Cytomics goes 3D: Toward tissomics

Authors


CYTOMICS AND MICROSCOPY

Conducting a Human Cytome Project will require state-of-the-art techniques and technologies in different fields. Methods arising from molecular biology (genome level) and protein chemistry (proteome level) will be necessary but not sufficient for this project. Although the cytome contains the genome and the proteome, the cytome level comprises additional and essential information. The art of combining molecular, morphologic, and phenotypic information at the single-cell level and at the biological locus where cellular dysfunction appears and formation of disorders takes place—an art known as microscopy—represents the key framework of methods for scrutinizing the cytome (1, 2).

Although two high-throughput and high-content methods are currently at hand for multiparametric and multicolor cell analysis, i.e., flow and image cytometry, spatial relations and the molecular morphology of cells and tissues are accessible only to microscopic analysis (3–5).

In a cytomics context, multiparameter fluorescence methods and hyperchromatic staining are of particular interest because they permit the sequential analysis of DNA status, protein expression and protein interaction, cellular distribution, and morphologic characteristics (6, 7).

THE WORLD IS NOT FLAT—NEITHER IS THE CELL

In most applications, microscopic images appear as two-dimensional representations of what is actually well extended in three dimensions. Hence, if we want to understand the interaction of molecules in the spatial context, the third dimension is highly relevant. Computer-aided microscopy represents an important improvement in understanding structural relations. It may comprise automated image recording, sophisticated image processing and analysis, three-dimensional (3D) reconstruction (Fig. 1), and data evaluation using bioinformatics. To facilitate data exchange within a Human Cytome Project, international procedures for data standardization in microscopy such as common file standards (8) and database architectures for the storage of image information (9, 10) are required to facilitate the establishment of public domain databases similar those for the Human Genome Project.

Figure 1.

Confocal 3D reconstruction of a mouse skin sheet. Using a Zeiss LSM 510 with a 63× oil-immersion objective lens at a zoom factor of 1, a 3D reconstruction of optical slices was obtained showing a cytoskeletal protein in green and anti-CD45 (pan-leukocyte) in red. The two longitudinal red features in the upper left quadrant are two hairs that are visible due to autofluorescence. The two prominent green structures are hair follicles.

CYTOMICS AND PATHOLOGY

Single-cell–oriented microscopy provides the differential analysis of dysfunctional cells in pathologic samples versus cells in normal adjacent tissue, localized within or between specimens and patients (11, 12). Microscopic analysis further ensures the direct observation of disease-affected cells, whereas flow cytometric analysis of immunologic indicator cells in peripheral blood frequently represent only cellular and molecular correlates for tissue- or organ-associated diseases.

Therefore, predictive medicine by cytomics can be considered a “cell-as-a-chip” test system, to provide multiparametric data of diseased versus healthy conditions, whereas in tissues the structural context of cell-to-cell relations can be additionally acknowledged. The tissue cytometric approach adds novel, as yet hardly appreciated information that will permit the fine tuning of clinical diagnosis and help to precisely predict disease courses in many common and socially important diseases such as cancer and cardiac and infectious diseases.

TISSUE CYTOMETRY AS AN IMPORTANT GOAL

The essential goal is to achieve a reliable and standardized process of quantitative and cell-based tissue cytometry that has recently been called tissomics (Fig. 2), a term that acknowledges the cytometric (i.e., cell-based and stoichiometric) analysis and the fact that the analysis is performed on tissues (13). By intent, the term has been selected to contrast with the traditional approach of histology, being reflected by the term histomics, because histology is generally associated with standard histopathologic work that represents nonfluorescence techniques and manual analysis. The histopathologic approach may further be nonquantitative and complicate standardization efforts (14).

Figure 2.

Tissomics analysis. In a thick tissue section (in this example, 120-μm section of brain tissue), cytometric analysis is performed in different depths of focus (five in this case) and the position of the cells in z direction is determined. This leads to the display of the cells in three dimensions (left three plots). For the entire analyzed block and for different layers, the quantitative expression of multiplexed data can be displayed (as a dot plot, center, or as histograms: e.g., DNA content; 1, G0/G1; 2, S phase; 3, G2/M phase). The position of an individual cell within the tissue (arrows) and those of its neighboring cells can be retrieved with its expression profile. (Brain section was provided by Prof. T. Arendt and stained by B. Mosch, Paul Flechsig Institute of Brain Research, Department of Neuroanatomy, University of Leipzig, Leipzig, Germany. Measurements and data analysis were performed by Dr. D. Lenz, Department of Pediatric Cardiology, Cardiac Center, University of Leipzig and A. Mittag, Interdisciplinary Centre for Clinical Research, Coronary Unit Z10, University of Leipzig.)

The close interaction of various specialists including molecular biologists, protein chemists, pathologists, clinicians, instrument developers, and software engineers will be required to develop tissomics as a standard routine procedure for predictive medicine by tissue biopsy analysis.

WHERE ARE WE HEADING?

The 3D analysis of cytomes requires high-throughput techniques that cover at least two orders of magnitude: 3D within the cell to address the molecular basis of cellular functionality and the cell in its 3D tissue environment to seize the organizational structure of cytomes. Addressing both “dimensions” simultaneously constitutes a technological challenge and will require entirely new approaches within the scope of a Human Cytome Project. The task will open new ways to look at diseases. It has the potential for individualized disease course prediction and, hence, for the development of novel, individualized strategies for curative therapy and new therapeutic concepts.

Ancillary