Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2018 May 29:7:e34257.
doi: 10.7554/eLife.34257.

Routine single particle CryoEM sample and grid characterization by tomography

Alex J Noble  1 , Venkata P Dandey #  1 , Hui Wei #  1 , Julia Brasch  1   2 , Jillian Chase  3   4 , Priyamvada Acharya  1   5 , Yong Zi Tan  1   2 , Zhening Zhang  1 , Laura Y Kim  1 , Giovanna Scapin  1   6 , Micah Rapp  1   2 , Edward T Eng  1 , William J Rice  1 , Anchi Cheng  1 , Carl J Negro  1 , Lawrence Shapiro  2 , Peter D Kwong  5 , David Jeruzalmi  3   4   7   8 , Amedee des Georges  3   4   8   9 , Clinton S Potter  1   2 , Bridget Carragher  1   2
Affiliations

Routine single particle CryoEM sample and grid characterization by tomography

Alex J Noble et al. Elife. .

Abstract

Single particle cryo-electron microscopy (cryoEM) is often performed under the assumption that particles are not adsorbed to the air-water interfaces and in thin, vitreous ice. In this study, we performed fiducial-less tomography on over 50 different cryoEM grid/sample preparations to determine the particle distribution within the ice and the overall geometry of the ice in grid holes. Surprisingly, by studying particles in holes in 3D from over 1000 tomograms, we have determined that the vast majority of particles (approximately 90%) are adsorbed to an air-water interface. The implications of this observation are wide-ranging, with potential ramifications regarding protein denaturation, conformational change, and preferred orientation. We also show that fiducial-less cryo-electron tomography on single particle grids may be used to determine ice thickness, optimal single particle collection areas and strategies, particle heterogeneity, and de novo models for template picking and single particle alignment.

Keywords: air-water; cryoET; fiducial-less; molecular biophysics; none; protomo; single particle; structural biology; tomography.

PubMed Disclaimer

Conflict of interest statement

AN, VD, HW, JB, JC, PA, YT, ZZ, LK, GS, MR, EE, WR, AC, CN, LS, PK, DJ, Ad, CP, BC No competing interests declared

Figures

Figure 1.
Figure 1.. Schematic diagrams of grid hole cross-sections containing regions of ideal particle and ice behavior for single particle cryoEM collection.
(A) A grid hole where all regions of particles and ice exhibit ideal behavior. (B) Grid holes where there are areas that exhibit ideal particle and ice behavior. Green arrows indicate areas with ideal particle and ice behavior. The generic particle shown is a low-pass filtered holoenzyme, EMDB-6803 (Yin et al., 2017). The particles were rendered with UCSF Chimera (Pettersen et al., 2004).
Figure 2.
Figure 2.. Depictions of potential ice and particle behavior in cryoEM grid holes, based on Figure 6 from (Taylor and Glaeser, 2008).
A region of a hole may be described by a combination of one option from (A) for each air-water interface and one or more options from (B). An entire hole may be described by a set of regions and one or more options from (C). (A) Each air-water interface might be described by either (1), (2), or (3). Note that cryoET might only be able to resolve tertiary and secondary protein structures/network elements at the air-water interface. (B) Particle behavior between air-water interfaces and at each interface might be composed of any combination of (1) through (5), with or without aggregation. B3 is different from B4 if, for example, a particle prone to denaturation is frozen before or after denaturation has begun, thus potentially changing the set of preferred orientations. At high enough concentrations additional preferred orientations might become available in B3 and B4 due to neighboring protein-protein interactions. (C) Ice thickness variations through a central cross-section of hole may be described by one option for one air-water interface and one option for the apposed interface. Note that in C1 the particle's minor axis may be larger than the ice thickness. In both C1 and C4, the particle may still reside in areas thinner than its minor axis if the particle is compressible. Phenomenon such as bulging or doming (Brilot et al., 2012) may be represented as a combination of C1-4.
Figure 3.
Figure 3.. Schematic diagrams of the average ice thickness (solid lines) ± (1 standard deviation and measurement error) (dashed lines) using the minimum measured values, average particle layer tilt (solid lines) ± (1 standard deviation and measurement error) (dashed lines), and percentage of samples with single and/or double particle layers (‘1’ and/or ‘2’ as defined in Table 1) at the centers of holes (A) and about 100 nm from the edge of holes (B).
Figure 4.
Figure 4.. A selection of cross-sectional schematic diagrams of particle and ice behaviors in holes as depicted according to analysis of individual tomograms.
The relative thicknesses of the ice in the cross-sections are depicted accurately. Each diagram is tilted corresponding to the tomogram from which it is derived; i.e. the depicted tilts represent the orientation of the objects in the field of view at zero-degree nominal stage tilt. If the sample concentration in solution is known, then it has been included below the sample name. Black lines on schematic edges are the grid film. The cross-sectional characteristics depicted here are not necessarily representative of the aggregate. An asterisk (*) indicates that a Video of the schematic diagram alongside the corresponding tomogram slice-through video is included for the sample. A dagger (†) indicates that a dataset is deposited for sample. A generic particle, holoenzyme EMDB-6803 (Yin et al., 2017), is used in place of some confidential samples (samples #40, 41, and 46).
Figure 5.
Figure 5.. Slices of tomograms, about 7 nm thick, showing variations in particle orientation of adsorbed and non-adsorbed particles for several samples.
Cross-sectional schematic diagrams showing the approximate locations of the slices are shown on the right. (A) HIV-1 trimer complex 1 shows a high degree of preferred orientation for particles adsorbed to the air-water interface and no apparent preferred orientation for non-adsorbed particles. (B) Rabbit muscle aldolase shows several views for adsorbed particles and non-preferred views for non-adsorbed particles. (C) DnaB helicase-helicase loader shows no apparent preferred orientation for adsorbed particles. (D) T20S proteasome shows predominantly one view for adsorbed particles, the same view for particles adsorbed to the primary layer of particles, and less preferred views for non-adsorbed particles. Scale bars are 100 nm.
Figure 6.
Figure 6.. Slices of tomograms, about 10 nm thick, at air-water interfaces of samples that show clear protein fragments (examples indicated with blue arrows) and/or partial particles (examples indicated with green arrows), presented roughly in order of decreasing overall fragmentation.
(A) Neural receptor shows a combination of fragmented 13 kDa domains consisting primarily of β-sheets and partial particles. (B) Apoferritin shows apparent fragmented strands and domains along with partial particles. (C) Hemagglutinin shows a clear dividing line, marked with blue, where the ice became too thin to support full particles, but thick enough to support protein fragments. (D) HIV-1 trimer complex one shows several protein fragments on the order of 10 kDa; however, these might be receptors intentionally introduced to solution before plunge-freezing. (E) GDH shows protein fragments interspersed between particles. (F) T20S proteasome shows partial particles, determined by measuring their heights in the z-direction, on an otherwise clean air-water interface (see the end of Video 10 for sample #42). For the examples shown here, it is not clear whether the protein fragments and partial particles observed are due to unclean preparation conditions, protein degradation in solution, or unfolding at the air-water interfaces, or a combination; all cases are expected to result in the same observables due to competitive and sequential adsorption. Scale bars are 100 nm.
Figure 7.
Figure 7.. Collection and processing limits imposed by variations in ice thickness (A) and particle layer tilt (B), given that the vast majority of particles in holes on conventionally-prepared cryoEM grids are adsorbed to an air-water interface.
(A) Variations in ice thickness within and between holes might limit the number of non-overlapping particles in projection images (efficiency of collection and processing), the accuracy of whole image and local defocus estimation (accuracy in processing), the signal-to-noise ratio in areas of thicker ice (efficiency of collection and processing), and the reliability of particle alignment due to overlapping particles being treated as a single particle. (B) Variations in the tilt angle of a given particle layer might affect the accuracy of defocus estimation if the field of view is not considered to be tilted, yet will increase the observed orientations of the particle in the dataset if the particle exhibits preferred orientations. Dashed black lines indicate the height of defocus estimation on the projected cross-section if sample tilt is not taken into account during defocus estimation. Particles are colored relative to their distance from the whole image defocus estimation to indicate the effects of ice thickness and particle layer tilt. Gray particles would be minimally impacted by whole-image CTF correction while red particles would be harshly impacted by whole-image CTF correction. Particles that would be uniquely identifiable in the corresponding projection image are circled in green.
Figure 8.
Figure 8.. Examples of typical single particle and ice behavior as might be revealed by fiducial-less cryoET and how such characterization might influence strategies for single particle collection.
Left: For a sample that exhibits thick ice near the edges of holes and ice in the center of holes that is thin enough for a single layer of particles to reside, single particle micrographs would optimally be collected a distance, d, away from the edges of holes. Middle: A sample that exhibits a high degree of preferred orientation may require tilted single particle collection by intentionally tilting the stage by a set of angles, α, in order to recover a more isotropic set of particle projections (Tan et al., 2017). Right: For a sample that consists of multiple layers of particles across holes, the sample owner may decide to proceed with collection with the knowledge that the efficiency will be limited by the particle saturation in each layer and that the resolution will be limited by the decrease in signal due to the ice thickness, t, and the accuracy of CTF estimation and correction. The results of cryoET on a given single particle cryoEM grid might also result in the sample owner deciding that the entire grid is not worth collecting on, potentially due to the situations described here or due to observed particle degradation. Due to depiction limitations, the single orientation of the particle in the middle column is depicted as being only in one direction, when in practice the particles may rotate on the planes of the air-water interfaces.
Figure 9.
Figure 9.. De novo initial model from fducial-less SPT.
(A) Gaussian picking of single particle datasets of DnaB helicase-helicase loader was not able to identify many low contrast side-views of the particle and 2D classification of the top-views incorrectly suggested C6 symmetry, resulting in unreliable initial model generation and stymying efforts to process the datasets further. (B) Fiducial-less single particle tomography (SPT) on the same grids used for single particle collection was employed to generate a de novo initial model, which was then used both as a template for picking all views of the particle in the single particle micrographs and as an initial model for single particle alignment, resulting in a 4.1 Å isotropic structure of DnaB helicase-helicase loader (manuscript in preparation). This exemplifies the novelty of applying this potentially crucial fiducial-less SPT workflow on cryoEM grids. Scale bars are 100 nm for the micrographs and tomogram, 10 nm for the 2D classes, and 5 nm for the 3D reconstructions.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Adrian M, Dubochet J, Lepault J, McDowall AW. Cryo-electron microscopy of viruses. Nature. 1984;308:32–36. doi: 10.1038/308032a0. - DOI - PubMed
    1. Agulleiro JI, Fernandez JJ. Fast tomographic reconstruction on multicore computers. Bioinformatics. 2011;27:582–583. doi: 10.1093/bioinformatics/btq692. - DOI - PubMed
    1. Agulleiro JI, Fernandez JJ. Tomo3D 2.0--exploitation of advanced vector extensions (AVX) for 3D reconstruction. Journal of Structural Biology. 2015;189:147–152. doi: 10.1016/j.jsb.2014.11.009. - DOI - PubMed
    1. Arnold SA, Albiez S, Bieri A, Syntychaki A, Adaixo R, McLeod RA, Goldie KN, Stahlberg H, Braun T. Blotting-free and lossless cryo-electron microscopy grid preparation from nanoliter-sized protein samples and single-cell extracts. Journal of Structural Biology. 2017;197:220–226. doi: 10.1016/j.jsb.2016.11.002. - DOI - PubMed
    1. Billsten P, Wahlgren M, Arnebrant T, McGuire J, Elwing H. Structural changes of T4 lysozyme upon adsorption to silica nanoparticles measured by circular dichroism. Journal of Colloid and Interface Science. 1995;175:77–82. doi: 10.1006/jcis.1995.1431. - DOI

Publication types

MeSH terms