BAD sensitizes breast cancer cells to docetaxel with increased mitotic arrest and necroptosis

Abstract

Breast cancer patients are commonly treated with taxane (e.g. docetaxel) chemotherapy, despite poor outcomes and eventual disease relapse. We previously identified the Bcl-2-associated death promoter (BAD) as a prognostic indicator of good outcome in taxane-treated breast cancer patients. We also demonstrated that BAD expression in human breast carcinoma cells generated larger tumors in mouse xenograft models. These paradoxical results suggest that BAD-expressing tumors are differentially sensitive to taxane treatment. We validated this here and show that docetaxel therapy preferentially reduced growth of BAD-expressing xenograft tumors. We next explored the cellular mechanism whereby BAD sensitizes cells to docetaxel. Taxanes are microtubule inhibiting agents that cause cell cycle arrest in mitosis whereupon the cells either die in mitosis or aberrantly exit (mitotic slippage) and survive as polyploid cells. In response to docetaxel, BAD-expressing cells had lengthened mitotic arrest with a higher proportion of cells undergoing death in mitosis with decreased mitotic slippage. Death in mitosis was non-apoptotic and not dependent on Bcl-XL interaction or caspase activation. Instead, cell death was necroptotic, and dependent on ROS. These results suggest that BAD is prognostic for favourable outcome in response to taxane chemotherapy by enhancing necroptotic cell death and inhibiting the production of potentially chemoresistant polyploid cells.

Figure 1

BAD increases sensitivity to docetaxel. (a) MDA-MB-231 cells expressing vector or BAD were treated with 125 nM docetaxel for 5 days. Cells were stained with Annexin V-647 and PI and analyzed via flow cytometry daily. Cell death in control group were subtracted from the docetaxel treated group. Annexin V+/PI+ population is depicted. Student’s t-test; n = 3. (b) MDA-MB-231 cells expressing vector or BAD were injected into the mammary fat pads of Taconic nude mice. Red arrows indicate docetaxel or vehicle injection time points. Tumor volume was measured weekly. One-way ANOVA with Dunnett’s post-hoc test; Vector + vehicle = 8, vector + docetaxel = 6, BAD + vehicle = 7, BAD + docetaxel = 7. (c) Scatter plot of tumor volume at day 49. (d) Representative images of tumors in the mammary gland of nude mice. Arrows indicate tumor location. (e) Kaplan-Meier survival curve of mice treated with vehicle or docetaxel.

Figure 2

BAD increases length in mitotic arrest with docetaxel treatment. (a) Representative cell fates of MDA-MB-231 cells expressing vector or BAD treated with 125 nM docetaxel for 72 hours. Scale bar = 20 μM. (b) Immunofluorescence images taken 48 hours after docetaxel treatment and stained with DAPI and α-Tubulin. Scale bar = 20 μM. (c) Cell fates of cells treated with docetaxel for 72 hours. Each horizontal line represents an individual cell. Line endpoint represents the time at which the indicated cell fate occurred. (d) Scatter-plot representation of individual cell which either slipped or died in mitosis, and the corresponding hours of mitotic arrest. Median and interquartile range are shown. Mann-Whitney statistical test. (e) Length of mitotic arrest for all cell fates. Mann-Whitney statistical test. (f) Cells were treated with 125 nM docetaxel for 16, 24, or 48 hours. DMSO control treatment was for 48 hours. Right: Quantification of cyclin B1 levels normalized to tubulin. Student’s t-test; n = 3. (g) Cell cycle phases of docetaxel treated cells were analyzed daily for 5 days with PI staining and flow cytometry. Student’s t-test; n = 3.

Figure 3

BAD and Bcl-XL binding is not required for docetaxel killing. (a) Left: MDA-MB-231 cells stably expressing vector or BAD were treated with 125 nM docetaxel for the indicated time points. Control is DMSO for 48 hours. Right: Quantification of protein band density for pBcl-XL-Ser62 over total Bcl-XL. Student’s t-test; n = 3. (b) Immunoprecipitation (IP) of BAD antibody after 24 hours of 125 nM docetaxel treatment. GST antibody was used as a negative control. (c) Immunoprecipitation with Bcl-XL antibody after 125 nM docetaxel for 5 days. Vimentin antibody was used as a negative control. Right: The flow throughs (FT) from the IPs were retained and subjected to western blot. (d) Annexin V+/PI+ staining and flow cytometry analysis of 125 nM docetaxel treated cells after 5 days. One-way ANOVA with Dunnett’s post-hoc test; n = 5.

Figure 4

Docetaxel is killing the cells through a non-apoptotic mechanism. (a) MDA-MB-231 cells stably expressing vector or BAD were treated with DMSO control (left) or 125 nM docetaxel (middle) and subjected to western blot. Right: Protein band quantification was performed. Student’s t-test; no statistical significance. (b) Cells were treated with 125 nM docetaxel for 3 days prior to immunoprecipitation (IP) with 6A7 BAX antibody. Vimentin antibody was used as a negative control. Input lanes were also probed with cleaved caspase-3 antibody (right). (c) Cells were treated with 125 nM docetaxel for 5 days and stained with Annexin V-647 and PI daily and analyzed via flow cytometry. The Annexin V+/PI− population is graphed. Student’s t-test; no statistical significance between vector and BAD. Right: Dot plots from flow cytometric analysis. Annexin V positive cells are on the x-axis, PI positive cells are on the y-axis. Time, in days, is increasing to the right.

Figure 5

Docetaxel is killing the cells through a necroptotic mechanism. MDA-MB-231 cells stably expressing vector or BAD were treated with 125 nM docetaxel (Dtxl), 20 μM Z-VAD-FMK (ZVAD), or 5 μM necrosulfonamide (NSA) for 5 days. Cells were stained with Annexin V-647 and PI and analyzed via flow cytometry. The Annexin V+/PI+ population is represented in a bar graph. Two-way ANOVA with Tukey’s post-hoc test; n = 3.

Figure 6

BAD-mediated mitotic arrest is downstream of AMPK. (a) MDA-MB-231 cells stably expressing vector or BAD were treated with 125 nM docetaxel for 48 hours prior to high-resolution respirometry. One-way ANOVA with Dunnett’s post-hoc test; n = 3. (b) Cells were treated with 125 nM docetaxel for 16, 24, or 48 hours. DMSO control treatment was for 48 hours. Right: Quantification of pAMPK-Thr172 levels over AMPK total protein were normalized to tubulin. One-way ANOVA with Dunnett’s post-hoc test; n = 3.

Figure 7

BAD requires ROS for docetaxel cell death. (a) MDA-MB-231 cells stably expressing vector or BAD were treated with 125 nM docetaxel or DMSO control for 48 hours prior to staining with CM-H₂DCFDA to measure reactive oxygen species (ROS). Mean fluorescence intensity (MFI) was measured via flow cytometry. Fold increase of docetaxel/control is graphed. Student’s t-test; n = 3. (b) MDA-MB-231 cells stably expressing vector or BAD were treated with 125 nM docetaxel or 10 mM N-acetyl cysteine (NAC) for 5 days. Annexin V+/PI+ staining flow cytometry analysis of cells is graphed. Two-way ANOVA with Tukey’s post-hoc test; n = 3. (c) MDA-MB-231 cells expressing BAD were treated with 125 nM docetaxel or 50 nM rotenone for 5 days. Annexin V+/PI+ staining flow cytometry analysis of cells is graphed. Student’s t-test; n = 3. (d) Schematic representation of docetaxel-mediated cell death. BAD increases length in mitotic arrest by inhibiting cyclin B1 degradation, leading to a ROS-dependent necroptotic cell death.