Down-regulation of Androgen Receptor by 3,3′-Diindolylmethane Contributes to Inhibition of Cell Proliferation and Induction of Apoptosis in Both Hormone-Sensitive LNCaP and Insensitive C4-2B Prostate Cancer Cells

Prostate cancer is the most frequently diagnosed cancer and is the leading cause of cancer death in men in the U.S. with an estimated 234,460 new cases and 27,350 deaths in 2006 (1). Despite the initial efficacy of androgen deprivation therapy, most patients with advanced prostate cancer eventually develop resistance to this therapy and progress to hormone-refractory prostate cancer (HRPC), for which there is no curative therapy (2). Chemotherapy for prostate cancer has been used for a number of years (3), however, only limited improvement in survival was recently observed in HRPCs when treated with docetaxel-based combination treatment (4). Therefore, novel targeted therapeutic approaches must be developed for the treatment of HRPC.

During the progression of prostate cancers from androgen-sensitive status to an androgen-independent stage, prostate cancer cells still contain androgen receptor (AR), suggesting that AR signaling plays a critical role in the development and progression of prostate cancer (5). AR is a member of the steroid receptor superfamily and is a nuclear transcription factor. Upon binding to AR, androgen activates AR, which, in turn, interacts with androgen response elements (ARE) in the promoter of target genes including prostate-specific antigen (PSA), regulating the transcription of target genes. PSA is a clinically important marker used to monitor diagnosis, treatment response, prognosis, and progression in patients with prostate cancer (6). In addition to androgen, the activity of AR may be modified by molecules in other cell signaling pathways. It has been reported that Akt and nuclear factor κB (NF-κB) regulate the AR signaling pathway by phosphorylation of AR or transcriptional regulation of AR (7, 8). Akt specifically binds to AR and phosphorylates serines 213 and 791, thereby activating AR (7). Blocking the Akt pathway by a dominant-negative Akt or an inhibitor of Akt abrogates the HER-2/neu-induced AR signaling (7). These results suggest that Akt is an activator of AR required for androgen-independent survival and growth of prostate cancer cells mediated by HER-2/neu signaling. It has been known that there are NF-κB binding sites in the promoter of AR (8), suggesting that NF-κB may regulate the expression of AR. The activation of Akt and NF-κB has been involved in the progression of prostate cancer from androgen dependence to independence (9, 10). In HRPC, promiscuous function of AR, together with the activation of Akt and NF-κB pathways, promotes cancer cells to become resistant to androgen deprivation therapy (9–12). In addition, androgen is also known to produce oxidative stress resulting in the production of reactive oxygen species, that, in turn, activate NF-κB and contribute to the induction of tumor cell proliferation (13). Therefore, AR, Akt, and NF-κB could be potential targets for the treatment of prostate cancer, especially HRPC.

3,3′-Diindolylmethane (DIM), an in vivo dimeric product of indole-3-carbinol (I3C), exhibits potent antiproliferative activities against various cancers including prostate cancer (14–16). We have reported that I3C significantly induced apoptosis and inhibited NF-κB and Akt activation in breast and prostate cancer cells, suggesting that I3C could be a chemopreventive and/or therapeutic agent for breast and prostate cancers (17–19). In addition, DIM has been shown to be an androgen antagonist (20), suggesting that the growth-inhibitory effects of DIM on prostate cancer could be due to the inactivation of multiple signaling pathways including AR signaling. To enhance the effects of DIM, Anderton et al. reported a formulated DIM (B-DIM from BioResponse, Boulder, CO), which showed approximately 50% higher bioavailability in vivo (21). However, no studies have been reported to date to elucidate the effect and molecular mechanisms of action of B-DIM on prostate cancer cells, especially on HRPC cells. Moreover, B-DIM is currently undergoing phase I studies in our cancer center; thus, we sought to investigate the molecular effects of B-DIM on paired androgen-sensitive LNCaP and androgen-insensitive C4-2B (derived from LNCaP cells) prostate cancer cell lines. Here, we report that B-DIM is a potent agent for inducing proliferation inhibition and apoptotic cell death of both androgen-sensitive LNCaP and androgen-insensitive C4-2B prostate cancer cells. This effect of B-DIM was partly mediated through the down-regulation of AR, Akt, and NF-κB signaling pathways.

Cell lines, reagents, and antibodies. Human prostate cancer cell lines including LNCaP, C4-2B, PC-3, and PC-3 stably transfected with AR were maintained in RPMI 1640 (Invitrogen, Carlsbad, CA) with 10% fetal bovine serum (FBS), 100 units/mL of penicillin, and 100 μg/mL of streptomycin in a 5% CO₂ atmosphere at 37°C. B-DIM was generously provided by Dr. Michael Zeligs (BioResponse) and was dissolved in DMSO to make a 50 mmol/L stock solution. Dihydrotestosterone (DHT; Sigma, St. Louis, MO) was dissolved in ethanol to make 100 μmol/L stock solutions. Anti-AR (Santa Cruz Biotechnology, Santa Cruz, CA), anti-pAR Ser²¹³ (Imgenex, San Diego, CA), anti-PSA (Lab Vision, Fremont, CA), anti-Akt (Santa Cruz Biotechnology), anti-pAkt Ser⁴⁷³ (Cell Signaling, Danvers, MA), anti-NF-κB p65 (Upstate, Charlottesville, VA), and anti-β-actin (Sigma) primary antibodies were used for Western blot analysis or confocal microscopic study.

Cell proliferation inhibition studies by MTT assay. Human LNCaP and C4-2B prostate cancer cells were seeded in 96-well plates. After 24 hours, the cells were treated with 0.1, 1, 10, 25, and 50 μmol/L B-DIM for 48 to 72 hours. Control cells were treated with 0.1% DMSO (vehicle control). After treatment, the cells were incubated with MTT (0.5 mg/mL, Sigma) in medium at 37°C for 2 hours and then with isopropanol at room temperature for 1 hour. The spectrophotometric absorbance of the samples was determined by using Ultra Multifunctional Microplate Reader (Tecan, Durham, NC) at 595 nm.

Histone/DNA ELISA for detection of apoptosis. The Cell Death Detection ELISA Kit (Roche, Palo Alto, CA) was used to detect apoptosis in prostate cancer cells treated with B-DIM according to the manufacturer's protocol. Briefly, the cytoplasmic histone/DNA fragments from LNCaP and C4-2B cells treated with 0, 10, and 25 μmol/L B-DIM for 24, 48, and 72 hours were extracted and incubated in microtiter plate modules coated with antihistone antibody. Subsequently, the peroxidase-conjugated anti-DNA antibody was used for the detection of immobilized histone/DNA fragments, followed by color development with 2,2′-azinobis(3-ethylbenzothiazoline-6-sulfonic acid) substrate for peroxidase. The spectrophotometric absorbance of the samples was determined by using Ultra Multifunctional Microplate Reader (Tecan) at 405 nm.

Western blot analysis. LNCaP, C4-2B, PC-3, and PC-3 cells stably transfected with AR were cultured in RPMI 1640 with 10% FBS or 10% dextran-coated charcoal-stripped FBS (DCC-FBS). Cells were then treated with B-DIM at various concentrations for different time periods followed by treatment with and without DHT (0.1 and 1 nmol/L) for 2 hours. After treatment, cells were lysed and protein concentrations were then measured using bicinchoninic acid protein assay (Pierce, Rockford, IL). The proteins were subjected to SDS-PAGE and electrophoretically transferred to nitrocellulose membrane. The membranes were incubated with various primary antibodies, and subsequently incubated with secondary antibody conjugated with peroxidase. The signal was then detected using the chemiluminescent detection system (Pierce).

PSA concentration detection. LNCaP and C4-2B cells were grown in six-well plates in complete RPMI 1640. When cells were 60% confluent, the monolayers were washed with serum-free medium and maintained in serum-free and phenol red–free RPMI 1640 with or without 25 μmol/L of B-DIM treatment for 24 and 48 hours. The conditioned medium was then collected and the protein concentration in the conditioned medium was quantified. The conditioned medium with equal amounts of protein for each sample was subjected to PSA detection using Human PSA ELISA Kit (Anogen, Mississauga, Ontario, Canada) according to the manufacturer's protocol.

Real-time reverse transcription-PCR analysis. LNCaP, C4-2B, and AR stably transfected PC-3 cells were treated as described above. Total RNA was extracted using Trizol (Invitrogen) and purified by using RNeasy Mini Kit (Qiagen, Valencia, CA). Total RNA (2 μg) from each sample was subjected to reverse transcription using the SuperScript First-Strand cDNA Synthesis Kit (Invitrogen) and the cDNAs were subjected to real-time PCR analysis for AR and PSA expression. Real-time PCR reactions were carried out in SmartCycler II (Cepheid, Sunnyvale, CA). The primers for AR were as follows: 5′-AGCCATTGAGCCAGGTGTAG-3′ and 5′-CGTGTAAGTTGCGGAAGCC-3′. The primers for PSA were as follows: 5′-GTGGGTCCCGGTTGTCT-3′ and 5′-AGCCCAGCTCCCTGTCT-3′. PCR amplification efficiency and linearity for each gene including targeted and control genes were tested. Data was analyzed according to the comparative cycle threshold (Ct) method and were normalized by β-actin or glyceraldehyde-3-phosphate dehydrogenase expression in each sample. Melting curves for each PCR reaction were generated to ensure the purity of the amplification product.

NF-κB DNA-binding activity measurement. LNCaP and C4-2B cells were treated as described above. Nuclear extracts were prepared according to the method described by Chaturvedi et al. (22), and the protein concentration was measured. Nuclear protein was then subjected to electrophoretic mobility shift assay (EMSA). EMSA was done by incubating 4 μg of nuclear proteins with IRDye-700 labeled NF-κB oligonucleotide and 2 μg of poly(dI-dC) for 30 minutes at room temperature in the dark. The DNA-protein complex formed was separated from free oligonucleotide on an 8% native polyacrylamide gel followed by scanning with the Odyssey Imaging System (LI-COR, Lincoln, NE).

Immunofluorescence staining and confocal imaging. NCaP and C4-2B cells were plated on coverslips in each well of a six-well plate containing 10% DCC-FBS. Cells were then treated with B-DIM (10 and 25 μmol/L) for different time periods followed by incubation with and without DHT (0.1 and 1 nmol/L) for 2 hours. Cells were then fixed with 10% formalin for 10 minutes. Then, coverslips were rinsed with PBS, treated with 0.2% bovine serum albumin in PBS for 45 minutes and with 0.5% Triton X-100 in PBS for 10 minutes, and incubated with anti-AR monoclonal antibody (1:50; Santa Cruz Biotechnology) at 37°C for 2.5 hours in PBS with 0.5% Triton X-100. After washing with PBS, the cells were incubated with FITC-conjugated anti-mouse antibody (1:100; Molecular Probes, Eugene, OR) along with 0.1 μg/mL of 4′,6-diamidino-2-phenylindole (DAPI; Sigma) at 37°C for 1 hour and washed with PBS. Cell images were captured on a Zeiss 310 laser-scanning inverted confocal microscope system, using 63× 1.2 objective and 488/364 nm laser wavelengths to detect FITC and DAPI, respectively.

Transient transfection with Akt cDNAs and/or reporter constructs. pLNCX-Akt (wild-type Akt), pLNCX-Myr-Akt (constitutively activated Akt), pLNCX-Akt-K179M (dominant negative), and pLNCX (control empty vector) were generously provided by Dr. Sellers (Dana-Farber Cancer Institute, Boston, MA). NF-κB-Luc (Stratagene, La Jolla, CA) contains six repeated copies of the NF-κB DNA-binding site and a luciferase reporter gene. pSV-β-gal reporter vector (Promega, Madison, WI) transfection was used for normalization of transfection efficiency. The pLNCX-Akt, pLNCX-Myr-Akt, pLNCX-Akt-K179M, or pLNCX was transiently cotransfected with NF-κB-Luc and pSV-β-gal into LNCaP and C4-2B cells using ExGen 500 (Fermentas, Hanover, MD). After 5 hours, the transfected cells were washed and incubated overnight with complete RPMI 1640, followed by treatment with 50 μmol/L of B-DIM for 48 hours. Subsequently, the luciferase activities in the samples were measured by Steady-Glo Luciferase Assay System (Promega) and ULTRA Multifunctional Microplate Reader (Tecan). β-Galactosidase activities were measured using the β-Galactosidase Enzyme Assay System (Promega). The nuclear proteins from transfected cells were also extracted and subjected to measurement of NF-κB DNA-binding activity using the EMSA method described above. The protein expressions of AR, p-AR, PSA, Akt, p-Akt(Ser⁴⁷³), and NF-κB in transfected LNCaP and C4-2B cells treated with or without B-DIM were measured by Western blot analysis.

LNCaP and C4-2B cells were also transiently cotransfected with PSA-Luc promoter construct containing ARE and pSV-β-gal vector by ExGen 500 (Fermentas). PSA-Luc construct was generously provided by Dr. Charles Young (Mayo Clinic, Rochester, MN). After 5 hours, the transfected cells were washed and incubated overnight with complete RPMI 1640 followed by treatment with 10 and 25 μmol/L of B-DIM for 24 and 48 hours. Subsequently, the luciferase and β-galactosidase activities in the samples were measured as described above.

Animal studies. The severe combined immunodeficiency (SCID)-human prostate cancer model of experimental bone metastasis used for our study was described previously (23, 24). Briefly, male homozygous CB-17 scid/scid mice, aged 4 weeks, were purchased from Taconic Farms (Germantown, NY). Human male fetal bone tissue was obtained by a third-party, nonprofit organization (Advanced Bioscience Resources, Alameda, CA) and written informed consent was obtained from the donor, consistent with regulations issued by each state involved and the federal government. After 1 week of acclimatization, the mice were implanted with a single human fetal bone fragment as described previously (23, 24). Suspensions of C4-2B cells (1 × 10⁶ cells in a volume of 20 μL of RPMI 1640) were injected intraosseously by insertion of a 27-gauge needle through the mouse skin directly into the marrow surface of the previously implanted bone. The mice were divided into two groups: control and B-DIM treatment groups. In the B-DIM treatment group, the mice were treated with B-DIM (1 mg/d) by gavage every day for a total of 7 weeks. The volume of the bone tumor in each group was determined by weekly caliper measurements according to the formula ab²/2: where a, length; b, cross-sectional diameter.

Inhibition of cell proliferation by B-DIM. Prior to our molecular experiments, we tested the effect of androgen (DHT) on the proliferation of LNCaP and C4-2B cells. As expected, LNCaP cells, which are known to be sensitive to androgen, showed growth stimulation of 25% and 35% at 48 and 72 hours, respectively, after treatment with 1 nmol/L of DHT. In contrast, C4-2B cells, which are insensitive to androgen, showed minimal or no growth stimulation (data not shown). In our pilot studies, we tested the effects of DIM and B-DIM on cell growth and found no significant difference at similar concentrations in vitro (data not shown). However, because B-DIM is known to have 50% greater bioavailability in vivo and no molecular mechanism, especially in HRPC cells, has been determined, we tested the effects of B-DIM on cell proliferation in prostate cancer cells by MTT assay. We found that the treatment of LNCaP and C4-2B prostate cancer cells with B-DIM resulted in a dose- and time-dependent inhibition of cell proliferation (Fig. 1A and B) with maximal inhibition seen at 50 μmol/L, demonstrating a potent growth-inhibitory effect of B-DIM on both androgen-sensitive and -insensitive prostate cancer cells. Because the inhibition of cell proliferation by B-DIM could also be due to the induction of apoptosis, we next tested the apoptosis-inducing effects of B-DIM in prostate cancer cells.

Induction of apoptosis by B-DIM in prostate cancer cells. To investigate whether the growth-inhibitory effect of B-DIM is due to the induction of apoptosis, a histone/DNA ELISA was conducted in B-DIM-treated LNCaP and C4-2B prostate cancer cells. We observed a significant induction of apoptosis in both androgen-dependent and -independent prostate cancer cells (Fig. 1C and D). This induction of apoptosis was time-dependent and directly correlated with the inhibition of cell proliferation, suggesting that the growth-inhibitory activity of B-DIM was partly mediated through an increase in apoptotic cell death. These results are in direct agreement with those observed earlier in I3C-treated breast and prostate cancer cells (17–19). Because AR and PSA play critical roles in the initiation and progression of prostate cancer, we detected the effects of B-DIM on AR and PSA expression by reverse transcription-PCR and Western blot analysis.

Inhibition of AR and PSA expressions in prostate cancer cells. By Western blot analysis, we found that B-DIM significantly inhibited the expression levels of AR and PSA proteins, as well as the secretion of PSA in LNCaP prostate cancer cells in both time-dependent and dose-dependent manners (Fig. 2A). Similarly, B-DIM also showed significant time-dependent and dose-dependent inhibition of AR and PSA protein expression and PSA secretion in C4-2B prostate cancer cells (Fig. 2B). Furthermore, B-DIM abrogated the DHT-induced up-regulation of AR and PSA proteins in both LNCaP (Fig. 2A) and C4-2B cells (Fig. 2B).

In order to further investigate whether down-regulation of AR and PSA proteins by B-DIM is a transcriptional event, we examined the expression levels of both AR and PSA mRNA by real-time reverse transcription-PCR. The results showed that B-DIM significantly down-regulated the expression of both AR and PSA mRNA in the presence and absence of DHT in both LNCaP (Fig. 2C) and C4-2B cells (Fig. 2D), indicating that B-DIM could inhibit the basal and DHT-induced AR and PSA transcriptional activities. These results are consistent with B-DIM-regulated AR and PSA expression at the protein level, suggesting that B-DIM-regulated AR and PSA gene expressions are transcriptional events.

In order to further prove this point, we tested the effects of B-DIM on AR stable clones of PC-3 prostate cancer cells in which the expression of AR is driven by an artificial promoter. We did not find down-regulation of AR by B-DIM in AR stably transfected PC-3 cells (Fig. 3A). We also transfected PSA-Luc vector, which contains PSA promoter with ARE and luciferase reporter gene, into LNCaP and C4-2B cells and treated the cells with B-DIM. We found that luciferase activity was significantly induced after PSA-Luc transfection and 10 to 25 μmol/L of B-DIM treatment for 24 to 48 hours significantly decreased luciferase activity in PSA-Luc-transfected LNCaP and C4-2B cells (Fig. 3B and C). These results further showed that the down-regulation of AR and PSA expression by B-DIM is a transcriptional event. Because it has been reported that AR may be regulated by Akt and NF-κB (7, 8), we next tested the effects of B-DIM on Akt and NF-κB signaling.

Down-regulation of Akt by B-DIM leading to inhibition of NF-κB, AR, and PSA. By Western blot analysis, we found that B-DIM down-regulated the protein levels of p-Akt, nuclear NF-κB, AR, p-AR, and PSA in both LNCaP and C4-2B cells (Fig. 4). To further investigate the relationship between Akt, NF-κB, AR, and PSA, and the effects of B-DIM on these molecules, we cotransfected Akt cDNA and NF-κB-Luc into LNCaP and C4-2B cells. We found that p-Akt, nuclear NF-κB, AR, p-AR(Ser²¹³), and PSA were up-regulated after wild-type Akt and Myr Akt transfections in LNCaP and C4-2B cells. However, the up-regulation of p-Akt, nuclear NF-κB, AR, p-AR(Ser²¹³), and PSA by Akt transfection were significantly abrogated in both LNCaP and C4-2B cells (Fig. 4).

Using transfection and luciferase assays, we found that luciferase activity was significantly increased after cotransfection with NF-κB-Luc and wild-type Akt or Myr Akt in both LNCaP and C4-2B cells (Fig. 5A), suggesting the activation of NF-κB by Akt transfection. Moreover, B-DIM treatment significantly abrogated the up-regulation of luciferase activity caused by Akt transfection. Furthermore, we conducted EMSA to test NF-κB DNA binding activity in Akt-transfected cells. The results showed that NF-κB DNA binding activity was significantly increased by wild-type Akt and Myr Akt transfection and this was inhibited by B-DIM treatment (Fig. 5B), which is consistent with the data from luciferase assays. Moreover, we also found that B-DIM at 10 or 25 μmol/L significantly inhibited NF-κB DNA binding activity in the presence and absence of DHT in both LNCaP and C4-2B cells (Fig. 5C).

Combined with the alteration in AR as observed by Akt transfection studies, our results suggest that there could be a crosstalk between p-Akt, NF-κB, and AR, and that the inhibition of Akt activation by B-DIM could lead to the down-regulation of NF-κB, AR, and PSA. Because the functions of AR in the regulation of its target genes mainly occur in the nucleus, we next tested the localization of AR before and after B-DIM treatment.

Inhibition of AR nuclear translocation by B-DIM. By Western blot analysis, we found that B-DIM reduced AR protein levels in both the cytosol and nuclear extracts (Fig. 4). However, the AR protein level was much more down-regulated by B-DIM in the nucleus than in the cytosol, suggesting that B-DIM could inhibit AR nuclear translocation. Therefore, we conducted immunofluorescent staining and confocal imaging to examine the effect of B-DIM on AR nuclear translocation in LNCaP and C4-2B prostate cancer cells. We found that both LNCaP and C4-2B cells treated with 25 μmol/L of B-DIM for 24 hours showed much less AR staining in the nucleus compared with control cells (Fig. 6A and B). These results suggest that B-DIM significantly inhibited AR translocation into the nucleus and, in turn, down-regulated AR target genes including PSA, consistent with reverse transcription-PCR and Western blot data showing the down-regulation of PSA mRNA and protein levels after B-DIM treatment (Figs. 2 and 4).

Inhibition of C4-2B bone tumor growth in vivo by B-DIM. To test whether B-DIM has any antitumor activity in vivo, we conducted a pilot animal experiment as “proof-of-concept” using a SCID-human model of experimental prostate cancer bone metastasis. After injection of C4-2B prostate cancer cells into fetal bone that was previously implanted, we found significant tumor growth, and the bone was severely destroyed by the infiltrating prostate cancer cells as observed previously (24); however, the residual bone also showed active new bone formation (Fig. 6C). Importantly, we found that B-DIM significantly inhibited C4-2B bone tumor growth in this model of prostate cancer bone metastasis, demonstrating the inhibitory effect of B-DIM on prostate cancer bone metastasis (Fig. 6C). The body weight of mice in each group did not show any significant difference, suggesting no obvious toxicity due to B-DIM treatment.

Androgen exerts its biological effects by binding to AR and activating AR transcriptional activity, promoting the growth of prostate epithelial cells. Therefore, androgen and AR play important roles in the normal development and maintenance of the prostate (5). More importantly, AR signaling also plays critical roles in the development and progression of prostate cancer. In the early stage of prostate cancer, androgen deprivation therapy shows inhibition of cancer cell growth, suggesting that androgen still controls functions of AR signaling at this stage. However, in HRPC, androgen deprivation therapy fails, although AR expression is still maintained in most cases (5). Actually, 30% of HRPCs show amplification and overexpression of the AR gene, indicating that the AR signaling pathway is malfunctioning in HRPC (25, 26). Studies on the progression of prostate cancer have indicated that an increase in AR mRNA and protein is both necessary and sufficient to convert prostate cancer from a hormone-sensitive to a hormone-refractory stage (27), and that overexpressed AR linked to p21^waf1 silencing may be responsible for androgen independence and resistance to apoptosis (28). These studies suggest that overexpression of AR is an important factor for acquired resistance of prostate cancer to androgen deprivation therapy. Therefore, AR is a key target for the treatment of both early stage prostate cancer and HRPC, and the inactivation of AR expression should be an important approach for the successful treatment of HRPC. In this study, we found that B-DIM significantly inhibited cell proliferation, induced apoptosis, down-regulated the expression of AR mRNA and protein, and abrogated the activation of AR by DHT in both androgen-sensitive and -insensitive prostate cancer cells. These results suggest that cell proliferation inhibition and induction of apoptosis by B-DIM is partly mediated by the down-regulation of AR. Although the degree of alteration of AR expression is higher than that of cell growth inhibition by B-DIM at the same time point, we believe that these phenomena could be due to the fact that B-DIM may have caused alterations of AR and other gene expressions before we observed changes in the biological behaviors of cells such as cell proliferation. B-DIM also inhibited AR nuclear translocation and subsequent expression and secretion of PSA, one of the important markers for progression of prostate cancer. Moreover, B-DIM significantly inhibited C4-2B prostate cancer cell growth in the implanted bone in SCID-human models of prostate cancer bone metastasis. These results suggest that B-DIM could be a potent agent for the treatment of androgen-sensitive, but most importantly, androgen-insensitive prostate cancer.

Several cell signal transduction pathways have been involved in the progression of HRPC by the interaction with AR signaling (29–31). Among them, the Akt pathway is an important cell signaling pathway for the survival of prostate cancer cells (32). It is believed that increased AR activity is caused by a crosstalk between AR, phosphoinositide-3-kinase (PI3K)/Akt, and mitogen-activated protein kinase pathways (29), although the relationship between Akt and AR remains controversial. It has been reported that Akt directly phosphorylates AR at Ser²¹³ or indirectly interacts with AR through GSK3β and β-catenin, and then enhances AR transactivation, promoting the growth of prostate cancer cells (7, 30, 33). However, a recent study by Yang et al. showed that inhibition of the PI3K/Akt pathway by LY294002 could result in the activation of FOXO3a, which could then induce AR expression to protect prostate cancer cells from apoptosis caused by the inhibition of the PI3K/Akt pathway (34). Therefore, we believe that the inhibition of both Akt and AR signaling pathways could be a powerful approach for the treatment of both androgen-dependent prostate cancer and HRPC. In this study, we found that transfection of Akt caused the activation of NF-κB and increased AR expression and phosphorylation, suggesting that there could be a direct crosstalk between Akt, NF-κB, and AR. We also found that B-DIM inhibited both Akt activation and AR transactivation, suggesting that B-DIM could be a potent agent for the treatment of both androgen-dependent prostate cancer and HRPC.

It has been well known that the NF-κB pathway plays an important role in the control of cell growth, differentiation, apoptosis, inflammation, stress response, and many other physiological processes in cellular signaling. The NF-κB signaling pathway is also involved in the development and progression of prostate cancer. NF-κB is overexpressed in prostatic intraepithelial neoplasia and prostate adenocarcinoma (35). Constitutive activation of NF-κB has been found in androgen-independent prostate cancer cells, whereas less activity of NF-κB has been observed in androgen-dependent prostate cancer cells (36, 37). Like Akt and AR, the relationship between NF-κB and AR activation remains controversial. Palvimo et al. reported that elevated expression of NF-κB p65 repressed AR-mediated transactivation in a dose-dependent manner, whereas NF-κB p50 did not influence AR transactivation (38). However, other investigators show that IL-4-induced NF-κB is required for AR activation (31), and that there are NF-κB binding sites in the promoter of AR (8), suggesting that the activation of NF-κB could enhance AR transactivation. Therefore, inhibition of both NF-κB and AR could be another powerful approach to treat both androgen-dependent prostate cancer and HRPC. Indeed, in this study, we found that B-DIM inhibited NF-κB, AR, and PSA, resulting in the cell proliferation inhibition and apoptotic cell death in both androgen-sensitive and -insensitive prostate cancer cells. It has been reported that NF-κB activates PSA expression by direct binding to the enhancer of PSA (39). Therefore, the inhibition of PSA expression and secretion by B-DIM could be mediated through the down-regulation of both NF-κB and AR.

In HRPC, the failure of androgen deprivation therapy is believed to be due to the AR modifications including AR mutations, AR amplification, and ligand-independent activation of AR through crosstalk with other signaling pathways (25, 26, 40, 41). Growing evidence shows the critical role of AR activation by nonandrogens in the development of androgen-independent prostate cancer (7, 29–31). Therefore, it is important to discover other nonandrogen molecules which activate AR. By Akt transfection, we observed increased AR expression and phosphorylation accompanied with increased p-Akt, NF-κB, and PSA, suggesting a crosstalk between Akt, NF-κB, AR, and PSA (Fig. 6D). It has been reported that Akt regulates NF-κB activation through IKK phosphorylation (42) and that NF-κB may activate AR signaling (31). Therefore, we believe that in the crosstalk, p-Akt could activate NF-κB and, in turn, activate AR which transactivates PSA expression, promoting the growth of prostate cancer cells. In addition, AR could also be activated directly by activated Akt (7), whereas PSA could be up-regulated directly by NF-κB (39). More importantly, we found that B-DIM could inhibit Akt, NF-κB, AR, p-AR, and PSA in both androgen-sensitive and -insensitive prostate cancer cells, demonstrating its effects on the interruption of these crosstalks. Taken together, these results along with our findings on the inhibition of cell growth in vitro, antitumor activity in vivo, and the induction of apoptosis by B-DIM in both LNCaP and C4-2B prostate cancer cells, clearly suggest that B-DIM could be a promising nontoxic agent for the treatment of prostate cancer, especially HRPC, for which there is no curative therapy. However, further in-depth studies, including further animal experiments and clinical trials, are needed to fully appreciate the value of B-DIM in the fight against prostate cancer.

Grant support: Department of Defense (Prostate Cancer Research Program, DAMD17-03-1-0042 awarded to F.H. Sarkar) and the National Cancer Institute, NIH (5R01CA108535 to F.H. Sarkar).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.