Cytomics in characterizing Toponomes: Towards the biological code of the cell

摘要

Progress in genomics, transcriptomics, proteomics, and molecular imaging has created enormous opportunities to analyze the molecular components of biological systems (1-6). Simultaneously, a new challenge arose: identifying the molecular networks that are formed by these components in cells, and understanding their “biological code” (the toponome) encompassing all cell functionalities (7, 8). Assuming that every molecular network is specifically characterized by a unique “mosaic image” of its molecular components, and assuming further that hundreds, or possibly thousands, of different components are assembled in one network, it would not be sufficient to localize only few of these components to “capture/recognize” the mosaic as a whole and identify it clearly in situ. To address this challenge we need imaging technologies enabling us to analyze the spatial arrangements of an extremely large number of molecular cell components in morphologically intact fixed cells in a given condition, on the single cell level, in one experiment. A robot system termed multi-epitope-ligand-kartography (MELK) (7), based on a 17-parameter fluorescence imaging technique (9), offers a solution to this problem, by bleaching each dye completely after each labelling and imaging step, and then applying a further set of dye-conjugated tags (antibodies or any other affinity reagent) to image the distribution of additional molecular cell components. A large series, or in situ chains, of those incubation–imaging–bleaching cycles leads to toponome maps of cells or tissues showing directly which proteins are colocalized or anti-colocalized in different compartments, revealing the functional organization of cellular structures (Fig. 1). In other words, we gain for the first time insight into the cell as a protein pattern formation machine on a high-dimensional light microscopic scale, and most importantly, can identify the proteins which are always colocalized in certain compartments of the cell or in certain cell types, while others are never colocalized, and how this system is altered under different biological conditions. For example, by using a muscle regeneration marker (10) as a land mark for orientation in situ, a nine-dimensional MELK analysis was performed on one single human muscle tissue section providing direct evidence for the existence of single transdifferentiating endothelial cells forming myogenic cells in situ (11), a prediction that was subsequently confirmed by an animal model (12). While these data suggested that new cellular mechanisms can be directly discovered by MELK in situ, it may be considered that, in principle, toponome patterns obtained by a proteome-wide MELK-toponome screening approach (fraction by fraction, and tissue by tissue), will allow us to systematically extract the rules, e.g. the grammar, of the whole protein co-compartmentalization apparatus of the cell in health and disease. Once the modes and rules of this system are known (which would be already a huge step), how can we examine which of the colocalized proteins are interdependent as a network and which are not? Subcellular, three-dimensional toponome fingerprints of two rhabdomyosarcoma cells. Different colors (except green) indicate different combinatorial arrangements of proteins within a variety of subcellular compartments, most of which are vesicles. Cell on the left: migratory cell; cell on the right: quiescent state. Note that these cells states are dinstinguishable by (i) type of vesicles (different combinatorial protein arrangements); (ii) numbers of vesicles of a given type, and (iii) location/relative arrangements of different vesicle types. The three-dimensional structures were superimposed on the original phase contrast images. Green: cell nuclei. Bar: 10 μm. They are mostly organized into topologically confined subpatterns, which are clearly separated from each other (7, 8). These subpatterns are termed combinatorial molecular phenotypes (briefly CMPs