|Home | About | Journals | Submit | Contact Us | Français|
Endothelin-1 (ET-1), produced by the prostate epithelia, likely plays an important role in the progression of prostate cancer. ET-1 can bind two receptor subtypes; generally, binding of the endothelin receptor A (ETA) induces a survival pathway, whereas binding of the endothelin receptor B (ETB) mediates clearance of circulating ET-1 as well as promotes apoptosis. In prostate carcinoma, hypermethylation of the ETB promoter results in repression of ETB expression, thereby eliminating the negative growth response that ET-1 binding elicits through this receptor. Therefore, activation of ETA exclusively provides a pathway for survival advantage. Our current studies examine the mechanisms by which activation of the ETA may allow growth/survival. ET-1 treatment of prostate tumor cells significantly decreased paclitaxel-induced apoptosis through activation of the ETA subtype. The anti-apoptotic effects of ET-1 are mediated, at least in part, through the Bcl-2 family. Although no significant changes in Bcl-2 expression occurred with ET-1 treatment, the pro-apoptotic family members Bad, Bax, and Bak all decreased significantly. Further analysis of the survival pathway demonstrated that phosphorylation of Akt occurs with ET-1 treatment in a time- and dose-dependent manner through phosphatidyinositol 3-kinase activation. These data support the combination of ETA antagonists and apoptosis-inducing therapies for prostate cancer treatment.
Prostate cancer is characterized by low rates of cell proliferation coupled with diminished rates of cell death . This pattern has made prostate cancer among the most resistant of malignancies to cytotoxic chemotherapeutic agents. Furthermore, the cornerstone of the management of advanced prostate cancer, androgen deprivation therapy, relies on the effective induction of apoptotic pathways. The emergence of androgen refractory prostate cancer, which leads to the lethal form of the disease, indicates that these cells likely have developed survival mechanisms to escape death.
The potent vasoconstrictor endothelin-1 (ET-1) has been implicated in prostate cancer disease progression [2–4]. ET-1 expression occurs in almost every human prostate cancer tissue studied [4,5]. Moreover, patients with metastatic prostate cancer have elevated levels of plasma ET-1 compared with patients with organ-confined cancer as well as healthy individuals . ET-1 binds to two receptor subtypes. Activation of the endothelin receptor A (ETA) can lead to induction of a survival pathway, whereas activation of the endothelin receptor B (ETB) can result in clearance of circulating ET-1 as well as in stimulation of apoptosis. However, the response to the binding of either receptor remains cell type-dependent. In prostate cancer, the expression of the endothelin receptors, ETA and ETB, is altered compared to the pattern seen in normal prostatic tissues [6,7]. The ETB, predominant on benign prostatic epithelial cells, has a much lower expression on prostate cancer cells, owing, at least in part, to frequent hypermethylation of the ETB gene, EDNRB . Increased ET-1 expression, coupled with the increased ETA expression that occurs with higher prostate tumor stage and grade, may produce a survival advantage for the prostate cancer cells. Indeed, in a phase II clinical trial of the ETA antagonist, atrasentan, there was a significant delay in time to disease progression compared to placebo in men with hormone refractory disease [9,10].
In studies of endothelial and stromal cell populations, ET-1 acting through the ETA inhibited apoptosis induced by a cytotoxic agent . Given that endothelin receptor expression in prostate cancer favors the ETA and the compelling results from the atrasentan clinical trials, it is our hypothesis that ET-1 can act as a survival factor for prostate cancer. Therefore, we studied ET-promoted survival in prostate cancer, and demonstrated that ET-1, acting through ETA and the phosphatidyinositol 3-kinase (PI3-kinase)/Akt pathway, inhibited paclitaxel-induced apoptosis.
Prostate cell lines DU145, PC3, LNCaP (American Type Culture Collection, Manassas, VA) PPC-1 , and TSU  were grown in RPMI 1640, and LAPC4 (gift from Dr. Robert Reiter, UCLA, Los Angeles, CA) cells were grown in Iscove's modified Dulbecco's medium supplemented with 10% FBS and penicillin/streptomycin.
Prostate cell lines were pretreated with 1.0 µM ABT-627 or A127722 (ETA antagonists; Abbott Laboratories, Abbott Park, IL), 1.0 µM RES-701 (ETB antagonist; American Peptide, Sunneyvale, CA), or A-192621 (ETB antagonist; Abbott Laboratories), 200 nM wortmannin (Sigma Chemical Co., St. Louis, MO), 10 µM LY294002 (Sigma Chemical Co.), or 20 µM PD98059 (Calbiochem, La Jolla, CA) for 1 hour prior to ET-1 treatment in serum-free medium. ET-1 (100 nM) was added followed by 100 nM paclitaxel (Bristol-Myers Squibb, Princeton, NJ) or an antibody to fas (10 ng/ml; Signal Transduction Laboratories, Lexington, KY) and the cells were incubated for 4 to 24 hours. The cells were scraped from the plates and pelleted by centrifugation (200g). The cell pellet was then lysed using apoptosis lysis reagent and centrifuged at 10,000g for 10 minutes. A spectrophotometric ELISA-based assay was used to quantify histone-associated DNA fragments present in the cell lysates according to the manufacturer's instructions (Roche Diagnostics, Indianapolis, IN).
Prostate cells were plated in 100-mm dishes and treated with ET-1 (0.1 nM–1.0 mM) for 5 minutes to 24 hours in the presence or absence of: 100 nM ABT-627 or A127722; PI3-kinase inhibitors, 200 nM wortmannin and 10 µM LY294002; MEK inhibitor, 20 µM PD98059; and p70 S6 kinase inhibitor, 5 nM rapamycin. The cells were lysed in 20 mM Tris-HCl buffer (pH 8.0) containing 10% glycerol, 1% Triton X-100, and 135 mM NaCl with fresh protease inhibitors. The proteins (40 µg) were separated by 10% or 12% SDS-PAGE and electrotransferred onto PVDF membranes. The membranes were blocked and incubated in primary antibody [phospho-Akt, Akt, phospho-p44/42 MAP kinase, BAD, 1:1000 dilution (NEB, Beverly, MA); Bcl-2, BclXL phospho-Raf, 1:500 dilution (Transduction Laboratories); Bax, Bak, caspase3, caspase 9, 1:200 dilution (Oncogene, Boston, MA)] in TBST (20 mM Tris-HCl, pH 8.0, 150 mM NaCl, 0.1% Tween 20) overnight at 4°C. After washing, the blots were incubated in secondary antibody (goat anti-mouse or goat anti-rabbit HRP, 1:20,000; Roche Diagnostics) for 1 hour and washed in TBST. Immunoreactive proteins were visualized by ECL (Amersham, Piscataway, NJ) and Kodak (Rochester, NY) XAR film. Densitometry was performed using a Molecular Imager FX phosphorimaging system with Quantity One 4.1 quantitation software (Bio-Rad, Hercules, CA). Band densities were normalized to corresponding β-actin bands. EC50 for ET-1-induced Akt phosphorylation was determined from the nonlinear regression analysis of the densitometry data using WinNonlin Pro Academic v4.0.1 (Mountain View, CA).
PPC-1 cells were grown in 100-mm tissue culture dishes (Becton Dickinson, San Jose, CA). When cells were sub-confluent, growth media were removed, cells were washed with RPMI, and serum-free media were added. Six groups were formed according to administered treatment: 1) serum-free media only; 2) ET-1 (10-7 M); 3) paclitaxel (10-7 M); 4) ET-1 for 1 hour followed by paclitaxel; 5) ABT-627 (10-6 M), followed by ET-1 for 1 hour, and then followed by paclitaxel; 6) A192621 (10-6 M), followed by ET-1 for 1 hour, and then followed by paclitaxel. Each group was treated for a total of 4 hours. In addition to paclitaxel, these studies were carried out using a fas antibody (10 ng/ml for 4 hours) to induce cell death with and without ET-1 treatment. Cells were collected in 24 hours and evaluated for apoptosis by Annexin V staining.
For flow cytometry using the Annexin V assay, cells were collected and double-stained with fluorescein isothiocyanate-conjugated Annexin V (PharMingen, San Diego, CA) and propidium iodide (PI) (Sigma Chemical Co.). Cells were counted and 105 cells for each condition (in 100 µl of Annexin V binding buffer) were placed in 5-ml round-bottom tubes (Becton Dickinson). Each condition was done in duplicate. Annexin V was added according to the manufacturer's recommendations. PI was used at a final concentration of 5 µg/ml. Annexin V-positive cells were considered apoptotic and their percentage of the total number of cells was calculated. Cells taking up vital dye PI were considered dead. Samples of 10,000 cells were analyzed by FACScan flow cytometer with LYSYS II software package (Becton Dickinson).
Treatment of the prostate cancer cell PPC-1 with paclitaxel induces apoptosis as demonstrated by an ELISA-based apoptosis assay as well as by flow cytometric analysis of Annexin V staining (Figures 1 and and2).2). In the presence of ET-1, however, there is a significant reduction in the amount of cell death. The addition of the selective ETA antagonist, A-127722, or its racemic compound, ABT-627, reversed the ability of ET-1 to inhibit apoptosis. The selective ETB antagonist, A192621, did not significantly affect the ability of ET-1 to inhibit apoptosis (Figure 2). The cells treated with A192621 + ET-1 + paclitaxel showed a slightly higher rate of apoptosis (31.63%) compared with ET-1 + paclitaxel (23.44%); however, this increase was not statistically significant (P = .558) and may be the result of partial nonspecific blockade of ETA by using 10 µM ETB antagonist in these studies. A similar pattern was seen in the other prostate cancer cell lines tested, all of which have no detectable ETB expression due to EDNRB methylation. The anti-apoptotic actions of ET-1 were also blocked by LY294002, an inhibitor of the PI-3 kinase pathway, but not by PD98059, an inhibitor of the MEK1 pathway (Figure 1). To confirm these results, prostate cancer cell lines were exposed to agonistic antibodies to fas, a potent inducer of apoptosis [14,15]. The addition of ET-1 significantly inhibited fas-induced apoptosis, whereas ETA antagonists blocked this effect (data not shown).
To understand the mechanisms whereby ET-1 inhibits apoptosis in prostate cancer cells, the expression of both anti-apoptotic and pro-apoptotic proteins was examined after exposure of PPC-1 cells to ET-1 for 0 to 6 hours. The expression of the anti-apoptotic proteins, Bcl-2 (Figure 3) and Bcl-xL (data not shown), was not altered by exposure to ET-1. However, the expression of the pro-apoptotic protein, Bad, was significantly reduced by ET-1, as were the pro-apoptotic proteins, Bax and Bak, although to a lesser degree than Bad (Figure 3). Density measurements of the immunoblot bands were obtained using a phosphorimaging system and normalized to corresponding β-actin bands. Densitometric analysis indicated a 3.2-fold reduction in Bak, a 14.3-fold reduction in Bax, and an undetectable expression of Bad after 6 hours of ET-1 treatment. Expression of caspase 3 did not change (Figure 3). In addition, expression of BclXL, caspase 9, phospho-p44/42 MAP kinase, and Raf remained constant in cells treated with ET-1 from 4 to 72 hours (data not shown).
The serine/threonine kinase Akt (protein kinase B) is a point of signaling convergence for growth and survival pathway signaling through PI3-kinase and PDK1. Once activated, the kinase Akt inhibits apoptosis by phosphorylating a variety of substrates including pro-apoptotic proteins Bad, forkhead transcription factors (FKHR), and glycogen synthase kinase (GSK-3) [16–18]. Phosphorylation of Akt also induces MDM2 activity, which leads to the degradation of p53 . In both a dose- and time-dependent fashion, ET-1 induced phosphorylation of Akt in the prostate cancer cell line PPC-1 (Figure 4A). Immunoblots were reprobed with β-actin (to confirm that equal loading was achieved) (Figure 4) and Akt antibody (to ensure that ET-1 treatment did not affect overall Akt expression levels) (data not shown). Densitometry was carried out on the immunoblots, and EC50 for ET-1 was calculated using WinNonlin software (Pro Academic v4.0.1; Mountain View). The EC50 was 220 pM for ET-1-induced Akt phosphorylation calculated from the dose response experiments. A 2.33-fold increase in Akt activation occurred within 30 minutes of ET-1 treatment and remained elevated for the duration of the experiment (3 hours). In both androgen-sensitive (LNCaP and LAPC-4) and androgen-insensitive cell lines (PC3 and PPC-1), ET-1 increased phosphorylation of Akt, whereas ET-1 did not produce a significant effect on Akt phosphorylation in the bladder cell line, TSU (Figure 4B).
Cell survival requires active repression of apoptosis through the inhibition of pro-apoptotic proteins or the induction of anti-apoptotic factors. Many cell survival signaling cascades trigger the activation of PI-3 kinase, leading to phosphorylation of Akt. Akt can regulate cell growth through its effects on the p70 S6 kinase pathway [20,21] as well as by negative regulation of p21WAF1 and p27KIP1 cell cycle inhibitors [22–25] and indirect effects on cyclin D1 and p53 [18,19]. To investigate which pathways were involved in the ET-1 activation of Akt in prostate cancer cells, we used inhibitors of PI3-kinase (wortmannin, LY294002), MEK1 (PD98059), and p70 S6 kinase (rapamycin) (Figure 5). Incubation with PI3-kinase inhibitors prior to ET-1 stimulation resulted in loss of Akt phosphorylation. The p70 S6 and MEK1 inhibitors, rapamycin and PD98059, had no effect on ET-1 induced phosphorylation of Akt, further indicating that ET-1 survival signaling occurs through a PI3-kinase pathway and not a p44/42 MAP kinase pathway and does not involve p70 S6 kinase signaling.
Following the initial description in prostate cancer [7,8], loss of ETB expression through hypermethylation of EDNRB has been reported in a variety of malignancies [26–31]. As the receptor responsible for ligand clearance and pathways counterregulatory to ETA signaling, a decrease in ETB expression results in unfettered ET-1/ETA activity. However, it remains unknown whether silencing of the ETB is crucial for carcinogenesis and/or disease progression. Although ETA and ETB are coupled to many of the same signaling molecules, ETB activation can uniquely mediate divergent responses such as vasodilation through nitric oxide production and apoptosis in opposition to the vasoconstriction and cell survival response of ETA signaling. This study demonstrated an anti-apoptotic effect of ET-1/ETA in prostate cancer cell lines that occurs through a PI3-kinase/Akt dependent pathway. Other studies have demonstrated that ET-1 may also signal through Erk1/2 to induce Akt phosphorylation and promote cell survival in other cell types [32–36]. The Erk1/2 pathway appears to be involved in ET-1-induced responses primarily in stromal cell populations or through ETB-mediated signaling. Similar to our findings, Del Bufalo et al.  demonstrated that ET-1 signaling through the ETA leads to activation of Akt through PI3-kinase in ovarian carcinoma cells. Clearly, ET-1 has diverse signaling cascades and cellular responses depending on cell type and receptor subtype expression. Interestingly, a recent study demonstrated a correlation between high levels of Akt phosphorylation and an increased probability for recurrence of prostate cancer . Inactivation of pro-apoptotic proteins through phosphorylation by Akt activation may represent a direct mechanism by which ET-1 promotes cell survival. Akt can also induce MDM2 activity, resulting in enhanced degradation of p53 and subsequent abrogation of p53-mediated gene expression . Indeed, ET-1 treatment of prostate cancer cells resulted in a decrease in the pro-apoptotic proteins Bad, Bax, and Bak, all of which are members of the Bcl-2 family of proteins whose expression is positively regulated by p53. The profile of expression of the ET axis in prostate cancer cells may therefore promote a unique survival pathway. Although we previously demonstrated that ET-1/ETA can act as a weak mitogen for prostate cancer cells , we believe that the inhibition of apoptosis through the upregulation of Akt activity accounts for the majority of the effects seen with ET-1 treatment. In this study, we saw a clear dose response with a half-maximal effect of ET-1 on Akt phosphorylation at a 220 pM concentration. These results are similar to the EC50 values of ET-1 on other functional responses in cells, including arachidonic acid release, Ca2+ mobilization, and glucose uptake ranging from 300 pM to 2.9 nM ET-1 [39–43]. We had found previously that circulating plasma ET was elevated in men with hormone refractory prostate cancer (13.2 pg/ml) compared with normal controls (5.1 pg/ml) . It is also likely that ET-1 concentrations in the circulation would be lower than local tissue concentrations wherein the prostate cancer cells are eliciting an autocrine mechanism of ET-1 signaling through the ETA. The identification of the ETA survival signaling mechanisms may lead to new insights into the function of ET-1 and its receptors in malignant prostate tissues, as well as provide a potential therapeutic target.
Although ET-1 is produced by a variety of malignancies, including those arising from breast, lung, colon, pancreas, kidney, and ovary cancers, many normal tissues also produce abundant ET-1. Indeed, endothelial cells perhaps make the majority of ET-1 found in the body. If the findings in prostate cancer extend to other malignancies expressing the ETA, ET-1 from any source may act as a survival factor for these malignancies. Thus, blockade of the ETA may enhance the efficacy of therapies based on cytotoxic agents as well as hormone ablation. For androgen-sensitive tissues, such as the prostate, androgen deprivation induces a wave of apoptosis. Androgen deprivation may also stimulate the expression of the ETA receptor. In canine and rat models, expression of the endothelin receptors increased following castration [44,45]. Increased ETA expression, coupled with the hypermethylation of the ETB promoter, may provide a mechanism whereby prostate cancer cells can survive androgen deprivation. The use of ETA antagonists at the initiation of androgen deprivation may potentiate prostate epithelial cell death by removing a survival pathway. Given the minimal toxicity profile of ETA antagonists, the combination of these agents with other therapeutic agents merits investigation.
The authors are very grateful to Moira Hitchens for the critical review of the manuscript and the members of the University of Pittsburgh flow cytometry core facility for assistance with the apoptosis studies.
1Support for the data analysis was provided by the Pharsight Academic License Program (for the WinNonlin software). This work was supported, in part, by the Mellam Family Foundation.