How to make tubular crystals by reconstitution of detergent-solubilized Ca2(+)-ATPase

2D crystallization of membrane proteins: rationales and examples

The difficulty in crystallizing channel proteins in three dimensions limits the use of X-ray crystallography in solving their structures. In contrast, the amphiphilic character of integral membrane proteins promotes their integration into artificial lipid bilayers. Protein-protein interactions may lead to ordering of the proteins within the lipid bilayer into two-dimensional crystals that are amenable to structural studies by electron crystallography and atomic force microscopy. While reconstitution of membrane proteins with lipids is readily achieved, the mechanisms for crystal formation during or after reconstitution are not well understood. The nature of the detergent and lipid as well as pH and counter-ions is known to influence the crystal type and quality. Protein-protein interactions may also promote crystal stacking and aggregation of the sheet-like crystals, posing problems in data collection. Although highly promising, the number of well-studied examples is still too small to draw conclusions that would be applicable to any membrane protein of interest. Here we discuss parameters influencing the outcome of two-dimensional crystallization trials using prominent examples of channel protein crystals and highlight areas where further improvements to crystallization protocols can be made.

One-headed kinesin derivatives move by a nonprocessive, low-duty ratio mechanism unlike that of two-headed kinesin

A single molecule of the "two-headed" motor enzyme kinesin can move along a microtubule continuously for many enzymatic turnovers (processive movement), and the velocity produced by one kinesin molecule is the same as that produced by many kinesin molecules (high duty ratio). We studied the microtubule movement driven at 1 mM ATP by biotinated N-terminal fragments of Drosophila kinesin heavy chain attached to streptavidin-coated coverslips at various surface densities. K448-BIO has velocity at a high density of vmax = 750 nm s-1 and is dimeric (hence two-headed); K365-BIO (vmax = 200 nm s-1) and K340-BIO (vmax = 90 nm s-1) are monomeric. Escape of microtubules from the surface was prevented by methylcellulose so that continuous trajectories of microtubules not continuously attached to motor molecules could be recorded by video microscopy. The component of instantaneous velocity parallel to the microtubule axis (v) was analyzed in trajectories with a mean velocity 0.4-0.7 times vmax. In K448-BIO trajectories, the distribution of v was bimodal with peaks near 0 and 750 nm s-1. Temporal autocorrelation analysis of v detected lengthy episodes of high-velocity movement consistent with isolated processive microtubule runs driven at vmax by single K448-BIO dimers. K365-BIO and K340-BIO trajectories had unimodal distributions of v and autocorrelation times much shorter than those for K448-BIO. Therefore the monomeric motors have duty ratio < 55% (i.e., no forward movement is generated for at least 45% of the enzymatic cycle time) or processivity below the detection limit of approximately 300 turnovers even in methylcellulose. Continuous movement at maximal velocity thus requires more than one kinesin head.

Projection map of cytochrome b6 f complex at 8 A resolution

The structure of the cytochrome b6 f complex has been investigated by electron microscopy and image analysis of thin three-dimensional crystals. Electron micrographs of negatively stained specimens were recorded and showed optical diffraction peaks to 10 A resolution. A projection map was calculated at 8 A resolution and showed the presence of cytochrome b6 f dimers. The extramembrane part of each monomer featured a C shape, with mean external diameter approximately of 53 A and an internal groove approximately 14 A long and approximately 9 A wide. Within each monomer, strong features were clearly resolved and tentatively attributed to some of the subunits of the cytochrome b6 f complex. The data are consistent with the Rieske iron-sulfur protein lying close to the monomer-monomer interface and the heme-bearing domain of cytochrome f far from it.

Electron and atomic force microscopy of membrane proteins

Electron crystallography is becoming a powerful tool for the resolution of membrane protein structures. The past year has seen the production of a bacteriorhodopsin model at 3.5 A and the structure of aquaporin 1 approaching atomic resolution. Determination of surface topographies of 2D crystals using the atomic force microscope is similarly advancing to a level that reveals submolecular details. As the latter is operated in solution, membrane proteins can be observed at work.