Challenges and opportunities in cryo-EM single-particle analysis

Abstract

Cryogenic electron microscopy (cryo-EM) enables structure determination of macromolecular objects and their assemblies. Although the techniques have been developing for nearly four decades, they have gained widespread attention in recent years due to technical advances on numerous fronts, enabling traditional microscopists to break into the world of molecular structural biology. Many samples can now be routinely analyzed at near-atomic resolution using standard imaging and image analysis techniques. However, numerous challenges to conventional workflows remain, and continued technical advances open entirely novel opportunities for discovery and exploration. Here, I will review some of the main methods surrounding cryo-EM with an emphasis specifically on single-particle analysis, and I will highlight challenges, open questions, and opportunities for methodology development.

Keywords: atomic resolution; cryo-electron microscopy; protein structure; protein–protein interaction; single-particle analysis; structural biology.

Figure 1.

General workflow for single-particle analysis. The main steps in the SPA workflow are depicted and will be referred to throughout the text. Although the workflow is depicted as approximately linear, often times the process is iterative, and it may be necessary to go back and optimize individual steps prior to proceeding forward.

Figure 2.

Preferred particle orientation and specimen tilting. A, top view of a cryo-EM grid (schematic) and selected cross-section within a particular hole. Particles within each hole are typically adsorbed to one of two air–water interfaces, causing them to adopt a preferred orientation on grids (87). Preferred particle orientation leads to anisotropic resolution in the reconstructed map. B, to overcome the preferred orientation and anisotropic resolution problem, the grid can be tilted inside the electron microscope (electron beam is in green). This results in more even coverage of Fourier space voxels and an improvement in the reconstructed volume.

Figure 3.

Experimental results showing the utility of a tilted data collection strategy. Comparison of HA trimer reconstructions refined and reconstructed independently from 0°-untilted images (top) or from 40°-tilted images (bottom). A, left to right, side view of the experimental reconstruction (in gray, the direction of preferred orientation is indicated by the red arrowhead) superimposed onto a projection of the envelope of the HA trimer crystal structure (in pink), displayed alongside a top view and a close-up of a particular region. The black arrowhead indicates streaking in the unsharpened map, which results from misalignment of orientations and/or overfitting. B, experimental half-map and map-to-model 3D FSC isosurfaces. The map-to-model 3D FSCs, which measure correlation to the true structure, are much worse than the half-map 3D FSCs, which measure the internal correlation of two randomly selected half-subsets of the data during refinement. This indicates that some amount of overfitting is present during refinement, since the resulting maps are worse than their apparent reported resolution. Figure adapted from Ref. .

Figure 4.

Advances in cryo-EM map depositions and resolutions by year. A, new depositions into the Electron Microscopy Data Bank by year. B, resolutions of deposited maps in 2018. C, highest resolutions for non-icosahedral specimens by year. D, highest resolutions for icosahedral specimens by year.

Figure 5.

High-resolution cryo-EM maps. A, β-gal, resolved to 1.90 Å (125). B, adeno-associated virus type 2, resolved to 1.86 Å (the reconstruction indicated the presence of hydrogen atoms) (124). C, human apoferritin, resolved to 1.65 Å (126). D, mouse apoferritin, resolved to 1.62 Å (R. Danev, H. Yanagisawa, and M. Kikkawa, manuscript in preparation), represent the state-of-the-art in cryo-EM reconstructions. The codes (EMD-####) refer to the deposition numbers within the electron microscopy databank (EMD).

Figure 6.

Validation procedures for cryo-EM maps and models. Steps for helping to ensure correctness of the derived cryo-EM map (A) and associated atomic model (B). Specific recommendations can be found in Refs. , .