Resolving Macromolecular Structures via Cryo-Electron Microscopy

193 concepts148 papers3 proposals

For decades, structural biology relied primarily on X-ray crystallography to map the atomic coordinates of life. But this approach has a fundamental limitation: it requires biological molecules to pack into rigid, highly uniform crystal lattices. For many of the most critical players in biology—such as flexible membrane proteins, dynamic molecular machines, and disordered amyloid filaments—crystallization is either impossible or forces the molecule into an artificial conformation.

Cryo-electron microscopy (cryo-EM) bypasses crystallization entirely by preserving molecules in their native liquid state through vitrification. By flash-freezing a thin aqueous layer of a sample using a cryogen like liquid ethane, water molecules are frozen instantly into an amorphous, glass-like ice before they can form a crystalline lattice. As highlighted by Xianwu Jing et al. in Synthesis and drag reduction properties of a hydrophobically associative polymer containing ultra-long side chains, vitrification avoids the structural distortions and artifacts typically introduced by traditional chemical dehydration or freeze-drying methods, allowing researchers to observe delicate microstructures exactly as they exist in solution.

Single-Particle Analysis: Averaging Out the Noise

Once a sample is vitrified, it is imaged under an electron microscope. However, a single raw cryo-EM image is incredibly noisy. Because high-energy electrons damage delicate biological bonds, researchers must use a very low electron dose, resulting in a grainy image with extremely low contrast.

To overcome this noise and reconstruct a high-resolution 3D structure, cryo-EM relies on Single-Particle Analysis (SPA). The core intuition of SPA is that a single micrograph contains hundreds or thousands of individual, identical protein particles suspended in random orientations. If you can determine the relative 3D orientation of each 2D projection, you can computationally align and average them. The random background noise cancels out, while the true structural signal amplifies.

[Vitrified Sample with Random Orientations] 
       │
       ▼ (Low-dose Electron Imaging)
[Noisy 2D Projection Images] 
       │
       ▼ (2D/3D Classification & Alignment)
[Averaged, High-Contrast Conformational States]
       │
       ▼ (Angular Back-Projection)
[High-Resolution 3D Density Map]

A major strength of this approach is its ability to handle biological flexibility. Because the molecules are frozen mid-motion, a single grid captures the entire structural ensemble. Instead of averaging everything together and getting a blurry map, advanced image processing algorithms sort the particles using 2D and 3D classification.

As demonstrated by Fan Yang in Single-particle cryo-EM structural studies of the β2AR–Gs complex bound with a full agonist formoterol, sorting heterogeneous protein particles computationally allows researchers to isolate distinct conformational states from a single noisy dataset. Similarly, Mengyuan Xu et al. utilized 3D classification in CryoEM structures of the human CLC-2 voltage-gated chloride channel reveal a ball-and-chain gating mechanism to separate the channel's native state from its inhibitor-bound state, resolving a flexible gating peptide that would otherwise be obscured.

Helical Reconstruction: Exploiting Intrinsic Symmetry

When dealing with elongated, filamentous structures—such as the amyloid aggregates responsible for neurodegenerative diseases—standard single-particle alignment fails because the filaments cannot be easily isolated into discrete, isotropic particles. Instead, researchers turn to helical reconstruction.

Helical reconstruction exploits the mathematical properties of a helix. Because a filament is composed of identical protein subunits arranged with a regular twist and rise, a single image of a filament contains multiple views of the same subunit from a continuous gradient of angles. By extracting overlapping segments along the filament and applying symmetry constraints, image processing software can treat these segments as highly ordered particle ensembles.

This methodology, refined through software packages like RELION as discussed by Yue‐De Yang et al. in Cryo-EM structures of tau filaments from Alzheimer’s disease with PET ligand APN-1607, has transformed our understanding of proteopathies. For example, Benjamin Falcon et al. demonstrated in Tau filaments from multiple cases of sporadic and inherited Alzheimer's disease adopt a common fold. that tau aggregates across diverse clinical cases share a remarkably consistent atomic "fold."

Conversely, Yang Yang et al. used helical reconstruction in Structures of α-synuclein filaments from human brains with Lewy pathology to reveal that $\alpha$-synuclein forms a distinct "Lewy fold" that differs fundamentally from structures generated in artificial in vitro environments. This capability allows researchers to classify neurodegenerative diseases at the molecular level based on the precise geometry of their filament cores.

Cryo-Electron Tomography: Structural Biology In Situ

While SPA and helical reconstruction excel when imaging purified proteins, many macromolecular complexes only function when embedded within their broader cellular architecture. For these systems, researchers use Cryo-Electron Tomography (cryo-ET).

Instead of averaging thousands of separate physical particles, cryo-ET takes a single vitrified specimen—such as a whole cell or a viral particle—and physically tilts it under the electron beam, capturing a "tilt series" of images from multiple angles. These 2D projections are then mathematically back-projected to calculate a 3D volume called a tomogram.

Because the sample is continuously exposed to electrons throughout the tilt series, managing radiation damage is paramount. To preserve high-resolution details, You Fu et al. utilized a dose-symmetric tilt scheme in Architecture of CTPS filament networks revealed by cryo-electron tomography. This scheme begins imaging at a zero-degree tilt and alternates back and forth between positive and negative angles (e.g., $0^\circ, 3^\circ, -3^\circ, 6^\circ, -6^\circ$). This ensures that the highest-resolution structural information is captured in the earliest exposures, before radiation damage accumulates, allowing the detailed mapping of self-assembling cellular networks in situ.

Pushing the Resolution Frontier

As the field moves toward true atomic resolution, researchers must correct for subtle physical and geometric distortions that occur during imaging.

For massive macromolecular complexes, the depth of focus of the electron microscope becomes a limiting factor. The assumption that the electron wave remains perfectly planar breaks down across thick samples—a phenomenon known as the Ewald sphere effect. To mitigate this, Zhennan Zhao et al. applied block-based reconstruction in Near-atomic architecture of Singapore grouper iridovirus and implications for giant virus assembly, mathematically breaking a giant viral capsid into smaller, localized sub-volumes that can be refined independently to maintain sharp resolution across the entire structure.

Global Motion Blur: Flexible domains move relative to the whole complex.
       │
       ▼ (Apply Local Masking)
Focused Alignment: Algorithm tracks only the region of interest.
       │
       ▼ (Local Refinement)
Near-Atomic Map: Small-molecule interactions become clearly visible.

Conversely, when imaging small, highly flexible regions within a massive complex, global alignment algorithms can become confused by the movements of the larger structure. In these scenarios, local refinement techniques are applied. By computationally masking out the rest of the protein and focusing the alignment algorithm exclusively on the region of interest, researchers can resolve fine details that would otherwise be blurred by global motion.

Zephan Melville et al. leveraged this strategy in A drug and ATP binding site in type 1 ryanodine receptor to map the precise coordinates of a small-molecule therapeutic within a highly flexible binding pocket. As noted by Chuan Hong in Maximizing the value of cryoEM in drug discovery, this ability to visualize ligand-target interactions without the need for crystallization has bridged a massive gap in structural biology, turning cryo-EM into an essential engine for rational drug design.

History & citationVersion 1 · 2026-10-04

Current · Permanent link

View frozen version 1

Cite this direction

Latest version (a living page):

@misc{lacuna_cryo_electron_microscopy_59f8755e,
  title = {cryo-electron microscopy},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/cryo-electron-microscopy/rdr_59f8755e749e48079e9ec43c4cbafae1},
  note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-20},
}

This exact version (version 1, frozen text):

@misc{lacuna_cryo_electron_microscopy_59f8755e_v1,
  title = {cryo-electron microscopy},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/cryo-electron-microscopy/rdr_59f8755e749e48079e9ec43c4cbafae1/v/1},
  note = {Version 1, frozen 2026-10-04. Accessed 2026-11-20},
}
About

These directions overlap with this one through shared papers.

See something wrong? Report a mistake