Development of a Multi-Step Leukemogenesis Model of MLL-Rearranged Leukemia Using Humanized Mice

Abstract 1311

Mixed-lineage-leukemia (MLL) fusion oncogenes are involved in de novo and secondary therapy-related acute leukemia in humans. To understand the pathogenesis of MLL-rearranged leukemia, several murine models for this disease have been established. However, the leukemia derived from murine hematopoietic stem cells (HSCs) may not be fully comparable with human leukemia. In vivo studies, in which primary human leukemia cells were transplanted into immunodeficient mice, provided us significant advances to understand the pathogenesis of human leukemia. However, these models are not suitable for studying physiological leukemogenesis because the initiation and progression processes are hard to be evaluated. Therefore, a new in vivo experimental model for analyzing leukemogenesis derived from primary human HSCs and by genetic hits is required.

Here we developed a humanized murine model for human leukemia by transplanting human cord blood-derived CD34-positive HSCs transduced with an MLL-AF10 oncogene and second hits into a supra-immunodeficient mouse strain, NOD/Shi-scid, IL-2Rγ^−/− (NOG) mice.

Injection of the MLL-AF10-transduced HSCs into the liver of NOG mice enhanced multilineage hematopoiesis, however, MLL-AF10 alone could not induce leukemic transformation. Accumulating evidence points to a multistep pathogenesis for leukemia development and progression. In the multi-genetic step models of leukemogenesis, particularly in the two-hit model of leukemia, the initial genetic hit often leads to abnormal cell differentiation (Type II mutation), while subsequent mutations may activate specific signaling pathways that are involved in cell growth, such as the Ras/MAP kinase pathway (Type I mutation). Because active mutations in Ras genes are often found in MLL-related leukemia, we next transduced constitutively active form of K-Ras mutant along with the MLL-AF10 oncogene. Enforced expression of the mutant K-Ras alone was insufficient to induce leukemia. Eight weeks after transplantation, all of the recipient mice transduced both MLL-AF10 and mutant K-Ras developed acute monoblastic leukemia (the M5 phenotype in French-American-British classification). We could also observe organ infiltration in liver and spleen in vivo that mimicked clinical symptoms observed in human leukemia.

We thus successfully established a murine model for human MLL-rearranged leukemia derived from human HSCs. The present model is a useful experimental platform for the multi-step leukemogenesis of human leukemia and for new therapeutic approaches that should be evaluated in vivo.