Genomic Instability Originates From Leukemia Stem Cells in a Mouse Model of CML-CP

Abstract 445

For decades, chronic myeloid leukemia (CML) has served not only as a paradigm for understanding the evolution and multi-step process of carcinogenesis but also for studying cancer stem and progenitor cells responsible for the initiation and/or maintenance of the disease. CML is initiated by BCR-ABL1 tyrosine kinase transformation of hematopoietic stem cells into leukemia stem cells (LSCs) to induce CML-chronic phase (CML-CP). The deregulated growth of LSC-derived leukemia progenitor cells (LPCs) leads to manifestation of the disease. It is unclear if LSCs and/or LPCs are able to acquire additional genetic changes that confer resistance to tyrosine kinase inhibitors (TKIs) and induce more aggressive CML blast phase (CML-BP). In addition, the mechanisms and consequences of genomic instability may differ substantially among these cells. For example, the effects of genetic aberrations acquired in quiescent LSCs may be dormant, but if the aberrations induce proliferation or appear in LSCs that are already cycling, they may generate TKI-resistant and/or more malignant clones. Alternatively, genomic instability in LPCs must be accompanied by the acquisition of LSC-like properties to prevent mutations from disappearing before they undergo terminal maturation.

Previously, we reported that BCR-ABL1–transformed cell lines accumulate reactive oxygen species (ROS)-induced oxidative DNA damage [8-oxoguanine (8oxoG), double strand breaks (DSBs)] resulting in genomic instability in vitro, which was responsible for acquired imatinib-resistance and accumulation of chromosomal aberrations (Nowicki et al., Blood, 2005; Koptyra et al., Blood, 2006; Koptyra et al., Leukemia, 2008). To determine which populations of CML-CP cells, LSCs and/or LPCs, accumulate genomic instability we employed the SCLtTA/BCR-ABL1 tetracycline-inducible (tet-off) transgenic mouse model of CML-CP with targeted expression of p210BCR-ABL1 in hematopoietic stem and progenitor cells (Koschmieder et al., Blood, 2005). Mice exhibiting CML-CP-like disease resulting from BCR-ABL1 induction demonstrated splenomegaly and Gr1⁺/CD11b⁺ myeloid expansion in bone marrow, spleen and peripheral blood. BCR-ABL1 mRNA expression was higher in the Lin⁻c-Kit⁺Sca1⁺ murine leukemia stem cell–enriched population (muLSCs) than in the Lin⁻c-Kit⁺Sca1⁻ murine leukemia progenitor cell–enriched population (muLPCs), thus reminiscent of human CML-CP (Lin⁻CD34⁺CD38⁻ LSCs > Lin⁻CD34⁺CD38⁺ LPCs). BCR-ABL1 induction increased levels of ROS (hydrogen peroxide, hydroxyl radical) and oxidative DNA damage (8-oxoG, DSBs) in muLSCs, but not in muLPCs. In addition, CFSE^max/eFluor670^max quiescent muLSCs displayed more ROS (superoxide, hydrogen peroxide) and oxidative DNA damage (8oxoG, DSBs) than non-induced counterparts. Currently, we are examining genomic instability in the most primitive long-term muLSCs (Lin⁻c-Kit⁺Sca1⁺CD34⁻Flt3⁻). Lastly, single nucleotide polymorphism (SNP) arrays detected a variety of genetic aberrations (addition, deletions) in BCR-ABL1–induced Lin⁻ BM cells. Individual mice displayed a great degree of diversity in the intensity of genetic instability accumulating between 31 to 826 aberrations, which recapitulate heterogeneity of sporadic aberrations detected in CML-CP patients. These aberrations include deletions in Trp53 and Ikzf1, and additions in Zfp423 and Idh1 genes, which have been linked to progression from CML-CP to CML-BP.

In summary, by using the SCLtTA/BCR-ABL1 inducible transgenic mouse model of CML-CP we showed that muLSCs, but not muLPCs, displayed elevated levels of ROS-induced oxidative DNA damage likely resulting in the accumulation of extensive genetic aberrations. This observation supports the hypothesis that genomic instability in CML-CP originates in LSCs. Current analysis of microarrays may shed some light on the mechanisms leading to enhanced ROS production and accumulation of oxidative DNA damage in muLSCs.