We all know finding a needle in a haystack is a sheer impossible task, especially when it is buried deep within. If the needle happens to rest on the top of the haystack, one may be fortunate enough to spot it shining in contrast to the surrounding hay. But that rarely happens. Similarly, scientists often face the challenge of locating small targets within large volumes, like specific proteins or specific organelles within the whole volume of a cell. But fear not, correlative light and electron microscopy (CLEM) comes to the rescue.

CLEM combines the strengths of light microscopy (LM) and electron microscopy (EM) techniques, such as transmission electron microscopy (TEM) and focused ion beam scanning electron microscopy (FIB-SEM). First, LM provides a broad view of the sample, capturing specific events or structures with fluorescent labels. Then, the sample undergoes EM processing, allowing nanoscale information to be correlated with LM data using localization markers (Figure 1). CLEM can be performed at a surface level, for example, correlating LM data with a 2D TEM image, as it is a well-established procedure in life sciences. However, sometimes your biological question cannot be answered looking at the surface level of a 2D TEM section, requiring a more global visualization in 3D. Obtaining 3D information in large volumes, called volume imaging, is one of the seven technologies to watch in 2023 chosen by Nature¹, as it allows us to explore complex research questions, such as cell-cell or cell-matrix interactions. This capability will be revolutionary for our understanding of how life happens. However, to date, CLEM volume imaging has been obtained only for dried samples at room temperature²  and is still an unchartered territory when it comes to imaging under cryogenic temperatures.

Imaging under cryogenic conditions is advantageous as it preserves the sample in a fully hydrated state, keeping the biological structures in a vitrified near-native state. But this unprecedented advantage comes with the cost of requiring more complex protocols for both imaging and sample preparation. CryoCLEM has proven successful for certain samples directly placed and vitrified on a TEM grid, enabling atomic-level observations of protein structures within cells³. But for larger samples like organoids or tissues, direct placement and vitrification on a TEM grid are not feasible. These larger models require robust freezing techniques like high-pressure freezing (HPF) to safely vitrify samples up to 200 μm thick. Unfortunately, HPF-frozen samples yield featureless ice surfaces, hindering precise LM-EM correlation. It is as if luck runs out, and the "shining up" spots can no longer guide the way.

At the Electron Microscopy Center of Radboud University Medical Center in the Netherlands, we have developed a creative solution for performing cryogenic correlative imaging in 3D samples, recently published in Communications Biology⁴. We have developed a workflow, bridging 3D cryo LM and 3D cryo FIB-SEM. By imprinting a recognizable and reproducible pattern on the top of the ice using FinderTOP, we have regained our luck to find our way to 3D correlate LM and EM.

This innovative approach was driven by the research interests of our research group⁵, focused on understanding the formation of mineralized tissues. These tissues, like our bones, involve the deposition of collagen-rich extracellular matrices (ECM) impregnated with mineral crystals. To study this process comprehensively, a robust imaging platform is required, and cryogenic imaging is a must to preserve the delicate structures of ECM and mineral precursors. The group chose zebrafish scales as a model, where resident cells deposit a partially mineralized collagen matrix to balance rigidity and flexibility for protection and movement.

As a proof-of-concept of our workflow, regenerating zebrafish scales, with millimeter dimensions and a thickness of tens of micrometers, were harvested, stained, and placed in an HPF carrier. The carrier was sealed with a FinderTOP, which imprints a square grid pattern on the sample surface during vitrification. This pattern is visible in cryoFIB/SEM and reflection mode cryo-light microscopy (Figure 2).

By using the pattern for localization and fluorescent features to identify the region of interest (ROI) with a confocal microscope, we targeted specific locations within the 3D volumes. High-resolution imaging of these locations in the native hydrated state was then performed using cryoFIB/SEM. Computational alignment of the imaging modalities allowed precise localization of mitochondria within cells, in 3D (Figure 3).

We believe that this correlative cryogenic workflow will enable targeted high-resolution volume imaging of various tissues without compromising their integrity, opening new possibilities for groundbreaking discoveries in life sciences. From now on, finding a needle in a haystack can be much easier than ever before.

References

1. Seven technologies to watch in 2023. https://www.nature.com/articles/d41586-023-00178-y.
2. Hoffman, D. P. et al. Correlative three-dimensional super-resolution and block-face electron microscopy of whole vitreously frozen cells. Science (80-. ). 367, (2020).
3. Mahamid, J. et al. Visualizing the molecular sociology at the HeLa cell nuclear periphery. Science (80-. ). 351, 969–972 (2016).
4. de Beer, M. et al. Precise targeting for 3D cryo-correlative light and electron microscopy volume imaging of tissues using a FinderTOP. Commun. Biol. 6, 510 (2023).
5. Biochemistry of Mineralized Tissues Group. https://www.mineralizedtissues.com/.