Microscopical studies in biology have relied on two complementary microscope technologies - light (fluorescence) microscopy and electron microscopy. Light microscopy is used to study phenomena at a global scale and to look for unique or rare events, and it also provides an opportunity for live imaging, while the forte of electron microscopy is the high resolution. Observation of living cells under EM is still impossible. Traditionally light and electron microscopy (EM) observations are carried out in different populations of cells/tissues. The advent of true correlative light-electron microscopy has allowed high resolution imaging by EM of the same structure observed by light microscopy. Thus a rare event captured by low resolution imaging of a population or transient events captured by live imaging can now also be studied at high resolution by electron microscopy.
This chapter describes correlative light-electron microscopy with details and useful tricks, including the way to localize the same cell after its transfection with a protein fused with a fluorescent tag, examination under the microscope in living condition, fixation, immunolabelling, embedding, and observation under EM. We also illustrate here the kinds of questions that the correlative video-light-EM (CVLEM) approach was designed to address, as well as the particular know-how that is important for the successful application of this technique. The potential and difficulties of this approach, along with the most impressive breakthroughs obtained by these methods, are discussed.

1. Introduction

Although advanced optical microscopy techniques can push resolution to 50-100 nm and even below, it is still much less than the resolution of EM and far from the resolution needed for the study of, for instance, the organization of assemblies of proteins and lipids in biological specimens. Often, the analysis of immuno-fluorescently-labeled structures needs a better-than-light-microscopy resolution. On the other hand, although the spatial resolution of EM is superior, its field of view is limited: a resolution of 1 nm can only be realized when small (2x2 µm²) areas are imaged.

Consequently, the study of rarely occurring events in cells or tissues is extremely tedious and time consuming. This limitation has motivated researchers to embark on the development of imaging methods that combine, for example, LM and EM - correlative microscopy (CM). CM uses a combination of microscopy methods for the study of rare cellular events of unique samples.
Now CVLEM is a rather complex procedure with the possibility to use several techniques for the identification of the organelle of interest subsequently under LM and then under EM. The CLEM procedure includes several stages: 1) observation of the structures labeled with fluorescent protein (FP, i.e., green FP) or other fluorescent markers in living cells; 2) immobilization (fixation or freezing); 3) immuno- or other type of labeling with gold or other markers suitable for EM; 4) embedding; 5) identification of the just examined cell in the resin block or within the frozen sample; 6) sectioning of thin or thick serial sections and identification of the cell on the resin block and cutting of thin or thick serial sections; 7) EM analysis and structure identification [1]. Each of these steps could be performed by different ways and all techniques have their own advantages and disadvantages (reviewed in [2]). Here, we are presenting an example of only one from many existing methods of CVLEM. We describe only those protocols that are indispensable for the presented type of CVLEM. Protocols of transfection, observation under a confocal microscope, EM tomography, the reader can find in corresponding protocol books.

2. Procedure

2.1. Observation of living cells and fixation.
2.1.1. Grow i.e. HeLa cells in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum and 2 mM L-glutamine at 37°C and 5% CO₂.
2.1.2. Suspend HeLa cells using standard procedures and plate cells for CVLEM on a MatTek Petri dish with the CELLocate cover slip attached to its bottom. The CELLocate cover slip contains an etched grid with coordinates that allow the localization of the cell of interest at any step in the preparation.
2.1.3. Transfect HeLa cells with cDNA of the GFP fusion protein using any standard method of transfection or microinjection of cDNA into the nucleus.
2.1.4. By 6 h to 48 h (depending of the fusion protein and method of transfection used) after transfection place the dish under an inverted fluorescence microscope or laser scanning confocal microscope.
2.1.5. Select the transfected cell of interest, and identify its position related to the coordinates of the CELLocate grid.
2.1.6. Draw (or photograph) the position of the cell on the map of the CELLocate grid. For instance, the cell could be near the cross of horizontal line A and vertical line 3 (see arrow in Fig. 1A, 2A, see Note 2). Another way to map the cell of interest is to scrap cells around the cell of interest using wood stick and then cells remaining in the centre of ring could be visible (Fig. 1C, D), Finally, the pattern of cell culture could be used for the labeling of the cell position (Fig. 1B)
2.1.7. Observe the dynamics of the GFP-labeled structures in the selected living cell using a multiphoton-, a laser scanning confocal-, or a digitalized fluorescent-inverted microscope, which allows the grabbing of a time-lapse series of images by a computer.
2.1.8. At the moment of interest, add fixative A to the cell culture medium while still grabbing images (fixative A: medium volume ratio is 1:1). Fixation usually induces the fast fading of GFP fluorescence and blocks the motion of labeled structures in the cell.
2.1.9. Stop grabbing time-lapse images and keep the cells in fixative for 5-10 min (during this time it is useful to grab a Z-series of images of the cell).
2.1.10. Replace the mixture with the fixative 1 and keep the cells in the fixative 1 for 5 min.
2.1.11. Replace fixative 1 with fixative 2 and keep cells there for 30 min.
2.1.12. Wash with 0.2 M HEPES (pH 7.2-7.3) for 10 min (see Note 3).