RRR-γ-tocopherol induces human breast cancer cells to undergo apoptosis via death receptor 5 (DR5)-mediated apoptotic signaling

2.1. Chemicals

Reagents for morphological analyses of apoptosis were purchased from Boehringer Mannheim (Indianapolis, IN). αT and γT were purchased from TAMA Biochemical Company (LTD, Tokyo, Japan). TRAIL was purchased from Alexis Biochemicals (San Diego, CA).

2.2. Cell culture

Sources and culture conditions for maintenance of MDA-MB-435 and MCF-7 human breast cancer cells have been described previously [20]. Human mammary epithelial cells (HMEC; Cooperative Human Tissue Network, Birmingham, AL) were primary cultures of human mammary cells derived from normal mammoplasty specimens [20]. For experiments, FBS was reduced to 2%. Cells were plated at 3 × 10⁵ cells/100 mm dish or 6 × 10⁵/150 mm dish for apoptosis and western blot analyses, respectively. γT and αT (200 mM) were dissolved in DMSO and further diluted in ethanol to 40 μM. Equivalent levels of DMSO/ethanol (1:4) were used as vehicle control (VEH).

2.3. Quantification of apoptosis

Apoptosis was quantified based on nuclear morphology following staining of DNA with DAPI as described previously [20], or an Annexin V-FITC/PI assay (Invitrogen, Carlsbad, CA) following the manufacturer's instructions. Annexin V-FITC/PI assay measures amount of phosphatidylserine on the outer surface of the plasma membrane (a biochemical alteration unique to membranes of apoptotic cells) and amount of propidium iodide (PI), a dye that does not cross the plasma membrane of viable cells but readily enters dead cells or cells in the late stages of apoptosis and binds DNA. Fluoresence was measured using Fluorescence Activated Cell Sorter (FACS) analyses with a FACSCalibur flow cytometer and data were analyzed using CellQuest software (BD Biosciences, San Jose, CA). Cells displaying phosphatidylserine on their surface (positive for annexin-V fluorescence) were considered to be apoptotic.

2.5. Western blot analyses

Preparation of whole cell protein extracts and western blot analyses were conducted as described previously [20]. Protein bands were imaged and quantified after correcting for GAPDH loading control using Scion Image Software (Scion Corporation, Frederick, MD). Primary antibodies used and their sources were as follows: PARP (Santa Cruz Biotechnology, Santa Cruz, CA); Bid and Bax (Pharmingen, Rockville, MD); caspase-8 and caspase-9 (Cell Signaling Technology, Beverly, MA); DR5 (Cell Sciences, Canton, MA); and glyceraldehyde-3-phosphate dehydrogenase (GAPDH, made in house). Secondary antibodies used included: horseradish peroxidase conjugated goat-anti-rabbit and rabbit-anti-mouse (Jackson Immunoresearch, Rockford, IL); and bovine-anti-goat (Santa Cruz Biotechnology).

2.6. FACS analyses of cell surface expression of DR4 and DR5 and their ligand TRAIL

Cell surface expression of DR4, DR5 and TRAIL was determined using FACS analyses. Treated and control cells were detached and washed three times with 0.5% BSA in PBS. Next, cells (25 μl of 8 × 10⁶ cells/ml in 0.5% BSA in PBS) were incubated with 1 μg normal mouse IgG₁ or IgG_2b for 15 min at room temperature followed by incubation with phycoerythrin-conjugated antibodies to DR4, DR5 or TRAIL or phycoerythrin-conjugated isotype immunoglobulins (R&D Systems, Minneapolis, MN) for 45 min. Intensity of red fluorescence was detected using a FACSCalibur flow cytometer and data were analyzed using CellQuest software.

2.7. Detection of active (conformationally changed) Bax

Active (conformationally changed) Bax was detected as described previously [21]. Briefly, 500 μg protein from whole cell extracts/0.5 ml was incubated overnight at 4°C with 2 μg of anti-Bax 6A7 (Sigma-Aldrich, St. Louis, MO), an antibody that specifically recognizes conformationally changed Bax, followed by reaction with 20 μl of protein G-agarose beads (Calbiochem, San Diego, CA) for 2 hours at 4°C. The conformationally changed Bax protein was analyzed by SDS-PAGE and detected using regular anti-Bax antibody.

2.8. Detection of mitochondrial permeability transition

The effect of γT on mitochondrial membrane potential was measured using the dye JC-1 (5,5′,6,6′-tetrachloro-1,1′,3,3′ tetraethylbenzimidazolylcarbocyanine iodide/chloride; Molecular Probes, Carlsbad, CA ). The JC-1 dye bearing a positive charge enters intact mitochondrial membrane due to the negative charge established by the intact mitochondrial membrane potential, accumulates as aggregates producing a red fluorescence [22]. In the absence of intact mitochondrial membrane, JC-1 accumulates in the cytoplasm in monomeric form characterized by green fluorescence [22]. Cells were collected, incubated with JC-1 for 30 min at 37°C followed by one wash with PBS. Red and green fluorescence intensity was measured using a FACSCalibur flow cytometer and data were analyzed using CellQuest software.

2.9. Detection of cytochrome c release from mitochondria into cytosol

A rapid and reliable technique to detect cytochrome c release during drug-induced apoptosis by FACS analyses was used [23]. Briefly, cells were treated, collected (2 × 10⁵) and incubated with 100 μl digitonin (50 μg/ml in PBS with 100 μM KCl) for 5 min on ice followed by fixing the cells with 4% paraformaldehyde in PBS for 20 min at room temperature, and incubating the cells in blocking buffer (3% BSA and 0.05% saponin in PBS) for 1 h. Cells were incubated with anti-cytochrome c antibody (clone 6H2.B4, BD Pharmingen, San Diego, CA) at 1:200 in blocking buffer overnight at 4°C, washed three times with PBS, incubated with FITC-labeled secondary antibody in blocking buffer for 1 h at room temperature. Cytochrome c fluorescence intensity was determined using a FACSCalibur flow cytometer and data were analyzed using CellQuest software.

2.10. Semi-quantitative reverse transcription PCR analysis

Semi-quantitative analyses were conducted (following company instructions) to detect DR5 mRNA expression by reverse transcriptase-polymerase chain reaction (RT-PCR) with Superscript RTase (250 U, Invitrogen) and Duplex PCR with Taq PCR Master Mix Kit (Qiagen Inc), respectively. Primers for DR5 (478 bp) were: forward GCC TCA TGG ACA ATG AGA TAA AGG TGG CT and reverse CCA AAT CTC AAA GTA CGC ACA AAC GG, and for β-actin (202 bp) were: forward GGC GGC ACC ACC ATG TAC CCT and reverse AGG GGC CGG ACT CGT CAT ACT [24]. PCR conditions employed were: 94°C for 3 min for initial denaturation, 35 cycles at 94°C for 1 min for denaturation, 61°C for 1 min for annealing and 72°C for 2 min for elongation [24]. PCR fragments were separated by electrophoresis on a 1.5% agarose gel and visualized by staining with ethidium bromide.

2.11.Transfection with DR5 and FADD siRNA

Cells were transiently transfected with siRNAs to DR5 or FADD (Ambion, Austin, TX) using a reverse transfection protocol following company instructions. A negative control siRNA (Ambion, Austin, TX), which has no significant sequence similarity to any mouse, rat or human gene sequences, was used to control for any nonspecific effects on gene expression. Optimal conditions were determined using Silencer CellReady siRNA Transfection Optimization and GAPDH Kit provided by Ambion. Briefly, for each well of a 6 well plate, 9 μl transfection agent; siPORT NeoFX (Ambion, Austin, TX) was mixed with 0.5 ml serum free media (OPTI-MEM-I) and incubated at room temperature for 10 min followed by addition of 2.3 μl siRNA (20 μM). The mixture of transfection agent and siRNA was incubated at room temperature for 10 min followed by mixing with 1 ml cells at 3.75 × 10⁵/ml in culture media. Cells were cultured for 1 day followed by replacement of the transfection media with culture media and incubating for an additional day. Transfected cells were re-cultured in 100 mm dish at 3 × 10⁵ cells for 8 hrs prior to γT treatments for 2 days.

2.12. Statistical Analyses

Apoptosis data were analyzed using a one-way analysis of variance followed by ANOVA Tukey test or a two-tailed student t-test to determine statistical differences among treatments and control values. Differences were considered statistical significant at p<0.05.

3.1. γT, but not αT, induced human breast cancer cells to undergo apoptosis

The ability of γT to induce apoptosis in MDA-MB-435 and MCF-7 human breast cancer cells was evaluated based on enumeration of cells containing condensed and/or fragmented DNA (DAPI assay), western blot assessment of PARP cleavage, or monitoring of the phosphatidylserine (PS) content of the outer leaflet of the plasma cell membrane (PS externalization) with FITC-conjugated annexin V.

Photomicrographs of DAPI stained cells showing bright condensed or fragmented chromatin in nuclei from apoptotic cells in comparison to non-apoptotic nuclei exhibiting an overall dull staining pattern are provided in Fig. 1A. Condensed/fragmented DNA in MDA-MB-435 cells and condensed DNA in MCF-7 cells were observed following treatment with 40 μM γT for 3 days. Quantification of cells undergoing apoptosis as determined by the DAPI assay, showed that γT treatments (40 μM) for 2 and 3 days produced significant increases in the percentage of apoptosis in comparison to vehicle controls in both cell types (Fig. 1B). A dose-dependent increase in apoptosis, as quantified by the DAPI assay, was observed in both cell types (Fig. 1C); namely, 12%, 22% and 43% apoptosis in MDA-MB-435 cells and 10%, 21% and 39% apoptosis in MCF-7 cells after 3 day treatment with 10, 20 or 40 μM γT, respectively. In contrast, no γT-induced apoptosis was seen in HMECs (Fig. 1C); thus showing specificity of γT-induced apoptosis for cancer cells. Interestingly, αT, the most abundant vitamin E form in human plasma and tissues, at equivalent concentrations (namely, 10-40 μM) did not induce apoptosis in any of the three cell types (Fig. 1C); thus showing that pro-apoptotic effects by γT are not a general property of vitamin E compounds. Cleavage of PARP into 89 kDa fragments, a well established biomarker for apoptosis, provides further evidence for the dose-dependent induction of apoptosis by γT in both cell types (Fig. 1D). γT treatments (40 μM) for 2 and 3 days induced 26% and 31% annexin V positive cells in MDA-MB-435 cells versus 7% annexin V positive cells in VEH treated cells, and 28% and 48% annexin V positive cells in MCF-7 cells versus 16.4% annexin V positive cells in VEH treated cells, respectively (Fig. 1E & F).

3.2. γT sensitized human breast cancer cells to TRAIL-induced apoptosis

In preliminary experiments, both MCF-7 and MDA-MB-435 cells were shown to be insensitive to TRAIL treatment in comparison to MDA-MB-231 cells which have been reported previously to be responsive [25] (data not shown). Interestingly, pre-treatment of MDA-MB-435 cells with 40 μM γT for 1 day followed by removing the treatment media and culturing the cells with 10 ng/ml TRAIL for 2 days significantly enhanced TRAIL-induced apoptosis to 42% from 13% by pre-γT treated alone (Fig. 2A). Treatment of cells with 10 ng/ml of TRAIL alone did not induce apoptosis (Fig. 2A). Employing the same treatments to HMECs did not induce any apoptosis (Fig. 2A), showing TRAIL specificity for cancer cells, as predicted based on the literature [15]. Additional proof of the ability of γT pre-treatment to sensitize TRAIL-refractory MDA-MB-435 cells to TRAIL-induced apoptosis came from analyses of PARP cleavage (Fig. 2B). These analyses showed higher levels of cleaved PARP (89 kDa) in cells receiving pre-γT + TRAIL than in cells receiving TRAIL alone (VEH +TRAIL) or pre-γT treatment alone (Fig. 2B). Furthermore, γT exhibited a dose-dependent increase in ability to sensitize the TRAIL-refractory MDA-MB-435 cells to TRAIL-induced apoptosis (Fig. 2C) which was again confirmed by analyses of PARP cleavage (Fig. 2D). Quantification of apoptotic cells using the Annexin V-FITC/PI assay confirmed that pretreatment (40 μM) or co-treatment (20 μM) of γT significantly enhanced TRAIL-induced apoptosis in comparison to γT or TRAIL alone treated MDA-MB-435 cells (Fig. 2E & F).

3.3. γT induced increased levels of cell surface DR5

To understand how γT might overcome the human breast cancer cell's resistance to the apoptosis-inducing effects of TRAIL, membrane expression of the TRAIL receptors DR4 & DR5 as well as the ligand TRAIL were examined by indirect immunofluorescence staining/FACS analyses. Data show that following γT treatment (40 μM for 2 days), MDA-MD-435 cells and MCF-7 exhibited increased membrane-bound DR5 expression in comparison to VEH control treated cells (Fig. 3A). No change in DR4 expression was observed following γT treatment in either cell type (Fig. 3A). Expression of TRAIL on the cell surface was also examined and revealed no change in either cell type following γT treatment (Fig. 3A). These data suggest that γT sensitizes the TRAIL-refractory cancer cells to TRAIL-induced apoptosis via increased expression of membrane-bound DR5.

3.4. γT induced increased levels of DR5 mRNA and protein

To investigate the molecular basis for the increase in membrane expression of DR5, the effects of γT on DR5 mRNA and protein levels were examined. γT increased DR5 mRNA levels by 4.6 and 4.7 fold after 2 and 3 days of treatment in MDA-MB-435 cells, and 2.1, 3.6 and 4.1 fold after 1, 2 and 3 days of treatment in MCF-7 cells, respectively (Fig. 3B, panels 1 & 2). DR5 protein levels also increased in both cell lines after 2 and 3 days of treatment with γT (Fig 3B, panels 3 & 4).

3.5. siRNA knockdown of DR5 or FADD block γT-induced apoptosis

Functional knockdown of DR5 using siRNA was performed to address the necessity of these cellular components in γT-induced apoptosis. DR5 siRNA significantly blocked γT-induced apoptosis (36%) in comparison to γT treated cells transfected with negative control siRNA (Fig. 4A). Validation of the success of the DR5 siRNA knockdown by western immunoblotting showed a markedly reduction in DR5 (DR5L/S) protein expression (Fig. 4B). To confirm if FADD, a death receptor associated protein that triggers death receptor mediated casapse-8 dependent apoptosis is involved in γT-induced apoptosis we also blocked FADD protein expression using siRNA to FADD. Likewise, data show a significant attenuation of γT-induced apoptosis (46%) by FADD siRNA (Fig. 4A) and a marked decrease in FADD protein expression (Fig. 4B). These data document a role for DR5 and FADD in γT-induced apoptosis in MDA-MB-435 cells, suggesting that γT induces apoptosis via DR5/FADD death receptor mediated events.. Knockdown of DR5 using siRNA to DR5 also attenuated pre-γT + TRAIL-induced apoptosis by 60% in comparison to negative control siRNA (Fig. 4C), demonstrating that γT upregulation of DR5 plays a critical role in sensitizing breast cancer cells to TRAIL-induced apoptosis.

3.6. γT induced apoptosis is mitochondria dependent

Downstream events involved in prototypical DR5 apoptotic signaling were investigated to establish what occurs in human breast cancer cells following γT-triggered apoptosis including: activation of caspases, increase in mitochondrial permeability, cytochrome c release, and activation of two pro-apoptotic members of the Bcl-2 family that can contribute to increased mitochondrial permeability (truncation of Bid and conformational change of Bax).

The activation status (i.e. cleavage) of two initiation caspases; caspase-8 (involved in DR apoptotic signaling) and caspase-9 (involved in intrinsic mitochondrial-mediated apoptotic signaling) were measured by western blot analyses. Cleavage of both caspases occurred after γT treatments of 1, 2, or 3 days in MDA-MB-435 cells and after 2 or 3 days in MCF-7 cells (Fig. 5A).

Mitochondrial permeability was examined using JC-1 dye. As described in the Materials and Methods, in the absence of intact mitochondria this dye accumulates in the cytoplasm as a green fluorescing monomeric form thus serving as an indicator of mitochondria permeability [22]. γT (40 μM) induced a higher percentage of green stained MDA-MB-435 cells after 2 and 3 days treatments (40% and 55%, respectively) in comparison to VEH control (8% and 9%, respectively), indicating that γT induces increased mitochondrial permeability in breast cancer cells (Fig. 5B).

Indirect immunofluorescence staining of cytochrome c in permeabilized cells, to permit cytosolic cytochrome c diffuse from the cells was used as an indicator of cytochrome c release from mitochondria into the cytosol [23]. During this procedure cytoplasmic but not mitochondrial cytochrome c is lost, thus a decrease in staining intensity is correlated with release of cytochrome c from the mitochondria. Three day γT (40 μM) treatments produced marked reductions in cytochrome c fluorescence in MDA-MB-435 and MCF-7 cells (60% and 36%, respectively) in comparison to VEH treated control cells (5% and 4%, respectively; Fig. 5C). These data support the involvement of intrinsic mitochondria-mediated apoptosis in γT-induced apoptosis.

Activation of Bax and Bid, upstream contributors to mitochondria-mediated apoptotic signaling, was analyzed. Active Bax was detected using a two-step procedure of immunoprecipitating conformationally changed Bax followed by detection by western immunoblotting. γT at 40 μM increased protein levels of active Bax by 3.9 and 3.6 fold after 2 and 3 days treatment in MDA-MB-435 cells, and 1.6, 5.2, and 3.6 fold after 1, 2 and 3 days treatments in MCF-7 cells in comparison to cells treated with VEH for 3 days (Fig. 5D). Western immunoblotting was used to detect cleavage of intact Bid, a caspase-8 substrate, to truncated Bid (tBid), a pro-apoptotic active form of Bid that promotes Bax conformational change. Data show that γT treatment induced Bid cleavage from its intact 21 kDa to truncated 15 kDa (tBid) form (Fig. 5D).