Toward detecting and identifying macromolecules in a cellular context: Template matching applied to electron tomograms

Abstract

Electron tomography is the only technique available that allows us to visualize the three-dimensional structure of unfixed and unstained cells currently with a resolution of 6 -8 nm, but with the prospect to reach 2-4 nm. This raises the possibility of detecting and identifying specific macromolecular complexes within their cellular context by virtue of their structural signature. Templates derived from the high-resolution structure of the molecule under scrutiny are used to search the reconstructed volume. Here we outline and test a computationally feasible two-step procedure: In a first step, mean-curvature motion is used for segmentation, yielding subvolumes that contain with a high probability macromolecules in the expected size range. Subsequently, the particles contained in the subvolumes are identified by cross-correlation, using a set of three-dimensional templates. With simulated and real tomographic data we demonstrate that such an approach is feasible and we explore the detection limits. Even structurally similar particles, such as the thermosome, GroEL, and the 20S proteasome can be identified with high fidelity. This opens up exciting prospects for mapping the territorial distribution of macromolecules and for analyzing molecular interactions in situ. It has been a dream of cell biologists to catch a glimpse of the molecular architecture inside cells or cellular organelles, ideally by using a noninvasive technique (1-3). Rapid freezing techniques have been developed, which allow the ''vitrification'' of biological materials and thus ensure their close-to-life preservation (4, 5). With the advent of automated electron tomography (6-9) it has become possible to obtain three-dimensional (3D) data sets of whole ice-embedded cells or organelles (10, 11) with subcritical doses. Currently, the resolution obtained in electron tomography of cellular structures ( Fig. 1) is in the range of 6 to 8 nm. It is reasonable to expect that high-end instrumentation will bring us into the realm of molecular resolution (2-4 nm). The goal of cellular electron tomography is not to obtain a high-resolution structure of a particular macromolecule; the goal is to identify a molecule by virtue of its structural signature and to locate it in the context of its cellular environment. Inevitably, the electron tomograms will suffer from a low signal-to-noise ratio (SNR), and so-called denoising techniques (12, 13) can provide only partial remedy. However, if we have high-or medium-resolution structures of the molecule under scrutiny, furnished by x-ray crystallography, electron microscopy, or any combination of structural biology techniques, these can be used as templates to search the reconstructed cellular volume. This type of scan will make it possible not only to map the distribution of molecules within the cell; it also will reveal the spatial relationships of molecules in functional modules.The purpose of this paper is to outline a computationally feasible strategy for the detection and identification of macromolecules in to...

Connecting μ-fluidics to electron microscopy

“…Combined with microfluidic methods for cell lysis, protein separation and labelling, this method for total cell content analysis for structure and mass offers the potential to optimally complement other experimental system biology techniques such as MS (Picotti et al, 2009) and ET (Barcena and Koster, 2009;Förster et al, 2010). For this particular application, the raw images must be analyzed by feature matching methods (Bohm et al, 2000;Förster et al, 2010;Penczek and Huang, 2004) to obtain quantitative information. To aid the analysis, a visual library will be constructed containing the projections of a priori target molecules.…”

“…Cryo--electron tomography (cryo--ET) is the ultimate technique to reveal the spatial organisation of protein structures and macromolecular complexes in single cells (Ben--Harush et al, 2010;Lučić et al, 2005). Currently cryo--ET is restrained by several limitations, such as the size of the cell that can be analysed (maximum diameter of ~2 µm) (Leis et al, 2009), and by problems in data segmentation and in the template matching required for protein recognition (restricted to relatively large protein complexes) (Bohm et al, 2000). Indeed, many target structures can only be recognised if labelled with electron--dense markers (e.g., gold labels) (Nickell et al, 2006), which, despite recent progress (Kireev et al, 2008), often involves harsh preparative treatment of the cells.…”

A cell-centered database for electron tomographic data

“…As more examples of structures such as synapses are added to the repository, additional analyses of structural variability become possible and the relationship of this variability to processes such as synaptic functioning can be modeled. Tomography researchers interested in merging cellular and molecular data will also benefit from having a repository of cell-level structures to search for molecular signatures (Bohm et al, 2000;Koster et al, 1997). In this case, the CCDB can work in concert with molecular databases, e.g., the IIMS database (http:// msd.ebi.ac.uk/iims.html) or the PDB, which can provide templates for pattern searches in cellular images.…”

“…Researchers are employing tomography not only to investigate the structure of purified protein complexes, but also to investigate their structure in situ, fitting the lower resolution tomographic data with atomic models or molecular envelopes obtained by molecular microscopy and X-ray crystallography (McEwen and Frank, 2001;Taylor et al, 1999). Others are using the superior resolution of cell-level tomograms to search for molecular signatures of proteins in complex cellular environments based on an understanding of protein structure, without the need for protein-specific stains (Bohm et al, 2000;Koster et al, 1997).…”

Single particle cryoelectron tomography characterization of the structure and structural variability of poliovirus–receptor–membrane complex at 30 Å resolution

“…The ultimate goal is to be able to interpret tomographic reconstructions of crowded cell interiors in terms of their smaller components by matching a library of known higher-resolution structures to features in the low-resolution landscape. Several recent papers have demonstrated substantial progress in this direction, showing that is possible to map individual particles in cells (Bohm et al, 2000;Frangakis et al, 2002;Garvalov et al, 2006;Medalia et al, 2002;Murphy et al, 2006;Ortiz et al, 2006), and in favorable cases to derive reconstructions of macromolecular complexes such as the ribosome, to a few nanometers resolution (Ortiz et al, 2006).…”