Two-dimensional crystallization of proteins on lipid monolayers at the air–water interface and transfer to an electron microscopy grid

Abstract

The two-dimensional (2-D) crystallization of proteins on lipid monolayers at the air–water interface is a well established method for crystallizing soluble proteins. The transfer of 2-D crystals from the air–water interface to an electron microscopy (EM) grid constitutes a critical and ill-controlled step in the whole procedure, which is likely to be responsible for the high variability of results obtained with this method. In this paper, we address the following questions: (1) does the material observed on EM grids constitute a true representation of the material present at the air–water interface? (2) is there an optimal method of transfer to obtain well-ordered protein 2-D crystals? To answer these questions, we combine data obtained on three different protein systems, annexin V, streptavidin and cholera toxin, using two types of EM grids, coated with either holey carbon films or continuous carbon films. These combined observations help us draw a coherent picture of the state of the interfacial films at the air–water surface and provide new insight into the perturbing influence of the transfer step. The main conclusions are: (1) both annexin V and streptavidin form crystalline monolayers at the air–water interface, which are well preserved when transfer is performed by means of holey carbon films; (2) a major reorganization of the material present at the water surface accompanies transfer with continuous carbon films; the basal monolayer is extensively damaged, transforming into domains and vesicular structures, which do not pre-exist at the water surface; with the three protein systems studied here, these domains are often crystalline; (3) the most striking structural reorganization induced by transfer with continuous carbon films is observed with annexin V, for which the native p6 crystalline assembly is transformed into another crystal form more ordered, with p3 symmetry. It is most probable that these conclusions also apply to other protein 2-D crystals formed by the lipid monolayer method. The recent in situ observation of 2-D crystals of annexin V formed on solid-supported bilayers, by atomic force microscopy, supports our interpretation that monolayers transferred with holey carbon films represent the genuine material pre-exisiting at the air–water interface.