Abstract
The existence of leukemia stem cells has been demonstrated in acute myeloid and lymphoblastic leukemias (AML and ALL). The origins of these cells are unknown, but it has been suggested that they result from the transformation of adult hematopoietic stem cells (HSC). To challenge this hypothesis we tested the ability of representative leukemia oncogenes to transform committed myeloid progenitor cells that lack the capacity for self-renewal. Flow-sorted populations of common myeloid progenitors (CMP), and granulocyte-monocyte progenitors (GMP) were transduced with the fusion oncogenes MOZ-TIF2 and BCR-ABL, respectively and their self-renewal and leukemogenic potential were tested in in vitro and in vivo assays. Utilizing the same experimental design we were also able to address the poorly understood question of the contribution of the cell of transformation to the eventual leukaemia phenotype.
In contrast to CMP or GMP transduced with BCR-ABL or non-transduced control cells, CMP or GMP that were retrovirally transduced with MOZ-TIF2 could be serially replated in methylcellulose cultures, and continuously propagated in liquid culture media containing IL-3. In further contrast, transplantation of CMP or GMP transduced with MOZ-TIF2 into recipient mice also resulted in an acute myeloid leukemia (AML). This leukaemia could be transplanted to secondary recipients, documenting the long-term self-renewal properties of the leukemic stem cells, yet in limiting dilution experiments did not cause disease below a transplanted cell dose of 1 x104 cells, suggesting that the probability of transferring leukemia to secondary recipients relates to the frequency of self-renewing leukemic stem cells within the total leukemic population. This in turn suggests that our retroviral bone marrow transduction and transplantation models have the same hierarchical organization of self-renewal as has been shown for human AML. The phenotype of the leukemias were virtually indistinguishable regardless of whether the initially transduced cell population was CMP, GMP or the control populations of whole bone marrow mononuclear cells or HSC, suggesting that MOZ-TIF2 may also have a dominant effect upon the eventual leukaemia phenotype.
These observations indicate that MOZ-TIF2, but not BCR-ABL, can confer properties of leukemic stem cells to committed myeloid progenitors. Control experiments conducted with with MOZ-TIF2 point mutants that do not cause leukemia in the murine BMT system and with BCR-ABL, a fully active leukemogenic tyrosine kinase, were insufficient to cause in-vitro changes in self-renewal or leukaemia. Together, these data argue strongly that retroviral insertional mutagenesis alone cannot explain these results. However, we cannot exclude the possibility that an active MOZ-TIF2, but not BCR-ABL, can collaborate with mutations induced by retroviral mutagenesis to confer properties of leukemic stem cells to committed progenitors. These findings have important implications regarding the origin of leukemic stem cells, and provide tools for understanding the transcriptional programs that confer properties of self-renewal in malignant and non-malignant cells.
Cell-extrinsic effects in T cell acute lymphoblastic pre-leukemia stem cells mediated by EphA3
