The presence of an internal tandem duplication (ITD) in the FLT3 gene, unlike mutations in the tyrosine kinase domain (TKD), predicts an adverse outcome in young adults with AML (

Mead et al,
Blood
,
2007
;
110
:
1262
–70
). The reason for this difference is not understood but may relate to distinct functional consequences of the different mutations. In order to explore this, the impact of exogenous FLT3 expression was examined in murine and human cell lines and primary human CD34+ cells using lentiviral IRES-GFP constructs containing either GFP only, wild-type (WT) FLT3, FLT3/TKD (D835Y) or FLT3/ITD (60bp insertion). Transduced cell lines (Ba/F3, 32Dcl3 or NB4) were sorted according to equal GFP expression by FACS to obtain an equivalent level of FLT3 expression, and this was confirmed by immunoblotting. In the absence of cytokines, Ba/F3-vector cells died within 48 hours. Ba/F3-WT cells retained viability but proliferated very slowly (doubling time [DT] 35.3 hours). Mutant-transduced cells proliferated in the absence of cytokines, but the DT was longer in Ba/F3-TKD cells than Ba/F3-ITD cells (18.7 hours versus 14.4 hours, P.05). Addition of FLT3-ligand (FL) increased the proliferation of Ba/F3-WT cells [DT 22.6 hours] but only had a marginal effect on mutant-transduced cells. Similar data were obtained with 32Dcl3 cells. To determine whether downstream intracellular signalling events reflected these differences, Ba/F3 cells were examined by immunoblotting. FL-stimulated Ba/F3-WT and cytokine-free Ba/F3-TKD and Ba/F3-ITD cells all demonstrated increased phosphorylation of MAPK and the PI3-kinase pathway compared to Ba/F3-vector cells. These increases were inhibited by the addition of the FLT3 inhibitor lestaurtinib, indicating their dependence on FLT3-signaling. Ba/F3-ITD cells showed strong phosphorylation of STAT5 to a similar level to mIL3-stimulated Ba/F3-vector cells, but only weak STAT5 phosphorylation was observed in Ba/F3-TKD and FL-stimulated Ba/F3-WT cells. These results were confirmed by immunostaining and flow cytometry of fixed and permeabilised cells and suggest that the level of STAT5 phosphorylation may, in part, account for the increased proliferation of FLT3/ITD-transduced cells. The impact of FLT3 mutants on hematopoietic cell differentiation was examined in the promyelocytic cell line NB4 transduced with the different FLT3 constructs and cultured with 0.037, 0.111, 0.333 or 1μM ATRA for 4 days. NB4-vector cells showed an ATRA concentration-dependent increase in CD11b expression, and a similar effect was observed in NB4-WT cells. At intermediate ATRA concentrations CD11b expression was reduced in NB4-ITD cells relative to NB4-vector cells (median fluorescence intensity [MFI] 64±5% of NB4-vector cells at 0.111 μM ATRA) consistent with impaired differentiation. In contrast, NB4-TKD cells showed increased ATRA-induced CD11b expression (MFI 170±24% of NB4-vector cells at 0.111 μM ATRA). These effects were ablated by lestaurtinib, indicating that the differences in differentiation in FLT3/TKD or ITD-transduced cells were due to signaling through the mutant receptor. To determine whether the effects observed in cell lines could be reproduced in primary hematopoietic cells, purified CD34+ cells were transduced with the different lentiviral constructs and cultured in cytokine-free liquid medium for 2 weeks. Transduction with either FLT3/TKD or ITD did not lead to factor-independent proliferation. Surviving cells were then plated in methylcellulose containing SCF, GM-CSF and G-CSF for a further 2 weeks. Of the initial 2×105 CD34+ cells, the estimated number of colony forming cells remaining after 2 weeks in cytokine-free liquid culture was 74±24 for vector-transduced cells. This was not different for FLT3/WT- (114±35, P.4) or TKD-transduced cells (124±46, P.4) but was markedly increased for FLT3/ITD-transduced cells (642±222, P.03). These data provide evidence that different FLT3 mutations have distinct consequences on cell survival, proliferation, differentiation and downstream signaling pathways. Given the clinical differences between the two types of mutation, these observations are relevant to the understanding of chemoresistance in AML.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author

Sign in via your Institution