration can affect the crystallization process, the sensitivity of the crystallization process to variations in the concentration has seldom been addressed. Owing to the difficulty of directly preparing solutions with very small concentration variations, it is hard to carry out an investigation of their effect on the crystallization process. In this paper, a simple but novel method for studying the effect of minute concentration variations on the success rate of protein crystallization is presented. By evaporating the crystallization droplet, a fine concentration gradient could be created. With this fine-tuned concentration gradient, it was possible to observe the effects of minute variations in the concentration or supersaturation on the crystallization. A very minor change in concentration (as low as 0.13% of the initial concentration, i.e. 0.026 mg ml(-1) for lysozyme and 0.052 mg ml(-1) for NaCl in the current study) or a very minor change in supersaturation (as small as 0.018) could cause a clear difference in the crystallization success rate, indicating that the crystallization of proteins is very sensitive to the concentration level. Such sensitive behaviour may be one reason for the poor reproducibility of protein crystallization.
© 2012 International Union of Crystallography