Identification of novel circulating human embryonic blood stem cells.

Using murine models, primitive hematopoietic cells capable of repopulation have been shown to reside in various anatomic locations, including the aortic gonad mesonephros, fetal liver, and bone marrow. These sites are thought to be seeded by stem cells migrating through fetal circulation and would serve as ideal targets for in utero cellular therapy. In humans, however, it is unknown whether similar stem cells exist. Here, we identify circulating hematopoeitic cells present during human in utero development that are capable of multilineage repopulation in immunodeficient NOD/SCID (nonobese diabetic/severe combined immunodeficient) mice. Using limiting dilution analysis, the frequency of these fetal stem cells was found to be 1 in 3.2 x 10(5), illustrating a 3- and 22-fold enrichment compared with full-term human cord blood and circulating adult mobilized-peripheral blood, respectively. Comparison of in vivo differentiation and proliferative capacity demonstrated that circulating fetal stem cells are intrinsically distinct from hematopoietic stem cells found later in human development and those derived from the fetal liver or fetal bone marrow compartment at equivalent gestation. Taken together, these studies demonstrate the existence of unique circulating stem cells in early human embryonic development that provide a novel and previously unexplored source of pluripotent stem cell targets for cellular and gene-based fetal therapies. (Blood. 2000;96:1740-1747)