The inherent differentiation program of short-term hematopoietic repopulating cells changes during human ontogeny.

Human umbilical cord blood (CB) could be an attractive source of hematopoietic repopulating cells for clinical stem cell therapy because of its accessibility and low propensity for unwanted immune reaction against the host. However, CB recipients suffer from severely delayed and often chronically deficient platelet recovery of unknown cause. Here we show that human short-term repopulating cells (STRCs), which predominantly carry early hematopoietic reconstitution after transplantation, display an intrinsically fixed differentiation program in vivo that changes during ontogeny. Compared to adult sources of hematopoietic cells, CB myeloidrestricted STRC-M showed a markedly reduced megakaryocytic and erythroid cell output in the quantitative xenotransplantation of human short-term hematopoiesis in NOD/SCID-beta2m(-/-) mice. This output in vivo was not altered by pre-treating CB cells before transplantation with growth factors that effectively stimulate megakaryocytopoiesis in vitro. Moreover, injecting mice with granulocyte colony-stimulating factor did not affect the differentiation of human STRC. These findings demonstrate that the differentiation capacity of human STRCs is developmentally regulated by mechanisms inaccessible to currently available hematopoietic growth factors, and explain why thrombopoiesis is deficient in clinical CB transplantation.

Short-term repopulating cells with myeloid potential in human mobilized peripheral blood do not have a side population (SP) phenotype.

Clinical use of purified hematopoietic stem cells in myeloablated patients requires cotransplantation of short-term repopulating cells (STRCs) to ensure timely count recovery. Here, we investigated the flow fluorescence-based side population (SP) phenotype of mobilized human peripheral blood (mPB) cells that rapidly repopulate the highly permissive nonobese diabetic/severe combined immunodeficient (NOD/SCID)-beta2 microglobulin(-)/- mouse. No SP cells from this source regenerated detectable progeny in these mice before 8 weeks, although by 12 weeks human B-lymphoid cells were seen in some recipients of SP mPB cells. All myeloid reconstituting activity, including that seen within 3 weeks after transplantation, was associated with the non-SP fraction. Isolation of SP cells depletes human mPB of the rapid myeloid reconstitution capacity provided by myeloid-restricted STRCs which are vital for early hematologic recovery in clinical transplant recipients.

Efficient marking of human cells with rapid but transient repopulating activity in autografted recipients.

Short-term hematopoietic reconstituting cells have been identified in mice, nonhuman primates, and among human cells that engraft xenogeneic hosts. We now present clonal marking data demonstrating a rapid but unsustained contribution of cultured human autografts to the initial phase of hematologic recovery in myeloablated patients. Three patients received transplants of granulocyte colony-stimulating factor-mobilized autologous peripheral blood (PB) cells, of which a portion (8%-25% of the CD34+ cells) had been incubated in vitro with growth factors (5 days) and clinical grade LN retrovirus (3-5 days). More than 9% of the clonogenic and long-term culture-initiating cells harvested were transduced. Semiquantitative and linear amplification-mediated polymerase chain reaction analyses of serial PB samples showed that marked white blood cells appeared in all 3 patients within 11 days and transiently constituted up to 0.1% to 1% of those produced in the first month. However, within another 2 to 9 months, marked cells had permanently decreased to very low levels. Analysis of more than 50 vector insertion sites showed none of the clones detected in the first month were active later. Eighty percent of inserts were located within or near genes, 2 near CXCR4. These findings provide direct evidence of cells with rapid but transient repopulating activity in patients and demonstrate their efficient transduction in vitro.

Evidence of similar effects of short-term culture on the initial repopulating activity of mobilized peripheral blood transplants assessed in NOD/SCID-beta2microglobulin(null) mice and in autografted patients.

OBJECTIVE: Human mobilized peripheral blood (mPB) is known to contain high numbers of cells with rapid but short-term repopulating activity in NOD/SCID-beta2microglobulin(-/-) mice. Here we assessed the effect of short-term culture on these cells and compared the levels of retained activity with the pace of hematologic recovery in myeloablated patients transplanted with similarly cultured autografts of the same cells. PATIENTS AND METHODS: In a phase 1 clinical study, mPB cells were collected from 6 advanced cancer patients. CD34(+) cells were then harvested, cultured for 3 days in the presence of early-acting growth factors, and transplanted, and posttransplant recovery of blood cell parameters monitored. Assays for primitive hematopoietic activity using both in vivo (in NOD/SCID-beta2microglobulin(-/-) mice) and in vitro (CFC and LTC-IC) endpoints were also performed on the cells pre- and posttransplant. RESULTS: All patients showed event-free, timely leukocyte recoveries but slightly delayed platelet recoveries in some cases. During the 3-day period of culture, the CFCs doubled but the LTC-IC activity decreased (twofold), as did the short-term repopulating activity in NOD/SCID-beta2microglobulin(-/-) mice. CONCLUSION: Patients can be transplanted with 3-day cultured autografts with minimal effects on hematologic recovery. This is associated with a variable but, on average, modest loss of short-term repopulating activity detectable in NOD/SCID-beta2microglobulin(-/-) mice.