Journal of Structural Biology

Structure elucidation of biological macromolecules by single particle cryogenic electron microscopy (SPA cryo-EM) or cryogenic electron tomography (cryo-ET) relies on low-dose imaging on cryogenic transmission electron microscopes (cryo-TEMs). Routine microscope setup remains technically demanding and can be time-consuming, particularly for inexperienced or infrequent users. We present LowDoseWizard, a guided workflow implemented in SerialEM that enables rapid and standardised setup of cryo-TEM imaging conditions. From minimal user input, the workflow configures microscope optics, camera parameters and image shift settings for all low-dose imaging states, and guides the user through key daily alignment procedures including beam shift offset calibration, objective lens astigmatism correction and coma-free alignment. The workflow is organised into modular routines that can be executed sequentially or independently, while microscope-specific acquisition parameters are defined in editable configuration files, allowing flexible adaptation to different instruments without modification of the core scripts. Across user sessions on three microscopes at EMBL Heidelberg, the complete setup required on average less than 15 minutes. To assess whether predefined imaging conditions generated by the workflow are compatible with high-resolution data collection, we acquired apoferritin data on a 200 kV Glacios and a 300 kV Titan Krios. These datasets yielded reconstructions at 1.62 [A] and 1.09 [A] resolution, respectively, demonstrating that rapid, guided setup can support near-atomic and atomic-resolution single particle cryo-EM. LowDoseWizard lowers the barrier to robust cryo-TEM setup, reduces the time spent on routine parameter selection and alignment, and helps users focus on sample-specific aspects of data acquisition such as target selection. The workflow should be particularly valuable in shared instrumentation environments, where accessibility, reproducibility and efficient microscope use are critical.

Due to recent technological advances, in situ structural cell biology is becoming a high throughput microscopy technique as all the steps of the workflow, from sample preparation to data analysis, are executed faster, more reliable and more reproducible. Sample thinning by cryoFIB-SEM is an essential tool in preparing electron transparent lamellae of biological specimens suitable for further characterization by cryoET. Modern cryoFIB-SEM instruments can be operated remotely and are capable of automated and unsupervised lamellae preparation. To take full advantage of these developments they need a constant supply of LN2 to maintain cryogenic conditions inside the microscope chamber. Here, we introduce a custom automated LN2 refill system that is compatible with gas cooled cryostages, supports long-term cryoFIB-SEM operations and liberates the user from highly repetitive and manual work. We believe this solution can be utilized with other cryoSEM or cryoFIB-SEM devices requiring N2 gas-flow cooling.

Preparing electron-transparent cryo-lamellae is inherently a serial and low-throughput process. Once the lamellae are milled, these thin structures endure both mechanical and thermal stress, and as a result many valuable lamellae crack or even disintegrate entirely. This loss is often regarded as a "lamella tax", i.e. an unavoidable cost of working with such fragile specimens. In this work, we introduce two modifications to the standard lamella-preparation workflow aimed at improving lamella mechanical resistance to crack formation and external stress. The first modification involves milling arrays of perforations directly within the lamella body. These perforations are designed to function as crack-arrest holes, intercepting cracks as they appear and preventing, or at least delaying their further propagation. By slowing crack growth, these features increase the likelihood that the lamella remains intact long enough to complete cryo-TEM imaging. The second modification replaces the conventional rigid attachment of the lamella to the surrounding cellular bulk material with a softer suspension using ring-shaped springs formed by ion beam milling. Mounting the lamella on smooth annular springs provides mechanical compliance both across and along the lamella axis, as well as at intermediate angles and in the out-of-plane direction. This flexibility allows the lamella to accommodate larger stresses and deformations without reaching its mechanical failure threshold. We fabricated a series of test lamellae incorporating different crack-arrest hole geometries, as well as lamellae suspended on soft annular springs. We performed high-resolution cryo-TEM imaging to characterise the perforations themselves and characterised the captured crack geometry within the lamellae at the highest level of detail achieved to date. TEM imaging shows crack interception and guided, non-catastrophic failure paths, while simulations confirm lowered stress in suspended lamellae.

Single-particle cryogenic electron microscopy (cryoEM) is a widely used technique for structure determination of biomacromolecules to near-atomic resolution. Random distributions of these molecules in vitrified ice are necessary to accumulate enough two-dimensional views to generate a complete three-dimensional (3-D) reconstruction. However, interactions between the sample and the air-water interface (AWI) that occur during vitrification often bias the views of the sample, a phenomenon termed preferred orientation, limiting our ability to obtain 3-D reconstructions. Surfactants are often used as sample additives to prevent AWI-induced deterioration, but no general strategy exists for surfactant choice, requiring laborious screening for each sample. To circumvent these issues, we developed SurfACT, a cocktail of diverse surfactants with distinct physicochemical properties that limits AWI-dependent sample denaturation and orientation bias, while mitigating individual surfactant-specific drawbacks. Here we demonstrate SurfACTs effectiveness with four proteins plagued by AWI-induced issues, including two species of hemagglutinin (HA), molybdenum-iron protein (MoFeP) from the nitrogenase enzyme, and aldolase. All four samples show drastically improved viewing distribution and map completeness when SurfACT is applied. Cryogenic electron tomography demonstrates that SurfACT redistributes particles from the AWI into the bulk ice, driving signal recovery and inhibiting denaturation. This versatile sample additive minimizes sample-specific screening and expands the capabilities and range of suitable samples for cryoEM.

Bacteria microcompartments (BMCs) are pseudo-organelles comprised of a self-assembling, semi-permeable protein shell, most commonly enclosing components of enzymatic pathways. -Carboxysomes (-CBs) are anabolic BMCs known for their role in sequestering Rubisco, the enzyme responsible for carbon fixation in plants, algae and bacteria, along with an upstream enzyme and an assembly protein. Rubisco has low selectivity for its substrate, CO2, and has a slow enzymatic turnover rate, resulting in an inefficient metabolic pathway. Within the -CB, Rubisco has been observed at a range of concentrations and with either a liquid-like assembly or a pseudo-lattice of polymerized fibrils. The biophysical origins of the fibril ultrastructure organization are unclear; however, it is only observed inside -CBs. Quantitative knowledge of the binding constants and energies for assembly and maintenance of these fibrils is critical for understanding this organization and Rubisco regulation, but quantitative methods for in situ analysis of Rubisco polymerization have been lacking. Here, we present an approach to convert tomography-derived -CB volumes and Rubisco particle positions into polymerization binding curves. We used this procedure to determine the Rubisco polymerization constants, including the nucleus size (n) and equilibrium polymerization constant (Kpol). The adopted modeling approach is consistent with in situ constraints, such as concentration-dependent binding interactions and confinement. This approach offers a powerful tool to evaluate both in vitro and potentially in vivo biomolecular interactions, both of Rubisco and of other proteins and polymers suitable for analysis by cryo-electron tomography. Significance StatementCryogenic electron tomography (cryoET) is a powerful method to resolve structures of proteins in their native environment at subnanometer-level resolution. Because tomography data retains spatial relationships of all particles, it intrinsically contains information about component (e.g., protein) binding interactions. Here, we use Rubisco polymerization in -carboxysomes as a model system to demonstrate that quantitative, biochemical binding analysis is possible with cryoET.

Affinity chromatography is a powerful and therefore popular method for the purification of proteins for structural studies. The success of the technique relies on the specificity of the interaction between the target protein and the affinity resin. Here, we present the identification of two protein contaminants isolated from HEK293 cell lysate following affinity purification of twin Strep-tagged or FLAG-tagged proteins. The contaminants were identified as human propionyl-coenzyme A carboxylase (hPCC) and protein arginine methyltransferase 5 in complex with methylosome protein 50 (PRMT5:MEP50) via a combination of cryo-EM data processing and proteomic analyses. This report serves to illustrate how these contaminants may appear in cryo-EM datasets and to highlight the paramount importance of affinity chromatography resin specificity for efficient protein purification.

Calculation of density maps from atomic models is essential for structural studies using crystallography and electron cryo-microscopy (cryoEM). These maps serve various purposes, including atomic model building, refinement, visualization, and validation. However, accurately comparing model-calculated maps to experimental data poses challenges, particularly because the resolution of cryoEM experimental maps varies across the map. Traditional crystallography methods generate finite-resolution maps with uniform resolution throughout the unit cell volume, while most modern software in cryoEM employ Gaussian-like functions to generate these maps, which does not adequately account for atomic model parameters and resolution. Recent work by Urzhumtsev & Lunin (2022, IUCr Journal, 9, 728-734) introduces a novel method for computing atomic model maps that incorporate local resolution and can be expressed as analytically differentiable functions of all atomic parameters. This approach enhances the accuracy of matching atomic models to experimental maps. In this paper, we detail the implementation of this method in CCTBX and Phenix. SynopsisNew tools implemented in CCTBX and Phenix allow the calculation of variable-resolution maps through a sum of atomic images expressed as analytic functions of all atomic parameters, along with their associated local resolution.

Mitochondria are dynamic signaling organelles that transduce metabolic and biochemical cues to facilitate cellular adaptation. Their complex structure and dynamics are essential for integrating metabolic pathways, responding to stressors, and communicating inter- and intra-cellular signals. While optimal mitochondrial activity is frequently linked to cellular and organismal health--influencing processes ranging from metabolism and regulated cell death to differentiation and growth--the mechanistic links between mitochondrial dysfunction and cellular defects leading to human disease remain incompletely understood. Understanding how mitochondrial shape and function are linked is crucial for deciphering the regulatory mechanisms of cell survival and fate. Here, we present a molecular resolution cryo-electron tomography (cryo-ET) imaging and image analysis platform to investigate the structure of isolated human mitochondria under different conditions. We describe optimized protocols for isolating mitochondria from human cells, vitrifying these samples with high-pressure freezing (HPF) using the waffle method, cryo-focused ion beam (cryo-FIB) milling to generate thin sections (lamellae), and imaging with cryo-transmission electron microscopy (cryo-TEM). This is complemented by a robust downstream processing pipeline for tilt-series alignment, tomogram reconstruction, and three-dimensional (3D) segmentation of tomograms using the latest state-of-the-art algorithms. With some variations, this versatile workflow is adaptable to other subcellular compartments for structural studies in isolation or within intact cells. Furthermore, our protocols provide a critical foundation for investigating the in-situ structure of protein machineries that govern key cellular processes.

Surface ice contamination is a persistent challenge in cryo-electron tomography (cryo-ET) workflows, where it can obscure regions of interest and contribute to curtaining artefacts during focused ion beam (FIB) milling. We demonstrate using high-pressure frozen yeast cells that a sublimation step within the scanning electron microscope (SEM) chamber before lamella milling visually removes surface ice and reduces sample roughness without detectable devitrification. While sublimation has been widely applied in cryo-SEM and volume imaging, it is not common on cryo-ET samples due to concerns about devitrification. Using tomographic reconstructions, we show that controlled sublimation improves lamella quality by reducing surface roughness and minimizing curtaining without compromising sample vitrification. Furthermore, subtomogram averaging of the 80S ribosome confirmed lamellae quality are preserved after sublimation. This approach offers a practical refinement to existing cryo-ET preparation protocols, requiring no additional instrumentation or workflow modifications.

Cryo-electron tomography (cryo-ET) is fast developing from a tool primarily used to investigate structures of individual macromolecular complexes in situ into a high-resolution probe for molecular processes within diverse functional contexts in intact cells. It is thus increasingly necessary that the data are analyzed and quantified as completely as possible. But annotating and structurally characterizing macromolecular complexes with a high degree of completeness is a significant challenge, especially for smaller molecular targets. In particular, it is difficult to avoid incomplete localizations of complexes, false identifications, or losses during computational classification. To address these issues, we assessed parameters in data processing, including the role of voxel size in template matching, the effects of Volta phase plate imaging on localization, classification, and map refinement, and the extent to which multi-particle-based refinement of tiltseries improves these data processing steps. Our analyses provide practical guidelines that help maximize completeness in cellular cryo-ET data; accurate description of the sample is crucial for visual proteomics experiments, and these optimizations help ensure that data annotation and analysis are comprehensive.

Cryo-Focused Ion Beam Scanning Electron Microscopy (cryoFIB-SEM) using samples fixed by high-pressure freezing uniquely enables high resolution cryo-volume Electron Microscope (cvEM) images of cell ultrastructure to be obtained from whole cells and complex tissues in their near native state. As the freezing process also preserves fluorescence, the link between three-dimensional (3D) ultrastructure and biological process is also enabled by targeted cryo-Correlative Light and Electron Microscopy (CLEM). However, the overall viability of cvEM is challenged by sample preparation, charge balance during imaging, sample sensitivity to beam damage, contamination, and very long acquisition times. Here we detail new experimental workflows to significantly reduce each of these effects and demonstrate the improvement in resolution possible with results from the nematode Caenorhabditis elegans and the ciliated protozoon Paramecium bursaria containing many endosymbiotic algae. These results demonstrate the versatility and potential wide-ranging utility of cvEM for 3D ultrastructural imaging of whole multicellular and unicellular organisms.

Standardized biological specimens are essential for optimizing cryoEM workflows and benchmarking instrument performance. While apoferritin fulfills this role for single-particle analysis, no equivalent exists for cryo-electron tomography. Ribosomes are frequently used but require large datasets due to C1 symmetry and structural heterogeneity, limiting rapid optimization and standardized comparison of workflows. Here, we present PP7 virus-like particles (VLPs) overexpressed in E. coli as a scalable in situ benchmark. VLPs have high orders of symmetry enabling rapid, high-resolution validation of tomographic pipelines from minimal datasets, while their distinct structural features across low to high resolutions provide a practical resolution metric.

Coccolithophores are unicellular marine phytoplankton that produce complex and intricately shaped mineralised scales called coccoliths. Coccoliths are produced in an intracellular vesicle where crystal nucleation occurs, from which several individual calcite units develop with anisotropic crystallographic facets, prompting studies into the cellular mechanisms which control crystal growth within the cell. Here, we characterise those morphological developments in 3D that occur during the formation of coccoliths by the species Gephyrocapsa huxleyi using cryo-ptychographic X-ray computed tomography. This technique is ideally suited to study coccolith mineral development, as intracellular structures can be imaged intact in their native state without needing to disrupt cells. Combined with additional imaging of developing coccoliths using cryo-transmission electron microscopy and scanning electron microscopy, we report the developmental stages involved in coccolith growth across the complete mineralisation period, while also showing that the constrained space created by individual crystal units growing in close confinement affects the final crystal morphology and overall mineral structure. These findings provide clarification on the mineralisation pathways that coccolithophores and other biomineralising organisms use to control the formation of highly functionalised crystalline structures, particularly relevant in the design of materials with tunable properties.

Cryo-electron tomography (cryo-ET) enables in situ three-dimensional visualization of many protein complexes and other macromolecular assemblies such as ribosomes in cells, yet automated macromolecule particle identification in 3D cryo-ET tomograms remains a major bottleneck due to dose-limited low signal-to-noise ratios, missing-wedge artifacts, and densely crowded cellular backgrounds. We present TomoSwin3D, an end-to-end three-dimensional (3D) macromolecule particle identification and classification pipeline centered on a Swin Transformer-based U-Net that performs particle identification and classification and outputs particle centroid coordinates. TomoSwin3D leverages a multi-channel input representation that augments raw tomogram densities with complementary 3D feature maps capturing edge strength (Sobel gradients), local contrast enhancement (morphological top-hat), and multiscale blob responses (Difference-of-Gaussians), improving detectability of small and low-contrast targets. To better preserve particle geometry and avoid hand-crafted shape assumptions, it adopts occupancy-preserving supervision that directly uses available 3D instance masks rather than heuristic Gaussian/spherical labels and applies scalable patch-wise inference followed by lightweight post-processing (connected-component analysis, size filtering, centroid extraction) for robust centroid coordinate extraction. Across diverse simulated and experimental cryo-ET tomogram benchmarks including SHREC 2021 and 2020 test datasets, EMPIAR dataset, and Cryo-ET data portal dataset, TomoSwin3D achieves strong and consistent performance in detecting proteins and other particles, outperforming existing methods, with a pronounced advantage in picking hard, small protein particles. These results establish TomoSwin3D as a scalable and accurate solution for high-throughput cryo-ET macromolecule particle picking and downstream subtomogram averaging.

We present new tools for the structure determination of amyloid filaments from electron cryo-microscopy (cryo-EM) images. We introduce a new algorithm for automated filament picking, based on their characteristic 4.75 [A] repeat signal; we implement the new auto-picker in a fully automated procedure for on-the-fly pre-processing of cryo-EM data sets of amyloid filaments; we present a graphical tool to select filament types based on bi-hierarchical clustering of filaments and 2D class assignments; and we introduce a denoising neural network for Blush regularisation that is re-trained on amyloid reconstructions. The implementation of these tools in release 5.1 of our open-software package RELION ensures broad applicability. We demonstrate their usefulness on two experimental data sets, including a previously described data set on recombinant human islet amyloid protein (hIAPP) with the S20G mutation for which we identify two new filament types.

In root nodule symbiosis, symbiosome compartments accommodate nitrogen-fixing rhizobia inside the plant cell. Differentiated into bacteroids, the rhizobia are surrounded by a peribacteroid space and a plant-derived peribacteroid membrane, which separates them from the plant cytoplasm but allows signal and nutrient exchange between host and microbe. The morphological features of symbiosomes are primarily determined by ultrastructural single focal plane imaging, with limited information about spatial details. This study combines 2D and 3D imaging, using transmission electron microscopy and focused ion beam scanning electron microscopy as complementary techniques to analyse the symbiosome ultrastructure and organisation in Lotus japonicus wild-type plants. The 3D model of a mature colonised root nodule cell region demonstrates a dense, puzzle-like arrangement of symbiosomes relative to one another and adjacent plant organelles. The symbiosome shape and size depends on the orientation and number of bacteroids within the compartment and features connective tubular structures. Furthermore, vesicular structures, some likely of bacterial origin, were present at the interface. The study presents a multi-angled analysis of symbiosome-related structures, highlighting their volumes, spatial distribution, and pronounced compactness. Interface associated vesicles, protrusions and connective structures hint towards a dynamic and flexible system that contributes to the plant-microbe crosstalk.

1.Congenital heart disease is commonly studied using genetically modified mouse models, but characterization of complex three-dimensional (3D) cardiac anomalies remains technically challenging. Histology of fixed paraffin-embedded samples, typically used for structural and molecular analysis, is limited to two dimensions (2D). Synchrotron radiation-based phase-contrast micro-computed tomography (SRPC-{micro}CT) enables rapid, high-resolution, 3D imaging but had yet to be fully integrated with molecular tissue analysis. Importantly, SRPC-{micro}CT is nondestructive and compatible with paraffin-embedded tissue, potentially allowing integration with downstream molecular analyses such as multiplexed spatial transcriptomic profiling on tissue sections. Here, we present a pipeline combining SRPC-{micro}CT with complementary molecular approaches for rapid and accurate phenotyping of mouse hearts and for investigating disease mechanisms. We demonstrate the successful integration of high-resolution 3D imaging with spatial transcriptomics, fluorescent stainings, and histochemistry. Notably, all 2D modalities were applied sequentially to a single tissue section and registered within the 3D volume. This multimodal framework provides a powerful approach for linking structural and molecular information that is broadly applicable across biomedical research.

The structural and functional characteristics of membrane proteins can be influenced by the composition of the membrane. Consequently, native membranes are most relevant for the study of receptors and other membrane proteins. In this study, we investigated two types of cell-derived vesicles: natively shed extracellular vesicles (EVs) and mechanically derived vesicles (MVs). To this end, we utilized the human breast cancer cell line SKBR3, which strongly overexpresses the receptor HER2. We designed a protocol based on designed ankyrin repeat proteins (DARPins) to purify EVs and MVs enriched in HER2, and to ensure the native orientation of the HER2 receptors within the vesicle. The isolated HER2-containing EVs and MVs were characterized by cryo-EM, cryo-electron tomography (cryo-ET) and mass spectrometry (MS), which revealed fundamental differences between the different vesicle types. Our study highlights the greater structural diversity of EVs over MVs. A single particle cryo-EM analysis and classification of all visible receptors on the vesicle surface yielded electron density consistent with HER2 at modest resolution. Taken together, our results suggest that MVs can serve better than EVs as a suitable platform for the structure determination of membrane proteins within their native membrane environments.

Cryo-electron tomography (cryo-ET) enables in situ structural analysis of macromolecular complexes within native cellular environments. However, the limited field of view required for high-resolution structure determination necessarily restricts a wider assessment of the broader cellular context. We present a multi-magnification cryo-ET acquisition strategy that integrates low- and high-magnification information from coincident sample regions during the same tilt-series. By interleaving acquisition of the magnifications at each tilt angle, this strategy enables simultaneous collection of large field-of-view, low-resolution tomograms and high-resolution, small field-of-view tomograms while minimising the impact of the increased electron dose. We demonstrate that we can capture cellular organisation across tens of microns, while still enabling subtomogram averaging to resolutions below 4 [A]. This integrated acquisition framework establishes a practical route to multi-scale cryo-ET, bridging molecular and cellular scales for more comprehensive biological insight.

Cryo-electron tomography (cryo-ET) is a powerful approach for visualizing macromolecular structures directly within cells, but its broader application is limited by the difficulty of reliably targeting specific structures for imaging. In particular, capturing small or rare objects within FIB-milled lamellae remains a major bottleneck. Here, we establish fluorescence-guided cryo-FIB milling workflows that overcome key sources of targeting error and enable routine capture of structures across a wide size range. For larger objects (>500 nm), we develop a single step registration-based targeting strategy that combines FIB-milled fiducials with physically grounded depth correction to account for focal shifts arising from refractive index mismatch. For smaller targets (150-500 nm), we implement real-time fluorescence-guided milling on a commercially available FIB SEM instrument with an integrated cryo fluorescence microscope allowing dynamic monitoring and precise termination of milling at the onset of target ablation. Using this strategy, we achieve consistent recovery of lamellae containing the targeted structure, including small single-copy organelles such as centrioles and cilia. Together, these workflows expand the accessible target space for cryo-ET and provide practical solutions for studying cellular structures that have previously been difficult to capture.