Correlative Light and Electron Microscopy (CLEM)

Article By Industries Needs

In modern biological and materials science research, visualizing complex structures across spatial scales—from macroscopic tissue architecture down to sub-nanometer molecular machinery—presents a fundamental challenge. Traditional light microscopy (LM) excels at live-cell imaging, multi-color spectral profiling, and specific molecular tagging, but it is bounded by the Abbe diffraction limit (~200 nm laterally). Conversely, transmission electron microscopy (TEM) and scanning electron microscopy (SEM) deliver sub-nanometer spatial resolution and reveal cellular ultrastructure in rich detail, but they lack molecular specificity and cannot image live, dynamic processes.

Correlative Light and Electron Microscopy (CLEM) bridges this gap (van den Dries et al., 2022). By combining fluorescence specificity with electron microscopy ultrastructure, CLEM allows researchers to locate rare or dynamic events in living cells or intact tissues and resolve those exact sites with nanometer-scale precision (van den Dries et al., 2022).

1. Fundamental Principles and Conceptual Rationale

CLEM combines the unique capabilities of light and electron imaging:

• Light Microscopy (LM / FM): Relies on fluorescent proteins (GFP, mCherry), organic dyes, or quantum dots to target specific proteins or organelles with high molecular contrast. FM can record functional dynamics, ion fluxes, and transient membrane fusion events in real time before sample preservation.
  
  
• Electron Microscopy (EM): Uses accelerated electron beams focused by electromagnetic lenses to achieve sub-nanometer resolution. Contrast relies on differential electron scattering by heavy metals (such as osmium tetroxide, uranyl acetate, or lead citrate) bound to cellular components. EM yields comprehensive context, showing organelles, lipid bilayers, macromolecular complexes, and the cytoskeleton without requiring pre-targeted probes.
  
  

CLEM overlaying strategies map fluorescence signals directly onto electron micrographs. This allows researchers to answer specific structural questions—such as determining the membrane morphology at a protein accumulation site or identifying the organelle carrying a tagged target—without relying on contrast guesswork.

       [ Live Cell / Tissue Specimen ]
                      │
        ┌─────────────┴─────────────┐
        ▼                           ▼
[ Fluorescence LM ]        [ Chemical / Cryo Fixation ]
 (Molecular context &       (Structure preservation)
  dynamic localization)             │
        │                           ▼
        │                  [ Electron Microscopy ]
        │                   (Sub-nanometer ultrastructure)
        └─────────────┬─────────────┘
                      ▼
        [ Image Registration & CLEM Overlay ]

2. Typical CLEM Workflows and Methodologies

Executing a CLEM experiment requires balancing sample preservation, fluorescence retention, and target relocation across instruments.

Pre-Embedding CLEM

In pre-embedding CLEM, light microscopy images are captured before the sample undergoes heavy metal staining, dehydration, and resin embedding for EM.

1. Cell Culturing & Live Imaging: Cells are cultured on specialized substrates (e.g., glass-bottom dishes etched with gridded coordinate patterns or finder grids).
   
   
2. Fluorescence Acquisition: Target structures or dynamic events are recorded under widefield, confocal, or super-resolution fluorescence illumination.
   
   
3. Fixation & Processing: The sample is fixed with aldehydes, stained with osmium tetroxide, dehydrated through an ethanol series, and embedded in epoxy resin (such as Epon or Durcupan).
   
   
4. Relocation & Ultramicrotomy: Using the recorded grid coordinates, the region of interest (ROI) is located, trimmed, and sectioned into 50–90 nm ultrathin slices for TEM or block-face SEM.
   
   

In-Resin / Post-Embedding CLEM

In-resin CLEM preserves fluorophore activity after resin embedding, allowing light and electron microscopy on the exact same physical section.

• Specialized Resins: Hydrophilic acrylic resins (such as Lowicryl or LR White) preserve fluorescent protein emission during polymerisation at low temperatures.
  
  
• Fixation Resistance: Fixation protocols must minimize osmium tetroxide concentrations, as heavy metals quench fluorescence (Iwasaki et al., 2022). Photoactivatable or osmium-resistant fluorescent proteins (e.g., mEosEM) are frequently used (Iwasaki et al., 2022).
  
  

Cryo-CLEM

Cryo-CLEM eliminates chemical fixatives and dehydration artifacts by rapidly freezing samples through plunge freezing or high-pressure freezing (HPF), trapping them in a vitrified (non-crystalline) ice state.

1. Cryo-Fluorescence LM: Samples on EM grids are imaged at liquid nitrogen temperatures (below $-150^\circ\text{C}$) on specialized cryo-stages.
   
   
2. Cryo-Focused Ion Beam (Cryo-FIB) Milling: Fluorescence targets guide a focused ion beam to trim ice-embedded cells into thin lamellae ($<200\text{ nm}$).
   
   
3. Cryo-Electron Tomography (Cryo-ET): The lamella is transferred to a cryo-TEM to reconstruct 3D molecular maps of macromolecular complexes in native conditions.
   
   

3. High-Precision Image Registration and Fiducial Markers

A primary technical challenge in CLEM is aligning two image datasets acquired at different resolutions, magnifications, and contrast mechanisms.

• Grid Patterns: Etched coordinate grids on glass substrates allow manual tracking of target fields from light to electron microscopes.
  
  
• Exogenous Fiducial Markers: Fluorescent beads embedded with heavy metal cores (such as gold-fluorophore nanoparticles or FluoroNanogold) appear in both fluorescence channels and electron micrographs (Iwasaki et al., 2022). These landmarks enable affine or non-rigid transformation algorithms to align images with precision down to a few nanometers.
  
  
• Laser-Induced Branding: Near-Infrared Laser Branding (NIRB) uses laser pulses to burn fiducial marks directly into fixed tissue surrounding an ROI, facilitating alignment in complex 3D tissue volumes (Iwasaki et al., 2022).
  
  

4. Key Applications Across Scientific Disciplines

Field	CLEM Application	Primary Outcome
Cell Biology	Mapping membrane trafficking, endocytosis, and autophagosome formation.	Correlates dynamic cargo markers with vesicle coat assembly and membrane curvature.
Neuroscience	Connectomics and synaptic plasticity research (Iwasaki et al., 2022).	Identifies functional neural circuits via fluorophores before tracing unlabelled synaptic connections in 3D EM (Iwasaki et al., 2022).
Virology	Tracking viral entry, assembly, and budding mechanisms.	Connects single virion fluorescent signals to host membrane remodelling events.
Geology & Materials	Characterizing phase boundaries, mineral inclusions, and synthetic materials (Cognigni et al., 2023).	Maps photoluminescence or cathodoluminescence to microstructural defects and crystal lattices (Cognigni et al., 2023).

5. Super-Resolution CLEM (SR-CLEM)

Conventional diffraction-limited light microscopy (~200 nm resolution) creates a resolution gap when correlated with electron microscopy ($<1\text{ nm}$). Super-Resolution CLEM (SR-CLEM) addresses this mismatch by integrating nanometer-scale fluorescence techniques (van den Dries et al., 2022):

• STED-CLEM: Uses Stimulated Emission Depletion (STED) microscopy to achieve sub-50 nm optical resolution before sectioning for EM.
  
  
• SMLM-CLEM: Combines Single-Molecule Localization Microscopy (PALM/STORM) with EM (van den Dries et al., 2022). Because SMLM reconstructs images from single fluorophore positions, it yields targeting precision comparable to EM slice thickness.
  
  

6. Current Challenges and Future Directions

Despite its utility, CLEM presents technical trade-offs:

1. Protocol Incompatibilities: Fixatives like glutaraldehyde and heavy metals like osmium preserve ultrastructure but can quench fluorophores or increase background autofluorescence (Iwasaki et al., 2022). Conversely, soft fixatives preserve fluorescence at the cost of EM structural preservation.
   
   
2. Sample Deformation: Resin polymerization, dehydration, and high-vacuum exposure can cause sample shrinkage, requiring non-rigid registration models to correct image distortion.
   
   
3. Workflow Integration: Emerging integrated platforms embed optical light paths directly inside the vacuum chambers of SEMs or TEMs. This hardware integration allows simultaneous light and electron collection, removing the need for manual sample transfers and reducing registration errors.
   
   

Advances in automated volume electron microscopy (such as FIB-SEM and Serial Block-Face SEM), artificial intelligence-assisted image registration, and fluorophore design continue to streamline CLEM, cementing its role as an essential tool for multi-scale biological imaging.

References

Cognigni, F., Miraglia, L., Contessi, S., Biancardi, F., & Rossi, M. (2023). Correlative Light and Electron Microscopy (CLEM): A Multifaceted Tool for the Study of Geological Specimens. Journal of Experimental and Theoretical Analyses, 1(2), 74–85. https://doi.org/10.3390/jeta1020006

Cited by: 7

Iwasaki, H., Ichinose, S., Tajika, Y., & Murakami, T. (2022). Recent technological advances in correlative light and electron microscopy for the comprehensive analysis of neural circuits. Frontiers in Neuroanatomy, 16. https://doi.org/10.3389/fnana.2022.1061078

Cited by: 9

van den Dries, K., Fransen, J., & Cambi, A. (2022). Fluorescence CLEM in biology: historic developments and current super‐resolution applications. FEBS Letters, 596(19), 2486–2496. https://doi.org/10.1002/1873-3468.14421

Cited by: 54

AI and Machine Learning in Microscopy

Quality Control and Calibration Standards