Validation of cryo-EM structure of IP₃R1 channel

Abstract

About a decade ago, three electron cryomicroscopy (cryo-EM) single-particle reconstructions of IP3R1 were reported at low resolution. It was disturbing that these structures bore little similarity to one another, even at the level of quaternary structure. Recently, we published an improved structure of IP3R1 at ∼1 nm resolution. However, this structure did not bear any resemblance to any of the three previously published structures, leading to the question of why the structure should be considered more reliable than the original three. Here, we apply several methods, including class-average/map comparisons, tilt-pair validation, and use of multiple refinement software packages, to give strong evidence for the reliability of our recent structure. The map resolution and feature resolvability are assessed with the gold standard criterion. This approach is generally applicable to assessing the validity of cryo-EM maps of other molecular machines.

Figure 1. Class-average self-consistency test

Projections of the 3D map (A), reference-based class-averages (B), reference-free class-averages (unaligned, C), and selected (unaligned) individual particle images (D) are shown for comparison. Only a representative subset of the full set of projection orientations is shown, but clear qualitative agreement between the four types of images can be observed in each row.

Figure 2. Results of tilt-pair analysis

(A, B) IP₃R1 tilt-pair validation plots for two image tilt-pairs. The grey circles denote particle pairs that cluster around the experimental tilt geometry, thus validating our IP₃R1 map (EMDB-5278)(Ludtke et al., 2011). A cross indicates the center of the cluster and each point represents a single pair of particles. The radial value indicates the amount of tilt determined between the pair of particles, and azimuthal value indicates the direction of tilt. Ideally all points would fall at exactly the experimental tilt/direction. Some spread indicates the relative uncertainty in orientation determination. Note that the radial axis extends only to 40 degrees in these plots. Corresponding statistics for used tilt-pair images is given in Table S1. (C) The same plot as in (A) is shown at larger scale. Validation plots for the same tilt-pair images as in (A) were calculated against different 3D maps, (D) EMDB-1061 (Sato et al., 2004), (E) from (Serysheva et al., 2003), and (F) from (Jiang et al., 2002). Note that the three previously published maps produce a completely random distribution with no clustering. See also Table S1 and Figure S2.

Figure 3. 3D reconstructions of IP₃R1 generated with different software packages

Maps were calculated using the same set of cryo-EM images and a matching filter was applied to bring each to the same ~20 Å resolution, with a comparable contour level. (A) 3D reconstructions performed reference-free; (B) 3D structure reconstructed using an initial model generated from an EMAN2 structure that had been low pass filtered by RELION; (C) FSCs between the reconstructions from EMAN1 and the maps generated using other software packages. See also Figure S1 and Figure S2.

Figure 4. ‘Gold standard’ FSC plot for the EMAN2 reconstruction

The EMAN2 refinement of the full data set produced a ‘gold standard’ resolution of about 17 Å. Splitting the data into more homogeneous subsets resulted in an additional map (not shown) with a ‘gold standard’ resolution of 14.7 Å. This demonstrates that resolution is limited by structural variability for this data set. If the data quality were limiting the resolution, splitting the data would not improve the resolution.