Combining BET and HDAC inhibitors synergistically induces apoptosis of melanoma and suppresses AKT and YAP signaling

Abstract

Histone acetylation marks have an important role in controlling gene expression and are removed by histone deacetylases (HDACs). These marks are read by bromodomain and extra-terminal (BET) proteins and novel inhibitiors of these proteins are currently in clinical development. Inhibitors of HDAC and BET proteins have individually been shown to cause apoptosis and reduce growth of melanoma cells. Here we show that combining the HDAC inhibitor LBH589 and BET inhibitor I-BET151 synergistically induce apoptosis of melanoma cells but not of melanocytes. Induction of apoptosis proceeded through the mitochondrial pathway, was caspase dependent and involved upregulation of the BH3 pro-apoptotic protein BIM. Analysis of signal pathways in melanoma cell lines resistant to BRAF inhibitors revealed that treatment with the combination strongly downregulated anti-apoptotic proteins and proteins in the AKT and Hippo/YAP signaling pathways. Xenograft studies showed that the combination of inhibitors was more effective than single drug treatment and confirmed upregulation of BIM and downregulation of XIAP as seen in vitro. These results support the combination of these two classes of epigenetic regulators in treatment of melanoma including those resistant to BRAF inhibitors.

Keywords: I-BET151; bromodomain; epigenetic; melanoma; panobinostat.

Figure 1. Combination of I-BET151 and LBH589 synergistically induces apoptosis in melanoma cells

A. Me1007 melanoma cells were treated with 2 μM I-BET151, 30 nM LBH589, combination or control for 48 h. Induction of apoptosis was determined by staining with Annexin-V/PI and flow cytometry analysis. B. Histogram represents mean (± SEM) of n = 3 experiments of different melanoma cell lines and melanocytes (HEM) drug-treated as described above. Combination treatment significantly induced apoptosis (p < 0.05) compared to single drug treatment in all tested melanoma cell lines. C. Combination index (CI) of the I-BET151 and LBH589 co-treatment are plotted at increasing drug concentration and fractional effect. CI < 1.0 indicates synergistic interaction. A representative Fa-CI plot (Chou-Talalay plot) for Patient-1-post cells is shown. D. CI values for different melanoma cell lines at a fractional effect (Fa) of 0.5 (dose required to kill 50% of cells). CI experiments were performed twice.

Figure 2. Combination-induced apoptosis is triggered by enhanced mitochondrial depolarization and caspase expression

A. Loss of mitochondrial outer membrane potential (ΔMOMP) monitored by flow cytometry of JC-1 staining of melanoma cells either treated with 2 μM I-BET151, 30 nM LBH589, combination or DMSO for 48 h is shown (n = 3, bars : ± SEM). B. Total protein lysates of drug-treated cells for 24 h were analyzed for cleavage of caspases by western blot. α-Tubulin was used as loading control. C. Patient-1-post and Me1007 cells were treated with drugs as described before for 48 h. To inhibit caspase activity, melanoma cells were pretreated with 10 μM of caspase inhibitor Q-VD-OPh for 30 minutes before drug treatment. Extent of cell death was assessed using Annexin-V/PI staining. D. Mitochondrial depolarization was measured using JC-1. Mean (± SEM) of n = 3 experiments is shown.

Figure 3. Combination-induced apoptosis requires expression of pro-apoptotic BIM

A. Induction of BIM mRNA expression level in Patient-1-post and Me1007 cells was determined by qRT-PCR. Mean expression levels (± SEM) of n = 2 experiments are presented. B. Knockdown of BIM protein expression by siRNA was performed in Patient-1-post and Me1007 cells. After 24 h, transfected cells were treated with drugs as described before and incubated for a further 48 h. Expression of the BIM_EL isoform is shown and α-Tubulin served as internal control. C. Knockdown of BIM reduces apoptosis in combination-treated in both cell lines. Mean (± SEM) of n = 3 experiments is shown. D. FOXO3a and BIM mRNA expression level analyzed by qRT-PCR are reduced upon siRNA-mediated knockdown of FOXO3a in Me1007 cells. E. Knockdown of FOXO3a reduces percentage of apoptosis in combination-treated Me1007 cells.

Figure 4. Identification of significant protein expression changes in response to combination treatment

A. RPPA data show genes that were most changed by combination of I-BET151 and LBH589 after 24 h. Downregulated proteins are represented in blue, whereas upregulated proteins are marked in red. Total protein lysates of Patient-1-post and Me1007 cells, harvested after 24 h upon treatment, were analyzed for changes in protein expression of B. pro-apoptotic and C. anti-apoptotic proteins by western blot. The large and small isoforms of MCL-1 are indicated by size markers. Expression of α-Tubulin served as internal control. D. Changes in protein expression of HEM were analyzed by western blot after 48 h and α-Tubulin was used as loading control. E. Fold change of BIM protein expression (± SEM) was quantified from three independent western blots by using ImageJ.

Figure 5. Co-treatment of I-BET151 and LBH589 downregulates AKT and Hippo/YAP signaling pathway

A. Total protein lysates of drug treated cells (24 h) were analyzed for the expression levels of AKT, its downstream target p-PRAS40 and Hippo/YAP signaling pathway in Me1007, Patient-3-post cells (left panel) and in melanocytes (right panel) by western blot. The phospho-YAP antibody detected an additional smaller splice variant that was not detected by the total-YAP antibody. GAPDH and α-Tubulin served as internal control, respectively. B. Downregulation of YAP1 mRNA level was determined by qRT-PCR in Patient-1-post and Me1007 cells after 24 h of drug treatment. Mean (± SEM) of n = 2–3 experiments is shown.

Figure 6. I-BET151 and LBH589 significantly inhibits melanoma growth and prolongs survival in a xenograft mouse model

A. Tumor volumes and B. Kaplan-Meier survival analysis following treatment of mice bearing Patient-1-post tumors with I-BET151 or LBH589 (panobinostat) alone or in combination is shown (n = 10 per treatment arm). Average tumor volume is shown until two mice in that treatment arm reached 1200 mm³. Compared with treatment of either agent alone, combination of I-BET151/LBH589 treatment significantly reduced tumor growth (vehicle vs. combination 65.4%, p < 0.001) and B. prolonged survival of mice (where survival of mice is defined as time to reach a tumor volume of 1200 mm³). C. Immunohistochemical analysis of BIM and XIAP levels in xenografts from the survival study are shown. D. Expression levels of BIM, XIAP and cleaved PARP were quantified in four separate xenografts per treatment (mean ± SEM). Compared to control mice, combination treatment lead to high tumor expression of BIM ( p = 0.015) and cPARP (0.047) and reduced XIAP (p = 0.0007) expression. E. A separate short term experiment measured BIM and YAP mRNA expression level of xenograft tumor tissue following 3 h of treatment. Mean expression levels (± SEM) of n = 3–4 mice per group is presented.