Probing the future of correlative microscopy

Abstract

Instrumental developments in microscopy, such as confocal scanning light microscopy and super resolution light microscopy, have firmly established imaging as a key technology in modern life science research. However, this would not have been possible without the availability of the right probes—the concurrent emergence of green fluorescent protein (GFP) completely transformed the field. We can surely assume that GFP and its derivatives have now penetrated most biomedical research labs that use imaging as a tool. The importance of fluorescent probes in light microscopy can further be measured by the award of the Nobel Prize for Chemistry in 2008 to Tsien, Shimomura and Chalfie. The role of probes is as pertinent in the emerging field of correlative microscopy, though arguably far more complex. In correlative microscopy, two or more imaging modalities are applied to a single sample, with the combined images yielding more information than when each modality is used independently. To reflect this, we often use the phrase ‘1+1=3’. The most established correlative microscopy technique is correlative light electron microscopy (CLEM). Usually confocal LM and transmission EM (TEM) are correlated, though recently there have been major developments in both light and electron microscopy leading to other combinations. Correlative light and scanning EM in particular is a growing area due to the availability of new 3D automated microscopes like the serial block face SEM, the focused ion beam SEM, and array tomography [5]. However, whereas fluorescent molecules (either as an expression construct or through antibody or ligand coupling) are the most commonly used to detect a protein of interest in LM, this fluorescence is not directly visible in the electron microscope. Here, an electron-dense moiety such as a gold particle is generally required. Thus, the most direct way to produce a marker/probe (these terms are used interchangeably in the field and in this issue) is to couple both a fluorescent molecule and an electron-dense particle to the protein of interest. Unfortunately though, it is not that simple, and in most instances, the direct coupling of, e.g. a gold particle next to a fluorescent molecule decreases or completely masks the fluorescence emission (see, e.g. [2]). Hence, there have been major efforts to create the ‘optimal’ CLEM probe, one that is visible directly in the LM and EM and can be easily tagged to a protein of interest. Most likely there will never be a single optimal CLEM probe that can be used for all applications, and one has to carefully analyse which probe is best suited for each application. This special issue deals with these latter two cases: the development of new probes and the application of a probe to a specific biological question. The issue starts with a paper by the group of John Robinson discussing possibly the oldest CLEM probe—fluoronanogold. Developed at the beginning of the 1990s, it combines a fluorescent moiety with a very small nanogold particle. As the quenching of fluorescence is strongly influenced by the size (and distance) of the gold particle to the fluorophore [2], this probe does not seem to suffer serious quenching. A disadvantage is that the 1.4-nm gold is too small to be directly visible in the EM against the background contrast of cells or tissues. Hence, silver enhancement has been a necessity. Originally used on Tokuyasu sections, the application of fluoronanogold in pre-embedment * Paul Verkade bixpv@bristol.ac.uk