The epithelial-cell component of the thymic stroma is essential for intrathymic T-cell development. Populations of thymic epithelial cells (TECs) that are functionally and phenotypically distinct are found in the cortical and medullary regions of the thymus. Whether these distinct cell types originate from a common progenitor has long been a subject of debate. Now, two complementary papers published in Nature unequivocally identify bipotent progenitors that can give rise to both cortical and medullary TECs in embryonic, as well as neonatal, mice.

Rossi et al. used an elegant clonal assay to investigate the developmental potential of single TEC progenitors in the mouse thymic rudiment at embryonic day 12 (E12). The TEC progenitors were isolated from whole thymic rudiment and phenotypically characterized. The authors showed that at this stage in development the epithelial-cell population was homogeneous with regard to expression of the epithelial-cell markers EpCAM1, MTS24 and cytokeratins, which are expressed at varying levels following differentiation.

staining for markers of cortical and medullary TECs confirmed that a single progenitor cell could give rise to both types of TEC

To track the developmental potential of individual TEC progenitors and to provide them with the appropriate environmental cues, Rossi et al. isolated progenitors from E12 rudiments of mice that were transgenic for enhanced yellow-fluorescent protein (eYFP) and microinjected single progenitor cells into individual E12 thymic lobes from wild-type mouse embryos. These lobes were then grafted under the kidney capsule of syngeneic wild-type mice. After 4 weeks of growth, cortical and medullary regions in the grafted lobes could be clearly defined using antibodies specific for MHC class II molecules. Notably, in all microinjected grafts that contained eYFP-positive cells, the cells were found to be distributed in both the cortical and medullary regions. Further staining for markers of cortical and medullary TECs confirmed that a single progenitor cell could give rise to both types of TEC.

Bleul et al. also used eYFP to follow the fate of TEC progenitors in vivo but, in this study, eYFP expression was only switched on randomly in a few very rare TEC progenitors after birth. Analysis of thymic lobes from these mice showed the presence of rare eYFP-positive single cells, as well as cell clusters that contained both cortical and medullary TECs. Each cell cluster probably arose from a single cell, owing to the rare genetic recombination event that was required to trigger eYFP expression. In addition to the presence of bipotent progenitors, Bleul et al. identified, in some postnatal thymi, unipotent progenitors that gave rise to clusters consisting of only medullary or only cortical TECs. These data indicate that, even after birth, the thymus can regenerate its stromal component.

Bleul et al. also went on to test whether a single epithelial-cell progenitor could generate functional thymic tissue in the alymphoid thymic rudiment of nude mice. Nude mice lack the transcription factor forkhead box N1 (FOXN1), which is crucial for the maturation of TECs, and they therefore develop a primitive thymus consisting of alymphoid cysts. By creating nude mice in which Foxn1 gene expression was restored randomly in the rare epithelial-cell progenitor, the authors showed that, adjacent to the alymphoid cysts, areas of thymic tissue developed from single progenitors. Remarkably, this de novo thymic tissue could support the development of functional T cells and their export to the periphery.

Does identification of the elusive 'stem cell' for TECs bring us closer to cell-based therapies for thymic disorders?