Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Mar 1.
Published in final edited form as: Methods Mol Biol. 2021;2302:137–151. doi: 10.1007/978-1-0716-1394-8_8

Studying Membrane Protein Structures by MicroED

Michael W Martynowycz 1,2,3, Tamir Gonen 1,2,3
PMCID: PMC9974864  NIHMSID: NIHMS1782095  PMID: 33877626

Abstract

Microcrystal electron diffraction (MicroED) enables atomic resolution structures to be determined from vanishingly small crystals. Soluble proteins typically grow crystals that are tens to hundreds of microns in size for X-ray crystallography. But membrane protein crystals often grow crystals that are too small for X-ray diffraction and yet too large for MicroED. These crystals are often formed in thick, viscous media that challenge traditional cryoEM grid preparation. Here, we describe two approaches for preparing membrane protein crystals for MicroED data collection: application of a crystal slurry directly to EM grids, and focused ion beam milling in a Scanning Electron Microscope (FIB-SEM). We summarize the case of preparing an ion channel, NaK, and the workflow of focused ion-beam milling. By milling away the excess media and crystalline material, crystals of any size may be prepared for MicroED. Finally, an energy filter may be used to help minimize inelastic scattering leading to lower noise on recorded images.

Keywords: CryoEM, Microcrystal electron diffraction, MicroED, Membrane proteins, FIB milling, Energy filter

1. Introduction

Several membrane proteins have been studied by electron diffraction using 3D crystals. The very first example of a novel structure of a membrane protein determined solely by electron diffraction was that of the water channel aquaporin-0 which was determined from crystals that contained only two layers [1]. This study laid the foundations for using molecular replacement with electron diffraction data [2] and from which the field of MicroED grew a decade later. Shortly after AQP0, the gap junction protein connexin 26 was determined using crystals only three layers thick [3]. In 2015 the structure of calcium ATPase was solved from much thicker 3D crystals [4] and the structure of the ion channel NaK was solved in 2018 using crystals with only ~1000 diffracting units [5] by continuous rotation MicroED [6]. MicroED coupled with focused-ion beam milling allowed structure determination of difficult membrane proteins crystallized with lipids: the voltage dependent anion channel (VDAC) in bicelles and the G-protein coupled receptor A2a in lipidic-cubic phase [7, 8]. These examples illustrate the utility of electron diffraction in determining structures of membrane proteins from vanishingly small crystals to atomic resolution.

Electrons interact with matter more strongly than X-rays [9]. This strong interaction means that adequate signal can be generated from smaller specimens. However, this also suggests that specimens can easily become too large for electrons to penetrate. Membrane protein crystals often grow too small for synchrotron X-ray diffraction studies, and therefor make ideal candidates for microcrystal electron diffraction (MicroED) [6, 10-18]. To investigate membrane protein crystals using MicroED, they first have to be transferred from their crystallization drop to an electron microscopy (EM) grid, have the excess solvent removed by blotting, and are vitrified to preserve their hydrated, crystalline state [19]. Under ideal conditions, a drop of crystal slurry can be prepared using procedures typical of cryoEM applications [16, 19, 20]. However, the liquid in the slurry can often be viscous and difficult to blot [7, 8]. Furthermore, the crystals may not survive the forces incurred during the blotting and plunging process [10, 22, 23]. Focused ion beam (FIB) milling under cryogenic conditions involves shaving large specimens down to sizes amenable for cryoTEM investigation [24-27]. Samples are thinned by ablating away material from the specimen using a focused beam of gallium or other ions [28]. CryoFIB milling has garnered attention as a method to prepare thin lamellae for electron cryo-tomography and MicroED investigation [25, 29-33]. MicroED is ideally suited for crystals smaller than 500 nm thick [34-37], whereas crystals for synchrotron X-ray diffraction are typically >20 μm thick [34, 38]. Many crystals grow in this range, including many membrane proteins. Even membrane protein crystals smaller than 500 nm thick may be difficult to investigate by MicroED using traditional blotting methods. Blotting away the viscous detergents present in these slurries is challenging. Here, we demonstrate how protein crystals that are too large for MicroED can be milled down to thin lamellae, and how protein crystals that are of appropriate size may be transferred onto EM grids for MicroED data collection. We summarize the case of the MicroED study of the ion channel, NaK [5], and discuss a general workflow for focused ion-beam milling of protein crystals [29-33].

These results demonstrate the applicability of both standard blotting and cryo-FIB milling to prepare membrane protein crystals that are challenging to handle for synchrotron X-ray diffraction. These approaches pave the way for the investigation of considerably more difficult systems to be amenable to MicroED data collection. Several groups are working in this area, and have all made contributions to this growing field [4, 31-33, 39-43].

2. Materials

A NaK construct with the first 19 residues removed was expressed and purified as described [5, 44, 45]. Purified NaK protein was concentrated to 10 mg/mL for crystallization. Crystals were grown as hanging drops in 96 well plates by mixing 0.2 μL protein solution with 0.2 μL 50–80% MPD, 100 mM HEPES pH 7.0. Small crystalline slurries appeared in the drops after 3–5 days at room temperature. Proteinase K (E. Album) was purchased from Sigma and used without further purification. Protein was dissolved in 100 mM Tris–HCl pH 7.5 to a concentration of 20 mg/mL. Crystals of proteinase K were grown in sitting drops by combining 1 μL protein solution with 1 μL reservoir solution containing 1.25 M ammonium sulfate, 100 mM Tris–HCl pH 7.5. Experiments were conducted using Quantifoil Cu200 R2/1 or Cu300 R2/2 EM grids, clipped using c-rings and autogrid clips purchased from Thermo-Fisher. Blotting was conducted using a Thermo-Fisher Vitrobot Mark IV. Milling experiments were conducted using a Thermo-Fisher Aquilos dual beam FIB-SEM. MicroED data and TEM images were collected using either a Thermo-Fisher Talos Arctica or TF-20 TEM at cryogenic temperatures, as described [6, 16, 46, 47].

List of critical supplies and equipment:

  1. Membrane protein crystals.

  2. TEM grids (any).

  3. Plunge freezer for vitrification of samples (any).

  4. Cryo-TEM with stage capable of continuous rotation along a single axis with a suitably fast camera for accurately reading out collected data.

  5. (Optional) Cryo-FIB/SEM for milling protein crystals along with a suitable cryo-transfer stage to shuttle the grids safely between the FIB/SEM and the TEM

  6. Personal computer for processing data (any).

3. Methods

3.1. Preparing the Grids

  1. Wrap a glass cover slide in Parafilm.

  2. Place EM grids with their carbon sides facing up on the paraffin covered glass cover slide (see Note 1).

  3. Glow-discharge for 15 s using a current of 20 mA and negative polarity.

  4. Remove glass cover slide and place near Vitrobot.

3.2. Blotting

3.2.1. Blotting Conditions for the Standard MicroED Workflow

  1. Set Vitrobot to 4–8 °C and 80–100% humidity at least 20 min before you plan to begin blotting.

  2. Load the Vitrobot with fresh filter papers on both sides.

  3. Pick up a grid by its edge with mountable tweezers and lock them in place.

  4. Load the tweezers into the Vitrobot using either the touch screen or the foot pedal.

  5. Plan out a wide range of blot times and forces to screen for ideal conditions, for example 4 s at force 4–16 s at 16 force over 4 grids.

  6. Set the number of blots to one and do not skip application.

  7. Once grid is in the application position, open the side of the Vitrobot and pipette ~1–3 μL of crystal slurry onto the carbon side of the glow-discharged EM grid (see Note 2).

  8. Allow liquid to incubate on the carbon side of the grid for approximately 30 s (Fig. 1).

  9. Blot the grid.

  10. Plunge the grid into liquid ethane.

  11. Transfer the grid to a grid box in liquid nitrogen.

  12. Store grids in a large LN2 dewar prior to use.

Fig. 1.

Fig. 1

Open in a new tab

Vitrobot with loaded tweezers incubating with sample prior to plunge freezing

3.2.2. Blotting Conditions for the CryoFIB Workflow

  1. Set Vitrobot to 4–8 °C and 80–100% humidity at least 20 min before you plan to begin blotting.

  2. Cut long strips of filter paper ~10–20 mm wide from Whatman #1 discs.

  3. Pick up a grid by its edge with mountable tweezers and lock them in place.

  4. Place tweezers inside Vitrobot.

  5. Select number of blots to zero, and do not skip application.

  6. Once grid is in the application position, open the side of the Vitrobot and pipette ~3 μL of crystal slurry onto the carbon side of the glow-discharged EM grid.

  7. Allow liquid to incubate on the carbon side of the grid for approximately 30 s.

  8. Secure a long strip of filter paper into a bow shape at the end of a large pair of forceps (Fig. 2).

  9. Gently touch the copper side of the EM grid with the filter paper.

  10. Wait until the filter paper wets, trying not to drag the paper against the grid.

  11. When the wet spot stops growing, remove the forceps and filter paper, closing the opening to the chamber.

  12. Plunge freeze the grid into the liquid ethane.

  13. Transfer the grid to a grid box in the liquid nitrogen.

  14. Repeat for as many grids as needed.

  15. Store grids in LN2 dewar.

Fig. 2.

Fig. 2

Open in a new tab

Schematic diagram of blotting grids for FIB milling inside the Vitrobot. Blotting is done by gently touching the back (copper side) of the EM grid using strips of filter paper loaded from the side of the Vitrobot. The Vitrobot is kept at constant settings of 4 °C and ~100% humidity

3.3. (Optional) Transfer to FIB/SEM

  1. Cool down FIB/SEM transfer station to LN2 temperature.

  2. Remove grids from LN2 dewar, and place in transfer station.

  3. Take out as many clips and c-rings as necessary for your experiments.

  4. With a sharpie or permanent marker, place a small dot on the outside rim of the flat clip ring side (Fig. 3a) (see Note 3).

  5. Clip the grids making sure the carbon side of the grid is on the flat side of the clip ring, with the copper side facing the C-ring (Fig. 3b).

  6. Load the clipped EM grids into the cryo-FIB transfer shuttle with the carbon side of the grid facing up, and the marked dot aligned to the top or bottom (Fig. 3c).

  7. Pump down the transfer rod, and load the cryo-FIB shuttle into the FIB/SEM.

Fig. 3.

Fig. 3

Open in a new tab

(a) The correct clipping of a TEM grid for milling crystals, showing the carbon face plush with the flat face of the autogrid clip ring, and (b) the copper side of the grid facing the C-clip side of the autogrid clip. (c) Correct loading of the clipped TEM grid into the cryoFIB shuttle for a Thermo-Fisher Aquilos FIB-SEM dual beam instrument. The crystals are on the carbon face of the TEM grid, facing toward the observer. The blue dot is aligned with the milling direction of the gallium beam. The blue dot is indicated by a blue arrow in either panel

3.4. (Optional) FIB Milling Crystals

  1. In the mapping position, take a high-resolution image of each grid to check if they are of acceptable quality for milling (Fig. 4). This image should use a low magnification of 60–100×, an accelerating voltage of 5 kV or less, and a dwell time of 1 μs or less. Resolutions of 1k–4k are acceptable.

  2. Visually asses the quality of each grid from this initial image. Good grids will have a visible distinction between the carbon film and the grid bars with visible holes through the vitreous ice layer. Crystals should be on the carbon area, not within 5–10 μm of the grid bars. If grids are not badly damaged and visible crystals are in the grid square areas, move the FIB-shuttle to the sputtering position.

  3. Sputter coat with 10–100 nm of fine platinum. Optionally, coat with an additional layer of carbon rich platinum using the gas injection system (GIS) for extra protection from the damaging gallium beam.

  4. Recover the instrument from sputtering and move the grid into the milling position. Alternatively, the MAPS software (Thermo-Fisher) can be used to take a high magnification montage of the grid at mapping position. Crystals can be identified in this montage similar to viewing an all-grid atlas in standard TEM software. Here, sites can be added and aligned ahead of time, to increase the throughput of milling crystal lamellae.

  5. Change the tilt angle from 18° to 0°.

  6. Move about the grid looking for sharp edges of crystals, or dunes of solvent-covered crystals. A good crystal is near the center of the grid square as close to the center of the grid as possible.

  7. Center the crystal in the SEM image. Tilt the sample to 18° and adjust the stage Z height to move the crystal back into the center of the SEM image. Link the SEM/FIB Z height.

  8. Take images at increasing magnifications using the FIB to locate and center the crystal in the FIB image. Increase the magnification until the crystal is just entirely visible in the FIB image. A well-calibrated stage will have the crystal centered in both the SEM and FIB images.

  9. Draw a box in the FIB image of the crystal that covers from the crystal’s center of mass to the top-most point of the crystal. Horizontally, the box should extend to cover all of the potential areas of the crystal. Change the mode from clearing to cleaning cross-section and change the direction of milling from bottom to top, to top to bottom. The goal is to slowly remove the crystalline material and have it leave this area. Exposing an area constantly rather than rastering away material can cause additional damage to the specimen.

  10. Draw a second box that vertically extends from the carbon film below the crystal to ~1–5 μm below the first box. This box should leave at least 1–5 μm on either side of the crystal or solvent. This extra space is necessary to hold the lamella and support the specimen after the material has been milled away. Change the milling mode from clearing to cleaning cross section, and assure this area is cleaning from bottom to top. This area will be removed and jettisoned below the crystal after the carbon film has been broken open.

  11. For both boxes, roughly approximate the z thickness of the specimen using the SEM image. Ideally, the amount of milling time required for a specific thickness of each crystal would be calibrated for your instrument. This is difficult to ascertain for every possible sample and solvent type. A rough approximation would be to set the milling profile to Silicone, and divide the actual thickness of the specimen by 2.5. This is a rough approximation made by scaling the density of solid silicone to that of solid water ice (2.32 g/cm3 to 0.9–1.0 g/cm3). This will give relatively accurate milling times for protein crystals. The time estimate the system gives should inform the ion beam current you use for the first few steps of milling. The first one or two steps are usually done using currents of ~500–100 pA, depending on the size of the crystals, and each set of milling takes approximately 10 min.

  12. Mill these two boxes using the ion beam.

  13. Switch the ion beam current back to 1–10 pA and take both an SEM and a FIB image (Fig. 4). Inspect the lamellae in both images to assess the success of the rough-milling (Fig. 4).

  14. In the Aquilos software, change the distance between the two boxes from 1–1.5 μm, to 500 nm. Adjust the size of both drawn boxes such that, vertically, they just encompass the crystal lamella. Horizontally, bring in the boxes by a small amount, perhaps 100 nm. Change the ion beam current to 10–100 pA.

  15. Mill these two boxes using the ion beam.

  16. Repeat step 13, and then step 14 changing the distance between boxes to 300 nm and the current to 10 pA.

  17. These general steps can be broken down into smaller changes and additional steps down in ion beam current. However, we have found that the platinum coated crystals are more sensitive to the final stages of milling (polishing) than the rough first few stages.

  18. Repeat steps 6–18 for all the desired crystals on each grid. We recommend doing 4–6 per grid when operating manually. The contamination rate in the Aquilos is typically 10–30 nm/h of redeposited ice contamination. A diligent operator can mill six lamellae in 2 h. Alleviating the contamination can be accomplished by rough-milling all the lamellae on a grid, and then going back for the final low-current polishing step for each crystal after (see Note 4).

  19. Remove the cryoFIB shuttle from the FIB/SEM.

  20. Load the shuttle into a LN2 cooled sample loading station. Move the grids from the shuttle to a grid box.

  21. Store the clipped grids with milled lamellae in a dewar until it is time to transfer them into a cryo-cooled TEM.

Fig. 4.

Fig. 4

Open in a new tab

The cryo-FIB/SEM workflow on microcrystals. (a) Images taken during the milling process by the ion beam, and (b) images taken by the SEM beam at similar stages. From left to right: overview of grid at low magnification, identification of protein crystal embedded in solvent, initial rough-milling of the crystal lamella, and final polished lamella after final milling step

3.5. Transfer Grids into the TEM

  1. Remove grids from dewar and load into autoloader cassette. For traditional grids, the rotational orientation of the grid does not matter. For milled grids, the blue dot should be rotated 90° in either direction from where it will be held by the grid transfer forceps. When the grid is loaded into the cassette, you should not be able to see the blue dot (Fig. 5).

  2. Insert cassette into autoloader. Perform an inventory of the grids. Load the grid into the TEM column.

  3. Identify crystals using low-dose search mode calibrated to either over-focused diffraction mode or low magnification imaging (~155×). Crystals typically appear as black, sharp-edged shapes in over-focused diffraction mode. In low-dose imaging, crystals will appear semitransparent with similar shape (Fig. 6). In milled grids, lamellae sites appear as long strips of empty space with material suspended between the edges (Fig. 4).

  4. Using either over-focused diffraction or ~900× LM imaging mode, adjust the stage to the eucentric height using the wobbler. Image the crystal identified in step 3 to inspect the specimen. In over-focused diffraction mode, the difference in contrast between the crystalline and amorphous area of the crystals is apparent for lamellae (Fig. 4).

  5. Insert and center the SA aperture to assure data is collected only from the desired area of the crystal. Insert the beam stop.

  6. Switch to parallel beam diffraction mode, assuring the beam is well aligned and the direct beam is behind the beam stop.

  7. Take a single diffraction image to assess the resolution and quality of the crystal lamellae (Fig. 5).

Fig. 5.

Fig. 5

Open in a new tab

Milled crystals in the TEM. (a) Loading of the clipped grid into the autoloader cassette at 90° relative to the loading of the cryoFIB shuttle. (b) Low-magnification grid atlas. Milled lamellae are visible as strips missing grid carbon and marked by blue arrows. (c) Over-focused diffraction image of the crystal milled in Fig. 2. (d) Diffraction pattern extending to ~2 Å resolution from this lamella

Fig. 6.

Fig. 6

Open in a new tab

Structure determination of the ion channel NaK. (a) Over-focused diffraction image of square NaK crystals with visible, sharp edges on a Quantifoil Cu300 R2/2 grid. Crystal slurries were applied directly to the grid and blotted in the Vitrobot. (b) Typical MicroED frame from a continuous rotation movie from a NaK microcrystal. (c) Final structure solution of NaK (PDB–6CPV) looking down the fourfold axis, with sodium atoms colored in orange and cartoon ribbons of the protein in gray

3.6. Collection of MicroED Data from Crystals

  1. Data are collected as described. Briefly, a wedge between 30° and 120° is selected for collection and aligned to the eucentric height in either imaging or over-focused diffraction modes.

  2. The camera and stage are set such that the stage begins rotating at a constant rate and the camera reads out frames at regular intervals of time. Typically, this is a rotation rate of 0.1–1.0°/s with frames being read out every 1–5 s, aiming for 0.5° wedges per frame for most proteins (Fig. 6).

  3. (Optional) MicroED data may be improved by using a zero-loss energy filter to minimize contributions from inelastic scattering (Fig. 7).

  4. Data are converted using the MicroED tools available at cryoEM.ucla.edu from TVIPS, MRC, or SER to SMV format [20, 29].

  5. Data processing from this stage follows the same procedures as X-ray crystallography data reduction pipelines. We recommend using XDS [48], iMOSFLM [49, 50], or DIALS [51, 52] to reduce the data, and then SHELXT [53] SHELXD [54], phaser [55], or MOLREP [56] to phase the data, depending on resolution. Structures can be solved using ab initio direct methods or molecular replacement depending on resolution [6, 57]. Recent MicroED studies have also demonstrated ab initio phasing by radiation damage [58].

  6. Structure refinement and model building similarly follow X-ray crystallographic methods as described [14, 20].

Fig. 7.

Fig. 7

Open in a new tab

Comparison of MicroED data collection with and without an energy filter. (a) Proteinase K MicroED data collected on a Thermo-Fisher F-20 at 200 keV without an energy filter, and (b) MicroED data from proteinase K collected on a JEOL 3200-FSC at 300 keV with an in-column omega energy filter. (c) Plot of 100 × 100 pixels around 1.6 Å resolution from (a), and a 100 × 100 pixel selection at 1.6 Å resolution from (d) showing the difference in measured intensities. Energy-filtered MicroED data were collected using an in-column Omega energy filter operating with a slit-width of 20 eV

4. Notes

  1. The copper side of the grid appears visibly shiny with an orange hue, whereas the carbon side appears darker. Grids shipped in Quantifoil grid boxes typically have their carbon sides all facing the center of the box.

  2. Large or visibly single crystals can also be used, but must be fragmented into smaller shards suitable for MicroED investigations [16, 20]. This is typically done by either pipetting the crystals repeatedly or sonicating the solution as described.

  3. The blue dot is meant to identify the direction in which the grids were milled. There are also commercial autogrids available from Thermo-Fisher that include a dot, or thinned clipring areas that accomplish a similar goal. In our case, we rotate the grid to assure the TEM tilt axis is perpendicular to the milling direction. We do this because there are typically large areas of crystal remaining that occlude the view of the lamellae while tilting to high angle. Rotation to a position perpendicular to the milling direction allows us to achieve the largest possible tilt range for MicroED data collection.

  4. We typically use software such as MAPS (Thermo-Fisher) to save the eucentric position of each lamella prior to milling. This allows milling in batch, where the rough-milling of all crystals is done in series, followed by fine-milling of each, and finally the polishing steps. This results in overall reduced ice contamination to the lamellae than milling each lamella completely before moving on to the next position.

Acknowledgments

The Gonen lab is supported by funds from the Howard Hughes Medical Institute and the National Institutes of Health P41GM136508. We would like to thank the members of the Jensen and Gonen laboratories for helpful discussions.

References

  • 1.Gonen T, Cheng Y, Sliz P et al. (2005) Lipid-protein interactions in double-layered two-dimensional AQP0 crystals. Nature 438:633–638 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Wisedchaisri G, Gonen T (2013) Phasing electron diffraction data by molecular replacement: strategy for structure determination and refinement. Methods Mol Biol 955:243–272 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Oshima A, Tani K, Hiroaki Y et al. (2007) Three-dimensional structure of a human connexin26 gap junction channel reveals a plug in the vestibule. Proc Natl Acad Sci 104:10034–10039 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Yonekura K, Kato K, Ogasawara M et al. (2015) Electron crystallography of ultrathin 3D protein crystals: atomic model with charges. Proc Natl Acad Sci 112:3368–3373 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Liu S, Gonen T (2018) MicroED structure of the NaK ion channel reveals a Na+ partition process into the selectivity filter. Commun Biol 1:38. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Nannenga BL, Shi D, Leslie AGW et al. (2014) High-resolution structure determination by continuous-rotation data collection in MicroED. Nat Methods 11:927–930 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.MicroED structure of lipid-embedded mammalian mitochondrial voltage-dependent anion channel. [DOI] [PMC free article] [PubMed]
  • 8.MicroED structure of the human adenosine receptor determined from a single nanocrystal in LCP. [DOI] [PMC free article] [PubMed]
  • 9.Arnal F, Balladore JL, Soum G et al. (1976) Calculation of the cross sections of electron interaction with matter. Ultramicroscopy 2:305–310 [DOI] [PubMed] [Google Scholar]
  • 10.Shi D, Nannenga BL, Iadanza MG et al. (2013) Three-dimensional electron crystallography of protein microcrystals. elife 2013:e01345. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Nannenga BL, Shi D, Hattne J et al. (2014) Structure of catalase determined by MicroED. elife 3:e03600. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Hattne J, Shi D, Glynn C et al. (2018) Analysis of global and site-specific radiation damage in cryo-EM. Structure 26:759–766.e4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Gallagher-Jones M, Glynn C, Boyer DR et al. (2018) Sub-ångström cryo-EM structure of a prion protofibril reveals a polar clasp. Nat Struct Mol Biol 25(2):131–134 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Jones CG, Martynowycz MW, Hattne J et al. (2018) The cryoEM method microED as a powerful tool for small molecule structure determination. ACS Cent Sci 4:1587–1592 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Vergara S, Lukes DA, Martynowycz MW et al. (2017) MicroED structure of Au 146 (p-MBA) 57 at subatomic resolution reveals a twinned FCC cluster. J Phys Chem Lett 8:5523–5530 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.de l CMJ, Hattne J, Shi D et al. (2017) Atomic-resolution structures from fragmented protein crystals with the cryoEM method MicroED. Nat Methods 14:399–402 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Clabbers MTB, van GE, Wan W et al. (2017) Protein structure determination by electron diffraction using a single three-dimensional nanocrystal. Acta Crystallogr D Struct Biol 73:738–748 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Van GE, Clabbers MTB, Das PP et al. (2016) Ab initio structure determination of nanocrystals of organic pharmaceutical compounds by electron diffraction at room temperature using a Timepix quantum area direct electron detector. Acta Crystallogr A Found Adv 72:236–242 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Shi D, Nannenga BL, de l CMJ et al. (2016) The collection of MicroED data for macromolecular crystallography. Nat Protoc 11:895–904 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Hattne J, Reyes FE, Nannenga BL et al. (2015) MicroED data collection and processing. Acta Crystallogr A Found Adv 71:353–360 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Caffrey M (2003) Membrane protein crystallization. J Struct Biol 142:108–132 [DOI] [PubMed] [Google Scholar]
  • 22.Petsko GA (1975) Protein crystallography at sub-zero temperatures: cryo-protective mother liquors for protein crystals. J Mol Biol 96:381–392 [DOI] [PubMed] [Google Scholar]
  • 23.Kriminski S, Kazmierczak M, Thorne RE (2003) Heat transfer from protein crystals: implications for flash-cooling and X-ray beam heating. Acta Crystallogr D Biol Crystallogr 59:697–708 [DOI] [PubMed] [Google Scholar]
  • 24.Marko M, Hsieh C, Schalek R et al. (2007) Focused-ion-beam thinning of frozen-hydrated biological specimens for cryoelectron microscopy. Nat Methods 4:215–217 [DOI] [PubMed] [Google Scholar]
  • 25.Rigort A, Bauerlein FJB, Villa E et al. (2012) Focused ion beam micromachining of eukaryotic cells for cryoelectron tomography. Proc Natl Acad Sci 109:4449–4454 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Schertel A, Snaidero N, Han H-M et al. (2013) Cryo FIB-SEM: volume imaging of cellular ultrastructure in native frozen specimens. J Struct Biol 184:355–360 [DOI] [PubMed] [Google Scholar]
  • 27.Giannuzzi LA, Stevie FA (1999) A review of focused ion beam milling techniques for TEM specimen preparation. Micron 30:197–204 [Google Scholar]
  • 28.Sergey G, Denis K, Ava H et al. (2018) Oxygen plasma focused ion beam scanning electron microscopy for biological samples. bioRxiv 2018:457820. 10.1101/457820 [DOI] [Google Scholar]
  • 29.Martynowycz MW, Zhao W, Hattne J et al. (2019) Collection of continuous rotation microED data from ion beam-milled crystals of any size. Structure 27:545–548.e2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Martynowycz MW, Zhao W, Hattne J et al. (2019) Qualitative analyses of polishing and precoating FIB milled crystals for microED. Structure 27:1594–1600.e2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Duyvesteyn HME, Kotecha A, Ginn HM et al. (2018) Machining protein microcrystals for structure determination by electron diffraction. Proc Natl Acad Sci 115:9569–9573 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Li X, Zhang S, Zhang J et al. (2018) In situ protein micro-crystal fabrication by cryo-FIB for electron diffraction. Biophys Rep 4:339–347 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Zhou H, Luo Z, Li X (2019) Using focus ion beam to prepare crystal lamella for electron diffraction. J Struct Biol 205(3):59–64 [DOI] [PubMed] [Google Scholar]
  • 34.Martynowycz M, Glynn C, Miao J, et al. (2017) MicroED structures from micrometer thick protein crystals. 10.1101/152504 [DOI] [Google Scholar]
  • 35.Latychevskaia T, Abrahams JP (2019) Inelastic scattering and solvent scattering reduce dynamical diffraction in biological crystals. Acta Crystallogr B Struct Sci Cryst Eng Mater 75:523–531 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Clabbers MTB, Abrahams JP (2018) Electron diffraction and three-dimensional crystallography for structural biology. Crystallogr Rev 24:176–204 [Google Scholar]
  • 37.Clabbers MTB, Gruene T, van Genderen E et al. (2019) Reducing dynamical electron scattering reveals hydrogen atoms. Acta Crystallogr A Found Adv 75:82–93 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Flot D, Mairs T, Giraud T et al. (2010) The ID23-2 structural biology microfocus beam-line at the ESRF. J Synchrotron Radiat 17:107–118 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Palatinus L, Brázda P, Boullay P et al. (2017) Hydrogen positions in single nanocrystals revealed by electron diffraction. Science 355:166–169 [DOI] [PubMed] [Google Scholar]
  • 40.Brázda P, Palatinus L, Babor M (2019) Electron diffraction determines molecular absolute configuration in a pharmaceutical nanocrystal. Science 364:667–669 [DOI] [PubMed] [Google Scholar]
  • 41.Gruene T, Wennmacher JTC, Zaubitzer C et al. (2018) Rapid structure determination of microcrystalline molecular compounds using electron diffraction. Angew Chem Int Ed 57:16313–16317 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Spurlino JC, Lu GY, Quiocho FA (1991) The 2.3-A resolution structure of the maltose- or maltodextrin-binding protein, a primary receptor of bacterial active transport and chemotaxis. J Biol Chem 266:5202–5219 [DOI] [PubMed] [Google Scholar]
  • 43.Yonekura K, Maki-Yonekura S (2016) Refinement of cryo-EM structures using scattering factors of charged atoms. J Appl Crystallogr 49:1517–1523 [Google Scholar]
  • 44.Alam A, Jiang Y (2009) Structural analysis of ion selectivity in the NaK channel. Nat Struct Mol Biol 16:35–41 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Alam A, Jiang Y (2009) High-resolution structure of the open NaK channel. Nat Struct Mol Biol 16:30–34 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Rodriguez JA, Gonen T (2016) High-resolution macromolecular structure determination by microED, a cryo-EM method. Elsevier, Inc. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Nannenga BL, Gonen T (2014) Protein structure determination by microED. Curr Opin Struct Biol 27:24–31 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Kabsch W (2010) Xds. Acta Crystallogr D Biol Crystallogr 66:125–132 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Leslie AGW, Powell HR (2007) Processing diffraction data with mosflm. 10.1007/978-1-4020-6316-9_4 [DOI] [Google Scholar]
  • 50.Battye TGG, Kontogiannis L, Johnson O et al. (2011) iMOSFLM: A new graphical interface for diffraction-image processing with MOSFLM. Acta Crystallogr D Biol Crystallogr 67:271–281 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Winter G, Waterman DG, Parkhurst JM et al. (2018) DIALS: implementation and evaluation of a new integration package. Acta Crystallogr D Struct Biol 74:85–97 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Clabbers MTB, Gruene T, Parkhurst JM et al. (2018) Electron diffraction data processing with DIALS. Acta Crystallogr D Struct Biol 74:506–518 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Sheldrick GM (2015) SHELXT – integrated space-group and crystal-structure determination. Acta Crystallogr A Found Crystallogr 71:3–8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Schneider TR, Sheldrick GM (2002) Substructure solution with SHELXD. Acta Crystallogr D Biol Crystallogr 58:1772–1779 [DOI] [PubMed] [Google Scholar]
  • 55.McCoy AJ, Grosse-Kunstleve RW, Adams PD et al. (2007) Phaser crystallographic software. J Appl Crystallogr 40:658–674 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Vagin A, Teplyakov A (1997) MOLREP: an automated program for molecular replacement. J Appl Crystallogr 30:1022–1025 [Google Scholar]
  • 57.Sawaya MR, Rodriguez J, Cascio D et al. (2016) Ab initio structure determination from prion nanocrystals at atomic resolution by microED. Proc Natl Acad Sci 113:11232–11236 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Martynowycz MW, Hattne J, Gonen T (2020) Experimental phasing of microED data using radiation damage. Structure 28:458–464.e2 [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES