Divergent Effects of Notch Signaling On Leukemia Stem Cells and Hematopoietic Stem Cells.

Abstract 395

The Leukemia Stem Cell (LSC) hypothesis proposes that a subset of the cells in the bulk tumor cell population is responsible for leukemia maintenance and propagation. We tested the LSC hypothesis in murine T-cell acute lymphoblastic leukemia/lymphoma (T-ALL) induced by retroviruses expressing strong gain-of-function Notch1 alleles. This model is relevant to human disease, as Notch1 is activated in ∼70% of human T-ALLs. By transferring sorted leukemia cells to syngeneic secondary recipients at limiting dilution, we found that LSC activity was enriched in the CD8⁺CD4^-HSA^hi (Immature Single Positive or “ISP”) T-cell subset. The LSC frequency was ∼1 in 1000 cells, two orders of magnitude higher than in the CD8⁺CD4⁺ (Double positive or “DP”) T-cell subset. We found similar results in a Kras^G12D-driven T-ALL mouse model where activating Notch mutations occur spontaneously in ∼80% cases. Surprisingly, we were unable to isolate Notch-activated hematopoietic stem and progenitor cells (HSPCs) to test for LSC activity. Upon further analysis, we observed that activation of Notch in HSPCs caused loss of stem cell quiescence and progressive, complete loss of long-term hematopoietic stem cells (LT-HSCs) over several weeks. Notch signals had no significant effects on stem cell homing, apoptosis, or senescence. Similar results were obtained in both noncompetitive and competitive secondary transplants as well as in a mouse model in which Notch activation is conditionally induced in HSCs by Cre recombinase. We conclude that while Notch signaling promotes LSC activity, it extinguishes HSCs. These results provide a rationale for therapeutic targeting of the ISP-like T-cell subset in Notch-activated T-ALL while underscoring the potential difficulty of manipulating Notch signaling to expand normal, long-term stem cell populations for clinical applications.