Co-expression of wild-type FLT3 attenuates the inhibitory effect of FLT3 inhibitor on FLT3 mutated leukemia cells

Abstract

FLT3 mutation is found in about 30% of acute myeloid leukemia (AML) patients and is associated with a poor prognosis. Several FLT3 inhibitors are undergoing investigation, while their clinical efficacies were lower than expected and several resistant mechanisms to FLT3 inhibitors have been demonstrated. Although most AML cells harboring FLT3 mutation co-express wild-type (Wt)-FLT3, it is not fully understood how Wt-FLT3 expression is associated with the resistance to FLT3 inhibitors. In this study, we elucidated a resistant mechanism by which FL-dependent Wt-FLT3 activation reduced inhibitory effects of FLT3 inhibitors. We demonstrated that FL-stimulation much more strongly reduced growth inhibitory effects of FLT3 inhibitors on Wt- and mutant-FLT3 co-expressing cells than sole mutant-FLT3 expressing cells both in vitro and in vivo. It was also confirmed that FL impaired the anti-leukemia effects of FLT3 inhibitors on primary AML cells. We elucidated that FL impeded the inhibitory effects of FLT3 inhibitors mainly through the activation of Wt-FLT3, but not mutated FLT3, in the Wt- and ITD-FLT3 co-expressing cells. Furthermore, FL-induced activation of Wt-FLT3-MAPK axis was the dominant pathway for the resistance, and the glycosylation of Wt-FLT3 was also vital for FL-dependent kinase activation and following resistance to FLT3 inhibitors. Thus, we clarified the importance of co-expressing Wt-FLT3 in resistance to FLT3 inhibitors. These findings provide us with important implications for clinical application and new strategies to improve clinical outcomes of FLT3 inhibitors.

Keywords: AML; FLT3 inhibitor; FLT3 ligand; Wt-FLT3; resistance.

Figure 1. FL impairs effects of FLT3 inhibitors on Wt-FLT3 co-expressing cells

A. Mutant-FLT3 32D cells were treated with increasing concentrations of the indicated FLT3 inhibitors in the presence or absence of FL at 10 or 50ng/ml for 48 hours. Fold changes to GI₅₀ values without FL are shown. B. MV4;11 cells and MOLM14 cells were treated with increasing concentrations of quizartinib or lestaurtinib with or without 50ng/ml FL for 72 hours. Fold changes to GI₅₀ values without FL are shown. C. Sole ITD-FLT3 expressing 32D cells and Wt- and ITD-FLT3 co-expressing cells were treated with increasing concentrations of quizartinib with or without FL at 50ng/ml for 24 hours and then processed to measure the binding of Annexin V and PI by flow cytometry to assess induction of apoptosis. D. 32D cells were co-cultured with COS-7 or FL-COS-7 cells overnight, followed by exposure to varying concentrations of quizartinib for 48 hours to assess apoptosis induction. E. MTT assay was performed at 24 hours after exposure to varying concentrations of quizartinib in 32D and COS-7 or FL-COS-7 co-culture condition. Error bars represent the mean values ± S.D. from at least three independent experiments (*P < 0.05, **P < 0.01, ***P < 0.001).

Figure 2. FL attenuate anti-leukemia effect of FLT3 inhibitors in vivo

A. 1×10⁷ FLT3-ITD expressing cells or Wt- and ITD-FLT3 co-expressing cells were inoculated subcutaneously into a shaved area of six NOD-SCID mice for each cell line. At seven days after inoculations, six mice from each cell line were divided into vehicle and quizartinib treatment groups. Quizartinib (1mg/kg q.d.) was orally administrated for 14 days. The graph shows the average volume of tumors corresponding to each cell line. B. The relative ratio of quizartinib-treated tumor volume to control tumor volume on day 11 is represented. Inhibitory effects of FLT3 inhibitors on co-expressing cells were much more reduced by the presence of FL in plasma than sole ITD-FLT3 expressing cells. C. and D. FL also impaired in vivo effects of quizartinib on Wt- and cyITD-FLT3 co-expressing cells. Error bars represent the mean values ± S.D. from at least three independent experiments (*P < 0.05, **P < 0.01, ***P < 0.001).

Figure 3. FL-dependent Wt-FLT3 signals reduced the inhibitory effect of FLT3 inhibitors

A. The established 32D cells were treated with quizartinib for 2h, followed by 10ng/ml FL stimulation for 10 minutes. Wt-FLT3 and mutant FLT3 were immunoprecipitated by each tag as indicated and subjected to Western blot. Phosphorylation levels of FLT3, STAT5, AKT and MAPK were examined. B. 32D cells were treated with quizartinib at the indicated concentrations in the presence or absence of 10ng/ml FL for 2 hours. Phosphorylation levels of FLT3, STAT5, AKT and MAPK were detected by Western blot. C. Established kinase dead-FLT3 and ITD-FLT3 co-expressing 32D cells were stimulated with FL and phosphorylation levels of FLT3, AKT and MAPK was determined by Western Blot. D. kinase dead FLT3 co-expressing ITD-FLT3 cells were treated by quizartinib with of without FL. Fold changes to GI₅₀ values without FL are shown. All results are representative of at least three independent experiments.

Figure 4. FL impaired the in vivo effects of FLT3 inhibitors and reduced the cytotoxic effects of FLT3 inhibitors in primary AML cells

A. Detection of FLT3-ITD mutation in primary AML cells by PCR. THP-1 was served as a control for Wt-FLT3.B. Primary AML blast samples were incubated in MethoCult methylcellulose semisolid medium with increasing concentrations of the indicated FLT3 inhibitors for 7 days in the absence or presence of 50ng/ml FL. The data on day 7 are normalized to untreated controls on day 0. Different primary AML blast samples harboring both Wt-FLT3 and ITD-FLT3 (upper panels) and primary AML cells with a homozygous FLT3-ITD mutation (lower panels) were incubated with quizartinib. C. Inhibitory effect on AML blast samples harboring both Wt-FLT3 and ITD-FLT3 were compared between quizartinib and lestaurtinib. D. FLT3-ITD positive two AML patient samples were subjected to Western blot. Primary AML cells were incubated with the indicated concentration of quizartinib with or without FL for 2 hours and then the phosphorylation status of FLT3, STAT5, and MAPK was detected.

Figure 5. The FL-Wt-FLT3-MAPK axis is essential in reduced inhibitory effects of FLT3 inhibitors

A. Wt- and ITD-FLT3 co-expressing 32D cells were treated with increasing concentrations of quizartinib and 40μmol/L (Wt- and ITD-FLT3 co-expressing cells) or 50μmol/L (Wt- and cyITD-FLT3 co-expressing cells) of U0126 with or without 50ng/ml FL for 48 hours, the GI₅₀ values of quizartinib were calculated. B. Co-expressing 32D cells were treated with quizartinib and U0126 with or without FL for 2 hours, and then the phosphorylation status of STAT5, AKT, MAPK and BAD was detected. C. Two primary AML samples harboring both Wt- and ITD-FLT3 were subjected to evaluate the inhibitory effects of combination therapy with quizartinib and U0126 in the absence or presence of 50ng/ml of FL. The data on day 7 were normalized to untreated controls on day 0. Error bars represent the mean ± S.D. from four independent experiments (***P < 0.001).

Figure 6. Cell surface localization of Wt- FLT3 is crucial to FL induced resistance to FLT3 inhibitors

A. Transduced 32D cells, Wt- and ITD-FLT3 (left) and Wt- and cyITD-FLT3 (right) co-expressing cells were treated with Tunicamycin or BFA at the indicated concentrations for 16 hours. Then, cells were subjected to flow cytometer analysis for surface FLT3 expression. The graph shows representative results from 3 independent experiments. B. and C. Both Wt- and ITD-FLT3 co-expressing 32D cells were incubated with increasing concentrations of Tunicamycin B. or BFA C. for 16 hours, followed by FL stimulation for 10 minutes. Phosphorylation status of FLT3, STAT5, AKT, and MAPK were examined by Western blot. (D and E) Co-expressing 32D cells were incubated with various concentrations of quizartinib and 2μg/ml of Tunicamycin D. or BFA E. with or without 50ng/ml FL for 48 hours. The GI₅₀ values of quizartinib were calculated. Error bars represent the mean values ± S.D. from at least three independent experiments (**P < 0.01, ***P < 0.001).