Heat Shock Protein 90 (Hsp90) Inhibition Results in Apoptotic Killing of Primary AML Cells with FLT3/ITD Mutation.

Internal tandem duplication (ITD) of the FMS-like receptor tyrosine kinase 3 (FLT3) is the most frequently reported molecular defect in Acute Myeloid Leukaemia (AML), occurring in 25–30% of cases. ITD mutations of FLT3 constitutively phosphorylate the receptor and are associated with a poor prognosis. It has been postulated that FLT3 mutations play a critical role in the pathogenesis of AML. FLT3/ITD is a known client of the molecular chaperone heat shock protein 90 (Hsp90). Using 17-allylamino-17-demethoxygeldanamycin (17-AAG), an inhibitor of Hsp90, we studied the in-vitro effects of Hsp90 inhibition in FLT3/ITD positive AML. A murine myeloid IL-3 dependent cell line (32D) was transfected with FLT3/ITD and FLT3/WT (wild type). The FLT3/ITD mutation resulted in constitutive autophosphorylation of the receptor and IL-3 independence of the cell line. 17-AAG specifically induced cell death by apoptosis in the FLT3/ITD cells in a dose dependent fashion (IC₅₀ 300nM). The effect of 17-AAG was most marked in the FLT3/ITD cells in the absence of IL-3. The introduction of IL-3 rescued the cells from inhibition suggesting that the IL-3 intracellular signalling pathway remained intact after treatment. 17-AAG synergised with cytosine arabinoside (Ara C) in the FLT3/ITD cells (combination index 0.65). Immunoprecipitation of Hsp90 in primary AML cells demonstrated co-precipitation of mutant phosphorylated FLT3 in cells that were FLT3/ITD positive. FLT3 does not, however, appear to be chaperoned by Hsp90 in blasts from patients with unmutated FLT3. Consequently we studied diagnostic samples from 20 AML patients in which cells were treated with serial concentrations of 17-AAG. Cell killing was significantly increased (P= 0.016) in samples that were FLT3/ITD positive (mean IC₅₀ 0.62± 0.50 μM, n=10) compared with FLT3/ITD negative (mean IC₅₀ 1.22± 0.47 μM, n=10). Exposure of FLT3/ITD positive AML blasts to 17-AAG resulted in rapid dephosphorylation of phosphorylated FLT3 in the majority of samples. This was not seen in FLT3/ITD negative samples. The dephosphorylation of mutant FLT3 was associated with an upregulation of the stress responsive chaperones Hsp70 and BiP (GRP78) suggesting that this response is via the direct inhibition of Hsp90. Both the STAT5 pathway and the phosphatidylinositol 3-kinase/Akt pathway have been shown to be important in signalling downstream of mutant FLT3. We have demonstrated dephosphorylation of STAT5 in tandem with dephosphorylation of FLT3/ITD following treatment of cells with 17-AAG. In addition, Hsp90 inhibition also results in downregulation of Akt protein kinase. This suggests that the apoptosis seen following exposure to 17-AAG is a direct result of disrupted signalling via inhibition of mutant FLT3. Our results provide evidence that inhibition of the chaperone protein Hsp90 may have a role in the therapy of FLT3/ITD positive AML. We have also shown Ara C to synergise with Hsp90 inhibition and would suggest that this may be a feasible clinical approach for this difficult disease.