The PI3K/mTOR dual inhibitor BEZ235 suppresses proliferation and migration and reverses multidrug resistance in acute myeloid leukemia

Abstract

Aberrant activation of the PI3K/Akt/mTOR pathway contributes to the proliferation of malignant cells, and may confer resistance to chemotherapy in various malignancies, including acute myeloid leukemia (AML). Chemoresistance is the major reason for relapse in AML. RAD001 (everolimus) has been used at d1 and d7 of an induction chemotherapy regimen for AML, which has acceptable toxicity and may improve conventional chemotherapeutic treatment. Dual inhibitors of PI3K and mTOR overcome some of the intrinsic disadvantages of rapamycin and its derivatives. In this study, we evaluated the effects of BEZ235, a PI3K/mTOR dual inhibitor, on the multidrug-resistant AML cell lines HL-60/VCR and K562/ADR in vitro. BEZ235 dose-dependently inhibited the viability of HL-60/VCR and K562/ADR cells with the IC₅₀ values of 66.69 and 71.44 nmol/L, respectively. BEZ235 (25-100 nmol/L) dose-dependently inhibited the migration of the two AML cell lines, and it also significantly sensitized the two AML cell lines to VCR and ADR. After treatment with BEZ235, the miR-1-3p levels were markedly increased in HL-60/VCR cells. Using TargetScan analysis and luciferase assays, we showed that miR-1-3p targeted BAG4, EDN1 and ABCB1, the key regulators of cell apoptosis, migration and multidrug resistance, and significantly decreased their levels in the two AML cell lines. Transfection of HL-60/VCR and K562/ADR cells with miR-1-3p-AMO to inhibit miR-1-3p could reverse the anti-proliferation effects of BEZ235. In conclusion, the PI3K/mTOR dual inhibitor BEZ235 effectively chemosensitizes AML cells via increasing miR-1-3p and subsequently down-regulating BAG4, EDN1 and ABCB1.

Figure 1

BEZ235 regulates cell proliferation, migration, apoptosis, and chemosensitivity in AML cells. (A) After exposure to various concentrations (from 25 to 100 nmol/L) of BEZ235 for 24 h, the viability of HL-60/VCR and K562/ADR cells was determined using CCK8 assays, and the viability decreased in a dose-dependent manner (^**P<0.01 compared with the untreated control group). (B) HL-60/VCR and K562/ADR cells treated as above. We collected the cells in the bottom Boyden chamber and counted them using trypan blue exclusion assays. The migration cell count also decreased in a dose-dependent manner (^**P<0.01 compared with the untreated control group). (C) Under identical culture conditions, the apoptosis of HL-60/VCR and K562/ADR cells was detected by the FCM method. Opposite results were observed, as both increased in a dose-dependent manner. (D) BEZ235 elevated the sensitivity of HL-60/VCR and K562/ADR cells to VCR and ADR.

Figure 2

BEZ235 up-regulates miR-1-3p expression. (A) Unsupervised hierarchical clustering of miRNA expression profiles in HL-60/VCR cells treated with IC₅₀ of BEZ235 for 24 h. (B) According to the miR array results, we detected 7 types of up-regulated miRs. (C) HL-60/VCR and K562/ADR cells treated with the indicated concentrations of BEZ235 for 24 h. Total RNA was extracted, and the miR-1-3p expression level was measured by qRT-PCR. ^**P<0.01.

Figure 3A–3D

miR-1-3p negatively regulates BAG4, EDN1 and ABCB1. (A) Seed sequence of miR-1-3p combined with the 3′-UTRs of BAG4, EDN1 and ABCB1. (B) The reporter construct containing the predicted miR-1-3p binding site in the 3′-UTRs of the targets. Overexpression of miR-1-3p-MIMIC significantly attenuated luciferase activity of the WT-BAG4, EDN1 and ABCB1 3′-UTRs in 293T cells rather than the MUT-BAG4, EDN1 and ABCB1 3′-UTRs. miR-1-3p-MIMIC and miR-1-3p-AMO were transfected into the HL-60/VCR and K562/ADR cells for 24 h, and the mRNA (C) and protein (D) levels of BAG4, EDN1 and ABCB1 were measured by qRT-PCR and Western blot analyses, respectively. ^**P<0.01.

Figure 3E–3H

miR-1-3p negatively regulates BAG4, EDN1 and ABCB1. To evaluate the effect of miR-1-3p on AML cells, we transfected the miR-1-3p-MIMIC or NC into HL-60/VCR and K562/ADR cells for 24 h. (E) The viability of the miR-1-3p-MIMIC group in both HL-60/VCR and K562/ADR cells was significantly lower than that in the NC group (^**P<0.01). (F) The migration assays showed fewer cells in the bottom chamber in the miR-1-3p-MIMIC group than those in the NC group (^**P<0.01). (G) Apoptosis in the miR-1-3p-MIMIC group in both HL-60/VCR and K562/ADR cells was significantly higher than that in the NC group. (H) Chemosensitivity in miR-1-3p-MIMIC group in both HL-60/VCR and K562/ADR cells was increased.

Figure 4

miR-1-3p is involved in the effect of BEZ235 on AML cells. HL-60/VCR and K562/ADR cells were transfected with miR-1-3p-AMO or NC for 24 h and treated with IC₅₀ of BEZ235 for 24 h. (A) The cell viability was measured by CCK8 assays, and the viability of the miR-1-3p-AMO group in both HL-60/VCR and K562/ADR cells was significantly higher than that in the NC group (^**P<0.01). (B) Migration was detected using transwell assays, and the results were similar to those of the proliferation assays (^**P<0.01). (C) The inhibitory effects of BEZ235 on BAG4, EDN1, and ABCB1 mRNA expression was attenuated by miR-1-3p-AMO. (D) The effects of BEZ235 on BAG4, EDN1, and ABCB1 protein expression was attenuated by miR-1-3p-AMO. (E) Apoptosis assays were performed using Annexin V/PI staining, and the apoptotic ratio in the miR-1-3p-AMO group in both HL-60/VCR and K562/ADR cells was substantially lower than that in the NC group. (F) Chemosensitivity in the miR-1-3p-AMO group in both HL-60/VCR and K562/ADR cells was significantly lower than that in the NC group.