Retrieving High-Resolution Information from Disordered 2D Crystals by Single Particle Cryo-EM

Ricardo Righetto, Nikhil Biyani, Julia Kowal, Mohamed Chami, Henning Stahlberg

Preprint posted on 6 December 2018

Article now published in Nature Communications at

2D crystals revisited – High resolution structures of the MloK1 ion channel are solved from “imperfect” 2D crystals by treating them as a single particle cryoEM sample

Selected by David Wright


Membrane proteins, despite making up over half of small molecule drug targets (1), make up less than 2% of the Protein Data Bank (PDB). Historically, X-ray crystallography was the most popular technique for the solution of membrane protein structures, however cryogenic electron microscopy (CryoEM) is now the go-to technique for large, flexible membrane proteins. There are many reasons why cryoEM is particularly suited for membrane proteins, but one of the most important is that cryoEM copes particularly well with the multitude of conformations commonly found in a membrane protein sample. MloK1 is a homotetrameric prokaryotic ion channel, which is used in this preprint as a test case for membrane proteins more generally.

The very first membrane protein structures were generated from 2D crystals (2), formed from the proteins aligning side-by-side as if in a native bilayer. Due to the fact that there is only one protein molecule in the z-axis, the resolution in this plane is often very poor. With the resolution revolution (3) due to improved hardware and software, single particle cryoEM has emerged as a high resolution technique to rival X-ray crystallography. In single particle cryoEM of membrane proteins the sample is typically solubilised in detergent, which might lead to a non-native structure.  On the other hand, 2D crystals are grown in a lipidic environment, so might result in a more native-like structure. This preprint uses novel software for single particle reconstruction of 2D crystals deemed “too imperfect” for traditional 2D electron crystallography.


The first point worthy of note is that the overall resolution of the final model was 4 Å, which is higher than the previously available MloK1 structure (4). Further to this, the authors calculate the Brag reflections of these data to be 10 Å or worse, suggesting that this hybrid approach actually improves the resolution of the model generated from these data. It might be expected that the “single particle” selected in analysis was one tetramer; however it seems that there are at least 9 tetramers in each single particle. Even though this ion channel is homotetrameric, it might be possible for the structures to be asymmetric; however, using localized reconstruction, “no significant deviations from 4-fold symmetry” were detected.

The most interesting result, in my opinion, is that eight different structures are generated from one crystal system. If this were crystallography, we would expect a single “averaged” structure, much like the 4 Å structure described above. By generating 3D classes, however, the different conformations can be identified, resulting in structures to 4.4 – 5.6 Å resolution. The largest RMSD between states was less than 1 Å, suggesting that there were no vast global differences between states (this is illustrated in figure 4 included below of the preprint and can be visualised in supplementary movies 1-3). It was also noted that there was no correlation between structure and position on the crystal e.g. that state A was not always found on the edge of the crystal (see supplementary figure 8).

Conclusions and comments

This work resulted in a structure with a higher resolution than that previously seen with MloK1. 2D crystallography generally leads to a structure with good resolution in the x and y planes (as there are thousands of molecules here) but very poor resolution in the z plane (as only one molecule is present). This results in what is often known as the “missing cone” of data. As these crystals are not perfectly planar, the “single particles” are tilted relative to each other, which offers slightly different views i.e. not all from directly above. There is still significant missing data, but this technique, for imperfect crystals, is clearly an improvement on typical 2D crystallography. To the authors’ knowledge, this is the first example of 3D classification within a 2D dataset, and I think it will certainly not be the last. This technique could be very interesting for use in 2D arrays of membrane proteins that are sometimes natively observed, for example in mitochondria and chloroplasts.

Why I chose this preprint

This is a really exciting article where the authors rescue poorly ordered crystals and generate a really nice set of structures. It is really impressive that the imperfections in the crystal that led to non-planarity have been exploited to minimise the “missing cone” that usually affects resolution in the z-axis. Hopefully this work will be applicable to other projects that have stalled with poorly-diffracting 2D crystals. This workflow has also been incorporated into the FOCUS package for other users’ future benefit.

Outstanding questions

  1. How easy it is to make 2D crystals in reality? How much sample is required compared to standard single particle cryoEM?
  2. If purified protein is available, is there any benefit to forming 2D crystals over single particle cryoEM?
  3. How many examples of native 2D crystals are there and would this approach be able to handle crystals with multiple, potentially heterogeneous components?


  1. Santos R, Ursu O, Gaulton A, Bento AP, Donadi RS, Bologa CG, et al. A comprehensive map of molecular drug targets. Nature Reviews Drug Discovery. 2016 12/02/online;16:19.
  2. Henderson R, Unwin PN. Three-dimensional model of purple membrane obtained by electron microscopy. Nature. 1975 Sep 4;257(5521):28-32. PubMed PMID: 1161000. Epub 1975/09/04. eng.
  3. Kühlbrandt W. The Resolution Revolution. Science. 2014;343(6178):1443.
  4. Kowal J, Biyani N, Chami M, Scherer S, Rzepiela AJ, Baumgartner P, et al. High-Resolution Cryoelectron Microscopy Structure of the Cyclic Nucleotide-Modulated Potassium Channel MloK1 in a Lipid Bilayer. Structure. 2018 2018/01/02/;26(1):20-7.e3.


Tags: ion channel, lipid

Posted on: 17 December 2018


Read preprint (No Ratings Yet)

Have your say

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Sign up to customise the site to your preferences and to receive alerts

Register here

preLists in the biochemistry category:

CellBio 2022 – An ASCB/EMBO Meeting

This preLists features preprints that were discussed and presented during the CellBio 2022 meeting in Washington, DC in December 2022.


List by Nadja Hümpfer et al.

20th “Genetics Workshops in Hungary”, Szeged (25th, September)

In this annual conference, Hungarian geneticists, biochemists and biotechnologists presented their works. Link:


List by Nándor Lipták


The advances in fibroblast biology preList explores the recent discoveries and preprints of the fibroblast world. Get ready to immerse yourself with this list created for fibroblasts aficionados and lovers, and beyond. Here, my goal is to include preprints of fibroblast biology, heterogeneity, fate, extracellular matrix, behavior, topography, single-cell atlases, spatial transcriptomics, and their matrix!


List by Osvaldo Contreras

ASCB EMBO Annual Meeting 2019

A collection of preprints presented at the 2019 ASCB EMBO Meeting in Washington, DC (December 7-11)


List by Madhuja Samaddar et al.

EMBL Seeing is Believing – Imaging the Molecular Processes of Life

Preprints discussed at the 2019 edition of Seeing is Believing, at EMBL Heidelberg from the 9th-12th October 2019


List by Dey Lab

Cellular metabolism

A curated list of preprints related to cellular metabolism at Biorxiv by Pablo Ranea Robles from the Prelights community. Special interest on lipid metabolism, peroxisomes and mitochondria.


List by Pablo Ranea Robles


This list of preprints is focused on work expanding our knowledge on mitochondria in any organism, tissue or cell type, from the normal biology to the pathology.


List by Sandra Franco Iborra