Graphene sandwich–based biological specimen preparation for cryo-EM analysis

Jie Xu, Xiaoyin Gao, Liming Zheng, Xia Jia, Kui Xu, Yuwei Ma, Xiaoding Wei, Nan Liu, Hailin Peng, Hongwei WangProceedings of the National Academy of Sciences 2024 · DOI

TL;DR

Researchers have developed a novel "graphene sandwich" technique for preparing biological samples for cryo-electron microscopy (cryo-EM). This method, by enclosing samples between two layers of graphene, significantly improves data quality by reducing beam-induced charging and particle motion, leading to higher resolution structural determination of macromolecules.

What Problem Does This Paper Address?

Cryo-electron microscopy (cryo-EM) has revolutionized structural biology, enabling the determination of high-resolution structures of biological molecules. However, a critical bottleneck remains in the preparation of specimens. Traditional methods often struggle with issues like the air-water interface (AWI), which can lead to particle denaturation, and beam-induced charging and motion, which degrade image quality and limit resolution. While graphene has shown promise as a support film due to its conductivity and strength, existing methods using a single layer of graphene still leave one side of the sample exposed to potential AWI perturbations. This paper addresses the need for a more robust and reliable specimen preparation method that minimizes these detrimental effects.

Methods

The core innovation is the "graphene sandwich" method. This involves:

  1. Graphene Film Preparation: Synthesizing free-standing graphene films using chemical vapor deposition (CVD). The air-facing side is coated with stearic acid molecules to protect it during etching.
  2. Sample Loading: A sample solution is applied to a graphene-coated EM grid.
  3. Sandwich Formation: A second layer of free-standing graphene film (with stearic acid coating) is placed on top of the sample-loaded graphene grid. This creates a sealed environment for the sample solution.
  4. Ice Formation: The sandwiched sample is then rapidly plunge-frozen in liquid ethane to form a thin, vitreous ice layer containing the macromolecules.
  5. Cryo-EM Imaging: The prepared grids are then used for cryo-EM data collection.

The researchers also employed standard cryo-EM techniques for data processing, including motion correction, CTF estimation, particle picking, and 3D classification, to reconstruct the structures of apoferritin, 20S proteasome, and SARS-CoV-2 spike protein. They also used Atomic Force Microscopy (AFM) for surface characterization and room-temperature TEM to analyze the graphene sandwich structure and bubble formation under electron irradiation.

Datasets

The paper utilized three biological macromolecules for demonstrating the efficacy of the graphene sandwich method:

  • Apoferritin: A well-characterized protein used for assessing data quality and resolution.
  • 20S Proteasome: Another stable protein complex used for structural determination.
  • SARS-CoV-2 Spike Protein: A more flexible and less symmetric protein, chosen to demonstrate the method's applicability to challenging targets.

The authors also generated and analyzed cryo-EM datasets for both graphene-sandwiched and conventionally prepared (single-sided graphene support) samples of these proteins to enable direct comparison of data quality.

Key Insights and Ideas

  • Effective Encapsulation: The graphene sandwich successfully seals the sample solution, preventing contact with the air-water interface and thus mitigating a major source of sample damage.
  • Reduced Beam Effects: The double layer of graphene significantly reduces beam-induced charging and particle motion compared to single-sided graphene supports. This is attributed to graphene's superior electrical conductivity and increased effective Young's modulus.
  • Improved Particle Distribution: The method leads to a more uniform distribution of particles within the ice layer, facilitating higher particle counts in reconstructions.
  • Potential for Higher Resolution: By improving data quality, the graphene sandwich method opens doors to achieving higher resolution cryo-EM structures, as demonstrated by the successful reconstructions of the tested macromolecules.
  • Radiation Damage Mitigation is Limited: While charging and motion are improved, the study found no significant reduction in radiation damage itself, suggesting that the graphene layers' position at the specimen's periphery limits their ability to scavenge radicals at cryogenic temperatures.

Key Results

  • High-Resolution Structures: The graphene sandwich method enabled the determination of high-resolution cryo-EM structures of apoferritin (3.1 Å), 20S proteasome (2.6 Å), and SARS-CoV-2 spike protein (2.5 Å).
  • Reduced Particle Motion: Particle displacement in the final frame was reduced by approximately two-fold compared to single-sided graphene support.
  • Smaller Charging Footprint: The diameter of the charging footprint was smaller in graphene-sandwiched samples, indicating reduced magnification effects from charging.
  • Improved Overall B-factor: The overall B-factor for apoferritin was significantly lower for the graphene-sandwiched sample (70.9 Ų) compared to the graphene-supported sample (93.5 Ų), indicating better overall data quality.
  • Increased Particle Concentration: The method showed an enrichment of particles on the graphene surface before sandwich formation, potentially increasing the effective concentration of target molecules.

Limitations and Future Work

  • Radiation Damage: The study did not observe a significant mitigation of radiation damage, which remains a challenge.
  • Future Directions: The authors suggest modifying the interior surface of the graphene sandwich to be more structure-friendly and incorporating patterned spacers (similar to graphene liquid cells) to achieve more uniform ice thickness control. They also propose exploring this method for a wider range of protein complexes and sensitive samples.

Summary updated

Abstract

High-quality specimen preparation plays a crucial role in cryo-electron microscopy (cryo-EM) structural analysis. In this study, we have developed a reliable and convenient technique called the graphene sandwich method for preparing cryo-EM specimens. This method involves using two layers of graphene films that enclose macromolecules on both sides, allowing for an appropriate ice thickness for cryo-EM analysis. The graphene sandwich helps to mitigate beam-induced charging effect and reduce particle motion compared to specimens prepared using the traditional method with graphene support on only one side, therefore improving the cryo-EM data quality. These advancements may open new opportunities to expand the use of graphene in the field of biological electron microscopy.
BibTeX
@inproceedings{xu2024graphene,
  author = {Jie Xu and Xiaoyin Gao and Liming Zheng and Xia Jia and Kui Xu and Yuwei Ma and Xiaoding Wei and Nan Liu and Hailin Peng and Hongwei Wang},
  title = {{Graphene sandwich–based biological specimen preparation for cryo-EM analysis}},
  booktitle = {Proceedings of the National Academy of Sciences},
  year = {2024},
  doi = {10.1073/pnas.2309384121},
}
Figure