Mitochondria-mediated apoptosis by diallyl trisulfide in human prostate cancer cells is associated with generation of reactive oxygen species and regulated by Bax/Bak

Epidemiologic data continue to support the premise that dietary intake of Allium vegetables such as garlic may reduce the risk of different types of malignancies including cancer of the prostate (1–3). The anticarcinogenic effect of Allium vegetables is attributable to organosulfur compounds (4). Allium vegetable-derived organosulfur compounds (OSC), including diallyl sulfide, diallyl disulfide, and/or diallyl trisulfide (DATS), have been shown to inhibit cancer in animal models induced by a variety of carcinogens, including tobacco-derived chemicals (5–8). The mechanisms by which OSCs inhibit chemically induced cancer are well studied and involve an increase in expression of phase 2 carcinogen-detoxifying enzymes and inhibition of cytochrome P450–dependent monooxygenases (8–10). We have also shown previously that oral administration of diallyl disulfide significantly inhibits growth of H-ras oncogene–transformed tumor xenografts in athymic mice (11).

More recent studies including those from our laboratory have revealed that naturally occurring OSC analogues can inhibit growth of cancer cells by causing G₂ and M phase cell cycle arrest and apoptosis induction (12–20). For example, the diallyl disulfide–induced cell death in colon cancer cells correlated with an increase in the levels of intracellular calcium (12), whereas G₂-M phase cell cycle arrest was accompanied by an increase in cyclin B1 protein level, reduction in complex formation between cyclin-dependent kinase 1 (Cdk1) and cyclin B1, and hyperphosphorylation of Cdk1 (13). We have shown recently that DATS is much more potent than either diallyl sulfide or diallyl disulfide in inhibiting proliferation of human prostate cancer cells (15). We also found that the DATS-treated prostate cancer cells are arrested in both G₂ phase and mitosis (16, 18). The DATS-induced G₂ phase cell cycle arrest in human prostate cancer cells is caused by reactive oxygen species (ROS)–mediated degradation and Ser²¹⁶ phosphorylation of Cdc25C, leading to the inhibition of kinase activity of Cdk1/cyclin B1 complex (16), whereas the mitotic block is linked to the activation of DNA damage checkpoint effector checkpoint kinase 1 (18). The ROS generation by DATS correlates with the degradation of iron storage protein ferritin, leading to an increase in labile (chelatable) iron (20). Despite these advances, however, the mechanism of DATS-induced apoptosis is not fully defined. An understanding of the mechanism by which DATS causes apoptosis is critical for its further development as a clinically useful anticancer agent because this knowledge could lead to the identification of mechanism-based biomarkers potentially useful in future clinical trials.

We have shown previously that DATS causes apoptotic cell death in PC-3 and DU145 human prostate cancer cell lines, which are androgen independent and lack functional p53 (15). However, it is unclear if the apoptosis induction by DATS is restricted to androgen-independent prostate cancer cells. Further studies are also needed to identify the signal(s) that mediate apoptosis program in DATS-treated cells. Here, we provide experimental evidence to indicate that DATS targets mitochondria to trigger cell death in human prostate cancer cells, which correlates with ROS generation and dependent on Bax and Bak proteins.

DATS (purity ∼97%) was purchased from LKT Laboratories. Cell culture reagents and fetal bovine serum (FBS) were from Life Technologies. RNase A was from Promega, 6-carboxy-2′,7′-dichlorodihydrofluorescein diacetate (H₂DCFDA) was from Molecular Probes, and propidium iodide, 4′,6-diamidino-2-phenylindole (DAPI), and N-acetylcysteine were from Sigma. The antibodies against Bax, Bak, Bid, Smac/DIABLO, and Bcl-xL were from Santa Cruz Biotechnology; the antibody against actin was from Sigma; the antibody against Bcl-2 was from DakoCytomation; the antibodies against procaspase-3, cytochrome c, and cytochrome c oxidase IV (COX IV) were from Cell Signaling Technology, BD PharMingen, and Invitrogen, respectively.

The LNCaP (American Type Culture Collection), LNCaP-C4-2 (UroCor), and LNCaP-C81 (a generous gift from Dr. Ming-Fong Lin, University of Nebraska, Omaha, NE) were maintained in RPMI 1640 supplemented with 1 mmol/L sodium pyruvate, 10 mmol/L HEPES, 0.2% glucose, 10% (v/v) FBS, and antibiotics. Normal prostate epithelial cell line PrEC was purchased from Clonetics and maintained in PrEBM (Cambrex). The mouse embryonic fibroblasts (MEF) derived from wild-type (WT) and Bak and Bax double knock-out (DKO) mice were maintained as described by Wei et al. (21). Stock solution of DATS was prepared in DMSO, and an equal volume of DMSO (final concentration 0.2%) was added to the controls.

The DATS-induced apoptosis was determined by (a) fluorescence microscopy after staining the cells with DAPI, (b) analysis of cytoplasmic histone-associated DNA fragmentation, or (c) analysis of subdiploid fraction by flow cytometry after staining the cells with propidium iodide essentially as described previously (15, 16, 18, 19). Cell viability was determined by trypan blue dye exclusion assay as described previously (15).

Transmission electron microscopy in LNCaP cells treated with DMSO (control) or 40 μmol/L DATS for 3, 6, 9, or 16 h at 37°C was done as described previously (22). The effect of DATS treatment on mitochondrial membrane potential was measured using a potential-sensitive dye 5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethyl-benzimidazolylcarbocyanine iodide (JC-1). Stock solution of JC-1 (1 mg/mL) was prepared in DMSO and freshly diluted with the assay buffer. Briefly, LNCaP cells (5 × 10⁴) or PrEC cells (3 × 10⁵) were seeded in six-well plates or T25 flasks and allowed to attach overnight. The cells were exposed to DMSO (control) or DATS as described above. Subsequently, the cells were collected by trypsinization and incubated for 15 min at 37°C in medium containing 10 μg/mL JC-1. The cells were washed with PBS, resuspended in 0.5 mL assay buffer, and analyzed using a Coulter Epics XL Flow Cytometer.

Cells (5 × 10⁵) were seeded in 100-mm culture dishes, allowed to attach overnight, and exposed to DMSO (control) or desired concentrations of DATS for specified time periods. The cell lysate was prepared as described previously (15, 18, 19). The mitochondrial and cytosolic fractions from control and DATS-treated cells were prepared using a kit from BioVision according to the manufacturer's recommendations. Immunoblotting was done as described previously (15, 16, 18–20). Immunoreactive bands were visualized using the enhanced chemiluminescence method.

Desired cells (1 × 10⁵) were cultured on coverslips and treated with specified concentrations of DATS or DMSO (control) for the indicated time periods. Cells were stained with 200 nmol/L MitoTracker Red at 37°C for 30 min. After washing with PBS, the cells were fixed with 2% paraformaldehyde overnight at 4°C and permeabilized using 0.1% Triton X-100 in PBS for 10 min. The cells were washed with PBS, blocked with 2% bovine serum albumin in PBS for 1 h, and incubated with monoclonal anti–cytochrome c antibody or polyclonal anti-Smac/DIABLO, anti–Bcl-2 or anti–Bcl-xL antibody overnight at 4°C. The cells were washed with PBS, incubated with Alexa Fluor 488–conjugated secondary antibody (1:1,000 dilution, Molecular Probes) for 1 h at room temperature. Subsequently, the cells were washed with PBS and stained with DAPI (10 ng/mL) for 5 min at room temperature. The cells were washed twice with PBS and examined under a fluorescence microscope.

The LNCaP cells were transiently transfected with pSFFV–Bcl-2, pSFFV–Bcl-xL or pSFFV-neo plasmids generously provided by Dr. Stanley J. Korsmeyer (Dana-Farber Cancer Institute, Boston, MA). Briefly, the LNCaP cells were seeded in six-well plates and transfected at 60–70% confluency with pSFFV–Bcl-2, pSFFV–Bcl-xL or pSFFV-neo plasmid using FuGENE6 transfection reagent (Roche Applied Sciences) according to the manufacturer's instructions. The cells were cotransfected with pEGFP-N1 plasmid to determine transfection efficiency. Twenty-four hours after transfection, the cells were treated with DMSO or DATS for 24 h, collected, and processed for immunoblotting and immunocytochemistry to confirm overexpression and mitochondrial localization of Bcl-2 or Bcl-xL and analysis of cytoplasmic histone-associated DNA fragmentation.

RNA interference of Bax was done using a Signal Silence Bax small interfering RNA (siRNA) purchased from Cell Signaling Technology. RNA interference of Bak was done using a Bak-targeted siRNA (Santa Cruz Biotechnology). A nonspecific control siRNA was purchased from Qiagen. LNCaP cells (1 × 10⁴) were seeded in six-well plates and allowed to attach overnight. Cells were transfected with 100 nmol/L of desired siRNA using OligofectAMINE (Invitrogen) according to the manufacturer's recommendations. Twenty-four hours after transfection, cells were treated with DMSO (control) or desired concentrations of DATS for 24 h. Both floating and adherent cells were collected, washed with PBS, and processed for analysis of cytoplasmic histone-associated DNA fragmentation and immunoblotting.

Intracellular ROS generation was measured by flow cytometry following staining with H₂DCFDA. Briefly, LNCaP cells (1 × 10⁵ cells per well in six-well plates) or PrEC cells (3 × 10⁵ cells/flask in T-25 flasks) were plated, allowed to attach overnight and exposed to DMSO (control) or desired concentrations of DATS for specified time periods. The cells were stained with 5 μmol/L H₂DCFDA for 30 min at 37°C, and the fluorescence was measured by flow cytometry as described previously (22). In some experiments, cells were either pretreated for 2 h with 4 mmol/L N-acetylcysteine and washed twice with PBS before DATS exposure or pretreated for 2 h with N-acetylcysteine and then cotreated with DATS.

We have shown previously that DATS suppresses growth of androgen-independent PC-3 and DU145 human prostate cancer cells by causing apoptosis (15). However, it is unclear if the DATS-induced cell death is restricted to hormone-independent prostate cancer cells. In the present study, we addressed this question using an androgen-sensitive human prostate cancer cell line (LNCaP) and its androgen-independent variants LNCaP-C81 and LNCaP-C4-2. The C4-2 and C81 variants have been widely used to study the mechanism of prostate cancer progression from the androgen-responsive to androgen-independent state (23–26). The LNCaP-C4-2 cell line, which was generated through coculture of parental LNCaP cells with human bone fibroblasts in vivo in castrated male athymic mice, displays elevated prostate specific antigen expression and increased anchorage-independent growth in soft agar (27, 28). The LNCaP-C81 line was also derived from LNCaP cells through multiple passages in monolayer culture in RPMI 1640 supplemented with 5% FBS and 1% glutamine (29). As can be seen in Fig. 1A, DATS treatment caused an increase in fraction of apoptotic cells with condensed and fragmented DNA as judged by the DAPI assay. The DATS-induced apoptosis in LNCaP cells was confirmed by analysis of subdiploid fraction by flow cytometry (data not shown). Next, we compared sensitivities of parental LNCaP cells and its androgen-independent variants C81 (Fig. 1B) and C4-2 (data not shown) to DATS by trypan blue dye exclusion assay. Presence of 0.2% DMSO in the culture medium (solvent for DATS) did not have any appreciable effect on the viability of the cell lines examined in the present study (compare bars marked with the letters C and D in Fig. 1B representing untreated cells and cells treated with 0.2% DMSO, respectively). The viability of each cell line examined was decreased significantly in the presence of DATS in a concentration-dependent manner (Fig. 1B). Moreover, the cell-killing effect of DATS was more or less comparable between androgen-sensitive LNCaP cells and its androgen-refractory variants. Consistent with these results, DATS treatment caused a concentration-dependent and statistically significant increase in cytoplasmic histone-associated DNA fragmentation in LNCaP and LNCaP-C81 cells (Fig. 1C) as well as in LNCaP-C4-2 cells (data not shown). Interestingly, PrEC cell line, which has been used extensively as a representative normal prostate epithelial cell line (30, 31), was significantly more resistant to DATS-induced apoptosis even at 80 and 160 μmol/L concentrations (Fig. 1D). These results indicated that DATS selectively causes apoptosis in human prostate cancer cells irrespective of their androgen responsiveness.

We have shown previously that DATS-induced apoptosis in PC-3 cells is significantly attenuated in the presence of a pan-caspase inhibitor and a caspase-9 specific inhibitor (19), suggesting the involvement of the mitochondrial pathway in the execution of apoptosis program (32, 33). To systematically explore this possibility, initially, we determined the effect of DATS treatment on mitochondrial morphology by transmission electron microscopy. Representative electron micrographs of LNCaP cells treated for 16 h with either DMSO (control) or 40 μmol/L DATS are depicted in Fig. 2A. The DMSO-treated control LNCaP cells exhibited a normal complement of healthy-looking mitochondria having an intact cristae structure. The mitochondria in DATS-treated cells seemed to have an intact outer membrane but disrupted cristae structure. The mitochondria of DATS-treated cells resembled type III mitochondria with intestinal or sausage-shape structure as described by Scorrano et al. (34). The DATS-mediated loss of the cristae structure was evident as early as 6 h posttreatment (data not shown).

We determined the effect of DATS treatment on mitochondrial membrane potential by flow cytometry following staining with a potential sensitive dye JC-1. The JC-1 dye bearing a delocalized positive charge enters the mitochondrial matrix due to the negative charge established by the intact mitochondrial membrane potential (35). In the absence of intact mitochondrial membrane potential, JC-1 accumulates in the cytoplasm in monomeric form characterized by green fluorescence (35). As can be seen in Fig. 2B, the DMSO-treated control LNCaP cells predominantly exhibited red fluorescence with minimal green fluorescence, indicating intact mitochondrial membrane potential. The DATS treatment caused a time-dependent and statistically significant increase in the percentage of cells with green fluorescence, indicating collapse of the mitochondrial membrane potential, which was evident as early as 1 h posttreatment (Fig. 2B). In dose-response studies, after 2 h treatment, the DATS-mediated collapse of the mitochondrial membrane potential was observed at 10 and 20 μmol/L concentrations (data not shown). Interestingly, the PrEC cell line was resistant to the DATS-mediated disruption of the mitochondrial membrane potential at least for up to 6 h of treatment with 20 and 40 μmol/L DATS (Fig. 2C).

To test whether DATS-mediated disruption of the mitochondrial membrane potential was accompanied by cytosolic release of apoptogenic molecules, mitochondrial and cytosolic fractions were prepared from control and DATS-treated cells and subjected to immunoblotting for cytochrome c. As shown in Fig. 2D, treatment of LNCaP cells with 10, 20, and 40 μmol/L DATS for 8 h resulted in an increase in cytosolic level of cytochrome c, which was accompanied by a decrease in its level in the mitochondria. The blot was stripped and reprobed with anti-COX IV and anti-actin antibodies to ensure equal protein loading as well as to rule out cross-contamination of mitochondrial and cytosolic fractions. The DATS-mediated release of apoptogenic molecules from the mitochondria to the cytosol was also evident in the immunohistochemical analysis of cytochrome c and Smac/DIABLO (data not shown). Collectively, these results indicated that DATS-induced apoptosis in our model was accompanied by the collapse of the mitochondrial membrane potential and the translocation of apoptogenic molecules from the mitochondria to the cytosol.

The Bcl-2 family proteins have emerged as critical regulators of mitochondria-mediated apoptosis by functioning as either promoters (e.g., Bax and Bak) or inhibitors (e.g., Bcl-2 and Bcl-xL) of the cell death process (36). Because DATS treatment disrupted mitochondrial membrane potential leading to cytosolic release of apoptogenic molecules in LNCaP cells, we proceeded to test whether DATS-induced apoptosis was regulated by Bcl-2 family proteins. We determined the effect of DATS treatment on levels of Bcl-2 family proteins in LNCaP cells by immunoblotting, and representative blots are shown in Fig. 3A. The levels of antiapoptotic proteins Bcl-2 and Bcl-xL were decreased on treatment of LNCaP cells with DATS. In addition, DATS treatment resulted in a modest increase in levels of multidomain proapoptotic protein Bak. The DATS-mediated induction of Bak protein was evident as early as 2 h after treatment and was sustained for the duration of the experiment (Fig. 3A). The DATS treatment did not have any appreciable effect on Bid protein level (Fig. 3A). Figure 3B summarizes the effect of DATS treatment on Bax and Bak protein levels in PrEC cells. Expression of Bax or Bak proteins was not altered by treatment of PrEC cells with 40 μmol/L DATS for 4, 8, 16, or 24 h (Fig. 3B). These results indicated that DATS treatment altered the ratio of proapoptotic to antiapoptotic Bcl-2 family proteins in LNCaP cells.

Because DATS treatment caused a decrease in protein level of Bcl-2, we determined the effect of its overexpression on DATS-induced apoptosis. The LNCaP cells transiently transfected with the empty vector (hereafter abbreviated as LNCaP/Neo) were used as control for direct comparison. The cells were cotransfected with pEGFP-N1 plasmid to determine transfection efficiency, which varied between 50% and 70% in different transfection experiments. Overexpression of Bcl-2 in LNCaP cells transfected with pSFFV–Bcl-2 plasmid (hereafter abbreviated as LNCaP/Bcl-2 cells) was confirmed by immunoblotting. The level of Bcl-2 protein was markedly higher in LNCaP/Bcl-2 cells than in LNCaP/Neo cells (Fig. 4A). Immunocytochemical analysis confirmed the mitochondrial localization of Bcl-2 in LNCaP/Bcl-2 cells (data not shown). Interestingly, the LNCaP/Neo and LNCaP/Bcl-2 cells were equally sensitive toward DATS-induced cytoplasmic histone-associated DNA fragmentation (Fig. 4B). These results indicated that the DATS-induced apoptosis in LNCaP cells was probably not attributable to the down-regulation of Bcl-2.

Next, we determined the effect of ectopic expression of Bcl-xL, which is a Bcl-2 homologue and functions to inhibit cell death by different stimuli (36). The level of Bcl-xL protein was markedly higher in LNCaP cells transiently transfected with pSFFV–Bcl-xL plasmid (LNCaP/Bcl-xL cells) than in the empty vector-transfected control LNCaP/Neo cells (Fig. 4C). The DATS-induced cytoplasmic histone-associated DNA fragmentation was evident not only in vector-transfected control cells but also in LNCaP/Bcl-xL cells (Fig. 4D). Moreover, the cytoplasmic histone-associated DNA fragmentation resulting from a 24-h treatment with 40 μmol/L DATS did not differ significantly between LNCaP/Neo and LNCaP/Bcl-xL cells as judged by paired t test (Fig. 4D). Collectively, these results indicated that DATS-induced apoptosis in LNCaP cells was not influenced by Bcl-2 or Bcl-xL protein level.

Because DATS treatment caused an increase in the protein level of Bak (Fig. 3A), we hypothesized that this protein might play an important role in the regulation of DATS-induced apoptosis. We tested this hypothesis by comparing sensitivities of SV40-immortalized MEFs derived from WT and DKO mice toward DATS-induced apoptosis. Because of immortalization by transfection with SV40 genomic DNA, the MEFs cannot be regarded as normal fibroblasts. Initially, we used the MEFs from WT mice to determine the effect of DATS treatment on levels of Bak and Bax proteins. Similar to LNCaP cells, DATS treatment resulted in a modest induction of Bax and Bak protein levels in WT MEFs (data not shown). The DATS treatment caused concentration-dependent and statistically significant increase in apoptotic cells in WT MEFs as judged by the analysis of cytoplasmic histone-associated DNA fragmentation (Fig. 5A) and subdiploid fraction (data not shown). On the other hand, the MEFs derived from DKO MEFs mice were significantly more resistant to DATS-induced cytoplasmic histone-associated DNA fragmentation compared with WT MEFs (Fig. 5A). To test whether Bak and Bax proteins function upstream of cytosolic release of cytochrome c in our model, we determined the effect of DATS treatment on cytosolic release of cytochrome c using WT and DKO MEFs. As can be seen in Fig. 5B, the DATS-mediated cytosolic release of cytochrome c was approximately 2-fold higher in WT MEFs compared with DKO. Consistent with these results, DATS treatment caused proteolytic cleavage of procaspase-3 in WT MEFs as evidenced by the appearance of a 17-kDa cleaved intermediate, which was barely detectable in DKO even after DATS treatment (Fig. 5B). Collectively, these results indicated that Bax and Bak play an important role in the execution of DATS-induced apoptosis.

We used Bax- and Bak-targeted siRNA to confirm the involvement of these proteins in the regulation of DATS-induced apoptosis. As can be seen in Fig. 5C, the level of Bax protein was decreased by about 65% in LNCaP cells transiently transfected with Bax siRNA. Although transfection of LNCaP cells with the Bax-targeted siRNA also resulted in a marked decrease in protein level of Bak, the Bak-targeted siRNA was more specific and caused the knockdown of only Bak. Nonetheless, the cytoplasmic histone-associated DNA fragmentation resulting from a 24-h treatment with 20 and 40 μmol/L DATS was statistically significantly lower in LNCaP cells with combined knockdown of Bax and Bak proteins compared with cells transiently transfected with a control nonspecific siRNA (Fig. 5D). Together, these results indicated that Bax and Bak were targets of DATS-induced cell death.

ROS are implicated in apoptosis induction by different stimuli including ionizing radiation and hyperoxia (37, 38). We have shown previously that DATS-mediated G₂-M phase cell cycle arrest in PC-3 and DU145 cells correlates with ROS generation (16). We designed experiments to determine whether DATS-induced apoptosis was mediated by ROS. Intracellular ROS generation in control (DMSO treated) and DATS-treated LNCaP cells was assessed by flow cytometry following staining with H₂DCFDA (22). The H₂DCFDA is cleaved by nonspecific cellular esterases and oxidized in the presence of H₂O₂ and other peroxides to yield fluorescent 2‘,7’-dichlorofluorescein (DCF). The DATS-treated LNCaP cells exhibited a concentration-dependent and statistically significant increase in DCF fluorescence (Fig. 6A), which was significantly attenuated by a 2-h pretreatment with antioxidant N-acetylcysteine (Fig. 6B). Moreover, N-acetylcysteine pretreatment conferred near-complete protection against DATS-induced disruption of the mitochondrial membrane potential (Fig. 6C) and cytoplasmic histone-associated DNA fragmentation (Fig. 6D). Interestingly, ROS generation was not observed in PrEC cells treated with 40 μmol/L DATS for at least up to 16 h (data not shown). The N-acetylcysteine–mediated protection against DATS-induced apoptosis was maintained even if the cells were washed twice with PBS and suspended in N-acetylcysteine–free media before DATS exposure (data not shown). Together, these results indicated that DATS-induced apoptosis in LNCaP cells correlated with ROS generation.

Prostate cancer is a leading cause of cancer-related deaths among men in the United States. Because the majority of the prostate cancers are androgen dependent at the time of diagnosis, hormone withdrawal is a frequently prescribed treatment option for initial management of this malignancy (39). This treatment modality, however, is palliative and has limited scope for hormone-refractory prostate cancers, which are highly aggressive and metastatic (39). Chemotherapy and radiation therapy are largely ineffective against hormone-independent prostate cancer (40, 41). Prostate cancer is usually diagnosed in the sixth and seventh decades of life, which presents a large window of opportunity for intervention to prevent or slow progression of the disease. Therefore, clinical development of agents that are nontoxic to normal cells but can inhibit growth of both hormone-responsive and hormone-independent prostate cancers could have a significant impact on disease-related cost, morbidity, and mortality for a large segment of the population. The results of the present study indicate that DATS meets all these criteria. First, DATS inhibits proliferation of androgen-sensitive LNCaP cells and its androgen-independent variants (LNCaP-C81 and LNCaP-C4-2) by causing apoptosis. Second, the PrEC normal prostate epithelial cell line is significantly more resistant to growth inhibition and apoptosis induction by DATS even at concentrations that are highly cytotoxic to the prostate cancer cells. Third, we have shown recently that oral administration of DATS (6 μmol, thrice per week) significantly retards growth of PC-3 xenografts in vivo in athymic mice in association with apoptotic cell death (42).

The DATS-mediated cell growth arrest and apoptosis induction in human prostate cancer cells (refs. 15, 16, 18, 19 and the present study) is evident at concentrations (10–40 μmol/L) that are well within the pharmacologically achievable range based on a recent pharmacokinetic study (43). The peak plasma concentration of DATS in rats following treatment with 10 mg DATS was shown to be about 31 μmol/L (43). Although the pharmacokinetic parameters for DATS in humans have not yet been measured, oral administration of 200 mg of synthetic DATS (also known as allitridum) in combination with 100 μg selenium every other day for 1 month to humans did not cause any harmful side effects (44). It is therefore possible that the plasma concentrations of DATS required for cancer cell growth inhibition may be achievable in humans.

The present study indicates that the DATS-mediated apoptosis in LNCaP cells is accompanied by morphologic changes in mitochondria and the collapse of the mitochondrial membrane potential, leading to the release of apoptogenic molecules from the mitochondria to the cytosol. Interestingly, the DATS-mediated disruption of the mitochondrial membrane potential is not observed in PrEC cells. To the best of our knowledge, the present study is the first published report to show the involvement of the mitochondria in the execution of DATS-induced apoptosis. Antiapoptotic Bcl-2 family members (Bcl-2 and Bcl-xL) play an important role in the regulation of mitochondria-mediated apoptosis by different stimuli (36). The antiapoptotic Bcl-2 family members possess four conserved Bcl-2 homology domains (BH1-BH4) and mainly prevent the release of apoptogenic molecules (e.g., cytochrome c) from mitochondria to the cytosol by forming heterodimer complexes with proapoptotic family members such as Bax (36, 45, 46). Increased expression of Bcl-2 has been observed in a variety of hematologic malignancies and solid tumors, and overexpression of Bcl-2 renders cancer cells resistant to different apoptotic stimuli including chemotherapeutic agents (47). The DATS treatment causes a decrease in protein levels of both Bcl-2 and Bcl-xL in LNCaP cells. It is interesting to note, however, that the ectopic expression of either Bcl-2 or Bcl-xL fails to confer significant protection against DATS-induced cell death in LNCaP cells. Thus, DATS can overcome the antiapoptotic actions of Bcl-2 and Bcl-xL and promote apoptosis even in cells that overexpress these important oncoproteins.

The proapoptotic Bcl-2 family proteins, which can be subdivided into the Bax subfamily of multidomain proteins (e.g., Bax and Bak) or BH3-only subfamily (e.g., Bid), induce mitochondrial membrane permeabilization and release of apoptogenic molecules from mitochondria to the cytosol (36, 45, 46). The present study indicates that Bax and Bak proteins play an important role in the regulation of DATS-induced apoptosis because (a) the SV40-immortalized MEFs derived from DKO mice are significantly more resistant toward DATS-induced cell death compared with the WT MEFs; (b) the DATS-mediated cytosolic release of cytochrome c and cleavage of procaspase-3 is relatively more pronounced in the MEFs derived from WT mice than in the MEFs derived from DKO mice; and (c) Bax and Bak knockdown in LNCaP cells confers statistically significant protection against DATS-induced apoptosis. We have shown recently that the DATS-mediated inhibition of PC-3 xenograft growth in athymic mice in vivo correlates not only with increased apoptosis but also induction of Bax and Bak proteins in the tumor tissues (42). It is important to point out, however, that Bax and Bak cannot fully explain the cell death caused by DATS because combined knockdown of these proteins confers only partial protection against DATS-induced apoptosis. It is intriguing to note that DATS treatment causes only a modest increase in protein levels of Bax and Bak, yet knockdown of these proteins confers statistically significant protection against DATS-induced apoptosis. Although the precise mechanism by which Bax and Bak regulate DATS-induced cell death remains elusive, it is possible that DATS treatment causes conformation change and oligomerization of Bax/Bak leading to their translocation to the mitochondria. This possibility is likely based on following considerations: (a) Bax activation by certain apoptotic stimuli is dependent on ROS generation (38), which is observed in DATS-treated prostate cancer cells (present study), and (b) microtubule-damaging agents have been shown to cause Bax activation (48), and DATS treatment is known to disrupt tubulin network (18). However, further studies are needed to systematically explore this possibility.

The DATS-induced cell death in LNCaP cells correlates with ROS generation. Because ROS generation is implicated in the pathogenesis of many chronic diseases including cancer, the potential side effects of DATS-mediated ROS production cannot be ignored. However, we are tempted to speculate that ROS generation by DATS may not be deleterious because (a) DATS is derived from vegetables consumed by humans on a daily basis, yet epidemiologic studies continue to support the premise that dietary intake of Allium vegetables including garlic may reduce the risk of different types of malignancies, including cancer of the prostate (1–3); (b) the DATS-treated athymic mice do not exhibit weight loss or any other signs of toxicity (42); and (c) a normal prostate epithelial cell line (PrEC) seems resistant to DATS-mediated ROS generation, mitochondrial membrane potential collapse, and apoptosis induction (present study). It is possible that the DATS-mediated ROS generation in cancer cells is transient and serves to trigger the apoptosis signaling cascade.

In conclusion, the present study reveals that DATS causes apoptosis in human prostate cancer cells, but not in a normal prostate epithelial cell line (PrEC), which correlates with mitochondrial membrane potential collapse and ROS generation. The DATS-induced cell death is regulated by Bax/Bak but independent of Bcl-2 and Bcl-xL.

We thank Dr. Stanley J. Korsmeyer for the generous gifts of pSFFV–Bcl-2, pSFFV–Bcl-xL, and pSFFV-neo plasmids and the MEFs derived from WT, Bax, and/or Bak knock-out mice and Dr. Ming-Fong Lin for the generous gift of LNCaP-C81 cells.