Here, we describe a protocol for applying a single monolayer of graphene to electron microscopy grids and how to prepare them for use in cryoEM structure determination.
Cryogenic electron microscopy (cryoEM) has emerged as a powerful technique for probing the atomic structure of macromolecular complexes. Sample preparation for cryoEM requires preserving specimens in a thin layer of vitreous ice, typically suspended within the holes of a fenestrated support film. However, all commonly used sample preparation approaches for cryoEM studies expose the specimen to the air-water interface, introducing a strong hydrophobic effect on the specimen that often results in denaturation, aggregation, and complex dissociation. Further, preferred hydrophobic interactions between regions of the specimen and the air-water interface impact the orientations adopted by the macromolecules, resulting in 3D reconstructions with anisotropic directional resolution.
Adsorption of cryoEM specimens to a monolayer of graphene has been shown to help mitigate interactions with the air-water interface while minimizing the introduction of background noise. Graphene supports also offer the benefit of substantially lowering the required concentration of proteins required for cryoEM imaging. Despite the advantages of these supports, graphene-coated grids are not widely used by the cryoEM community due to the prohibitive expense of commercial options and the challenges associated with large-scale in-house production. This paper describes an efficient method for preparing batches of cryoEM grids that have nearly full coverage of monolayer graphene.
Single-particle cryogenic electron microscopy (cryoEM) is an increasingly applicable technology used to investigate the 3D structures of biomacromolecules. Technological advances in electron microscope optics, direct electron detection1, and computer algorithms2,3,4 over the past decade have enabled cryoEM users to determine the structures of biochemically stable macromolecular complexes to near-atomic resolution5,6,7,8. Despite these advances, there remain notable barriers to preserving samples for cryoEM imaging, which prevent the majority of biological specimens from being resolved to such high resolutions.
Sample preparation for high-resolution cryoEM analysis involves trapping macromolecules that are evenly distributed in a wide range of orientations within a thin layer of vitrified ice. The "blot and plunge" methods of freezing are the most widely used methods employed to generate thin films of biological samples on grids for cryoEM studies9,10. These methods involve applying a few microliters of sample solution to an EM grid containing a fenestrated film that has been made hydrophilic and subsequently blotting away the majority of the sample with filter paper before rapidly plunging the grid into a cryogen of liquid ethane or ethane-propane mixture9.
While this method has successfully been used to determine structures of a wide range of biological specimens, all commonly used cryoEM specimen preparation methods expose specimens to the hydrophobic air-water interface (AWI), which often introduces issues that limit high-resolution structure determination. It is established that biological specimens have a high propensity to denature when exposed to the AWI, which can lead to complex aggregation and disassembly11,12,13,14. Furthermore, hydrophobic patches on the surfaces of biological specimens cause particles to adopt preferred orientations in the ice12. In many scenarios, a single hydrophobic region of the sample forces all particles to adopt a singular orientation in the ice, thereby abolishing the ability to generate a reliable reconstruction. In addition to issues with the AWI, specimens may demonstrate an affinity for the surface of the fenestrated layer of film, limiting the number of particles suspended in the ice within the holes15.
Several methodological and technological solutions have been developed to diminish these issues arising from interactions with the AWI or the films16,17. One popular approach is to coat the fenestrated film of the EM grids with a thin (tens of nanometers) layer of amorphous carbon. This coating provides a continuous surface across the holes to which particles can adsorb, with the benefit of partially shielding the sample from interactions with the AWI15,18,19,20. However, the additional carbon layer elevates the amount of background signal in the imaged regions, introducing noise that can compromise attainable resolution, particularly for small (<150 kDa) specimens. In recent years, the use of graphene oxide (GO) flakes to produce supporting films on cryoEM grids has been shown to have advantages over traditional amorphous carbon. GO flakes are produced through the oxidation of graphite layers, resulting in pseudo-crystalline sheets of monolayer graphite that are hydrophilic due to their substantial oxygen content in the form of carboxyl, hydroxyl, and epoxy groups on the surfaces and edges. Commercial GO flakes in aqueous suspensions are inexpensive, and there are numerous published methods for applying GO flakes to EM grids18,21. However, these methods often result in grids that are only partially covered with GO flakes, as well as regions that contain multiple layers of GO flakes. Further, GO flakes contribute a noticeable background signal to cryoEM images that is close to that observed with thin amorphous carbon22,23.
Pristine monolayer graphene, which consists of a single 2D crystalline array of carbon atoms, is distinct from GO in that it does not produce phase contrast in the electron microscope. Monolayer graphene thus can be used to generate an invisible support layer for imaging biological samples. Monolayer graphene is also stronger than GO and can be applied as a single monolayer on an EM grid, and recent advancements in the fabrication of graphene-coated EM grids have made it possible to prepare high-coverage monolayer graphene grids in-house24,25,26,27,28,29,30. However, despite the benefits of using graphene-coated grids for cryoEM structure determination, they are not widely used due to the prohibitive expense of commercial options and the complexity of in-house production. Here, we describe a step-by-step guide to effectively producing EM grids covered with a monolayer of graphene for cryoEM structure determination of biological specimens (Figure 1). By following this detailed protocol, cryoEM researchers can reproducibly prepare dozens of high-quality graphene support grids in a single day. The quality of the graphene-coated grids can be readily examined using a low-end transmission electron microscope (TEM) equipped with a LaB6 filament.
The preservation of biological samples in a thin layer of vitreous ice is a critically important step for high-resolution cryoEM structure determination. However, researchers often encounter problems arising from interactions with the AWI, which introduces preferred orientation, complex disassembly, denaturation, and aggregation. Furthermore, samples cannot always be concentrated sufficiently to populate the thin ice suspended across the holes of a fenestrated film. Several research groups have developed methods to coat EM grids with a monolayer of graphene to help overcome some of these limitations24,25,26,27,28,29,30, and graphene grids have been used with great success. Here, we provide step-by-step instructions for effectively preparing batches of graphene grids in-house and examining the quality of the graphene grids by TEM. We emphasize that particular care should be exercised during some of the critical steps, which we outline below.
Graphene has a strong tendency to attract airborne contaminants. Therefore, during the graphene grid fabrication process, it is important to make sure all the tools that make contact with the graphene/Cu sheet or the grids are clean and dust-free. Glass coverslips used to transfer graphene can be cleaned by rinsing with ethanol and DI water or using an air-duster. It is also advised to work under a fume hood and keep graphene sheets and grids covered with foil or a clean glass plate at all times. Dust or contaminants on the grids may prevent graphene from thoroughly adhering to the EM grids. When handling graphene or graphene-coated grids, it is important to be electrically grounded to prevent damage to the graphene film from static discharge. Static discharge can be avoided by using a wrist grounding strap, touching a grounded metal object each time graphene or graphene grids are handled, and/or not wearing a glove on the hand holding the tweezers24.
Since a monolayer of graphene is very thin (the width of a carbon atom), it is important to support graphene with an organic layer such as MMA or poly-MMA (PMMA) during the transfer of graphene to grids. PMMA is the most widely used material for graphene transfer. However, PMMA has a strong affinity with graphene and can often result in polymer contamination on the graphene film. MMA is used in this protocol, as it leaves less residual contamination25. However, both PMMA and MMA have the disadvantage of forming wrinkles and cracks that can be observed in some areas of the graphene film (Figure 3B). It can be challenging to avoid these wrinkles as they commonly occur during graphene growth by the CVD method31. A method has been recently developed for growing ultra-flat graphene without wrinkles, whereby the copper foil is replaced by a Cu(111)/sapphire wafer as a growth substrate32.
Based on our experience, it is better to purchase graphene/Cu sheets and support the graphene with MMA in-house than purchasing polymer-covered Cu-graphene sheets from manufacturers, which become brittle after copper etching and are difficult to handle in subsequent steps. The spin coater we used for MMA coating can be cheaply built using parts from a local computer/hardware store, as previously described25.
During the step of MMA coating, it is important to cover the entirety of the graphene surface on the Cu-graphene sheet with MMA. After the Cu has been etched away, MMA-graphene will become semi-transparent, and areas lacking MMA coverage will look like empty holes. To prevent MMA coating on the copper side, it is important to place a small piece of blotting paper underneath it during coating such that it soaks up any excess MMA that spins out from the CVD film.
After etching and rinsing, the MMA/graphene sheet is ready to be transferred to EM grids by using a commercial or homemade trough system with a syringe or peristaltic pump to control the water level. Prior to the transfer step, it is important to thoroughly prerinse the grids in successive baths of chloroform, acetone, and IPA. Baking graphene-coated grids at 65 °C helps to preserve graphene integrity and promotes the adsorption of graphene to the grids. Lastly, to prevent MMA contamination on the grids, it is important to thoroughly remove MMA in an acetone bath and clean the grids in IPA. Any unwashed MMA residue will be observed on EM grids and diminish the signal-to-noise ratio of the images (Figure 3C). The acetone-IPA washing process can be repeated to further clean the graphene surfaces.
To render graphene grids hydrophilic, we exposed the grids to UV/Ozone. Different models of UV/ozone cleaners may require optimization to sufficiently oxygenate the graphene layer for cryoEM sample preparation without damaging the graphene. Regardless of the system, it is critical to use these grids for cryoEM sample application immediately after UV/Ozone treatment. Alternative methods to render graphene grids hydrophilic are described in other studies33,34.
The authors have nothing to disclose.
We thank Dr. Xiao Fan for helpful discussions while establishing these methods at Scripps Research. B.B. was supported by a postdoctoral research fellowship from the Hewitt Foundation for Medical Research. W.C. is supported by a National Science Foundation predoctoral fellowship. D.E.P is supported by the National Institutes of Health (NIH) grant NS095892 to G.C.L. This project was also supported by NIH grants GM142196, GM143805, and S10OD032467 to G.C.L.
70% EtOH | Pharmco (190 pf EtOH) | 241000190CSGL | |
Acetone | Sigma Aldrich | 650501-4L | |
Ammonium persulfate (APS) | Sigma Aldrich | 215589-500g | Hazardous; use extreme caution |
Chloroform | Sigma Aldrich | C2432-1L | |
Clamping TEM Grid Holder Block for 45 Grids | PELCO | 16830-45 | |
Computer fan | Amazon (Noctua) | B07CG2PGY6 | |
Cover slip | Bellco Glass | 1203J71 | Standard cover slips |
Crystallizing dish | Pyrex | 3140-100 | |
Electronics duster | Falcon Safety Products | 75-37-6 | |
Falcon Dust-off Air Duster | Staples | N/A | |
Filter papers | Whatman | 1001-055 | |
Fine tip tweezer | Dumont | 0508-L4-PO | |
Flask | Pyrex | 4980-500 | |
Fork | Supermarket | N/A | |
Glass pasteur pipette | VWR | 14672-608 | |
Graphene/Cu | Graphenea | N/A | CVD monolayer graphene cu |
Grid Coating Trough | Ladd Research Industries | 10840 | Fragile |
Isopropanol | Fisher Scientific | 67-63-0 | |
Kapton Tape | Amazon (MYJOR) | MY-RZY001 | Polyimide tape |
Kimwipes | Fisher Scientific | 06-666 | |
Long twzeer | Cole Parmer Essentials | UX-07387-15 | |
Metal grid holder | Ted Pella | 16820-81 | |
MMA(8.5)MMA EL 6 | KAYAKU Advanced Materials | M31006 0500L 1GL | Flammable |
Model 10 Lab Oven | Quincy Lab, Inc. | FO19013 | |
Petri dish | Pyrex | 3610-102 | |
Plasma cleaner (Solarus 950) | Gatan, Inc. | N/A | |
Scissors | Fiskars | 194813-1010 | |
Standard Analog Orbital Shaker | VWR | 89032-088 | |
UltrAuFoil R1.2/1.3 – Au300 | Quantifoil | N/A | Holey gold grids |
Ultraviolet Ozone Cleaning Systems | UVOCS | model T10X10/OES |