Generation of a FLT3 Kinase Domain Point Mutation (D835Y equivalent) Knock-in Mouse Confers Less Aggressive Disease Compared with FLT3 Internal Tandem Duplication (ITD) Mice

Abstract 924

FLT3 is a tyrosine kinase receptor that is frequently mutated in acute myeloid leukemia (AML). The most common mutations involve either an internal tandem duplication (ITD) in the juxtamembrane domain or a point mutation in the tyrosine kinase domain (TKD). The most common TKD mutation is a D835Y mutation in which aspartic acid is mutated to tyrosine at residue 835. ITD mutations are reported in approximately 25% of adult AML and 15% of pediatric leukemia. TKD mutations occur in approximately 8–10% of both adult and pediatric leukemia. Both of these mutations constitutively activate FLT3, but only patients with an ITD mutation have a significantly poorer prognosis. Small molecule inhibitors against FLT3 are currently in clinical trials with some success, but resistance to single agent treatment frequently occurs. A better understanding of the molecular mechanisms behind the prognostic difference, between the two types of mutations and the activation, inhibition, and resistance of FLT3 will hopefully lead to more tailored treatments of leukemia.

To help address some of these questions we have generated a knock-in mouse with an Aspartic acid to Tyrosine mutation at reside 838 (D838Y) which is homologous to the FLT3/D835Y mutation reported in the majority of human leukemia cases with a FLT3-TKD mutation. These mice have been compared to the mice with a FLT3-ITD mutation previously generated and characterized by the lab. Mice with either the FLT3/D838Y or FLT3-ITD mutation develop predominantly a myeloproliferative disease. Mice with the FLT3/D838Y mutation show a less aggressive phenotype when compared to mice with the FLT3/ITD mutation in terms of survival, spleen size, and histopathologic changes of the bone marrow, spleen and liver.

Flow cytometry analysis shows that bone marrow from 8–12 week old mice with the FLT3/D838Y mutation have a marginally expanded fraction of granulocytic/monocytic lineage cells in their bone marrow compared to their wildtype littermate controls, but a lower fraction when compared to age-matched FLT3/ITD mice.

B cell development is significantly blocked at the late pro-B to pre-B transition in the FLT3/ITD mice, with very few cells developing into pre-B or more mature B cells (Li et al. Blood. 117(11): 3131-9). In contrast, FLT3/D838Y mice demonstrate an expansion of early and late pro-B cell populations, with a significant fraction of cells maturing beyond the pre-B stage to become mature B cells.

In mice with the FLT3/D835Y mutation the long-term HSC (LT-HSC) compartment, responsible for the self-renewal and maintenance of the hematopoietic stem/progenitor cell pool, appears to have a greater than 2-fold expansion in the number of immunophenotypic LT-HSCs. This appears to be a functional expansion as the BM cells from FLT3/D835Y mice have long-term engraftment superiority compared with ITD and wildtype mice. The short term HSC and MPP compartments from both FLT3/D838Y and FLT3/ITD mice appear to be comparably expanded compared to wildtype controls.

In human leukemia, FLT3/ITD signaling results in the activation of downstream targets, including STAT5, MAP kinase and AKT. Consistent with the intermediate phenotype observed, intracellular levels of phospho-Stat5 in the FLT3-D838Y mice were lower than levels measured in the FLT3-ITD mice, but higher than those of wildtype littermate controls. RT-PCR measured expression levels of FLT3 and Pim1 in lineage negative cells of D838Y mice fall between those of the wildtype and ITD mice. (Ratios: FLT3 – ITD/wt 3.7; D838Y/wt 2.5; Pim1 – ITD/wt 3.4; D838Y/wt 2.5).

In summary, we have generated a knock-in mouse model with the FLT3 kinase domain D838Y mutation. The mice demonstrate a less aggressive phenotype when compared to knock-in mice with the FLT3/ITD mutation, analogous to the less aggressive AML seen in cases involving D835 mutations compared to ITD mutations. Both mouse models on the same mouse background provide a useful platform for understanding the molecular mechanisms of FLT3 mutations and for testing new therapeutic agents targeting FLT3.