Long-term repopulation of irradiated mice with limiting numbers of purified hematopoietic stem cells: in vivo expansion of stem cell phenotype but not function

Hematopoietic stem cells were isolated from normal adult mouse bone marrow based on surface antigen expression (Thy-1.1(low)Lin(neg)Ly-6A/E+) and further selected for low retention of rhodamine 123. This population of cells (Rh-123low) could mediate radioprotection and long-term (greater than 12 months) repopulation after transplantation of as few as 25 cells. Transfer of five genetically marked Rh-123low cells in the presence of 10(5) normal bone marrow cells resulted in reconstitution of peripheral blood by greater than 10% donor cells in 64% (30 of 47) of recipient mice. Of 46 animals surviving after 24 weeks, 10 had over 50% donor-derived cells in peripheral blood. Two general patterns of long-term reconstitution were observed: one in which many donor-derived cells were observed 5 to 6 weeks after reconstitution and another in which donor-derived cells were rare initially but expanded with time. This result suggests that two classes of long-term repopulating hematopoietic stem cells exist, differing in their ability to function early in the course of transplantation. Alternatively, distinct anatomic sites of engraftment may dictate these two outcomes from a single type of cell. As an approach to measure the extent of self-renewal by the injected cells, recipients of five or 200 stem cells were killed 8 to 13 months after the transplants, and Thy-1.1(low)Lin(neg)Ly-6A/E+ progeny of the original injected cells were isolated for a second transplant. While a numerical expansion of cells expressing the cell surface phenotype of stem cells was observed, along with activity in the colony-forming unit-spleen assay, the expanded cells were vastly inferior in radioprotection and long-term reconstitution assays when compared with cells freshly isolated from normal animals. This result demonstrates that in stem cell expansion experiments, cell surface antigen expression is not an appropriate indicator of stem cell function.