Distinct Human Cell Populations Produce Rapidly Detectable Levels of Neutrophils and Platelets in NOD/SCID-IL-2Receptor Gamma Chain Null Mice Transgenically Engineered to Produce Human Myeloid Growth Factors

Abstract 1892

Cord blood (CB) is becoming an increasingly utilized source of cells for cancer patients who are eligible for therapies that require a transplant to rescue them from toxic side effects on their own hematopoietic cells but lack a suitable HLA-matched donor. This strategy is now routinely used in children, but delayed neutrophil and platelet recovery remain unsolved problems and these problems are exacerbated in adults. To address this issue, we first surveyed the variability in 8 individual CB harvests of parameters routinely used to predict the utility of CB units as transplants (i.e., CD34+ and in vitro myeloid clonogenic progenitor cell frequencies). In addition, we compared their 3-week outputs of CD33/15/66+ cells (neutrophils and monocytes) in the marrow and CD41a+ platelets in the blood of sublethally irradiated NSG mice after the IV transplantation of ∼10⁴ CD34+ cells. These latter assessments were based on ongoing experiments in our lab demonstrating that, at this transplant dose, the outputs measured are linearly related to the number of CD34+ cells injected and detect transplantable progenitor cell types that are biologically distinct from cells with longer term repopulating activity. The results showed variation between CBs in all parameters, a marked lack of correlation between %CD34+ cells or % total CFCs in initial cells and %CD41a+ cells regenerated at 3 weeks/10⁴ CD34+ cells transplanted (R=-0.28 and 0.35, respectively), and a weak correlation between the %CD33/15/66+ cells regenerated at 3 weeks/10⁴ CD34+ cells transplanted and %CD34+ cells or % total CFCs in the initial CB cells (R values of 0.46–0.64). However, although engraftment of primitive human cells in NSG mice appears highly efficient, terminal differentiation of the myeloid lineages in these mice is poor. One possible explanation for this deficiency in mature cell output is that several of the murine growth factors responsible for regulating the production and release of these cells into the circulation in mice are not cross-reactive on human cells. We therefore hypothesized that engineering NSG mice to produce the human counterparts might significantly improve the detection of short term repopulating human cells whose maximum clone size might be limiting in NSG mice. Three potential relevant factors are IL-3, GM-CSF and Steel factor. We therefore backcrossed a line of transgenic NS mice we had created to express human IL-3, GM-CSF and Steel factor onto the NSG strain to produce homozygous NSG mice expressing all 3 of these human factors (NSG-3GS mice). We then compared these NSG-3GS mice with NSG mice in terms of their ability to stimulate the production within 3 weeks of human neutrophil-monocytes and platelets from intravenously transplanted CD34+ cells isolated from pooled CB harvests. The results showed that the levels of neutrophils and monocytes generated in the marrow of the NSG-3GS mice were elevated to levels of >50% of the marrow in 90% of the mice, even at the lowest number of CD34+ cells transplanted. Human neutrophils and monocytes were also elevated in the blood of the NSG-3GS mice where, despite the observed “saturation” of the marrow, there was a linear dose-response in the number of human neutrophils and monocytes present in the blood with increasing CD34+ cells infused. These findings are consistent with the reported activities of these molecules in vitro and in patients suggesting their physiological relevance in this murine xenograft model. We next utilized this assay to characterize the cells responsible for the neutrophil/monocyte and platelet repopulating activities detected in NSG-3GS mice. Preliminary assessment of the CD34+CD45RA- population on the basis of CD123 (IL-3 receptor alpha chain) expression indicates that the CD123+ fraction is enriched for short term (3-week) neutrophil/monocyte repopulating activity, while the CD123- fraction is enriched for short term (3-week) platelet repopulating activity. In summary, NSG-3GS mice significantly enhance the output of human cells with short term human myeloid repopulating ability thereby enabling neutrophil/monocyte outputs as well as platelet outputs to be assessed by analysis of peripheral blood samples. We have also used this tool to obtain evidence that these two outputs are derived from distinct cell types. Direct quantification of these may add to future predictions of graft quality.