Recent years have seen much progress in determining the developmental potential of murine stem cells found in the bone marrow. Meanwhile, events and cell types involved in human hematopoietic development, especially the earliest stages, are still poorly understood. This knowledge is paramount if we are to continue to develop innovative cellular therapies for malignancies and other pathologies. We use the NOD/SCID xenotranplantation mouse model to evaluate the potential of human hematopoietic precursors in vivo. The first specific aim of this proposal is to correlate engraftment in the mouse model with graft duration in cancer patients receiving non-myeloablative allogeneic transplants. This study will serve to validate the model as an accurate pre-clinical predictor of the performance of transplant products in patients by demonstrating that the same cells are responsible for engraftment in both species. The second specific aim of the proposal concerns using the mouse model as an accurate system to predict T cell recovery in patients following myeloablative therapies. Here we will assess the ability of the mouse thymus to produce normal human T cell development following bone marrow repopulation with human precursor fractions. In the final specific aim of this study we will characterize the potential of a newly defined, rare population of putative stem cells found in cord blood and in bone marrow, termed SP cells. We will subfractionate this population and identify fractions with long-term, multilineage potential, expand these populations in vitro, and characterize any additional potential they may have to supply precursors for other tissue types. If we are successful in these studies we will have demonstrated that the NOD/SCID mouse model, that is used increasingly as a pre-clinical assay, is capable of accurately predicting the performance of various human precursor fractions and cell products in human patients. Additionally, we should be able to demonstrate which human precusor fractions are capable of giving rise to a broad spectrum of T cells following myeloablation and stem cell transplant, that will be important in clinical strategies that rely on graft versus tumor effects. Finally, it should be possible to characterize the potential of human SP cells and may be possible to expand these putative stem cells in vitro. This would be an enormous gain for patients enrolled in future clinical trials that will rely on stem cell therapies.