Microsc. Microanal., page 1 of 20 doi:10.1017/S1431927614000099 © MICROSCOPY SOCIETY OF AMERICA 2014 Liquid Scanning Transmission Electron Microscopy: Imaging Protein Complexes in their Native Environment in Whole Eukaryotic Cells Diana B. Peckys 1 and Niels de Jonge 1,2, * 1 Leibniz Institute for New Materials (INM), 66123 Saarbrücken, Germany 2 Department of Molecular Physiology and Biophysics, Vanderbilt University School of Medicine, Nashville, TN 37232-0615, USA Abstract: Scanning transmission electron microscopy (STEM) of specimens in liquid, so-called Liquid STEM, is capable of imaging the individual subunits of macromolecular complexes in whole eukaryotic cells in liquid. This paper discusses this new microscopy modality within the context of state-of-the-art microscopy of cells. The principle of operation and equations for the resolution are described. The obtained images are different from those acquired with standard transmission electron microscopy showing the cellular ultrastructure. Instead, contrast is obtained on specific labels. Images can be recorded in two ways, either via STEM at 200 keV electron beam energy using a microfluidic chamber enclosing the cells, or via environmental scanning electron microscopy at 30 keV of cells in a wet environment. The first series of experiments involved the epidermal growth factor receptor labeled with gold nanoparticles. The labels were imaged in whole fixed cells with nanometer resolution. Since the cells can be kept alive in the microfluidic chamber, it is also feasible to detect the labels in unfixed, live cells. The rapid sample preparation and imaging allows studies of multiple whole cells. Key words: Liquid STEM, ESEM-STEM, gold nanoparticle, specific label, epidermal growth factor receptor (EGFR), whole cell, live cell microscopy, environmental scanning electron microscopy I NTRODUCTION Cellular function is governed by the molecular machinery of proteins, DNA, lipids, etc., dynamically assembling into functional macromolecular complexes. Over the last half century numerous technical advances in biomedical research achieved important insights into these complex molecular interactions (Sali et al., 2003; Robinson et al., 2007). Yet, many scientific questions about the interplay of macro- molecular complexes relating to cellular function are not fully understood, partly because methods are lacking to examine them within the context of intact cells. Biochemical methods for characterization of protein complexes first destroy the cells and then process the pooled material derived from originally discrete and separate intracellular compartments. As a consequence, biochemical studies sometimes create contradicting results (Mackay et al., 2007), in addition, they cannot provide information about the original location of the complexes within the cells. Imaging cellular components in the context of whole cells is challen- ging because the involved dimensions are at the nanoscale. The length scale at which protein interactions occur is exemplified in Figure 1 for the dimer formation of the epidermal growth factor (EGF) receptor (EGFR) (Coskun & Simons, 2011; Arkhipov et al., 2013), a transmembrane receptor playing a critical role in the pathogenesis and progression of many different types of cancer (Normanno et al., 2006). An important question is under which conditions and in which cellular regions dimerization occurs (Arkhipov et al., 2013; Endres et al., 2013). This example reflects only one of many questions related to the interplay of the approxi- mately 20,000 proteins at work in a typical eukaryotic cell. A biologist or biophysicist would ideally have access to a rapid microscopy technique with spatial resolution of a few nanometers to identify the locations of individual subunits in protein complexes, while the cell would remain intact and in its natural aqueous environment. As discussed below, fluor- escence techniques (Hell, 2007; Lippincott-Schwartz & Manley, 2009) do not have sufficient spatial resolution. Nanoscale resolution is traditionally achieved with electron microscopy but various available methods involve plastic embedding or preparation in amorphous ice, and sectioning, slicing, or fractionation of the cells (Robinson et al., 2007; Stahlberg & Walz, 2008; Pierson et al., 2009; Hoenger & Bouchet-Marquis, 2011; Kourkoutis et al., 2012). The ela- borative sample preparation prevents rapid imaging, and the cells do not remain intact nor in their native liquid water environment. In particular, membrane proteins representing many important drug targets remain difficult to study with the currently available microscopy methods. This paper describes a recently developed approach for the nanoscale study of proteins in whole eukaryotic cells in liquid that combines much of the functionality of light microscopy with the high resolution of electron microscopy, so-called Liquid scanning transmission electron microscopy *Corresponding author. niels.dejonge@inm-gmbh.de Received September 26, 2013; accepted January 2, 2014