Protein crystallization in confined geometries

We studied the crystallization of a globular protein, lysozyme, in the cubic phase of the lipid monoolein. The solubility of lysozyme in salt solution decreased by a factor of ∼ 4 when confined in cubic phase. Calculations and Monte Carlo simulations show that this can be explained by the confinement of lysozyme molecules to the narrow water cells in the cubic phase. We report a study of the crystallization of lysozyme, a globular protein, in the cubic phase of monoolein, an uncharged lipid. The protein solubility was found to decrease by a factor of ∼ 4. We account for this observation by the confinement of lysozyme molecules within nanoscopic water 'cells' in the cubic phase. The resulting entropy-driven rise in their chemical potential can explain our observation of solubility suppression compared to the lipid-free case. The generic nature of this explanation means that our results are of potential interest to three rather different audiences. First, and most generally , scientists interested in the basic physics of confinement [1] may find the lysozyme-monoolein system a useful model for detailed study. Secondly, there is growing interest in directing the self-assembly of synthetic or natural nanoparticles using various means of confinement (e.g. at an interface [2] or in emulsion droplets [3]). Our work suggests that lipid cubic phases can be useful in this regard. Thirdly, and most specifically, lipid cubic phases offer an environment for globular protein crystal-lization [4]. (Their use for crystallizing membrane proteins [5], which reside in the lipid bilayers, relies on different physics [6, 7].) Lysozyme crystallization in lipid cubic phase has been reported before [8], although to date there is no complete phase diagram or physical explanation of the mechanism. We have studied the phase behavior of lysozyme solution as a function of protein and salt concentration in a monoolein cubic phase, and compared it with the case without lipids. The cubic phase was found to lower the solubility dramatically. This result was then explored using simple calculations and Monte Carlo simulations. Lysozyme and monoolein (MO) were used as purchased from Sigma-Aldrich. Protein concentration in 0.05 M sodium acetate buffer was determined, after filtration through a 0.1 µm Millipore filter, by UV spectropho-tometry (specific absorbance A 280nm = 2.64 ml/mg.cm). NaCl solutions of known concentration were then added. We first determined the bulk solubility boundary of lysozyme/salt without lipid from the dissolution of 'seed' crystals [4], which occurs at …