Community recommendations on cryoEM data archiving and validation.

In January 2020, a workshop was held at EMBL-EBI (Hinxton, UK) to discuss data requirements for the deposition and validation of cryoEM structures, with a focus on single-particle analysis. The meeting was attended by 47 experts in data processing, model building and refinement, validation, and archiving of such structures. This report describes the workshop's motivation and history, the topics discussed, and the resulting consensus recommendations. Some challenges for future methods-development efforts in this area are also highlighted, as is the implementation to date of some of the recommendations.

Community recommendations on cryoEM data archiving and validation: Outcomes of a wwPDB/EMDB workshop on cryoEM data management, deposition and validation.

In January 2020, a workshop was held at EMBL-EBI (Hinxton, UK) to discuss data requirements for deposition and validation of cryoEM structures, with a focus on single-particle analysis. The meeting was attended by 47 experts in data processing, model building and refinement, validation, and archiving of such structures. This report describes the workshop's motivation and history, the topics discussed, and consensus recommendations resulting from the workshop. Some challenges for future methods-development efforts in this area are also highlighted, as is the implementation to date of some of the recommendations.

Structural Studies of Bacteriophage Φ6 and Its Transformations during Its Life Cycle.

From the first isolation of the cystovirus bacteriophage Φ6 from Pseudomonas syringae 50 years ago, we have progressed to a better understanding of the structure and transformations of many parts of the virion. The three-layered virion, encapsulating the tripartite double-stranded RNA (dsRNA) genome, breaches the cell envelope upon infection, generates its own transcripts, and coopts the bacterial machinery to produce its proteins. The generation of a new virion starts with a procapsid with a contracted shape, followed by the packaging of single-stranded RNA segments with concurrent expansion of the capsid, and finally replication to reconstitute the dsRNA genome. The outer two layers are then added, and the fully formed virion released by cell lysis. Most of the procapsid structure, composed of the proteins P1, P2, P4, and P7 is now known, as well as its transformations to the mature, packaged nucleocapsid. The outer two layers are less well-studied. One additional study investigated the binding of the host protein YajQ to the infecting nucleocapsid, where it enhances the transcription of the large RNA segment that codes for the capsid proteins. Finally, I relate the structural aspects of bacteriophage Φ6 to those of other dsRNA viruses, noting the similarities and differences.

The Ewald sphere/focus gradient does not limit the resolution of cryoEM reconstructions.

In our quest to solve biomolecular structures to higher resolutions in cryoEM, care must be taken to deal with all aspects of image formation in the electron microscope. One of these is the Ewald sphere/focus gradient that derives from the scattering geometry in the microscope and its implications for recovering high resolution and handedness information. While several methods to deal with it has been proposed and implemented, there are still questions as to the correct approach. At the high acceleration voltages used for cryoEM, the traditional projection approximation that ignores the Ewald sphere breaks down around 2-3 Å and with large particles. This is likely not crucial for most biologically interesting molecules, but is required to understand detail about catalytic events, molecular orbitals, orientation of bound water molecules, etc. Through simulation I show that integration along the Ewald spheres in frequency space during reconstruction, the "simple insertion method" is adequate to reach resolutions to the Nyquist frequency. Both theory and simulations indicate that the handedness information encoded in such phases is irretrievably lost in the formation of real space images. The conclusion is that correct reconstruction along the Ewald spheres avoids the limitations of the projection approximation.

Advances in understanding the molecular structure of retinoschisin while questions remain of biological function.

Retinoschisin (RS1) is a secreted protein that is essential for maintaining integrity of the retina. Numerous mutations in RS1 cause X-linked retinoschisis (XLRS), a progressive degeneration of the retina that leads to vision loss in young males. A key manifestation of XLRS is the formation of cavities (cysts) in the retina and separation of the layers (schisis), disrupting synaptic transmission. There are currently no approved treatments for patients with XLRS. Strategies using adeno-associated viral (AAV) vectors to deliver functional copies of RS1 as a form of gene augmentation therapy, are under clinical evaluation. To improve therapeutic strategies for treating XLRS, it is critical to better understand the secretion of RS1 and its molecular function. Immunofluorescence and immunoelectron microscopy show that RS1 is located on the surfaces of the photoreceptor inner segments and bipolar cells. Sequence homology indicates a discoidin domain fold, similar to many other proteins with demonstrated adhesion functions. Recent structural studies revealed the tertiary structure of RS1 as two back-to-back octameric rings, each cross-linked by disulfides. The observation of higher order structures in vitro suggests the formation of an adhesive matrix spanning the distance between cells (∼100 nm). Several studies indicated that RS1 readily binds to other proteins such as the sodium-potassium ATPase (NaK-ATPase) and extracellular matrix proteins. Alternatively, RS1 may influence fluid regulation via interaction with membrane proteins such as the NaK-ATPase, largely inferred from the use of carbonic anhydrase inhibitors to shrink the typical intra-retinal cysts in XLRS. We discuss these models in light of RS1 structure and address the difficulty in understanding the function of RS1.

The progressive spectral signal-to-noise ratio of cryo-electron micrograph movies as a tool to assess quality and radiation damage.

BACKGROUND AND OBJECTIVE: The quality of a cryoEM reconstruction is fundamentally a function of the signal-to-noise ratio (SNR) of the original micrographs. The SNR embodies multiple aspects of image formation, including microscope details and alignment, specimen composition and thickness, how it is recorded, and how the specimen degrades during imaging. With the advent of direct electron detectors and the recording of a series of images for each micrograph (a movie), we have an opportunity to count every electron and derive fully quantitative results. After alignment of the movie frames of a micrograph, we can calculate the SNR, or its spatial frequency equivalent, the spectral SNR (SSNR). This SSNR reflects residual movement between frames and the progressive effect of radiation damage. The goal is to develop a quantitative analysis of the SSNR and radiation damage to assess and improve the quality of micrographs. METHODS: Several test cases were selected from the EMPIAR database and ten micrograph movies downloaded for each case. The movie frames were aligned as rigid bodies to compensate for stage and support movement. The SSNR for subsets of frames was then calculated to assess the effect of residual movement. The progressive SSNR (PSSNR) was subsequently calculated to determine the decrease in signal accumulation as a result of radiation damage. RESULTS: In all cases the alignment of the movie frames compensated for global movement to the extent that the effect on the SSNR is negligible. The subset SSNR can be used as a tool to further confirm the extent of residual movement. The progressive SSNR indicates an increase in value up to an asymptote, consistent with the theory for radiation damage. Fitting these curves gives the inherent SNR before exposure, and the critical dose, which decreases with spatial frequency with an exponential parameter roughly between one and two. CONCLUSIONS: The implementation of the PSSNR for movie frames provides a tool for assessing micrograph quality and progression of radiation damage. The estimation of the critical dose further quantifies radiation damage and may shed some light on the mechanisms of damage. These are likely both a function of the specimen composition and the imaging parameters used.

Cryo-Electron Tomography of the Herpesvirus Procapsid Reveals Interactions of the Portal with the Scaffold and a Shift on Maturation.

Herpes simplex virus 1 (HSV-1) requires seven proteins to package its genome through a vertex in its capsid, one of which is the portal protein, pUL6. The portal protein is also thought to facilitate assembly of the procapsid. While the portal has been visualized in mature capsids, we aimed to elucidate its role in the assembly and maturation of procapsids using cryo-electron tomography (cryoET). We identified the portal vertex in individual procapsids, calculated a subtomogram average, and compared that with the portal vertex in empty mature capsids (A-capsids). The resulting maps show the portal on the interior surface with its narrower end facing outwards, while maintaining close contact with the capsid shell. In the procapsid, the portal is embedded in the underlying scaffold, suggesting that assembly involves a portal-scaffold complex. During maturation, the capsid shell angularizes with a corresponding outward movement of the vertices. We found that in A-capsids, the portal translocates outward further than the adjacent capsomers and strengthens its contacts with the capsid shell. Our methodology also allowed us to determine the number of portal vertices in each capsid, with most having one per capsid, but some none or two, and rarely three. The predominance of a single portal per capsid supports facilitation of the assembly of the procapsid.IMPORTANCE Herpes simplex virus 1 (HSV-1) infects a majority of humans, causing mostly mild disease but in some cases progressing toward life-threatening encephalitis. Understanding the life cycle of the virus is important to devise countermeasures. Production of the virion starts with the assembly of an icosahedral procapsid, which includes DNA packaging proteins at a vertex, one of which is the dodecameric portal protein. The procapsid then undergoes maturation and DNA packaging through the portal, driven by a terminase complex. We used cryo-electron tomography to visualize the portal in procapsids and compare them to mature empty capsids. We found the portal located inside one vertex interacting with the scaffold protein in the procapsid. On maturation, the scaffold is cleaved and dissociates, the capsid angularizes, and the portal moves outward, interacting closely with the capsid shell. These transformations may provide a basis for the development of drugs to prevent HSV-1 infections.

FSC-Q: a CryoEM map-to-atomic model quality validation based on the local Fourier shell correlation.

In recent years, advances in cryoEM have dramatically increased the resolution of reconstructions and, with it, the number of solved atomic models. It is widely accepted that the quality of cryoEM maps varies locally; therefore, the evaluation of the maps-derived structural models must be done locally as well. In this article, a method for the local analysis of the map-to-model fit is presented. The algorithm uses a comparison of two local resolution maps. The first is the local FSC (Fourier shell correlation) between the full map and the model, while the second is calculated between the half maps normally used in typical single particle analysis workflows. We call the quality measure "FSC-Q", and it is a quantitative estimation of how much of the model is supported by the signal content of the map. Furthermore, we show that FSC-Q may be helpful to detect overfitting. It can be used to complement other methods, such as the Q-score method that estimates the resolvability of atoms.

High resolution electron tomography and segmentation-by-modeling interpretation in Bsoft.

Bsoft offers many tools for the processing of tomographic tilt series and the interpretation of tomograms. Since I introduced tomography into Bsoft almost two decades ago, the field has advanced significantly, requiring refinement of old algorithms and development of new ones. The current direct detectors allow us to collect data more efficiently and with better quality, progressing towards automation. The goal is then to also automate alignment of tilt series and reconstruction. I added an estimation of the specimen thickness as well as fiducialless alignment, to augment the existing fiducial-based alignment. High-resolution work requires correction for the contrast transfer function, in tomography complicated by the tilted specimen. For this, I developed a method to generate a power spectrum using the whole micrograph, compensating for tilting. This is followed by routine determination of the contrast transfer function, and correction for it during reconstruction. The next steps involve interpretation of the tomogram, either by subtomogram averaging where possible, or by segmentation and modeling otherwise. Such interpretation actually constitutes the main time-consuming part of tomography and is less amenable to automation compared to the initial reconstruction.

The Mottled Capsid of the Salmonella Giant Phage SPN3US, a Likely Maturation Intermediate with a Novel Internal Shell.

"Giant" phages have genomes of >200 kbp, confined in correspondingly large capsids whose assembly and maturation are still poorly understood. Nevertheless, the first assembly product is likely to be, as in other tailed phages, a procapsid that subsequently matures and packages the DNA. The associated transformations include the cleavage of many proteins by the phage-encoded protease, as well as the thinning and angularization of the capsid. We exploited an amber mutation in the viral protease gene of the Salmonella giant phage SPN3US, which leads to the accumulation of a population of capsids with distinctive properties. Cryo-electron micrographs reveal patterns of internal density different from those of the DNA-filled heads of virions, leading us to call them "mottled capsids". Reconstructions show an outer shell with T = 27 symmetry, an embellishment of the HK97 prototype composed of the major capsid protein, gp75, which is similar to some other giant viruses. The mottled capsid has a T = 1 inner icosahedral shell that is a complex network of loosely connected densities composed mainly of the ejection proteins gp53 and gp54. Segmentation of this inner shell indicated that a number of densities (~12 per asymmetric unit) adopt a "twisted hook" conformation. Large patches of a proteinaceous tetragonal lattice with a 67 Å repeat were also present in the cell lysate. The unexpected nature of these novel inner shell and lattice structures poses questions as to their functions in virion assembly.

Local resolution estimates of cryoEM reconstructions.

The field of cryoEM has quickly advanced in last years with the new biochemical, technological, methodological and computational developments. It has allowed significant progresses in Structural Biology, typically reaching quasi-atomic resolutions in the reconstructed maps. However, this rapid advance has also generated new questions relevant to resolution estimates. The global resolution metrics and their criteria have been deeply discussed in the last decade, but despite that, it remains as an important issue in the field. Recently, the introduction of local resolution measurements has changed how cryoEM reconstructions are interpreted, providing information about the existence of heterogeneity, flexibility, and angular assignment errors, and using it as a tool to aid in modeling. In this review we revisit the concept of local resolution and the different algorithms in the current state of the art. However, the concept of local resolution is not uniquely defined, and each implementation measures different features. This may lead to inappropriate interpretation of local resolution maps. Hence, a set of good practices is provided in this review to avoid misleading and over-interpretation of the reconstructions.

Protocols for Processing and Interpreting cryoEM Data Using Bsoft: A Case Study of the Retinal Adhesion Protein, Retinoschisin.

The goal of cryoEM is to determine the structures of biomolecules from electron micrographs. In many cases the processing is straightforward and can be handled with routine protocols. In other cases, the properties and behavior of the specimen require adaptions to properly interpret the data. Here I describe the protocols for examining the higher order assemblies of the retinal adhesion protein, retinoschisin (RS1), using the Bsoft package. The protocols for micrograph preprocessing, 2D classification and 3D alignment and reconstruction follow the usual patterns for the majority of cryoEM specimens. The interpretation of the results is specific to the branched network of RS1 filaments. The 2D class averages are used to determine the relative positions of the RS1 molecules, thus defining the interacting interfaces in the network. The major interface of the linear filament is then further examined by reconstructing the "unit cell" and fitting the molecular models.

Atomic structures of an entire contractile injection system in both the extended and contracted states.

Contractile injection systems are sophisticated multiprotein nanomachines that puncture target cell membranes. Although the number of atomic-resolution insights into contractile bacteriophage tails, bacterial type six secretion systems and R-pyocins is rapidly increasing, structural information on the contraction of bacterial phage-like protein-translocation structures directed towards eukaryotic hosts is scarce. Here, we characterize the antifeeding prophage AFP from Serratia entomophila by cryo-electron microscopy. We present the high-resolution structure of the entire AFP particle in the extended state, trace 11 protein chains de novo from the apical cap to the needle tip, describe localization variants and perform specific structural comparisons with related systems. We analyse inter-subunit interactions and highlight their universal conservation within contractile injection systems while revealing the specificities of AFP. Furthermore, we provide the structure of the AFP sheath-baseplate complex in a contracted state. This study reveals atomic details of interaction networks that accompany and define the contraction mechanism of toxin-delivery tailocins, offering a comprehensive framework for understanding their mode of action and for their possible adaptation as biocontrol agents.

Carbon replicas reveal double stranded structure of tight junctions in phase-contrast electron microscopy.

Replica-based freeze-fracture and freeze-etching electron microscopy methods provide surface topography information, particularly suited to studying membrane protein complexes in their native context. The fidelity and resolution of metal replicas is limited by the inherent property of metal atoms to crystallize. To overcome the limitations of metal replicas, we combined amorphous carbon replicas with phase-contrast electron microscopy. Using this approach, tight junction intramembrane fibrils were shown to have a double stranded morphology.

Single-particle reconstruction statistics: a diagnostic tool in solving biomolecular structures by cryo-EM.

In single-particle analysis (SPA), the aim is to obtain a 3D reconstruction of a biological molecule from 2D electron micrographs to the highest level of detail or resolution as possible. Current practice is to collect large volumes of data, hoping to reach high-resolution maps through sheer numbers. However, adding more particles from a specific data set eventually leads to diminishing improvements in resolution. Understanding what these resolution limits are and how to deal with them are important in optimization and automation of SPA. This study revisits the theory of 3D reconstruction and demonstrates how the associated statistics can provide a diagnostic tool to improve SPA. Small numbers of images already give sufficient information on micrograph quality and the amount of data required to reach high resolution. Such feedback allows the microscopist to improve sample-preparation and imaging parameters before committing to extensive data collection. Once a larger data set is available, a B factor can be determined describing the suppression of the signal owing to one or more causes, such as specimen movement, radiation damage, alignment inaccuracy and structural variation. Insight into the causes of signal suppression can then guide the user to consider appropriate actions to obtain better reconstructions.

Cryo-EM of retinoschisin branched networks suggests an intercellular adhesive scaffold in the retina.

Mutations in the retinal protein retinoschisin (RS1) cause progressive loss of vision in young males, a form of macular degeneration called X-linked retinoschisis (XLRS). We previously solved the structure of RS1, a 16-mer composed of paired back-to-back octameric rings. Here, we show by cryo-electron microscopy that RS1 16-mers can assemble into extensive branched networks. We classified the different configurations, finding four types of interaction between the RS1 molecules. The predominant configuration is a linear strand with a wavy appearance. Three less frequent types constitute the branch points of the network. In all cases, the "spikes" around the periphery of the double rings are involved in these interactions. In the linear strand, a loop (usually referred to as spike 1) occurs on both sides of the interface between neighboring molecules. Mutations in this loop suppress secretion, indicating the possibility of intracellular higher-order assembly. These observations suggest that branched networks of RS1 may play a stabilizing role in maintaining the integrity of the retina.

The first single particle analysis Map Challenge: A summary of the assessments.

The recent successes of cryo-electron microscopy fostered great expectation of solving many new and previously recalcitrant biomolecular structures. However, it also brings with it the danger of compromising the validity of the outcomes if not done properly. The Map Challenge is a first step in assessing the state of the art and to shape future developments in data processing. The organizers presented seven cases for single particle reconstruction, and 27 members of the community responded with 66 submissions. Seven groups analyzed these submissions, resulting in several assessment reports, summarized here. We devised a range of analyses to evaluate the submitted maps, including visual impressions, Fourier shell correlation, pairwise similarity and interpretation through modeling. Unfortunately, we did not find strong trends. We ascribe this to the complexity of the challenge, dealing with multiple cases, software packages and processing approaches. This puts the user in the spotlight, where his/her choices becomes the determinant of map quality. The future focus should therefore be on promulgating best practices and encapsulating these in the software. Such practices include adherence to validation principles, most notably the processing of independent sets, proper resolution-limited alignment, appropriate masking and map sharpening. We consider the Map Challenge to be a highly valuable exercise that should be repeated frequently or on an ongoing basis.

Map Challenge assessment: Fair comparison of single particle cryoEM reconstructions.

Cryo-electron microscopy (cryoEM) is capable of achieving near-atomic resolution of biomolecular structures due to recent advances in hardware. Despite the long history of image processing software development for cryoEM, uncertainty about best practices and validation remains. The Map Challenge was therefore designed to test the current state of single particle reconstruction. As the first such challenge, the participants were given the freedom to analyze the cases in whichever way they wanted. Therefore, the maps submitted feature different sizes, sampling and orientations, making assessment non-trivial. To be fair, I developed a method to pose all maps in each case in the same configuration with minimal interpolation. I assessed the quality of these maps by visual inspection and Fourier shell correlation (FSC). Comparing the even-odd FSC with an FSC calculated against a reference structure analysis, I concluded that the quality of the maps related more to the user than to other factors, such as the software package used. Poor quality maps suffer either from lack of data or poor choices made by the user. Some maps appear significantly better than a reference or consensus of other maps, indicating overfitting. Best practices to avoid problems include an understanding of the effects of reference map modifications on particle image alignment, and generating appropriate masks. Ultimately, none of the issues revealed in the Map Challenge is insurmountable, as underscored by the excellent quality of reconstructions achieved by a significant number of participants.

Single particle reconstruction and validation using Bsoft for the map challenge.

The Bsoft package is aimed at processing electron micrographs for the determination of the three-dimensional structures of biological specimens. Recent advances in hardware allow us to solve structures to near atomic resolution using single particle analysis (SPA). The Map Challenge offered me an opportunity to test the ability of Bsoft to produce reconstructions from cryo-electron micrographs at the best resolution. I also wanted to understand what needed to be done to work towards full automation with validation. Here, I present two cases for the Map Challenge using Bsoft: ß-galactosidase and GroEL. I processed two independent subsets in each case with resolution-limited alignment. In both cases the reconstructions approached the expected resolution within a few iterations of alignment. I further validated the results by coherency-testing: i.e., that the reconstructions from real particles give better resolutions than reconstructions from the same number of aligned noise images. The key operations requiring attention for full automation are: particle picking, faster accurate alignment, proper mask generation, appropriate map sharpening, and understanding the amount of data needed to reach a desired resolution.