Characterization of murine hematopoietic stem cells with high self-renewal activity

Hematopoietic stem cells (HSCs) produce all blood cell types required throughout life. They are identified by their ability to sustain the production of at least 1% of the mature white blood cells for 4 or more months, as demonstrated in limiting dilution or single-cell transplants. Recently, methods for obtaining suspensions of highly purified HSCs from mouse bone marrow have been developed. This has made possible the design of experiments to address: (i) the nature and extent of their biological heterogeneity, (ii) whether maintenance of durable in vivo reconstituting activity is regulated separately from HSC survival and proliferative activity, and (iii) whether differences in gene expression distinguish HSCs with durable as compared to finite in vivo self-renewal potential. Assessment of the different types of donor-derived blood cells produced in ~100 mice transplanted with a single HSC (or a 4-day in vitro-derived clone) allowed identification of 4 subtypes, only 2 of which could propagate continuing blood formation in secondary and even tertiary recipients. To facilitate the investigation of HSC subtypes with durable self-renewal potential (as compared to the other two subtypes with large, but finite, self-renewal potential), I devised a strategy that achieves their simultaneous, but separate, isolation from each other prior to transplantation. From time course experiments that compared the effects of altered extrinsic cytokine stimulation on different HSC activities, I showed that low Steel factor concentrations can rapidly (within 16 hours) extinguish their in vivo regenerative ability without affecting their immediate subsequent survival or mitogenesis in vitro. Finally, from comparative gene expression analyses I identified 3 genes (Vwf, Rhob, and Pld3) that are consistently expressed at higher levels in HSCs with durable self-renewal potential than in several closely related cell types with less extensive or already extinct self-renewal potential. Together, these findings provide strong support for a model of HSC regulation that includes a degree of separation in the mechanisms that control HSC self-renewal from those influencing their survival, mitogenesis, and lineage commitment probabilities. Further investigations of this model using the tools and molecules herein identified should facilitate improvements in ex vivo HSC expansion and in understanding leukemogenesis.Medicine, Faculty ofMedical Genetics, Department ofGraduat