Abstract
In both health and disease, human tissues are composed of a multiplicity of cellular types, many of which unknown and uncharacterized. Accurate, quantitative measurements of tissue cell composition are difficult to perform, as classic analytical techniques, such as flow cytometry and immunohistochemistry, depend on the availability of antigen-specific monoclonal antibodies and allow only for a few markers to be tested in parallel on each cell. As a result, the full repertoire of cell types contained in a tissue, and their relative abundance in physiological and pathological situations, remain largely unexplored. We developed a novel analytical platform for the study of tissue cell composition, based on single-cell gene-expression RT-PCR (SINCE-PCR). We tested the precision, accuracy and sensitivity of the method, and we used it to investigate the cell composition of human colon epithelia, both normal and cancerous. Starting from primary surgical specimens, we obtained single-cell suspensions of colon epithelial cells and we used flow cytometry to sort them one-by-one, in arrays of hundreds of single cells. We then exploited a microfluidic platform to analyze each individual cell for the expression of 96 genes in parallel, and we used statistical clustering algorithms to associate cells with similar gene-expression profiles. This novel approach led to the discovery of new epithelial cell populations and novel biomarkers to differentially label them. Moreover, it showed that, in human colon tumors, the heterogeneity observed among cancer cells closely mirrors the cellular diversity observed in normal epithelia, where immature progenitor cells intermingle with distinct subsets of mature goblet cells and enterocytes. Finally, SINCE-PCR analysis of monoclonal colon cancer xenografts, obtained from injection of a single (n = 1) “cancer stem cell” into immunodeficient mice, formally proved that this diversity is epigenetic in nature and originates from multi-lineage differentiation processes, reminiscent of physiological stem-cell systems.