Derivation of myoepithelial progenitor cells from bipotent mammary stem/progenitor cells.

There is increasing evidence that breast and other cancers originate from and are maintained by a small fraction of stem/progenitor cells with self-renewal properties. Recent molecular profiling has identified six major subtypes of breast cancer: basal-like, ErbB2-overexpressing, normal breast epithelial-like, luminal A and B, and claudin-low subtypes. To help understand the relationship among mammary stem/progenitor cells and breast cancer subtypes, we have recently derived distinct hTERT-immortalized human mammary stem/progenitor cell lines: a K5(+)/K19(-) type, and a K5(+)/K19(+) type. Under specific culture conditions, bipotent K5(+)/K19(-) stem/progenitor cells differentiated into stable clonal populations that were K5(-)/K19(-) and exhibit self-renewal and unipotent myoepithelial differentiation potential in contrast to the parental K5(+)/K19(-) cells which are bipotent. These K5(-)/K19(-) cells function as myoepithelial progenitor cells and constitutively express markers of an epithelial to mesenchymal transition (EMT) and show high invasive and migratory abilities. In addition, these cells express a microarray signature of claudin-low breast cancers. The EMT characteristics of an un-transformed unipotent mammary myoepithelial progenitor cells together with claudin-low signature suggests that the claudin-low breast cancer subtype may arise from myoepithelial lineage committed progenitors. Availability of immortal MPCs should allow a more definitive analysis of their potential to give rise to claudin-low breast cancer subtype and facilitate biological and molecular/biochemical studies of this disease.