Breaking the biological code


  • Walter Schubert

    Corresponding author
    1. Molecular Pattern Recognition Research Group, Institute of Medical Neurobiology, Medical Faclty, Otto-von-Guericke University Magdeburg, Germany
    2. International Faculty, CAS-MPG Partner Institute for computational Biology, Shanghai, China
    • Institute of Medical Neurobiology, Molecular Pattern Recognition Research Group, ZENIT Building, Medical Faculty, Otto-von-Guericke-University Magdeburg, Germany
    Search for more papers by this author

Our present knowledge of the cell is the result of decades of cell biological research that has unravelled the biochemical reaction pathways and their compartmentalization in subcellular organelles. One of the biggest future challenges in cell biology and human medicine is to decipher the whole functional plan of a cell or a tissue (its biological code). How does the cell establish, organize and coordinate in time and space the myriads of different cellular functionalities involved, for instance, in migration, in the highly selective topologically confined cell-to-cell interactions during morphogenesis of tissues, organs and organisms? Moreover diseases are the result of the operation of large pathological molecular networks within cells and tissues. To detect and decipher these biological codes (entirety of all protein networks, also termed the toponome) of healthy and diseased organisms we must develop technologies that address the protein network structure and function directly in the cell and tissue in vivo/situ (in contrast to proteomics techniques, that rely on protein analyses ex vivo). Given, for example, the protein network architecture and function of synapses in the central nervous system (CNS), we are facing a significant bottleneck: Appr. 1,000 synaptic proteins have been found by proteomics defining the “average synaptic proteome”. However, while this number of proteins is extremely large, it is physiologically impossible that they are all expressed in every synapse. The important next step is to map all the protein clusters that are really expressed in individual synapses in the CNS to answer the question, which protein clusters are and which are not related to disease, and define their decisive role. Conceivably, the number of similar important problems to be solved in biology and medicine is quasi unlimited, but they all have in common that researchers rely on techniques allowing them to colcocalize a very large number of molecular components in the same biological structure.

Traditional fluorescence microscopy using multiple dyes for colocalization studies is limited to the spectral isolation of five to maximal ten dyes (1). This is not sufficient to identify and explore large molecular networks in the identical biological structure: for example Zipf's law (2) – a power law and measure of the hierarchical architecture of molecular systems (3) – does not apply when the number of molecular components localized simultaneously is too low, e.g. <15, but applies when the number exceeds 45 (3). A powerful way to overcome this spectral limitation of traditional multicolor fluorescence microscopy and address molecular networks is to bleach a dye after imaging and re-stain the same or other structures in the identical sample with the same dye coupled to a tag having the same or another specificity, and repeat similar cycles with other tags many times resulting in multidimensional colocation patterns.

In 1990 it was shown that 17 different proteins could be selectively colocalized in the same muscle tissue section by running many cycles of incubation, imaging and bleaching (4). Since this first demonstration of the feasibility and specificity of “re-staining” for the identification of more than 50 cellular phenotypes at a tissue site, the importance of re-staining techniques has been increasingly recognized (5–12). For example, investigators have combined laser scanning multicolor analysis and localization of various CD markers by performing 3 sequential cycles of de-and re-staining (7). Similar approaches were reported for the analysis of peripheral blood mononuclear cells with the emphasis that the non-consumptive nature of re-staining techniques is superior to flow cytometry (5, 6). A re-staining technique was used to colocalize 9 cellular marker proteins resulting in the identification of a new cellular transdifferentiation mechanism in skeletal muscle regeneration (12), a finding that has been confirmed by in vivo models (13) and apparently has stimulated new cell therapy approaches to dystrophic muscle disorders (14). The advent of high resolution CCD imaging in conjunction with the development of imaging robots that automatically perform a random number of incubation-imaging-bleaching cycles (at least 100), as well as powerful analytical software, has marked the entry of n-dimensional high-throughput analysis of protein cluster-networks in biological systems (3, 10).

Now, Huisman et al. in this issue (page 875) add a new chapter to re-staining approaches. The authors use the dye TO-PRO-3 to calculate DNA content in adrenal tissue sections. They acquire image stacks by confocal laser scanning microscopy, and after bleaching the dye, they re-stain the tissue again with TO-PRO-3 to image the same area again. The authors elaborated a protocol for optimal de-staining and re-staining with that dye. The technique can provide an important complement to current approaches of DNA content and feature analysis: while the clinical importance of DNA feature analysis in malignant tissues has been demonstrated (15), so far, the fading of TO-PRO-3 and unsatisfactory staining results often required several measurements in different visual fields or tissues sections rendering a quantitative analysis difficult. The present approach reported by Huisman et al. overcomes this limitation by re-staining the same tissue section/area thereby permitting a cross-quantification of repetitive staining and saving valuable clinical material. Given these improvements it is reasonable that this approach promises to become an important quantitative tool in clinical cytometry. Moreover, it appears to be possible that TO-PRO-3 nuclear chromatin quantification and large scale protein colocalization studies can be combined to analyze the functional organization of the cell nucleus. I anticipate that the next big challenge in the post genome era will employ the fortutious confluence of re-staining techniques to address the functional organization of cells and tissues on a large scale of molecular cell components. Given, for example the principle of the toponome, the detection of lead proteins (16) controlling protein networks in the one cell, has shown that proteins and protein clusters are hierarchically and topologically interlocked as networks (3): When a lead protein is functionally inhibited or blocked, the topologically confined protein clusters disassemble, and a loss of function is observed. To this end breaking the limitations of multispectral imaging microscopy by running a large number of cycles of de- and re-staining, performed and controlled by imaging robots, will pave the way to finally understand the rules governing as what we term the biological code of cells and tissues.