Structural analysis of membrane protein complexes by single particle electron microscopy

Single particle electron microscopy (EM) is an increasingly important tool for the structural analysis of macromolecular complexes. The main advantage of the technique over other methods is that it is not necessary to precede the analysis with the growth of crystals of the sample. This advantage is particularly important for membrane proteins and large protein complexes where generating crystals is often the main barrier to structure determination. Therefore, single particle EM can be employed with great utility in the study of large membrane protein complexes. Although the construction of atomic resolution models by single particle EM is possible in theory, currently the highest resolution maps are still limited to approximately 7-10A resolution and 15-30 A resolution is more typical. However, by combining single particle EM maps with high-resolution models of subunits or subcomplexes from X-ray crystallography and NMR spectroscopy it is possible to build up an atomic model of a macromolecular assembly. Image analysis procedures are almost identical for micrographs of soluble protein complexes and detergent solubilized membrane protein complexes. However, electron microscopists attempting to prepare specimens of a membrane protein complex for imaging may find that these complexes require different handling than soluble protein complexes. This paper seeks to explain how high-quality specimen grids of membrane protein complexes may be prepared to allow for the determination of their structure by EM and image analysis.     

Electron microscopic structural analysis of Photosystem I,Photosystem II,and the cytochromeb6/f complex from green plants and cyanobacteria

Electron microscopy (EM) in combination with image analysis is a powerful technique to study protein structure at low- and high resolution. Since electron micrographs of biological objects are very noisy, substantial improvement of image quality can be obtained by averaging individual projections. Crystallographic and noncrystallographic averaging methods are available and have been applied to study projections of the large protein complexes embedded in photosynthetic membranes from cyanobacteria and higher plants. Results of EM on monomeric and trimeric Photosystem I complexes, on monomeric and dimeric Photosystem II complexes, and on the monomeric cytochromeb6/f complex are discussed.     

Improving integrative 3D modeling into low‐ to medium‐resolution electron microscopy structures with evolutionary couplings

Electron microscopy (EM) continues to provide near‐atomic resolution structures for well‐behaved proteins and protein complexes. Unfortunately, structures of some complexes are limited to low‐ to medium‐resolution due to biochemical or conformational heterogeneity. Thus, the application of unbiased systematic methods for fitting individual structures into EM maps is important. A method that employs co‐evolutionary information obtained solely from sequence data could prove invaluable for quick, confident localization of subunits within these structures. Here, we incorporate the co‐evolution of intermolecular amino acids as a new type of distance restraint in the integrative modeling platform in order to build three‐dimensional models of atomic structures into EM maps ranging from 10–14 Å in resolution. We validate this method using four complexes of known structure, where we highlight the conservation of intermolecular couplings despite dynamic conformational changes using the BAM complex. Finally, we use this method to assemble the subunits of the bacterial holo‐translocon into a model that agrees with previous biochemical data. The use of evolutionary couplings in integrative modeling improves systematic, unbiased fitting of atomic models into medium‐ to low‐resolution EM maps, providing additional information to integrative models lacking in spatial data.     

Cloning, sequencing, and overexpression of genes for ribosomal proteins from Bacillus stearothermophilus

Although a low resolution model for the arrangement of the proteins of the small and large ribosomal subunits is known, a detailed mechanistic understanding of the function of the ribosome awaits a high resolution structure of its components. While crystals have been obtained of several ribosomal proteins from Bacillus stearothermophilus, determination of atomic resolution structures of these proteins is impeded by the difficulty of obtaining large amounts of native proteins for crystallographic or NMR studies. We describe here the cloning and overexpression in Escherichia coli of the genes for ribosomal proteins S5, L6, L9, and L18 from B. stearothermophilus. S5 is extremely toxic to E. coli when overexpressed, and we have taken advantage of a new tightly regulated expression system to obtain high yields (more than 100 mg of pure protein/liter of culture) of this protein. The B. stearothermophilus S5 produced in E. coli crystallizes, and the crystals are identical to those obtained from the native protein. The crystals diffract to 2-A resolution.     

Rotation isomerization of side chains of globular proteins and its effect on the x-ray scattering curves of proteins in a solution

A new method has been developed for averaging the intensity of X-ray diffuse scattering of proteins by different conformations of side groups. The method is based on the algorithm allowing to calculate statistical weights of rotation isomers of side chains. It is shown that for protein structures obtained from high resolution crystallographic data, conformations of the majority of surface groups correspond to rotation isomers with the greatest statistical weight. It has been found that for medium size proteins (with molecular weight varying from 15,000 to 30,000 dalton) whose structure has been determined at high resolution, the influence of rotation isomerization of side chains on the scattering indicatrices does not exceed 5%. The influence of side chains mobility on the scattering curves of large proteins is also small. For these two classes of proteins the rotation isomerization of side groups can be ignored when interpreting significant (exceeding 10%) divergences between experimental and theoretical scattering curves.     

A decade of progress in understanding the structural basis of protein synthesis

The key reaction of protein synthesis, peptidyl transfer, is catalysed in all living organisms by the ribosome - an advanced and highly efficient molecular machine. During the last decade extensive X-ray crystallographic and NMR studies of the three-dimensional structure of ribosomal proteins, ribosomal RNA components and their complexes with ribosomal proteins, and of several translation factors in different functional states have taken us to a new level of understanding of the mechanism of function of the protein synthesis machinery. Among the new remarkable features revealed by structural studies, is the mimicry of the tRNA molecule by elongation factor G, ribosomal recycling factor and the eukaryotic release factor 1. Several other translation factors, for which three-dimensional structures are not yet known, are also expected to show some form of tRNA mimicry. The efforts of several crystallographic and biochemical groups have resulted in the determination by X-ray crystallography of the structures of the 30S and 50S subunits at moderate resolution, and of the structure of the 70S subunit both by X-ray crystallography and cryo-electron microscopy (EM). In addition, low resolution cryo-EM models of the ribosome with different translation factors and tRNA have been obtained. The new ribosomal models allowed for the first time a clear identification of the functional centres of the ribosome and of the binding sites for tRNA and ribosomal proteins with known three-dimensional structure. The new structural data have opened a way for the design of new experiments aimed at deeper understanding at an atomic level of the dynamics of the system.     

HingeProt: automated prediction of hinges in protein structures

Proteins are highly flexible molecules. Prediction of molecular flexibility aids in the comprehension and prediction of protein function and in providing details of functional mechanisms. The ability to predict the locations, directions, and extent of molecular movements can assist in fitting atomic resolution structures to low-resolution EM density maps and in predicting the complex structures of interacting molecules (docking). There are several types of molecular movements. In this work, we focus on the prediction of hinge movements. Given a single protein structure, the method automatically divides it into the rigid parts and the hinge regions connecting them. The method employs the Elastic Network Model, which is very efficient and was validated against a large data set of proteins. The output can be used in applications such as flexible protein-protein and protein-ligand docking, flexible docking of protein structures into cryo-EM maps, and refinement of low-resolution EM structures. The web server of HingeProt provides convenient visualization of the results and is available with two mirror sites at http://www.prc.boun.edu.tr/appserv/prc/HingeProt3 and http://bioinfo3d.cs.tau.ac.il/HingeProt/.     

Real space refinement of acto-myosin structures from sectioned muscle

We have adapted a real space refinement protocol originally developed for high-resolution crystallographic analysis for use in fitting atomic models of actin filaments and myosin subfragment 1 (S1) to 3-D images of thin-sectioned, plastic-embedded whole muscle. The rationale for this effort is to obtain a refinement protocol that will optimize the fit of the model to the density obtained by electron microscopy and correct for poor geometry introduced during the manual fitting of a high-resolution atomic model into a lower resolution 3-D image. The starting atomic model consisted of a rigor acto-S1 model obtained by X-ray crystallography and helical reconstruction of electron micrographs. This model was rebuilt to fit 3-D images of rigor insect flight muscle at a resolution of 7 nm obtained by electron tomography and image averaging. Our highly constrained real space refinement resulted in modest improvements in the agreement of model and reconstruction but reduced the number of conflicting atomic contacts by 70% without loss of fit to the 3-D density. The methodology seems to be well suited to the derivation of stereochemically reasonable atomic models that are consistent with experimentally determined 3-D reconstructions computed from electron micrographs.     

Structural biology of MCM helicases

The eukaryotic MCM2-7 complex is recruited onto origins of replication during the G1 phase of the cell cycle and acts as the main helicase at the replication fork during the S phase. Over the last few years a number of structural reports on MCM proteins using both electron microscopy and protein crystallography have been published. The crystal structures of two (almost) full-length archaeal homologs provide the first atomic pictures of a MCM helicase. However one of the structures is at low resolution and the other is of an inactive MCM. Moreover, both proteins are monomeric in the crystal, whereas the activity of the complex is critically dependent on oligomerization. Lower resolution structures derived from electron microscopy studies are therefore crucial to complement the crystallographic analysis and to assemble the multimeric complex that is active in the cell. A critical analysis of all the structural results elucidates the potential conformational changes and dynamic behavior of MCM helicase to provide a first insight into the gamut of molecular configurations adopted during the processes of DNA melting and unwinding.     

Single particle electron microscopy in combination with mass spectrometry to investigate novel complexes of membrane proteins

Large data sets of molecular projections of the membrane proteins Photosystem I and Photosystem II from cyanobacteria were analyzed by single particle electron microscopy (EM). Analysis resulted in the averaging of 2D projections from the purified complexes but also in the simultaneous detection and averaging of 2D projections from large contaminating complexes, which were present in frequencies as low as 0.1%. Among them T-shaped and L-shaped contaminants were found. The L-shaped particles could be assigned to Complex I just from the shape, although no Complex I from a cyanobacterium has been structurally characterized. A systematic comparison by single particle EM and mass spectrometry of two differently purified Photosystem II complexes resulted in the assignment of PsbZ, a small peripheral subunit of 6.8kDa, within the structure. Together these data suggest that screening for membrane protein structures by single particle EM and mass spectrometry may be a new approach to find novel structures of such proteins. We propose here a scheme for searching for novel membrane protein structures in specific types of membranes. In this approach single particle EM and mass spectrometry, after pre-fractionation using one- or multidimensional protein separation techniques, are applied to characterize all its larger components.     

Multi-resolution contour-based fitting of macromolecular structures

A novel contour-based matching criterion is presented for the quantitative docking of high-resolution structures of components into low-resolution maps of macromolecular complexes. The proposed Laplacian filter is combined with a six-dimensional search using fast Fourier transforms to rapidly scan the rigid-body degrees of freedom of a probe molecule relative to a fixed target density map. A comparison of the docking performance with the standard cross-correlation criterion demonstrates that contour matching with the Laplacian filter significantly extends the viable resolution range of correlation-based fitting to resolutions as low as 30 A. The gain in docking precision at medium to low resolution (15-30 A) is critical for image reconstructions from electron microscopy (EM). The new algorithm enables for the first time the reliable docking of smaller molecular components into EM densities of large biomolecular assemblies at such low resolutions. As an example of the practical effectiveness of contour-based fitting, a new pseudo-atomic model of a microtubule was constructed from a 20 A resolution EM map and from atomic structures of alpha and beta tubulin subunits.     

Imaging of organelles by electron microscopy reveals protein-protein interactions in mitochondria and chloroplasts

Ongoing progress in electron microscopy (EM) offers now an opening to visualize cells at the nanoscale by cryo-electron tomography (ET). Large protein complexes can be resolved at near-atomic resolution by single particle averaging. Some examples from mitochondria and chloroplasts illustrate the possibilities with an emphasis on the membrane organization. Cryo-ET performed on non-chemically fixed, unstained, ice-embedded material can visualize specific large membrane protein complexes. In combination with averaging methods, 3D structures were calculated of mitochondrial ATP synthase at 6 nm resolution and of chloroplast photosystem II at 3.5 nm.     

Towards a complete atomic structure of spectrin family proteins

The spectrin family of proteins represents a discrete group of cytoskeletal proteins comprising principally alpha-actinin, spectrin, dystrophin, and homologues and isoforms. They all share three main structural and functional motifs, namely, the spectrin repeat, EF-hands, and a CH domain-containing actin-binding domain. These proteins are variously involved in organisation of the actin cytoskeleton, membrane cytoskeleton architecture, cell adhesion, and contractile apparatus. The highly modular nature of these molecules has been a hindrance to the determination of their complete structures due to the inherent flexibility imparted on the proteins, but has also been an asset, inasmuch as the individual modules were of a size amenable to structural analysis by both crystallographic and NMR approaches. Representative structures of all the major domains shared by spectrin family proteins have now been solved at atomic resolution, including in some cases multiple domains from several family members. High-resolution structures, coupled with lower resolution methods to determine the overall molecular shape of these proteins, allow us for the first time to build complete atomic structures of the spectrin family of proteins.     

Computational reconstruction of multidomain proteins using atomic force microscopy data

Classical structural biology techniques face a great challenge to determine the structure at the atomic level of large and flexible macromolecules. We present a novel methodology that combines high-resolution AFM topographic images with atomic coordinates of proteins to assemble very large macromolecules or particles. Our method uses a two-step protocol: atomic coordinates of individual domains are docked beneath the molecular surface of the large macromolecule, and then each domain is assembled using a combinatorial search. The protocol was validated on three test cases: a simulated system of antibody structures; and two experimentally based test cases: Tobacco mosaic virus, a rod-shaped virus; and Aquaporin Z, a bacterial membrane protein. We have shown that AFM-intermediate resolution topography and partial surface data are useful constraints for building macromolecular assemblies. The protocol is applicable to multicomponent structures connected in the polypeptide chain or as disjoint molecules. The approach effectively increases the resolution of AFM beyond topographical information down to atomic-detail structures.     

Atomic contacts in protein structures. A detailed analysis of atomic radii, packing, and overlaps

A rigorous quantitative assessment of atomic contacts and packing in native protein structures is presented. The analysis is based on optimized atomic radii derived from a set of high-resolution protein structures and reveals that the distribution of atomic contacts and overlaps is a structural constraint in proteins, irrespective of structural or functional classification and size. Furthermore, a newly developed method for calculating packing properties is introduced and applied to sets of protein structures at different levels of resolution. The results show that limited resolution yields decreasing packing quality, which underscores the relevance of packing considerations for structure prediction, design, dynamics, and docking.     

Computational methods for ultrastructural analysis of synaptic complexes

Electron microscopy (EM) provided fundamental insights about the ultrastructure of neuronal synapses. The large amount of information present in the contemporary EM datasets precludes a thorough assessment by visual inspection alone, thus requiring computational methods for the analysis of the data. Here, I review image processing software methods ranging from membrane tracing in large volume datasets to high resolution structures of synaptic complexes. Particular attention is payed to molecular level analysis provided by recent cryo-electron microscopy and tomography methods.     

Crystallographic Refinement and Structure-Factor Time-Averaging by Molecular Dynamics in the Absence of a Physical Force Field

Abstract

The application of Molecular-Dynamics simulation in protein-crystallographic structure refinement has become common practice. In this paper, structure optimizations are described where the driving force is derived only from the crystallographic data and not from any physical potential energy function. Under this extreme condition ab initio structure refinement and the application of structure-factor time averaging was investigated using a small 9 atom test system. Success in ab initio refinement, where the starting atomic positions are randomly distributed, depends on the resolution of the crystallographic data used in the optimization. The presence of high resolution data introduces false minima in the X-ray energy profile, enhancing the search problem significantly. On the same system, we also tested the method of time-averaged crystallographically restrained Molecular Dynamics, again in the absence of a physical force field. In this method, the diffraction data is modelled by an ensemble of structures instead of one single structure. In comparison to conventional single-structure refinement, more reflections were required to determine a correct atomic distribution. A time-averaging simulation at 0.2 nm resolution (40 reflections) yielded an incorrect distribution, although a low R-factor was obtained. Simulations at 0.1 nm resolution (248 reflections) gave both low R-factors, 3 to 4%, and correct atomic distributions. The scale factor between the observed and time-averaged calculated structure factor amplitudes appeared to be unstable, when optimized during a time-averaging simulation. Tests of time-averaged restrained simulations with noise added to the observed structure-factor amplitudes, indicated that noise is modelled when no information in the form of constraints or restraints is available to distinguish it from real data.     

Atomic resolution structures of native copper nitrite reductase from Alcaligenes xylosoxidans and the active site mutant Asp92Glu

We provide the first atomic resolution (<1.20 A) structure of a copper protein, nitrite reductase, and of a mutant of the catalytically important Asp92 residue (D92E). The atomic resolution where carbon-carbon bonds of the peptide become clearly resolved, remains a key goal of structural analysis. Despite much effort and technological progress, still very few structures are known at such resolution. For example, in the Protein Data Bank (PDB) there are some 200 structures of copper proteins but the highest resolution structure is that of amicyanin, a small (12 kDa) protein, which has been resolved to 1.30 A. Here, we present the structures of wild-type copper nitrite reductase (wtNiR) from Alcaligenes xylosoxidans (36.5 kDa monomer), the "half-apo" recombinant native protein and the D92E mutant at 1.04, 1.15 and 1.12A resolutions, respectively. These structures provide the basis from which to build a detailed mechanism of this important enzyme.     

Deriving protein dynamical properties from weighted protein contact number

It has recently been shown that in proteins the atomic mean-square displacement (or B-factor) can be related to the number of the neighboring atoms (or protein contact number), and that this relationship allows one to compute the B-factor profiles directly from protein contact number. This method, referred to as the protein contact model, is appealing, since it requires neither trajectory integration nor matrix diagonalization. As a result, the protein contact model can be applied to very large proteins and can be implemented as a high-throughput computational tool to compute atomic fluctuations in proteins. Here, we show that this relationship can be further refined to that between the atomic mean-square displacement and the weighted protein contact-number, the weight being the square of the reciprocal distance between the contacting pair. In addition, we show that this relationship can be utilized to compute the cross-correlation of atomic motion (the B-factor is essentially the auto-correlation of atomic motion). For a nonhomologous dataset comprising 972 high-resolution X-ray protein structures (resolution <2.0 A and sequence identity <25%), the mean correlation coefficient between the X-ray and computed B-factors based on the weighted protein contact-number model is 0.61, which is better than those of the original contact-number model (0.51) and other methods. We also show that the computed correlation maps based on the weighted contact-number model are globally similar to those computed through normal model analysis for some selected cases. Our results underscore the relationship between protein dynamics and protein packing. We believe that our method will be useful in the study of the protein structure-dynamics relationship.