Imaging the membrane protein bacteriorhodopsin with the atomic force microscope.

The membrane protein bacteriorhodopsin was imaged in buffer solution at room temperature with the atomic force microscope. Three different substrates were used: mica, silanized glass and lipid bilayers. Single bacteriorhodopsin molecules could be imaged in purple membranes adsorbed to mica. A depression was observed between the bacteriorhodopsin molecules. The two dimensional Fourier transform showed the hexagonal lattice with a lattice constant of 6.21 +/- 0.20 nm which is in agreement with results of electron diffraction experiments. Spots at a resolution of approximately 1.1 nm could be resolved. A protein, cationic ferritin, could be imaged bound to the purple membranes on glass which was silanized with aminopropyltriethoxysilane. This opens the possibility of studying receptor/ligand binding under native conditions. In addition, purple membranes bound to a lipid bilayer were imaged. These images may help in interpreting results of functional studies done with purple membranes adsorbed to black lipid membranes.     

Molecular motions and hydration of purple membranes and disk membranes studied by neutron scattering

Fast stochastic equilibrium fluctuations (time scale: 10^–10–10^–13 seconds) in purple membranes (PM) and in disk membranes (DM) have been measured with quasielastic incoherent neutron scattering. The comparison of predominantly stochastic motions occurring in purple membranes and in disk membranes revealed qualitatively similar dynamical behaviour. Models of internal motions within restricted volumes have been shown to be useful to fit the spectra from both samples. From fits using these models we found “amplitudes” 15 to 20% larger for motions in DM samples compared to PM samples. This indicates a higher internal flexibility of the DM. Because the dynamical behaviour is very sensitive to the hydration of the protein-lipid complex, we also performed neutron diffraction experiments to determine lamellar spacings as a measure of level of hydration and as a function of temperature. From these studies the interaction of solvent molecules with the surface of the protein-lipid complex appears to be qualitatively similar for both types of membranes. Received: 12 February 1998 / Revised version: 18 March 1998 / Accepted: 27 March 1998     

Structure determination of asymmetric membrane profiles using an iterative Fourier method.

An iterative Fourier method is applied to solving and refining the electron density profile projected into the line perpendicular to a membrane surface. Solutions to the continuous X-ray scattering pattern derived from swelling of multilayer systems or from membrane dispersions can be obtained by this technique. The method deals directly with the observed structure factors and does not rely on deconvolution of the Patterson function. We used this method previously to derive the electron density profile for acetylcholine receptor membranes (Ross et al., 1977). The present paper is an analysis of the theoretical basis for the procedure. In addition, the technique is tested on artificially generated continuous-scattering data, on the data for frog sciatic nerve myelin derived from swelling experiments by Worthington and McIntosh (1974), and on the data for purple membrane (Blaurock and Stoeckenius, 1971). Although the method applies to asymmetric membranes, the special case of centrosymmetric profiles is also shown to be solvable by the same technique. The limitations of the method and the boundary conditions that limit the degeneracy of the solution are analyzed.     

The distribution of tetramethylammonium ions on the surface of purple membranes

The two-dimensional distribution of deuterated tetramethylammonium (TMA⁺) ions on the surface of purple membranes of Halobacterium halobium was determined by neutron diffraction. The measurements were performed on stacks of these membranes with a high concentration of TMA⁺ molecules in the water layer between the membranes. A difference Fourier analysis of samples with deuterated and undeuterated ions showed an excess of 8.5 TMA⁺ ions per elementary cell in the lipid areas compared to the protein areas. A total number of 90 ions per elementary cell in the intermembrane space was estimated from the preparation procedure. The excess in the lipid domains may result from the higher affinity of TMA⁺ ions for the lipid head groups and/or from the fact that the protein (bacteriorhodopsin) protrudes slightly out of the lipid surface.Abbreviations BR bacteriorhodopsin - TMA Tetramethylammonium     

Moist and soft, dry and stiff: a review of neutron experiments on hydration-dynamics-activity relations in the purple membrane of Halobacterium salinarum

Twenty-five years of neutron experiments on hydration and thermal dynamics in purple membranes of Halobacterium salinarum are reviewed. Neutron diffraction, elastic and quasielastic scattering, allowed to map the distribution of water and lipids and to measure thermal fluctuations and correlation times in the membranes, under various conditions of temperature, hydration and lipid environment. Strong correlations were established between dynamics parameters and the activity of bacteriorhodopsin (the purple membrane protein), as a light driven proton pump supporting the hypothesis that the influence of hydration on activity is in fact due to its effects on membrane thermal dynamics. Hydrogen-deuterium labelling experiments highlighted stiffer and softer parts in the bacteriorhodopsin structure. The soft parts would allow the conformational changes involved in activity, while the stiffer ones may control a valve-like function in vectorial proton transfer.     

Structure and hydration of purple membranes in different conditions

The unit cell dimension of the bacteriorhodopsin lattice in purple membranes decreases by the same amount (2%) upon drying the membranes at room temperature as when they are cooled to liquid nitrogen temperatures. Neutron diffraction experiments with H2O:2H2O exchange, however, show that whereas in the dry membranes the lipid headgroups are dehydrated and the decrease in dimension is due to a smaller area occupied by the lipid molecules, the water of hydration remains in place in the cooled membranes, and the decrease in dimension is due to thermal contraction only. These data suggest a hypothesis that functional bacteriorhodopsin, in the wet state at room temperature, has a relatively soft environment that would allow large amplitude motions of the protein; in the dry membranes at room temperature (which are inactive), the amplitudes of protein motions would be inhibited by a more close-packed environment as they are reduced, due to thermal contraction, in the cold membranes.     

Analysis of high-resolution electron diffraction patterns from purple membrane labelled with heavy-atoms

Progress in the structure determination of bacteriorhodopsin, the protein component of purple membrane from Halobacterium halobium has been limited by the lack of three-dimensional phase information between 6 and 3 A resolution. By analogy with X-ray methods, it is possible that heavy-atom labelling of the membrane crystal may provide heavy-atom derivatives that can be used for phasing by the multiple isomorphous replacement method. This paper describes the screening of heavy-atom compounds as potential derivatives, and the evaluation of the data collected from these heavy-atom-labelled membranes. Improvements in the methods for collecting electron diffraction data and analysing and merging the data are presented. Diffraction patterns of purple membrane samples were taken at -120 degrees C to minimize radiation damage. About 30 heavy-atom compounds were tested for use as potential derivatives. The diffraction patterns from labelled membranes were analysed by examining 6.5 A difference Fourier maps. Two heavy-atom compounds were selected for three-dimensional data collection at 3 A resolution. In addition, a full set of native data at -120 degrees C was collected to 2.7 A resolution. The intensity merging, heavy-atom derivative evaluation, heavy-atom refinement and the calculation of phases are presented. Phases are compared to those determined by electron microscope imaging, and limitations of the method are discussed. It is concluded that, with the present accuracy of data collection and the present magnitude of delta F/F available for the derivatives, the phasing power is too small. The phases that are obtained are not sufficiently accurate to provide a reliably interpretable map. It may be possible, however, to use the heavy-atom derivative data in difference Fourier calculations in which the presence of a peak would confirm the phases calculated from a model or obtained by electron microscope imaging.     

X-ray diffraction reconstruction of the inverted hexagonal (HII) phase in lipid-water systems.

The structure of the inverted hexagonal (HII) phase in biological lipid-water systems is studied to examine the physical interactions which drive the polymorphic phase behavior and which are also thought to play a relevant role in biological membrane function. A method is derived which yields the complex phase factors of the HII phase diffraction amplitudes from examination of a single sample. This method is applied to a low-resolution Fourier reconstruction of the HII phase in dioleoylphosphatidylethanolamine (DOPE) + water, specifically to examine deviations from the presumed circular model of the HII phase. It is found that the average radius of the water core, Rw, as determined from a Fourier reconstruction, is in good agreement with previously measured values of Rw obtained from more time-consuming traditional methods [Tate, M. W., & Gruner, S. M. (1989) Biochemistry 28, 4245]. In addition to the average value of Rw, the Fourier reconstruction also can be used to determine the true shape of the water core. It is found that the water core is circular to within 5% of Rw when the unit cell size is less than approximately 75 A. Above 75 A, however, a definite shape deformation becomes apparent, with radial noncircularities of 5-10%, probably in response to the increased entropic cost of packing the hydrocarbon chains into the anisotropic environment of the HII unit cell [Kirk, G. L., Gruner, S. M., & Stein D. E. (1984) Biochemistry 23, 1093]. As a more direct probe of the packing anisotropy, Fourier reconstructions of DOPE + dodecane and DOPE + squalene systems were compared with the reconstruction of DOPE. These oils are known to promote the low temperature occurrence of the HII phase, presumably by a reduction in the hydrocarbon packing stress. In support of this hypothesis, the alkanes were observed to relax the water core to a circular shape for even large lattices. In addition, anisotropy of the electron density near the end of the lipid chains is reduced when alkane is added, implying a more uniform hydrocarbon packing environment, consistent with the results of neutron diffraction upon the addition of deuterated decane [Turner, D. C., Gruner, S. M., & Huang, J. (1992) Biochemistry (following paper in this issue)].     

Water molecules and exchangeable hydrogen ions at the active centre of bacteriorhodopsin localized by neutron diffraction. Elements of the proton pathway?

Neutron diffraction is used to localize water molecules and/or exchangeable hydrogen ions in the purple membrane by H2O/2H2O exchange experiments at different values of relative humidity. At 100% relative humidity, differences in the hydration between protein and lipid areas are observed, accounting for an excess amount of about 100 molecules of water in the lipid domains per unit cell. A pronounced isotope effect was observed, reproducibly showing an increase in the lamellar spacing from 60 A in 2H2O to 68 A in H2O. At 15% relative humidity, the positions of exchangeable protons became visible. A dominant difference density peak corresponding to 11 +/- 2 exchangeable protons was detected in the central part of the projected structure of bacteriorhodopsin at the Schiff's base end of the chromophore. A difference density map obtained from data on purple membrane films at 15% relative humidity in 2H2O, and the same sample after complete drying in vacuum, revealed that about eight of these protons belong to four water molecules. This is direct evidence for tightly bound water molecules close to the chromophore binding site of bacteriorhodopsin, which could participate in the active steps of H+ translocation as well as in the proton pathway across this membrane protein.     

Tertiary structure of bacteriorhodopsin. Positions and orientations of helices A and B in the structural map determined by neutron diffraction

Positions and rotations of two helices in the tertiary structure of bacteriorhodopsin have been studied by neutron diffraction using reconstituted, hybrid purple membrane samples. Purple membrane was biosynthetically 2H-labeled at non-exchangeable hydrogen positions of leucine and tryptophan residues. Two chymotryptic fragments were purified, encompassing either the first two or the last five of the seven putative transmembrane segments identified in the amino acid sequence of bacteriorhodopsin. The 2H-labeled fragments, diluted to variable extents with the identical, unlabeled fragment, were mixed with their unlabeled counterpart; bacteriorhodopsin was then renatured and reconstituted. The crystalline purple membrane samples thus obtained contained hybrid bacteriorhodopsin molecules in which certain transmembrane segments had been selectively 2H-labeled to various degrees. Neutron diffraction powder patterns were recorded and analyzed both by calculating difference Fourier maps and by model building. The two analyses yielded consistent results. The first and second transmembrane segments in the sequence correspond to helices 1 and 7 of the three-dimensional structure, respectively. Rotational orientations of these two helices were identified using best fits to the observed diffraction intensities. The data also put restrictions on the position of the third transmembrane segment. These observations are discussed in the context of folding models for bacteriorhodopsin, the environment of the retinal Schiff base, and site-directed mutagenesis experiments.     

Sulfur distribution in bacteriorhodopsin from multiple wavelength anomalous diffraction near the sulfur K-edge with synchrotron x-ray radiation.

Bacteriorhodopsin contains nine sulfur atoms from the nine methionine residues. The distribution of these sulfur atoms in the projected density map was determined from x-ray diffraction experiments using multiple wavelength anomalous diffraction (MAD) at the sulfur K-edge (5.02 A) with synchrotron radiation. The experiments were performed with uniaxial samples of oriented purple membranes at room temperature and 86% relative humidity. For such samples only the real part f' (lambda) of the resonant scattering amplitude of sulfur contributes to the observed scattering intensity. The sulfur density was determined from the difference in diffraction intensities detected at two wavelengths near the sulfur K-edge that were approximately 0.004 A apart. The measured change in f' between these two wavelengths corresponds to 6 electron units. This shows that large anomalous dispersion effects occur near the sulfur K-edge. The in-plane positions of the sulfur atoms of Met32, Met56, and Met209 were determined unambiguously. The difference density from Met20, Met60, Met118, and Met145 is concentrated in the interior of the seven alpha-helical bundle, overlaps strongly in the projected density map, and cannot be resolved at the resolution of these experiments (8.2 A). This method of localizing individual sulfur atoms can be applied to other two-dimensional protein crystals and is promising in conjunction with the site-directed introduction of sulfur atoms by the use of cysteine mutants.     

The location of retinal in the purple membrane profile by neutron diffraction.

The trans-membrane location of retinal in the purple membrane of Halobacterium halobium, has been determined by low-angle neutron scattering studies on aqueous dispersions of the membranes. The membrane was bleached and regenerated with deuterated and with hydrogen-containing retinal. The modified retinal was obtained by extraction from bacteria grown in a totally deuterated medium. The determination of the retinal position is based on the differences in neutron scattering between a purple membrane sample with normal, protonated retinal and another sample with deuterated retinal. A distinct scattering density difference between the two preparations was observed. A direct structure determination was used with the retinal localized from a Fourier difference density profile. We conclude that the β-ionone ring portion of the retinal is situated centrally in the membrane.     

Can the low-resolution structures of photointermediates of bacteriorhodopsin explain their crystal structures?

To understand the molecular mechanism of light-driven proton pumps, the structures of the photointermediates of bacteriorhodopsin have been intensively investigated. Low-resolution diffraction techniques have demonstrated substantial conformational changes at the helix level in the M and N intermediates, between which there are noticeable differences. The intermediate structures at atomic resolution have also been solved by x-ray crystallography. Although the crystal structures have demonstrated local structural changes, such as hydrogen bond network rearrangements including water molecules, the large conformational changes at the helix level are not necessarily observed. Furthermore, the two reported crystal structures of an intermediate accumulated using a common method were distinct. To reconcile these apparent discrepancies, low-resolution projection maps were calculated from the crystal structures and compared to the low-resolution intermediate structures obtained using native membranes. The crystal structures can be categorized into three groups, which qualitatively correspond to the low-resolution structures of the M1-type, M2-type, and N-type determined in the native membrane. Based on these results, we conclude that at least three types of intermediate structures play a role during the photocycle.     

X-ray diffraction from a single layer of purple membrane at the air/water interface

X-ray diffraction patterns have been recorded from a single layer of purple membrane ( approximately 50 A thickness) at the air/water interface in a Langmuir trough. Grazing-incidence X-ray diffraction is demonstrated to be a promising method for obtaining structural information on membrane proteins under physiological conditions. The method is so sensitive that diffraction can be measured from samples with only 10(13) protein molecules in the beam. Diffraction from hexagonal crystals of purple membrane with a lattice constant of 61. 3 A was observed up to the order {h,k}={4,3}, corresponding to a resolution of approximately 9 A. The work reported here is a first step towards a new way of protein crystallography using grazing-incidence X-ray diffraction at the air/water interface.     

Localization and orientation of functional water molecules in bacteriorhodopsin as revealed by polarized Fourier transform infrared spectroscopy.

Linear dichroic difference Fourier transform infrared spectra upon formation of the M photointermediate were recorded with oriented purple membranes. The purpose was to determine the angle of the directions of the dipole moments of 1) the water molecule whose O-H stretching vibration appears at 3643 cm-1 for the unphotolyzed state and 3671 cm-1 for the M intermediate, and 2) the C=O bond of protonated Asp85 in the M intermediate. The angle of 36 degrees we find for the C=O of the protonated Asp85 in the M intermediate is not markedly different from 26 degrees for unprotonated Asp85 in the model based on cryoelectron diffraction, indicating the absence of gross orientation changes in Asp85 upon its protonation. The O-H band at 3671 cm-1 of a water molecule in the M intermediate, although its position has not determined, is fixed almost parallel to the membrane plane. For the unphotolyzed state the angle of the water O-H to the membrane normal was determined to be 60 degrees. On the basis of these data and the structural model, we place the water molecule in the unphotolyzed state at a position where it forms hydrogen bonds with the Schiff base, Asp85, Asp212, and Trp86.     

A fourier tool for the analysis of coherent light scattering by bio-optical nanostructures

The fundamental dichotomy between incoherent (phase independent) and coherent (phase dependent) light scattering provides the best criterion for a classification of biological structural color production mechanisms. Incoherent scattering includes Rayleigh, Tyndall, and Mie scattering. Coherent scattering encompasses interference, reinforcement, thin-film reflection, and diffraction. There are three main classes of coherently scattering nanostructures-laminar, crystal-like, and quasi-ordered. Laminar and crystal-like nanostructures commonly produce iridescence, which is absent or less conspicuous in quasi-ordered nanostructures. Laminar and crystal-like arrays have been analyzed with methods from thin-film optics and Bragg's Law, respectively, but no traditional methods were available for the analysis of color production by quasi-ordered arrays. We have developed a tool using two-dimensional (2D) Fourier analysis of transmission electron micrographs (TEMs) that analyzes the spatial variation in refractive index (available from the authors). This Fourier tool can examine whether light scatterers are spatially independent, and test whether light scattering can be characterized as predominantly incoherent or coherent. The tool also provides a coherent scattering prediction of the back scattering reflectance spectrum of a biological nanostructure. Our applications of the Fourier tool have falsified the century old hypothesis that the non-iridescent structural colors of avian feather barbs and skin are produced by incoherent Rayleigh or Tyndall scattering. 2D Fourier analysis of these quasi-ordered arrays in bird feathers and skin demonstrate that these non-iridescent colors are produced by coherent scattering. No other previous examples of biological structural color production by incoherent scattering have been tested critically with either analysis of scatterer spatial independence or spectrophotometry. The Fourier tool is applied here for the first time to coherent scattering by a laminar array from iridescent bird feather barbules (Nectarinia) to demonstrate the efficacy of the technique on thin films. Unlike previous physical methods, the Fourier tool provides a single method for the analysis of coherent scattering by a diversity of nanostructural classes. This advance will facilitate the study of the evolution of nanostructural classes from one another and the evolution of nanostructure itself. The article concludes with comments on the emerging role of photonics in research on biological structural colors, and the future directions in development of the tool.     

Ab initio phasing of high-symmetry macromolecular complexes: successful phasing of authentic poliovirus data to 3.0 A resolution

A genetic algorithm-based computational method for the ab initio phasing of diffraction data from crystals of symmetric macromolecular structures, such as icosahedral viruses, has been implemented and applied to authentic data from the P1/Mahoney strain of poliovirus. Using only single-wavelength native diffraction data, the method is shown to be able to generate correct phases, and thus electron density, to 3.0 A resolution. Beginning with no advance knowledge of the shape of the virus and only approximate knowledge of its size, the method uses a genetic algorithm to determine coarse, low-resolution (here, 20.5 A) models of the virus that obey the known non-crystallographic symmetry (NCS) constraints. The best scoring of these models are subjected to refinement and NCS-averaging, with subsequent phase extension to high resolution (3.0 A). Initial difficulties in phase extension were overcome by measuring and including all low-resolution terms in the transform. With the low-resolution data included, the method was successful in generating essentially correct phases and electron density to 6.0 A in every one of ten trials from different models identified by the genetic algorithm. Retrospective analysis revealed that these correct high-resolution solutions converged from a range of significantly different low-resolution phase sets (average differences of 59.7 degrees below 24 A). This method represents an efficient way to determine phases for icosahedral viruses, and has the advantage of producing phases free from model bias. It is expected that the method can be extended to other protein systems with high NCS.     

Net Electric Charge on Photopigment Molecules and Frog Retinal Receptor Disk Membrane Structure

The photopigment molecules in frog retinal receptor disk membranes protude some 50-65% of their molecular diameter (~42 A) into the aqueous surface layer of the disk membrane, depending on whether the photopigment is bleached, while the remainder is embedded in the lipid core of the membrane. In order to determine whether the presumably polar groups covering this surface protruding into the aqueous phase possessed net electric charge, we collected X-ray diffraction data from the photopigment molecules in wet pellets of oriented disk membranes as a function of the pH and ionic strength of the sedimentation medium. The Fourier analysis applied to this data provided average nearest neighbor separations for the photopigment molecules for their planar arrangement in the disk membranes. Changes in the average separation of photopigment molecule nearest neighbors as a function of pH, ionic strength, and photopigment bleaching indicated that photopigment molecules possess negative net electric charge, that this net electric charge occurs in the aqueous surface layer of the disk membrane, and that this net charge is reduced on photopigment bleaching. This polar portion of the photopigment molecule may thereby determine the location of the photopigment molecules relative to the lipid core and other photopigment molecules in the disk membrane. In addition, the orientation (dichroism) of the photopigment relative to an axis normal to the plane of the disk membrane and the bleaching-dependent “sinking” of the photopigment molecule into the lipid core of the disk membrane may be accounted for.     

The susceptibility of pure tubulin to high magnetic fields: a magnetic birefringence and x-ray fiber diffraction study.

The orientational behavior of microtubules assembled in strong magnetic fields has been studied. It is shown that when microtubules are assembled in a magnetic field, they align with their long axis parallel to the magnetic field. The effect of several parameters known to affect the microtubule assembly are investigated with respect to their effect on the final degree of alignment. Aligned samples of hydrated microtubules suitable for low-resolution x-ray fiber diffraction experiments have been produced, and the results obtained from the fiber diffraction experiments have been compared with the magnetic birefringence experiments. Comparisons with earlier fiber diffraction work and small-angle x-ray solution scattering experiments have been made.     

Determining bilayer hydrocarbon thickness from neutron diffraction measurements using strip-function models.

Neutron diffraction methods provide information about the distribution of matter in biological and model membrane systems. The information is derived from plots (profiles) of scattering length density along an axis normal to the membrane plane. Without the use of specific deuteration, the generally low resolution of the profiles limits their interpretation in terms of specific chemical constituents (e.g., lipid headgroup, lipid hydrocarbon, protein, and water). A fundamental and useful structural assignment to make is the boundary between the headgroup and hydrocarbon regions of bilayers. We demonstrate here that strip-function model representations of neutron scattering length density profiles of bilayers are sufficient to determine accurately the position of the headgroup-hydrocarbon boundary. The resulting hydrocarbon thickness of the bilayer is useful for determining the area per lipid molecule and consequently the molecular packing arrangements of the membrane constituents. We analyze data obtained from dioleoylphosphatidylcholine (DOPC) bilayers at 66% RH using standard Fourier profile analyses and from DOPC deuterated specifically at the C-2 carbon of the acyl chains using difference Fourier analysis. We demonstrate that strip-function models accurately define the positions of the C-2 carbons and thus the hydrocarbon thickness (dhc) of the bilayer. We then show, using quasi-molecular models, that the strip-model analysis probably provides an accurate measure of dhc because of the exceptionally high scattering length density difference between the carbonyl and methylene groups.