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
. 2021 Mar 1;8(Pt 2):186-194.
doi: 10.1107/S2052252520016243.

Devitrification reduces beam-induced movement in cryo-EM

Jan-Philip Wieferig  1 , Deryck J Mills  1 , Werner Kühlbrandt  1
Affiliations

Devitrification reduces beam-induced movement in cryo-EM

Jan-Philip Wieferig et al. IUCrJ. .

Abstract

As cryo-EM approaches the physical resolution limits imposed by electron optics and radiation damage, it becomes increasingly urgent to address the issues that impede high-resolution structure determination of biological specimens. One of the persistent problems has been beam-induced movement, which occurs when the specimen is irradiated with high-energy electrons. Beam-induced movement results in image blurring and loss of high-resolution information. It is particularly severe for biological samples in unsupported thin films of vitreous water. By controlled devitrification of conventionally plunge-frozen samples, the suspended film of vitrified water was converted into cubic ice, a polycrystalline, mechanically stable solid. It is shown that compared with vitrified samples, devitrification reduces beam-induced movement in the first 5 e Å-2 of an exposure by a factor of ∼4, substantially enhancing the contribution of the initial, minimally damaged frames to a structure. A 3D apoferritin map reconstructed from the first frames of 20 000 particle images of devitrified samples resolved undamaged side chains. Devitrification of frozen-hydrated specimens helps to overcome beam-induced specimen motion in single-particle cryo-EM, as a further step towards realizing the full potential of cryo-EM for high-resolution structure determination.

Keywords: beam-induced movement; cryo-EM; devitrification.

PubMed Disclaimer

Figures

Figure 1
Figure 1
Motion-corrected micrographs of apoferritin (120–130 Å outer diameter) in cubic ice at 1.077 Å pixel size and a defocus of 1.93 µm (a) and 1.96 µm (b). Red boxes show the areas enlarged in (c) and (d). The ice lattice is visible by eye in both micrographs. (a) shows numerous small crystallites with different orientations, whereas in (b) the crystal lattice is uniform. This is also evident from the respective Fourier transforms (e, f). The lower right-hand quadrant in (e) has multiple reflections, whereas in (f) it shows only one single spot at 3.68 Å resolution. The highest resolution to which the CTF could be fitted is indicated by the thin white circles in (e) (3.19 Å) and (f) (2.96 Å). The top right quadrant shows the rotational average of the Fourier transform and the left half shows the fitted CTF.
Figure 2
Figure 2
Ice contamination upon devitrification for 5–6 s at −105°C. The image on the left shows a Quantifoil R2/2 foil hole. The image on the right shows another foil hole at higher magnification, with alcohol oxidase against a background of contaminating ice crystals. Fourier transforms of manually picked particles revealed that the aqueous film around the particles in this image had not devitrified.
Figure 3
Figure 3
Radial power spectra of apoferritin in devitrified (a, b) and vitreous (c, d) water. Average radial signal intensities were calculated from 1600–2600 particles and the same number of background regions of equal size. The blue line shows the difference between the power spectra of particles (red) and background (black) for regions of the transform where the particle signal is above background. For particles in cubic ice this was the case to 0.250 Å−1 (a) or 0.255 Å−1 (b), and for particles in vitreous water to 0.244 Å−1 (c) or 0.255 Å−1 (d). In the regions beyond, background was stronger except for sporadic higher frequencies. Particle and background intensities at low spatial frequency extend to 1.9–4.6 × 105.
Figure 4
Figure 4
Rosenthal plots of the inverse squared resolution versus the natural logarithm of the number of asymmetric units of horse spleen apoferritin (ApoF; 24-fold symmetry, octahedral) in cubic ice (a) and vitreous water (b). B factors were calculated from the slope, excluding the first data point of all data sets with vitreous water. FSC curves for ApoF vitreous 2 and ApoF devitrified 2 are shown in Supplementary Fig. S4.
Figure 5
Figure 5
Averaged per-frame motion of all particles used in the last refinement round of (a) alcohol oxidase (AOX) and apoferritin (ApoF) (b, c). The particle drift is calculated from the particle trajectories determined by Bayesian polishing in RELION-3 (Suppplementary Fig. S3). A systematic reduction of particle drift is evident for both apoferritin and alcohol oxidase in cubic ice (red) compared with vitreous water (blue) in the most important early frames.
Figure 6
Figure 6
Relative Bayesian polishing B factors determined in RELION-3 for apoferritin (ApoF) or alcohol oxidase (AOX) in cubic ice (a) or for apoferritin in vitreous water (b). The per-frame B factor of the first, minimally damaged frame is better by 28–60 Å2 for devitrified than for vitreous samples.
Figure 7
Figure 7
20 000 particles were randomly chosen from the final 3D refinement of the full data sets and re-extracted from individual frames for 3D reconstruction. In data sets acquired from apoferritin (ApoF) in cubic ice (left) the resolution of the reconstructions using only the first frame (1.3 e Å−2) were better than those of vitrified specimens (right). The overall trend in the reconstructions from apoferritin in cubic ice was a decrease in resolution with increasing fluence, as expected owing to accumulating radiation damage. With apoferritin in vitreous water, the highest resolution single-frame reconstruction was obtained with the fourth frame (4.6 e Å−2), at which point the initial burst of beam-induced movement had subsided. For vitrified samples, the resolution of the first-frame reconstruction (1.15–1.23 e Å−2) was consistently lower than for any of the first ten frames owing to the large initial beam-induced movement.
Figure 8
Figure 8
Density maps reconstructed from two data sets each of apoferritin in cubic ice (ApoF devitrified 1 and 2) and apoferritin in vitreous water (ApoF vitreous 1 and 2) with a fitted 1.5 Å resolution X-ray structure of horse apoferritin (PDB entry 2w0o). Rows 1 and 2 show reconstructions from 20 000 particles from the first or tenth movie frame. Density is drawn at the default contour level where 1% of the voxels are above the threshold.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Allegretti, M., Mills, D. J., McMullan, G., Kühlbrandt, W. & Vonck, J. (2014). eLife, 3, e01963. - PMC - PubMed
    1. Bartels-Rausch, T., Bergeron, V., Cartwright, J. H. E., Escribano, R., Finney, J. L., Grothe, H., Gutiérrez, P. J., Haapala, J., Kuhs, W. F., Pettersson, J. B. C., Price, S. D., Sainz-Díaz, C. I., Stokes, D. J., Strazzulla, G., Thomson, E. S., Trinks, H. & Uras-Aytemiz, N. (2012). Rev. Mod. Phys. 84, 885–944.
    1. Booy, F. P. & Pawley, J. B. (1993). Ultramicroscopy, 48, 273–280. - PubMed
    1. Brilot, A. F., Chen, J. Z., Cheng, A., Pan, J., Harrison, S. C., Potter, C. S., Carragher, B., Henderson, R. & Grigorieff, N. (2012). J. Struct. Biol. 177, 630–637. - PMC - PubMed
    1. Campbell, M. G., Cheng, A., Brilot, A. F., Moeller, A., Lyumkis, D., Veesler, D., Pan, J., Harrison, S. C., Potter, C. S., Carragher, B. & Grigorieff, N. (2012). Structure, 20, 1823–1828. - PMC - PubMed

Grants and funding

This work was funded by Max-Planck-Gesellschaft grant .