G-CSF upregulates the expression of aquaporin-9 through CEBPB to enhance the cytotoxic activity of arsenic trioxide to acute myeloid leukemia cells

BACKGROUND: Arsenic trioxide (ATO) is highly effective in acute promyelocytic leukemia (APL) patients, but it fails to show satisfactory efficacy in other acute myeloid leukemia (AML) patients with non-APL subtypes. Different from the APL cells, most non-APL AML cells express low levels of the ATO transporter Aquaporin-9 (AQP9) protein, making them less sensitive to ATO treatment. Recently, we found that granulocyte colony stimulating factor (G-CSF) can upregulate the expression of AQP9. We hypothesized that the pretreatment with G-CSF may enhance the antitumor effect of ATO in non-APL AML cells. In addition, we aimed to elucidate the underlying mechanisms by which G-CSF upregulates the expression of AQP9. METHODS: Non-APL AML cell lines including THP-1 and HL-60 were pretreated with or without G-CSF (100 ng/ml) for 24 h, followed by the treatment with ATO (2 μM) for 48 h. Cell morphology was observed under the microscope after Wright-Giemsa staining. Flow cytometry was performed to evaluate the cell apoptosis levels. The intracellular concentrations of ATO were determined by atomic fluorescence spectrometry. The mRNA and protein expression were respectively measured by quantitative reverse transcription PCR (RT-qPCR) and western blotting. Target genes were knocked down by transfection with small interfering RNA (siRNA), or overexpressed by transfection with overexpression plasmids. The cell line derived xenograft mouse model was established to confirm the results of the in vitro experiments. RESULTS: Compared with using ATO alone, the combination of G-CSF with ATO induced the cell apoptosis more dramatically. G-CSF upregulated the expression of AQP9 and enhanced the intracellular concentrations of ATO in AML cells. When AQP9 was overexpressed, it markedly enhanced the cytotoxic activity of ATO. On the other hand, when AQP9 was knocked down, it profoundly attenuated the combinational effect. Moreover, we found that the upregulation of AQP9 by G-CSF depends on the transcription factor CCAAT enhancer binding protein beta (CEBPB). We also demonstrated that the combination of G-CSF and ATO significantly inhibited tumor growth in the xenograft mouse model. CONCLUSIONS: The combination of G-CSF and ATO may be a potential therapeutic strategy for AML patients.