KIT mutations are the most common secondary mutations in inv(16) AML patients (10–45%), and their presence may suggest poor prognosis. It is therefore important to verify that KIT mutations cooperate with CBFB-MYH11, the fusion gene generated by inv(16), for leukemogenesis and to investigate the underlying mechanism. Here, we transduced bone marrow (BM) cells from wild type (WT) and conditional Cbfb-MYH11 knock-in (Cbfb+/56m; Tg(Mx1-Cre)) mice with retroviral vectors carrying wildtype (WT) or mutant (activation mutations D816V or D816Y) KIT variants. In colony forming assays, KIT (WT or D816 mutants) transduction led to significantly fewer colonies (> 7 fold decrease) from WT BM cells, whereas the Cbfb+/56m; Tg(Mx1-Cre) BM cells were only mildly affected (1.6 fold decrease in colony numbers). Further analysis of transduced BM cells indicated that KIT transduction significantly (p<0.05) increased cell death in transduced Lin− BM cells (both Cbfb+/56m; Tg(Mx1-Cre) and WT), as compared to untransduced Lin− BM cells, which could explain the decreased total colony numbers. Analysis of WT Lin− BM cells transduced with WT KIT and D816 mutants showed similar massive cell death (87% (wt); 87% (D816Y); 94% (D816V)(N=4 for each). On the other hand, in transduced Cbfb+/56m; Tg(Mx1-Cre) Lin− BM cells, the cell death rates were 74.3% (wt), 55.2% (D816Y) and 84.5% (D816V)(N=4 for each). This difference in the level of cell death could explain the differential effects of KIT on colony formation from transduced Cbfb+/56m; Tg(Mx1-Cre) BM cells vs. WT BM cells. These results also suggest that BM cells expressing Cbfb-MYH11 are more resistant to the toxic effects of KIT than WT BM cells.
Moreover, more mixed-lineage (CFU-GEMM) and fewer erythroid (BFU-E) colonies were obtained from Cbfb+/56m; Tg(Mx1-Cre) BM cells transduced with D816V/Y KIT than those transduced with WT KIT, suggesting differentiation defects in early myeloid and erythroid progenitors were induced by the mutant KIT.
We then transplanted the transduced BM cells and found that 60% and 80% of mice transplanted with Cbfb+/56m; Tg(Mx1-Cre) BM cells expressing D816V or D816Y KIT, respectively, died from leukemia within 9 months, while none of the control mice did. Results from limiting dilution transplantations using multiple donor leukemia cells (N=3) showed that mice transplanted with as little as 10 cells died from leukemia within two month, while mice transplanted with 106Cbfb+/56m; Tg(Mx1-Cre) leukemia cells died around two months. We also found a significant increase of mitotic cells in Cbfb+/56m; Tg(Mx1-Cre) leukemic spleen cells that carried the KIT mutations. These data indicate that the KIT D816 mutations not only facilitate the transformation of Cbfb+/56m; Tg(Mx1-Cre) BM cells to leukemia cells, but also help maintain these leukemia cells with higher leukemia initiating cells and proliferation.
We next explored the response of this aggressive leukemia to a novel small molecule, the kinase inhibitor PKC412, and found that Cbfb+/56m; Tg(Mx1-Cre) leukemia cells carrying KIT D816 mutations were sensitive to this kinase inhibitor, with significantly less survival than leukemia cells without mutant KIT and cells treated by vehicle only after overnight treatment. Signaling pathway analysis of these leukemia cells suggested that Stat3 and P44/42 MAPK signaling, which has been reported to be activated in cancer cells and is involved in cell proliferation, might be imported for these leukemia cells. Our data provide clear evidence for cooperation between mutated KIT and CBFB-MYH11 during leukemogenesis and show that acute myeloid leukemia cells carrying the inv(16) fusion gene and an activating KIT mutation respond to the small molecule PKC412.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.