Activation of FOXO3a is sufficient to reverse mitogen-activated protein/extracellular signal-regulated kinase kinase inhibitor chemoresistance in human cancer

Abstract

Drug resistance is a central challenge of cancer therapy that ultimately leads to treatment failure. In this study, we characterized a mechanism of drug resistance that arises to AZD6244, an established mitogen-activated protein/extracellular signal-regulated kinase kinase (MEK) 1/2 inhibitor currently being evaluated in cancer clinical trials. AZD6244 enhanced the expression of transcription factor FOXO3a, which suppressed cancer cell proliferation. In AZD6244-resistant cancer cells, we observed the impaired nuclear localization of FOXO3a, reduced FOXO3a-mediated transcriptional activity, and decreased the expression of FOXO3a target gene Bim after cell treatment with AZD6244. Resistant cells could be sensitized by phosphoinositide 3-kinase (PI3K)/AKT inhibitors, which are known to enhance FOXO3a nuclear translocation. Our findings define FOXO3a as candidate marker to predict the clinical efficacy of AZD6244. Furthermore, they suggest a mechanism of resistance to MEK inhibitors that may arise in the clinic yet can be overcome by cotreatment with PI3K/AKT inhibitors.

Figure 1

AZD6244 enhanced FOXO3a expression and induced suppression of cancer cell proliferation. A, tumor volume of the HCT116 xenografts treated with Placebo or AZD6244 was measured for 21 d. The tumor sections of four individual DMSO or AZD6244-treated HCT116 xenografts were subjected to immunohistochemistry with a FOXO3a antibody. Relative percentages of nuclear FOXO3a expression of individual xenograft tumors from B were analyzed and the mean values of FOXO3a expression in Placebo or AZD6244-treated group were indicated as bars. B, lysates from various cancer cell lines: breast cancer (MDA-MB-435), colon cancer (HCT116, SW620, and HT29), and melanoma (WM793) treated with DMSO or AZD6244 (10 μmol/L) for 4 h were subjected to immunoblotting with the indicated antibodies. C, after transfection with a GFP vector or GFP-FOXO3a, MDA-MB-435 cells were treated with DMSO or AZD6244 (10 μmol/L) for 24 h and subjected to cell cycle analysis. D, MDA-MB-435 cells transfected with control siRNA or FOXO3a siRNA were treated with AZD6244 (10 μmol/L) for 24 h with or without Taxol and subjected to cell cycle analysis.

Figure 2

AZD6244-resistant cancer cells show impaired FOXO3a activity and decreased Bim expression in response to AZD6244 treatment. A, lysates from AZD6244-sensitive cell lines and AZD6244-resistant cell lines were subjected to immunoblotting with FOXO3a and Bim antibodies. Relative fold of protein expressions were normalized by actin and the mean values of AZD-sensitive and AZD-resistant cells were indicated as bars. B, AZD6244-sensitive cell lines (HCT116, SW620, and WM793) and AZD6244-resistant cell lines (MDA-MB-468, SKOV3, and SKBR3) were treated with AZD6244 of indicated concentrations and the lysates were subjected to immunoblotting with indicated antibodies. C, AZD6244-sensitive cell lines (HCT116 and SW620) and AZD6244-resistant cell lines (DU145 and SKBR3) were treated with indicated concentrations of AZD6244 and the Bim mRNAs were subjected to real-time PCR analysis. D, after transfection with control siRNA or FOXO3a siRNA, SW620 cells were treated with AZD6244 of indicated concentrations for 6 h and subjected to real-time PCR analysis. After being transfected with GFP vector or GFP-FOXO3a, SKBR3 cells were treated with AZD6244 of indicated concentrations for 6 h and subjected to real-time PCR analysis. Bar graphs, mean values of the representative results from two experiments conducted in triplicates for each.

Figure 3

AZD6244-resistant cancer cells show retarded endogenous FOXO3a nuclear translocation and reduced Bim promoter association in response to AZD6244 treatment. A, AZD6244-resistant SKOV3 cells were treated with DMSO, AZD6244 (10 μmol/L), API-2 (10 μmol/L), or AZD6244 (10 μmol/L) along with API-2 (10 μmol/L) for 24 h and then subjected to immunoflourescence analysis. Bar graphs, the percentages of cells with high nuclear FOXO3a expression from two independent immunoflourescence experiments. B, AZD6244-sensitive SW620 cells and AZD6244-resistant SKBR3 cells were treated with DMSO, AZD6244 (10 μmol/L), or API-2 (10 μmol/L) for 6 h and then subjected to chromatin immunoprecipitation analysis. C, SKOV3, SKBR3, and SW620 were treated with DMSO, AZD6244 (10 μmol/L), and API-2 (10 μmol/L) for 6 h and subjected to real-time PCR analysis. Bar graphs, the mean values of the representative results from two experiments conducted in triplicates for each.

Figure 4

API-2 synergizes with AZD6244, suppressing cell proliferation and colony formation in AZD6244-resistant cancer cells. AZD6244-resistant cells (A) SKBR3 and SKOV3, and AZD6244-sensitive cells WM793 and SW620, were treated with DMSO, AZD6244, API-2, or AZD6244 along with API-2 for 48 h and then subjected to MTT assays. B, SKBR3 and SW620 cells were treated with DMSO, AZD6244 (5 μmol/L), API-2 (5 μmol/L), or AZD6244 (5 μmol/L) along with API-2 (5 μmol/L) for 48 h and then subjected to propium iodide staining and analyzed for sub-G₁ percentage. AZD6244-resistant cells (C) SKOV3 and SKBR3 cells were treated with DMSO, AZD6244 (10 μmol/L), API-2 (10 μmol/L), or AZD6244 (10 μmol/L), along with API-2 (10 μmol/L), while subjected to colony formation assays. Bar graphs, the mean values of the representative triplicate results from two experiments. D, MDA-MB-231 cells transfected with control or FOXO3a siRNA were treated with DMSO, AZD6244 (5 μmol/L), or AZD6244 (5 μmol/L), along with API-2 (5 μmol/L) for 48 h and then subjected to MTT assays. Bar graphs, the mean values of the representative triplicate results from two experiments.

Figure 5

AZD6244 combined with API-2 specifically kill cancer cells but not normal cells. A, a panel of normal breast and lung epithelial cell lines (MCF-10A, MCF-12A, and HBE4-E6E7) and several cancer lines were treated with DMSO, AZD6244 (5 μmol/L), API-2 (5 μmol/L), or AZD6244 (5 μmol/L) along with API-2 (5 μmol/L) for 48 h and then subjected to MTT assays. Bar graphs, the mean values of the representative results from two experiments conducted in triplicates for each. B, AZD6244-resistant cancer cells showed impaired endogenous FOXO3a nuclear translocation, reduced FOXO3a-Bim promoter association, and significantly decreased Bim expression in response to AZD6244. However, API-2 (bold arrow, AKT inhibitor) enhanced FOXO3a nuclear translocation, increased FOXO3a-Bim promoter association and enhanced Bim expression, and induced cell apoptosis. In AZD6244-sensitive cancer cells, both AZD6244 and API-2 were shown to induce FOXO3a activity and induced cell apoptosis.