Hematopoietic stem cells: inferences from in vivo assays.

Mice and humans both contain a population in their marrow which can permanently regenerate all of the hematopoietic lineages. This developmental potential was first demonstrated in myeloablated mice transplanted with genetically marked marrow obtained from congenic donors. More recently, this approach has been used to devise an in vivo limiting dilution assay for "competitive (lymphomyeloid) repopulating units" (CRU) that allows murine hematopoietic stem cells to be quantitated. Measurements of murine CRU have shown that this population expands concomitantly with the total hematopoietic system during ontogeny and to some extent post-transplant. During these periods of expansion, defective c-kit function can be seen to preferentially compromise CRU self-renewal more than early CRU detection (which requires differentiation and amplification of the progeny of CRU, but may not require extensive CRU self-renewal). In humans, a similar cell type with transplantable lymphomyeloid differentiation potential can be identified in cord blood on the basis of its ability to engraft sublethally irradiated immunodeficient nonobese diabetic/severe combined immunodeficient mice. Quantitation of these human CRU by limiting dilution analysis of unseparated, highly purified (CD34+CD38-) and cultured (CD34+CD38-) human cord blood cells indicates that their numbers (like the long-term culture-initiating cell [LTC-IC] population) can be slightly expanded in cytokine-supplemented serum-free media, although not as extensively as anticipated from analogous studies of human marrow LTC-IC cultured under the same conditions. Taken together, the results of our studies suggest that the self-renewal of mitotically activated hematopoietic stem cells can be enhanced by their interactions with particular cytokine combinations whose effectiveness in this regard may change during ontogeny.