Electron cryomicroscopy methods

Computational Methods and Deep Learning for Elucidating Protein Interaction Networks

Protein interactions play a critical role in all biological processes, but experimental identification of protein interactions is a time- and resource-intensive process. The advances in next-generation sequencing and multi-omics technologies have greatly benefited large-scale predictions of protein interactions using machine learning methods. A wide range of tools have been developed to predict protein-protein, protein-nucleic acid, and protein-drug interactions. Here, we discuss the applications, methods, and challenges faced when employing the various prediction methods. We also briefly describe ways to overcome the challenges and prospective future developments in the field of protein interaction biology.

Constructing atomic structural models into cryo-EM densities using molecular dynamics - Pros and cons

Accurate structure determination from electron density maps at 3-5 Å resolution necessitates a balance between extensive global and local sampling of atomistic models, yet with the stereochemical correctness of backbone and sidechain geometries. Molecular Dynamics Flexible Fitting (MDFF), particularly through a resolution-exchange scheme, ReMDFF, provides a robust way of achieving this balance for hybrid structure determination. Employing two high-resolution density maps, namely that of β-galactosidase at 3.2 Å and TRPV1 at 3.4 Å, we showcase the quality of ReMDFF-generated models, comparing them against ones submitted by independent research groups for the 2015-2016 Cryo-EM Model Challenge. This comparison offers a clear evaluation of ReMDFF's strengths and shortcomings, and those of data-guided real-space refinements in general. ReMDFF results scored highly on the various metric for judging the quality-of-fit and quality-of-model. However, some systematic discrepancies are also noted employing a Molprobity analysis, that are reproducible across multiple competition entries. A space of key refinement parameters is explored within ReMDFF to observe their impact within the final model. Choice of force field parameters and initial model seem to have the most significant impact on ReMDFF model-quality. To this end, very recently developed CHARMM36m force field parameters provide now more refined ReMDFF models than the ones originally submitted to the Cryo-EM challenge. Finally, a set of good-practices is prescribed for the community to benefit from the MDFF developments.

Predictive and Experimental Approaches for Elucidating Protein-Protein Interactions and Quaternary Structures

The elucidation of protein-protein interactions is vital for determining the function and action of quaternary protein structures. Here, we discuss the difficulty and importance of establishing protein quaternary structure and review in vitro and in silico methods for doing so. Determining the interacting partner proteins of predicted protein structures is very time-consuming when using in vitro methods, this can be somewhat alleviated by use of predictive methods. However, developing reliably accurate predictive tools has proved to be difficult. We review the current state of the art in predictive protein interaction software and discuss the problem of scoring and therefore ranking predictions. Current community-based predictive exercises are discussed in relation to the growth of protein interaction prediction as an area within these exercises. We suggest a fusion of experimental and predictive methods that make use of sparse experimental data to determine higher resolution predicted protein interactions as being necessary to drive forward development.

CHARMM-GUI MDFF/xMDFF Utilizer for Molecular Dynamics Flexible Fitting Simulations in Various Environments

X-ray crystallography and cryo-electron microscopy are two popular methods for the structure determination of biological molecules. Atomic structures are derived through the fitting and refinement of an initial model into electron density maps constructed by both experiments. Two computational approaches, MDFF and xMDFF, have been developed to facilitate this process by integrating the experimental data with molecular dynamics simulation. However, the setup of an MDFF/xMDFF simulation requires knowledge of both experimental and computational methods, which is not straightforward for nonexpert users. In addition, sometimes it is desirable to include realistic environments, such as explicit solvent and lipid bilayers during the simulation, which poses another challenge even for expert users. To alleviate these difficulties, we have developed MDFF/xMDFF Utilizer in CHARMM-GUI that helps users to set up an MDFF/xMDFF simulation. The capability of MDFF/xMDFF Utilizer is greatly enhanced by integration with other CHARMM-GUI modules, including protein structure manipulation, a diverse set of lipid types, and all-atom CHARMM and coarse-grained PACE force fields. With this integration, various simulation environments are available for MDFF Utilizer (vacuum, implicit/explicit solvent, and bilayers) and xMDFF Utilizer (vacuum and solution). In this work, three examples are shown to demonstrate the usage of MDFF/xMDFF Utilizer.

Molecular dynamics-based refinement and validation for sub-5 Å cryo-electron microscopy maps

Two structure determination methods, based on the molecular dynamics flexible fitting (MDFF) paradigm, are presented that resolve sub-5 Å cryo-electron microscopy (EM) maps with either single structures or ensembles of such structures. The methods, denoted cascade MDFF and resolution exchange MDFF, sequentially re-refine a search model against a series of maps of progressively higher resolutions, which ends with the original experimental resolution. Application of sequential re-refinement enables MDFF to achieve a radius of convergence of ~25 Å demonstrated with the accurate modeling of β-galactosidase and TRPV1 proteins at 3.2 Å and 3.4 Å resolution, respectively. The MDFF refinements uniquely offer map-model validation and B-factor determination criteria based on the inherent dynamics of the macromolecules studied, captured by means of local root mean square fluctuations. The MDFF tools described are available to researchers through an easy-to-use and cost-effective cloud computing resource on Amazon Web Services.

DOI:http://dx.doi.org/10.7554/eLife.16105.001

To understand the roles that proteins and other large molecules play inside cells, it is important to determine their structures. One of the techniques that researchers can use to do this is called cryo-electron microscopy (cryo-EM), which rapidly freezes molecules to fix them in position before imaging them in fine detail.

The cryo-EM images are like maps that show the approximate position of atoms. These images must then be processed in order to build a three-dimensional model of the protein that shows how its atoms are arranged relative to each other. One computational approach called Molecular Dynamics Flexible Fitting (MDFF) works by flexibly fitting possible atomic structures into cryo-EM maps. Although this approach works well with relatively undetailed (or ‘low resolution’) cryo-EM images, it struggles to handle the high-resolution cryo-EM maps now being generated.

Singharoy, Teo, McGreevy et al. have now developed two MDFF methods – called cascade MDFF and resolution exchange MDFF – that help to resolve atomic models of biological molecules from cryo-EM images. Each method can refine poorly guessed models into ones that are consistent with the high-resolution experimental images. The refinement is achieved by interpreting a range of images that starts with a ‘fuzzy’ image. The contrast of the image is then progressively improved until an image is produced that has a resolution that is good enough to almost distinguish individual atoms.

The method works because each cryo-EM image shows not just one, but a collection of atomic structures that the molecule can take on, with the fuzzier parts of the image representing the more flexible parts of the molecule. By taking into account this flexibility, the large-scale features of the protein structure can be determined first from the fuzzier images, and increasing the contrast of the images allows smaller-scale refinements to be made to the structure.

The MDFF tools have been designed to be easy to use and are available to researchers at low cost through cloud computing platforms. They can now be used to unravel the structure of many different proteins and protein complexes including those involved in photosynthesis, respiration and protein synthesis.

DOI:http://dx.doi.org/10.7554/eLife.16105.002

Molecular dynamics simulations of homo-oligomeric bundles embedded within a lipid bilayer

Using molecular dynamics simulations, we studied the structure, interhelix interactions, and dynamics of transmembrane proteins. Specifically, we investigated homooligomeric helical bundle systems consisting of synthetic α-helices with either the sequence Ac-(LSLLLSL)3-NH2 (LS2) or Ac-(LSSLLSL)3-NH2 (LS3). The LS2 and LS3 helical peptides are designed to have amphipathic characteristics that form ion channels in membrane. We simulated bundles containing one to six peptides that were embedded in palmitoyl-oleoyl-phosphatidylcholine (POPC) lipid bilayer and placed between two lamellae of water. We aim to provide a fundamental understanding of how amphipathic helical peptides interact with each other and their dynamical behaviors in different homooligomeric states. To understand structural properties, we examined the helix lengths, tilt angles of individual helices and the entire bundle, interhelix distances, interhelix cross-angles, helix hydrophobic-to-hydrophilic vector projections, and the average number of interhelix hydrophilic (serine-serine) contacts lining the pore of the transmembrane channel. To analyze dynamical properties, we calculated the rotational autocorrelation function of each helix and the cross-correlation of the rotational velocity between adjacent helices. The observed structural and dynamical characteristics show that higher order bundles containing four to six peptides are composed of multiple lower order bundles of one to three peptides. For example, the LS2 channel was found to be stable in a tetrameric bundle composed of a "dimer of dimers." In addition, we observed that there is a minimum of two strong hydrophilic contacts between a pair of adjacent helices in the dimer to tetramer systems and only one strong hydrophilic interhelix contact in helix pairs of the pentamer and hexamer systems. We believe these results are general and can be applied to more complex ion channels, providing insight into ion channel stability and assembly.

Networks of ProteinProtein Interactions: From Uncertainty to Molecular Details

Proteins are the bricks and mortar of cells. The work of proteins is structural and functional, as they are the principal element of the organization of the cell architecture, but they also play a relevant role in its metabolism and regulation. To perform all these functions, proteins need to interact with each other and with other bio-molecules, either to form complexes or to recognize precise targets of their action. For instance, a particular transcription factor may activate one gene or another depending on its interactions with other proteins and not only with DNA. Hence, the ability of a protein to interact with other bio-molecules, and the partners they have at each particular time and location can be crucial to characterize the role of a protein. Proteins rarely act alone; they rather constitute a mingled network of physical interactions or other types of relationships (such as metabolic and regulatory) or signaling cascades. In this context, understanding the function of a protein implies to recognize the members of its neighborhood and to grasp how they associate, both at the systemic and atomic level. The network of physical interactions between the proteins of a system, cell or organism, is defined as the interactome. The purpose of this review is to deepen the description of interactomes at different levels of detail: from the molecular structure of complexes to the global topology of the network of interactions. The approaches and techniques applied experimentally and computationally to attain each level are depicted. The limits of each technique and its integration into a model network, the challenges and actual problems of completeness of an interactome, and the reliability of the interactions are reviewed and summarized. Finally, the application of the current knowledge of protein-protein interactions on modern network medicine and protein function annotation is also explored.

Coarse-grained molecular dynamics of tetrameric transmembrane peptide bundles within a lipid bilayer

The conformations of model transmembrane peptides are studied to understand the structural and dynamical aspects of tetrameric bundles using a series of coarse grain (CG) molecular dynamics (MD) simulations since membrane proteins play a crucial role in cell function. In this work, two different amphipathic models have been constructed using similar hydrophobic/hydrophilic characteristics with two structurally distinct morphologies to evaluate the effect of roughness and hydrophilic topology on the structure of tetrameric bundles, one class that forms an ion-channel and one class that does not. Free energy calculations of typical amphipathic peptide topologies show that using a relatively smooth surface morphology allows for a stable conformation of the tetramer bundle in a diamond formation. However, the model with side chains attached to the core in order to roughen the surface has a stable square tetramer bundle which is consistent with experimental data and all-atom (AA) MD simulations. Comparisons of the CG simulations with AA MD simulations are in reasonable agreement with the formation of tetrameric homo-oligomers, partitioning within the lipid bilayer and tilt angle with respect to the bilayer normal. We concluded that a square or diamond shape tetrameric homo-oligomers could be stabilized by rational design of the peptide morphology and topology of the surface, thus allowing us to tune the permeability of the bundle or channel.

Two-dimensional crystallization of integral membrane proteins for electron crystallography

Although membrane proteins make up 30% of the proteome and are a common target for therapeutic drugs, determination of their atomic structure remains a technical challenge. Electron crystallography represents an alternative to the conventional methods of X-ray diffraction and NMR and relies on the formation of two-dimensional crystals. These crystals are produced by reconstituting purified, detergent-solubilized membrane proteins back into the native environment of a lipid bilayer. This chapter reviews methods for producing two-dimensional crystals and for screening them by negative stain electron microscopy. In addition, we show examples of the different morphologies that are commonly obtained and describe basic image analysis procedures that can be used to evaluate their promise for structure determination by cryoelectron microscopy.

Present and future of membrane protein structure determination by electron crystallography

Membrane proteins are critical to cell physiology, playing roles in signaling, trafficking, transport, adhesion, and recognition. Despite their relative abundance in the proteome and their prevalence as targets of therapeutic drugs, structural information about membrane proteins is in short supply. This chapter describes the use of electron crystallography as a tool for determining membrane protein structures. Electron crystallography offers distinct advantages relative to the alternatives of X-ray crystallography and NMR spectroscopy. Namely, membrane proteins are placed in their native membranous environment, which is likely to favor a native conformation and allow changes in conformation in response to physiological ligands. Nevertheless, there are significant logistical challenges in finding appropriate conditions for inducing membrane proteins to form two-dimensional arrays within the membrane and in using electron cryo-microscopy to collect the data required for structure determination. A number of developments are described for high-throughput screening of crystallization trials and for automated imaging of crystals with the electron microscope. These tools are critical for exploring the necessary range of factors governing the crystallization process. There have also been recent software developments to facilitate the process of structure determination. However, further innovations in the algorithms used for processing images and electron diffraction are necessary to improve throughput and to make electron crystallography truly viable as a method for determining atomic structures of membrane proteins.

Single particle electron microscopy

Electron microscopy (EM) in combination with image analysis is a powerful technique to study protein structures at low, medium, and high resolution. Since electron micrographs of biological objects are very noisy, improvement of the signal-to-noise ratio by image processing is an integral part of EM, and this is performed by averaging large numbers of individual projections. Averaging procedures can be divided into crystallographic and non-crystallographic methods. The crystallographic averaging method, based on two-dimensional (2D) crystals of (membrane) proteins, yielded in solving atomic protein structures in the last century. More recently, single particle analysis could be extended to solve atomic structures as well. It is a suitable method for large proteins, viruses, and proteins that are difficult to crystallize. Because it is also a fast method to reveal the low-to-medium resolution structures, the impact of its application is growing rapidly. Technical aspects, results, and possibilities are presented.

Spline-based image-to-volume registration for three-dimensional electron microscopy

This paper presents an algorithm based on a continuous framework for a posteriori angular and translational assignment in three-dimensional electron microscopy (3DEM) of single particles. Our algorithm can be used advantageously to refine the assignment of standard quantized-parameter methods by registering the images to a reference 3D particle model. We achieve the registration by employing a gradient-based iterative minimization of a least-squares measure of dissimilarity between an image and a projection of the volume in the Fourier transform (FT) domain. We compute the FT of the projection using the central-slice theorem (CST). To compute the gradient accurately, we take advantage of a cubic B-spline model of the data in the frequency domain. To improve the robustness of the algorithm, we weight the cost function in the FT domain and apply a "mixed" strategy for the assignment based on the minimum value of the cost function at registration for several different initializations. We validate our algorithm in a fully controlled simulation environment. We show that the mixed strategy improves the assignment accuracy; on our data, the quality of the angular and translational assignment was better than 2 voxel (i.e., 6.54 angstroms). We also test the performance of our algorithm on real EM data. We conclude that our algorithm outperforms a standard projection-matching refinement in terms of both consistency of 3D reconstructions and speed.

Modeling membrane proteins based on low-resolution electron microscopy maps: a template for the TM domains of the oxalate transporter OxlT

The availability of both EM and high-resolution crystallographic data for several membrane proteins (MPs) permits a detailed evaluation of the ability of molecular modeling techniques to complement EM data in the development of models of MPs. A protocol for this purpose is presented, consisting of (1) identifying transmembrane (TM) domains from sequence; (2) assigning buried and lipid-exposed faces of the TM domains; and (3) assembling the TM domains into a bundle, based on geometric restraints obtained from the EM data. The protocol is validated by predicting the structures of several 7- and 12-TM MPs to within 3-5 A r.m.s.d. from their crystal structures. The protocol is applied to generate a model of the oxalate transporter OxlT, for which a high-resolution structure is not yet available.

A multiresolution approach to orientation assignment in 3D electron microscopy of single particles

Three-dimensional (3D) electron microscopy (3DEM) aims at the determination of the spatial distribution of the Coulomb potential of macromolecular complexes. The 3D reconstruction of a macromolecule using single-particle techniques involves thousands of 2D projections. One of the key parameters required to perform such a 3D reconstruction is the orientation of each projection image as well as its in-plane orientation. This information is unknown experimentally and must be determined using image-processing techniques. We propose the use of wavelets to match the experimental projections with those obtained from a reference 3D model. The wavelet decomposition of the projection images provides a framework for a multiscale matching algorithm in which speed and robustness to noise are gained. Furthermore, this multiresolution approach is combined with a novel orientation selection strategy. Results obtained from computer simulations as well as experimental data encourage the use of this approach.

Distance-restrained docking of rifampicin and rifamycin SV to RNA polymerase using systematic FRET measurements: developing benchmarks of model quality and reliability

We are developing distance-restrained docking strategies for modeling macromolecular complexes that combine available high-resolution structures of the components and intercomponent distance restraints derived from systematic fluorescence resonance energy transfer (FRET) measurements. In this article, we consider the problem of docking small-molecule ligands within macromolecular complexes. Using simulated FRET data, we have generated a series of benchmarks that permit estimation of model accuracy based on the quantity and quality of FRET-derived distance restraints, including the number, random error, systematic error, distance distribution, and radial distribution of FRET-derived distance restraints. We find that expected model accuracy is 10 A or better for models based on: i), > or =20 restraints with up to 15% random error and no systematic error, or ii), > or =20 restraints with up to 15% random error, up to 10% systematic error, and a symmetric radial distribution of restraints. Model accuracies can be improved to 5 A or better by increasing the number of restraints to > or =40 and/or by optimizing the distance distribution of restraints. Using experimental FRET data, we have defined the positions of the binding sites within bacterial RNA polymerase of the small-molecule inhibitors rifampicin (Rif) and rifamycin SV (Rif SV). The inferred binding sites for Rif and Rif SV were located with accuracies of, respectively, 7 and 10 A relative to the crystallographically defined binding site for Rif. These accuracies agree with expectations from the benchmark simulations and suffice to indicate that the binding sites for Rif and Rif SV are located within the RNA polymerase active-center cleft, overlapping the binding site for the RNA-DNA hybrid.

An automatic method for predicting transmembrane protein structures using cryo-EM and evolutionary data

The transmembrane (TM) domains of many integral membrane proteins are composed of alpha-helix bundles. Structure determination at high resolution (<4 A) of TM domains is still exceedingly difficult experimentally. Hence, some TM-protein structures have only been solved at intermediate (5-10 A) or low (>10 A) resolutions using, for example, cryo-electron microscopy (cryo-EM). These structures reveal the packing arrangement of the TM domain, but cannot be used to determine the positions of individual amino acids. The observation that typically, the lipid-exposed faces of TM proteins are evolutionarily more variable and less charged than their core provides a simple rule for orienting their constituent helices. Based on this rule, we developed score functions and automated methods for orienting TM helices, for which locations and tilt angles have been determined using, e.g., cryo-EM data. The method was parameterized with the aim of retrieving the native structure of bacteriorhodopsin among near- and far-from-native templates. It was then tested on proteins that differ from bacteriorhodopsin in their sequences, architectures, and functions, such as the acetylcholine receptor and rhodopsin. The predicted structures were within 1.5-3.5 A from the native state in all cases. We conclude that the computational method can be used in conjunction with cryo-EM data to obtain approximate model structures of TM domains of proteins for which a sufficiently heterogeneous set of homologs is available. We also show that in those proteins in which relatively short loops connect neighboring helices, the scoring functions can discriminate between near- and far-from-native conformations even without the constraints imposed on helix locations and tilt angles that are derived from cryo-EM.

EMDep: a web-based system for the deposition and validation of high-resolution electron microscopy macromolecular structural information

This paper describes the design and implementation of a Web-based deposition system, EMDep, for macro-molecular volumes determined by electron microscopy and deposited at the European Bioinformatics Institute (EBI) for inclusion in the Electron Microscopy Data Base (EMDB). EMDep is a flexible and portable system (http://www.ebi.ac.uk/msd-srv/emdep/) that allows for the acceptance and validation of data, by an interactive depositor-driven operation. The system takes full advantage of the knowledge and expertise of the experimenters, rather than relying on the database curators, for the complete and accurate description of the structural experiment and its results.

Prospects of electron cryotomography to visualize macromolecular complexes inside cellular compartments: implications of crowding

Electron cryotomography has unique potential for three-dimensional visualization of macromolecular complexes at work in their natural environment. This approach is based on reconstructing three-dimensional volumes from tilt series of electron micrographs of cells preserved in their native states by vitrification. Resolutions of 5-8 nm have already been achieved and the prospects for further improvement are good. Since many intracellular activities are conducted by complexes in the megadalton range with dimensions of 20-50 nm, current resolutions should suffice to identify many of them in tomograms. However, residual noise and the dense packing of cellular constituents hamper interpretation. Recently, tomographic data have been collected on vitrified eukaryotic cells (Medalia et al., Science (2002) in press). Their cytoplasm was found to be markedly less crowded than in the prokaryotes previously studied, in accord with differences in crowding between prokaryotic and eukaryotic cells documented by other (indirect) biophysical methods. The implications of this observation are twofold. First, complexes should be more easily identifiable in tomograms of eukaryotic cytoplasm. This applies both to recognizing known complexes and characterizing novel complexes. An example of the latter-a 5-fold symmetric particle is-given. Second, electron cryotomography offers an incisive probe to examine crowding in different cellular compartments.

Deriving folds of macromolecular complexes through electron cryomicroscopy and bioinformatics approaches

Intermediate-resolution (7-9A) structures of large macromolecular complexes can be obtained by electron cryomicroscopy. This structural information, combined with bioinformatics data for the individual protein components or domains, can lead to a fold model for the entire complex. Such approaches have been demonstrated with the 6.8 A structure of the rice dwarf virus to derive models for the major capsid shell proteins.