Tumor Progression Revealed by Single Cell Sequencing

Genomic analysis provides insights into the role of copy number variation in disease, but most current methods are not designed to resolve mixed populations of cells. Methods for analysis of single cells at select loci1,2,3 are available but until recently there were no reliable methods for genome-wide assessment. The existing methods4,5,6 offer limited improvement over cytological techniques that have been available since the 1980’s. The lack is particularly acute in the analysis of genetically heterogeneous tumors, where multiple subpopulations each with distinct mutational histories co-exist7,8,9. Here we have combined whole genome amplification (WGA) and massively parallel DNA sequencing to achieve robust single cell copy number profiles with a resolution of nearly 50 kilobases in a method we call single nucleus sequencing (SNS). We apply this method first to single cells from cultures and show that there is only minor genetic variation. Then we use this technique to profile 100 single cells isolated from a basal-like breast cancer we know to be genetically heterogeneous10. We observe a few distinct subpopulations, an observation that supports a model for tumor progression in which a few sequential clonal expansions form the mass of the tumor. Moreover, we detect significant numbers of pseudodiploid cells that do not share genomic markers with the major expansions, and these cells may signify the presence of potentially large unstable diploid tumor subpopulation not revealed by previous genomic methods.