Search

Our experience in the growth of two-dimensional crystals of different proteins is presented. Polyethylene glycol was used to produce two-dimensional arrays of haemocyanin from O. vulgaris and of cholera toxin. The arrays showed a hexagonal close-packed structure of only randomly oriented molecules. The increase in protein concentration probably occurred too quickly to allow complete crystallization. Different two-dimensional arrays of hexameric haemocyanin molecules (from P. interruptus) were obtained by microdialysis through the specimen supporting film. A comparison was made with X-ray data. Two-dimensional tetrameric arrays of molecules, possibly rhodopsin, were seen in samples of bovine retinal rod outer segments in the presence of ammonium sulphate. Two-dimensional crystals of complex I (from bovine mitochondria) were prepared by dialysis in the presence of ammonium sulphate. A three-dimensional reconstruction was made from two tilt-series by computer filtration using the direct SIRT procedure. Finally, the possibility of computer crystallization using correlation techniques in combination with correspondence analysis is discussed.

We investigated the structure of two-dimensional crystals from bovine heart mitochondrial NADH: ubiquinone oxidoreductase. A detailed description of uranyl acetate-stained crystals demonstrated that they are composed of fragments in a spatial arrangement according to space group P4212 [J. Brink, S. Hovmöller, C.I. Ragan, M.W.J. Cleeter, E.J. Boekema and E.F.J. van Bruggen, European J. Biochem. 166 (1987) 287]. To gain more structural information on the crystal structure and to assess the effects of various negative stains on the structure preservation and appearance, we examined stained crystals by means of electron microscopy and image analysis. The space group P4212 appeared to be present for several stains tested, i.e. ammonium molybdate, uranyl acetate, uranyl nitrate and uranyl sulphate. Use of phosphotungstic acid and silicotungstate resulted in a reduction of symmetry to pseudo-P4212 or p4. Use of sodium tungstate led to a considerable loss of resolution to 3.8 nm at best, whereas otherwise 1.5 to 1.9 nm could be demonstrated. The lattice vectors were not affected by the stains; they were determined as a = b = 14.9 +/- 0.25 nm with gamma = 89.8 degrees +/- 0.6 degrees. Image analysis showed the presence of similar structures with the molybdate and uranyl compounds. Differences were observed in the case of the tungstate type of stains. Furthermore, the analysis revealed the complete absence of the four small pores of 2.0 nm diameter in the unit cell. This effect was observed irrespective of the type of stain and supporting film, and could be ascribed only to the glow-discharge treatment of the supporting film. The observed difference must be caused by changed interactions between the protein, stain and supporting film. Application of correspondence analysis and clustering algorithms to the various reconstructed images of the crystals showed that they could be separated into several clusters. Each of these clusters corresponded on the average to only one type of stain, whereas a further division according to the specific uranyl compounds was observed. This study therefore shows that under identical preparation conditions subtle differences between individual stains can be detected.

Two-dimensional crystalline arrays of NADH:Q oxidoreductase preparations have been obtained by microdiffusion of protein dissolved in detergent against a 15 mM sodium acetate buffer of pH 5.5 containing 10% (w/v) ammonium sulphate. Electron microscopy was used to study the structure of negatively stained crystals. Computer-reconstructed images were obtained by the Fourier peak filtering method. The crystals have p4 symmetry and a square unit cell with dimensions of 15.2 +/- 0.5 nm. The four asymmetric units in the unit cell form a single tetrameric molecule with a dimension in the third direction of 8.2 nm. It is concluded on the basis of the estimated molecular mass that each tetramer cannot contain more than only one FMN molecule. This implies that the tetramers possibly are only a part of Complex I, since there is much evidence that one functional enzyme molecule of Complex I contains two FMN molecules.

The results of Section IV can be summarized in a simple ATP synthase model. This model implies that either the alpha or the beta subunits must be closer to the membrane. The work of Gao and Bauerlein (1987) indicates that the alpha subunits are closer to the membrane. Although the overall structure is more or less clear, important questions need to be clarified. First, the number and the arrangement of the subunits in the F0 part must be known. Second, the exact shape of F1, and particularly the shape of the large subunits needs to be elucidated. On the basis of fluorescence resonance energy transfer measurements by McCarty and Hammes (1987), a model was presented showing large oblong subunits. Such 'banana-shaped' subunits, which are also presented in the many phantasy models (e.g. Walker et al., 1982), are very unlikely in view of the electron microscopical results, although the large subunits do not need to be exactly spherical. The third and most interesting central question is on the changes in the structure that take place during the different steps in the synthesis of ATP. It can now be taken as proven that the energy transmitted to the ATP synthase is used to induce a conformational change in the latter enzyme, in such a way as to bring about the energy-requiring dissociation of already synthesized ATP (Penefsky, 1985 and reviewed in Slater, 1987). But the way in which the three parts of the ATP synthase are involved is completely unknown. It is rather puzzling that such a long distance exists between the catalytic sites, which are on the interface of the alpha and beta subunits and the F0 part where the proton movements occur, which, according to Mitchell's theory (1961), is the driving force for the synthesis of ATP. Perhaps alternative mechanisms such as the collision hypothesis formulated by Herweijer et al. (1985) are more realistic in describing the mechanism of ATP synthesis. It would bring the complexes I and V close together, not only in the artificial way treated in this paper, but in a useful way for energy conversion.

We have investigated the structure of two-dimensional crystals from preparations of NADH:ubiquinone oxidoreductase from beef-heart mitochondria. The crystal structure of these crystals was previously determined to be equivalent with two native enzyme molecules per unit cell, i.e. a p2 symmetry [Boekema, E. J., Van Heel, M. G. & Van Bruggen, E. F. J. (1984) Biochim. Biophys. Acta 787, 19-26]. However, the optical diffraction patterns of the crystals displayed a clear fourfold symmetry. A Fourier analysis carried out on the calculated diffraction pattern proved unambiguously that the crystal symmetry was p42(1)2. Following crystallographic rules the unit cell therefore contained eight identical molecules. As a consequence, only a subcomplex of the enzyme rather than the intact enzyme formed the crystal. Electron microscopy of isolated, single molecules of the iron-sulphur protein, a dissociation product of complex I, revealed the presence of square complexes with sides of approximately 15 nm. Since these complexes were indistinguishable from the building blocks (unit cells) of the two-dimensional crystals, the crystals could be composed of Fe-S protein fragments only. The nature of the fragments in the unit cell was probed by immuno-labelling with monovalent antibodies (Fab's), raised against the 75-kDa subunit from the Fe-S protein, followed by image analysis. We found at least four binding sites for the anti-(75-kDa subunit) Fab per unit cell, indicating the presence of at least four copies of the antigen. In order to account for these observations we postulate the hypothesis that the two-dimensional crystals obtained from complex I are composed of iron-sulphur protein molecules in an octameric arrangement.

We have studied the structure of bovine heart mitochondrial NADH:ubiquinone (Q) oxidoreductase (EC 1.6.99.3) by image analysis of electron micrographs. A three-dimensional reconstruction was calculated from a tilt-series of a two-dimensional crystal of the molecule. Our interpretation of the position of the molecule in the unit cell of the crystal is supported by additional (low-resolution) analysis of images of single molecules. The three-dimensional reconstruction was calculated with the aid of an iterative real-space reconstruction algorithm. The various projections used as input to the algorithm were obtained by averaging the images of the tilted crystal through a Fourier-space peak-filtering procedure. The reconstructed unit cell measures 15.2 X 15.2 nm in the plane of the two-dimensional crystal and has a height of 10-11 nm. The unit cell contains one molecule consisting of four large subunits. At the present resolution of about 1.3 nm in the untilted projection, these four monomers are seen as two dimers related by a two-fold axis. Two views of the single particles have been recognized; they are the top and side view of the building block of the crystal. After computer image alignment and correspondence analysis, clusters of similar particles have been averaged. In the averages an uneven stain distribution is seen around the molecules, which may result from preferential staining of hydrophilic parts of the molecule. The molecular mass of the whole molecule was determined from scanning transmission electron microscopy measurements as (1.6 +/- 0.2) X 10(6) daltons.

Two-dimensional crystals of bovine NADH: ubiquinone oxidoreductase fragments were studied by cryo-electron microscopy and computer image analysis at two focus settings. The lattice parameters of the crystals were determined as a = 14.5 +/- 0.2 nm, b = 14.8 +/- 0.3 nm with gamma = 87.8 degrees +/- 0.2 degrees. Translation and rotation alignment resulted in a final resolution of 1.38 (close-to-focus) or 1.9 nm (1.3 microns underfocus). When only translational alignment was employed 1.6 nm resolution resulted (close-to-focus case). The application of a further correction procedure surprisingly lowered the achieved resolution to 2.2 nm. A combination of the projected structures for the two focus settings in Fourier space resulted in a noise-free image displaying enhanced image contrast for reciprocal spacings from 1/5.0 to 1/1.4 nm-1. The structure in amorphous ice comprises four separate quasi-identical motifs, each resembling a distorted triangle. Two symmetry-related (p2) motifs reveal a pore-like feature. In negative stain the structure has p4mm symmetry. The absence of in-plane symmetry elements (i.e. twofold rotation and screw axes) and other differences might be accounted for (at least in part) if the crystal studied was slightly tilted.