Background

Multiple protein targets can be imaged at once by fluorescence microscopy (FM), electron microscopy (EM), or correlative light and EM (CLEM) (Giepmans et al., 2005; Lucas et al., 2012; Philimonenko et al., 2014; Johnson et al., 2015; Kim et al., 2015). FM enables multi-color microscopy in both living and fixed cells, and acquiring data can be relatively fast and easy. However, EM offers better resolution for imaging nanoscale features, including cell receptors, membrane boundaries, neuronal connections (Hildebrand et al., 2017), chromatin organization (Ou et al., 2017), or the endocytic machinery (Sochacki et al., 2017). We anticipate an increased reliance on multi-color, cross-platform imaging for investigating proteins associated with normal cell function and human diseases (Megason and Fraser, 2007; Lichtman et al., 2008; Milne and Subramaniam, 2009; Muller and Heilemann, 2013; Plaza et al., 2014; Kremer et al., 2015; Lucocq et al., 2015; Karreman et al., 2016; Laine et al., 2016; Romero-Brey and Bartenschlager, 2017). Currently, most high-resolution imaging studies obtain protein-specific contrast with immunolabeling, which has known shortcomings (Berglund et al., 2008; Bordeaux et al., 2010; Baker, 2015; Bradbury and Pluckthun, 2015).

The central obstacle that has limited progress in multi-scale microscopy is the shortage of genetic tags for labeling proteins. There are a number of protein tags available for imaging proteins by FM, including fluorescent proteins and self-labeling enzyme tags (Sunbul and Yin, 2009; Hinner and Johnsson, 2010). However, there are few genetic tags for EM or CLEM (Ellisman et al., 2012). Most EM tags rely on the oxidation of diaminobenzidine (DAB) to form a polymer that is stained with osmium tetroxide to generate contrast. Examples include APEX (Martell et al., 2012; Lam et al., 2015), miniSOG (Shu et al., 2011), the tetracysteine tag (Gaietta et al., 2002), and others (Kuipers et al., 2015; Liss et al., 2015). Among the DAB-reliant EM tags, miniSOG, FLIPPER (Kuipers et al., 2015), and the tetracysteine tag are compatible with CLEM. However, DAB staining is finicky, and it can be difficult to localize the stain sufficiently to resolve targets. Further progress in multi-scale microscopy will require new methods for labeling proteins with EM- and CLEM-compatible reporters, such as quantum dots (Qdots) (Giepmans et al., 2005) or FluoroNanogold^TM particles (Takizawa et al., 2015).

We recently described a new class of tags for multiscale microscopy called Versatile Interacting Peptide (VIP) tags (Tane et al., 2017; Doh et al., 2018). VIP tags use heterodimerizing coiled-coil peptides to label proteins. One coil is expressed as a fusion to the protein of interest. This coil has a partner, the probe peptide, that is conjugated to a reporter molecule to deliver protein-specific contrast. The probe peptide can be conjugated to a number of reporters, such as fluorophores, small molecules (e.g., biotin), or nanoparticles.

In this Bio-Protocol, we outline the use of VIPER to image a transmembrane receptor by CLEM. VIPER is a VIP tag comprised of CoilE tag and CoilR probe peptide. In Procedure A, we describe the plating and transfection of cells to express a CoilE-tagged receptor: transferrin receptor 1 (TfR1-CoilE). Procedure B describes how to label a cell receptor with CoilR-biotin for subsequent detection with streptavidin-Qdot655. In Procedure C we have an illustrated guide on how to mount ITO coverslips to a slide holder for correlative fluorescence imaging. Procedure D describes image acquisition with a commercially available CLEM microscope, the FEI CorrSight^TM. In Procedures E and F we describe methods for preparing samples for EM. Procedure G details the acquisition of SEM micrographs on a Helios Nanolab^TM 660 EM. We additionally developed a quantitative image analysis pipeline for automated image segmentation on high magnification SEM images (see the Data Analysis section). This program runs in Matlab and reports the number of nanoparticles within a field of view in SEM micrographs.

This protocol is published in tandem with two supporting publications detailing the use of VIPER (Figure 1). Doh et al. (2019a) details how to generate CoilR probe peptides, including biotinylated CoilR. Doh et al. (2019b) describes imaging VIPER-labeled proteins in cells by FM.