Increasing evidences indicate that CXCR4/CXCL12 signaling pathway plays a pivotal role in the process of distant site metastasis that accounts for more than 90% of prostate cancer related deaths in patients. Thus, novel drugs that can downregulate CXCR4/CXCL12 axis have a great potential in the treatment of metastatic prostate cancer. In this report, we tested an agent, ursolic acid (UA) for its ability to modulate CXCR4 expression in prostate cancer cell lines and inhibit metastasis in vivo in transgenic adenocarcinoma of mouse prostate (TRAMP) model. We observed that UA downregulated the expression of CXCR4 in prostate cancer cells irrespective of their HER2 status in a dose- and time-dependent manner. Neither proteasome inhibitor nor lysosomal stabilization had any effect on UA-induced decrease in CXCR4 expression. When investigated for the molecular mechanisms, it was observed that the downregulation of CXCR4 was due to transcriptional regulation as indicated by downregulation of mRNA expression, inhibition of NF-κB activation and modulation of chromatin immunoprecipitation activity. Suppression of CXCR4 expression by UA further correlated with the inhibition of CXCL12-induced migration and invasion in prostate cancer cells. Finally, we also found that UA treatment can inhibit metastasis of prostate cancer to distal organs, including lung and liver and suppress CXCR4 expression levels in the prostate tissues of TRAMP mice. Overall, our experimental findings suggest that UA exerts its antimetastatic effects through the suppression of CXCR4 expression in prostate cancer both in vitro and in vivo.
The process of metastasis, driven by chemokines and its receptors is a major cause of death in patients with prostate cancer.1, 2 In addition to chemokines, a number of other molecules have been linked with cancer metastasis, including matrix metalloproteases (MMPs), vascular endothelial growth factor (VEGF), platelet-derived growth factor (PDGF) and transforming growth factor β (TGF-β).3 Chemokines are a super family of small, cytokine-like proteins that regulate adhesion and transendothelial migration of leukocytes especially during immune and inflammatory reactions.4 Based on the position of the first two conserved cysteine residues, the chemokines can be classified into four subfamilies, CXC, CC, C, and CX3C, and exert their biological effects through selective membrane-bound G protein-coupled receptors.2, 5 Among the large family of chemokines and their receptors, the most well studied is CXCR4/CXCL12 signaling cascade, which is expressed by various types of tumor cells, including breast6 and prostate,7 and plays a critical role in determining the metastatic destination of tumor cells.8 Moreover, CXCR4/CXCL12 has also been linked with leukocyte trafficking,9 B cell lymphopoiesis10 and HIV invasion of host cells.11
The effects of the CXCR4/CXCL12 axis on prostate cancer are considered to be multifactorial and it has been implicated in both the homing of tumor cells to specific organs, as well as the growth of tumor cells at specific locations, which are most likely mediated by the effects of CXCR4 on proliferation, adhesion and invasion.12–14 Moreover, neutralization of CXCR4 function with specific CXCR4 antibody has been found to reduce the metastatic load and also inhibit prostate tumor growth when cells were injected directly into the tibia.7, 15 Also, as CXCR4 expression has been correlated with metastasis and poor overall survival rate in patients with prostate cancer,16 CXCR4 is considered as a potential therapeutic target for inhibiting prostate cancer metastasis.14
In the present report, we studied the effect of ursolic acid (UA), (3β-hydroxy-urs-12-en-28-oic-acid), a pentacyclic triterpenoid derived from rosemary (Rosemarinus officinalis), apples, cranberries, Eriobotrya japonica, Calluna vulgaris, basil (Ocimum sanctum) and Eugenia jumbolana17–20 as a novel regulator of CXCR4/CXCL12 signaling axis. Although UA has been reported to exhibit significant antiproliferative, proapoptotic, antiangiogenic and chemopreventing activities in different tumor cells and animal models,20–23 but its effect on the malignant progression and metastasis of prostate cancer has not been tested before in animal models. The transgenic adenocarcinoma of mouse prostate (TRAMP) model closely mimics human prostate growth and progression and has been used extensively as a model for intervention studies with chemopreventive agents in prostate cancer.24, 25 In this model, expression of the SV40 large and small T oncoproteins is directed to the secretory cells of the prostate epithelium, under the control of the minimal rat probasin promoter.26, 27 The T antigen abrogates p53 and Rb function to spontaneously develop the progressive stages of prostate cancer from prostatic intraepithelial neoplasia (PIN) to the advanced stages of adenocarcinoma that subsequently leads to metastatic dissemination of tumor cells to distant organs.15
Because CXCR4 is known to mediate proliferation, invasion and metastasis of tumor cells, we postulated that UA may modulate the expression of CXCR4 and inhibit tumor metastasis in vivo. Our results indicate that UA can downregulate CXCR4 expression and CXCL12-induced migration and invasion in various prostate cancer cell lines through inhibition of NF-κB activation. We also report for the first time that UA treatment can reduce the metastasis of prostate cancer to distal organs, including lung, liver and gastric tract through the modulation of CXCR4 expression levels in TRAMP model, thereby suggesting that UA could be useful for intervention of prostate cancer diagnosed at relatively advanced stages.
Material and Methods
UA (98% pure) was purchased from Guangxi Changzhou Natural Products Development Company Ltd (China). UA was dissolved in dimethylsulfoxide as a 100 mM stock solution and stored at 4°C. Further dilution was done in cell culture medium. 3-(4,5-dimethylthiazol-2-Yl)-2,5-diphenyltetrazolium bromide (MTT), dimethyl formamide (DMF), Tris, glycine, NaCl, SDS, N-Acetyl-L-leucyl-L-leucyl-L-norleucinal (ALLN), chloroquine and β-actin antibody were purchased from Sigma-Aldrich (St. Louis, MO). RPMI 1640, fetal bovine serum (FBS), 0.4% trypan blue vital stain, antibiotic-antimycotic mixture and HRP-conjugated secondary antibodies were obtained from Invitrogen (Carlsbad, CA). Antibodies against CXCR4 and HER2 were obtained from Abcam (Cambridge, MA). CXCL12 was purchased from ProSpec-Tany TechnoGene (Rehovot, Israel).
Human androgen independent DU145 and androgen dependent LNCaP prostate cancer cell lines were kindly provided by Prof. Shazib Pervaiz of our university. Wild type murine embryonic fibroblasts (MEFs) and PC-3 cells were kindly provided by Prof. Bharat B. Aggarwal, M.D., Anderson Cancer Center, Houston. Prostate cancer cells were cultured in Rosewell Park Memorial Institute (RPMI 1640) medium containing 10% FBS and 1× antibiotic-antimycotic solution and were maintained at 37°C in a 5% CO2 incubator. MEFs cells were cultured in Dulbecco's Modified Eagle Medium (DMEM) containing 1× antibiotic-antimycotic solution with 10% FBS.
Western blot analysis
Untreated and UA-treated prostate cancer cells were lysed in lysis buffer [20 mM Tris (pH 7.4), 250 mM NaCl, 2 mM EDTA (pH 8.0), 0.1% Triton X-100, 0.01 mg/mL aprotinin, 0.005 mg/mL leupeptin, 0.4 mM PMSF, and 4 mM NaVO4]. TRAMP prostate tumor tissues (50 mg/mouse) were minced and incubated on ice for 30 min in 0.5 mL of ice-cold lysis buffer (10% NP-40; 5M, NaCl; 1M, HEPES; 0.1M, EGTA; 0.5M, EDTA; 0.1M, PMSF; 0.2M, sodium orthovanadate; 1 M, NaF; 2 μg/mL, aprotinin; 2 μg/mL, leupeptin) and homogenized for 5 min. Lysates were then spun at 14,000 rpm for 10 min to remove insoluble material and resolved on a 10% SDS gel. After electrophoresis, the proteins were electrotransferred to a nitrocellulose membrane, blocked with 5% nonfat milk to minimize nonspecific binding and probed with anti-CXCR4 or HER2 antibodies (1:3000) overnight at 4°C. The blot was washed, exposed to HRP-conjugated secondary antibodies for 1 h, and the CXCR4/HER2 expression was detected by chemiluminescence emission (ECL; GE Healthcare, Little Chalfont, Buckinghamshire, UK). The densitometric analysis of the scanned blots was done using Image J software and the results are expressed as fold change relative to the control.
The cytotoxic effects of UA were determined by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) uptake method as described previously.28
Nuclear extract preparation
To determine NF-κB activation, nuclear extracts were prepared using a nuclear extraction kit (Active Motif, Carlsbad, CA). Briefly, cells were washed, collected in ice-cold PBS in the presence of phosphatase inhibitors, and then centrifuged at 300g for 5 min. Cell pellets were resuspended in a hypotonic buffer, treated with detergent and centrifuged at 14,000g for 30s. After collection of the cytoplasmic fraction, the nuclei were lysed, and nuclear proteins were solubilized in lysis buffer and protein concentrations were determined by the Bradford protein assay (Bio-Rad Laboratories, Hercupetins, CA).
NF-κB reporter assay
DU145 cells were plated in 96-well plates with 1 × 104 cells per well in 10% FBS containing RPMI medium. After overnight incubation, cells were transfected with NF-κB reporter plasmid linked to a luciferase gene or with the dominant-negative IκBα (IκBα-DN) plasmid, cotransfected with β-galactosidase plasmid (Promega, Madison, WI). NF-κB luciferase plasmid was obtained from Stratagene (La Jolla, CA). Transfections were done according to the manufacturer's protocols using FuGENE® 6 obtained from Roche (Indianapolis, IN). At 24 hr post-transfection, cells were treated with UA for indicated time points and then washed and lysed in luciferase lysis buffer (Promega). Luciferase activity was measured by Tecan plate reader (Durham, NC) by using a luciferase assay kit (Promega) and was normalized to β-galactosidase activity. All luciferase experiments were done in triplicate and repeated three or more times.
NF-κB DNA-binding activity assay
NF-κB DNA-binding activity was analyzed using the TransAM NF-κB p65 transcription factor assay kit (Active Motif, Carlsbad, CA), following the manufacturer's instructions. Briefly, nuclear extracts (20 μg) from UA-treated cells were incubated in a 96-well plate precoated with oligonucleotide containing the NF-κB consensus-binding sequence 5′-GGGACTTTCC-3′ and incubated at room temperature for 1 hr with continuous shaking at 100 rpm. The plates were washed, incubated with p65 primary antibody for 1 hr at room temperature. After the final incubation with secondary antibody, the wells were then washed with washing buffer, and color was developed by addition of substrate solution. The enzymatic product was measured at 450 nm with a microplate reader (Tecan Systems, San Jose, CA).
RNA extraction and PCR analysis
Total RNA isolation was performed using TRIZOL® reagent (Invitrogen) according to the manufacturer's instructions. The relative expression of CXCR4 was analyzed using QIAGEN OneStep RT-PCR kit with GAPDH as an internal control. The RT-PCR reaction mixture contained 10 μL of 5× QIAGEN OneStep RT-PCR buffer, 1 μg of total RNA, 0.6 μM each of forward and reverse primers, 2 μL of dNTP mix and 2 μL of QIAGEN OneStep RT-PCR enzyme mix in a final volume of 50 μL. The reaction was performed at 50°C for 30 min, 95°C for 5 min, 95°C for 1 min, 58°C for 1 min and 72°C for 1 min for 33 cycles with a final extension at 72°C for 10 min. PCR products were run on 1% agarose gel containing 1× GelRed nucleic acid gel stain from Biotium (Hayward, CA). Stained bands were visualized under UV light and photographed. The following forward and reverse primer set was used: CXCR4, 5′-GAAGCTCTTGGTGAAAAGG-3′ and 5′-GAGTCGATGCTGATCCCAAT-3′; GAPDH, 5′-TGATGACATCAAGAAGGTGGTGAAG-3′ and 5′-TGATGACATCAAGAAGGTGGTGAAG-3′. For real time PCR, 1 μg of total RNA was transcribed to generate cDNA as described previously.29 For a 50 μL reaction, 10 μL of RT product was mixed with 1× TaqMan® Universal PCR Master mix, 2.5 μL of 20× TaqMan probes for CXCR4, 2.5 μL of 20× GAPDH TaqMan probe as the endogenous control. Relative gene expression was obtained after normalization with endogenous GAPDH and determination of the difference in threshold cycle (Ct) between treated and untreated cells using 2-ΔΔCt method. Primers and probes for human CXCR4, and GAPDH were purchased as assays-on-demand kits (Applied Biosystems).
Chromatin immunoprecipitation (ChIP) assay
DU145 cells were processed for the ChIP assay as per the protocol described previously.28, 30 The antibody used for the ChIP was NF-kB (p65) from Santa Cruz Biotechnology (SantaCruz, CA). The sequence for human CXCR4 gene promoter was follows: sense primer, 5′-ACAGAGAGACGCGTTCCTAG-3′ and antisense primer, 5′-AGCCCAGGGGACCC TGCTG-3′. The PCR products were analyzed on 2% agarose gel electrophoresis and documented.
Wound Healing Assay
DU145 and PC-3 cells were treated with UA for indiacte time points. Before plating the cells, two parallel lines were drawn at the underside of the wells, to serve as fiducial marks demarcating the wound areas to be analyzed. Before inflicting the wound, the cells should be fully confluent. The growth medium was aspirated off and replaced by calcium-free PBS to prevent killing of the cells at the edge of the wound by exposure to high calcium concentrations before two parallel scratch wounds were made perpendicular to the marker lines with a sterile 1000-μL automated pipette tip. Thereafter, the calcium-free medium was then changed to medium with or without UA. After incubation for 12 hr with UA, the growth medium was then changed to basal medium with CXCL12. Twenty-four hours later, the wounds were observed using bright field microscopy and multiple images were taken at areas flanking the intersections of the wound and the marker lines at the start and end of the experiment. Gap distance of the wound was measured at three different sites using Photoshop software, and the data were normalized to the average of the control. Graphs were plotted against the percentage of migration distance the cells moved before and after treatment, normalized to control.
The in vitro invasion assay was performed using Bio-Coat Matrigel invasion assay system (BD Biosciences, San Jose, CA), according to the manufacturer's instructions. DU145 and PC-3 cells (2 × 105 cells) were suspended in serum-free RPMI media and seeded into the Matrigel transwell chambers consisting of polycarbonate membranes with 8-μm pores. After preincubation with or without UA for 12 hr, the transwell chambers were then placed into appropriate wells of a 24-well plate in which either the basal medium only or basal medium containing CXCL12 had been added. After incubation for 24 hr, the upper surfaces of the transwell chambers were wiped with cotton swabs and the invading cells were fixed and stained with crystal violet solution. The invading cell numbers were counted in five randomly selected microscope fields (×200).
In vivo metastasis study
Animal experiments were conducted in accordance with Singapore NACLAR guidelines (Law as of November 2004) for laboratory animal use and care. Briefly, 4-week-old tramp female mice were purchased from The Jackson Laboratory and were mated with C57BL/6 male at NUS CARE (Singapore). The pups were genotyped at 3 weeks of age and the pups that were positive for the transgene were used for the experiments. The inbred male TRAMP mice were fed with normal diet until they reached 24 weeks. All mice were weighed before start of experiment. The mice were then randomized into the following treatment and control group (n = 5). Twenty-four week male TRAMP mice were fed with 1% w/w UA enriched diet for 12 weeks. Control group mice received normal diet. At the end of treatment period, animals were anaesthetized by CO2 inhalation. Blood samples collected by cardiac puncture were kept at 4°C overnight and centrifuged at 10,000 rpm for 20 min to separate serum, which was stored in aliquots at −80°C. Prostate gland was microdisected from the seminal vesicles under a stereomicroscope, fixed in 10% phosphate buffered formalin for immunohistochemistry analysis. Lung, liver and GI tract were collected, washed thoroughly in ice cold PBS and scored for metastatic colonies. Macroscopic examination of the organs and scoring for metastatic colonies were done using a stereomicroscope. Colonies were counted in both the treatment and control group.
At the end of treatment period, prostate tumor tissue was fixed with 10% phosphate buffered formalin, processed and embedded in paraffin. Tissue sections, 5 μm size, were cut and deparafinized as described previously.29 Sections were incubated with primary antibody anti-CXCR4 at 1:75 dilutions. Negative control sections were incubated in 2% bovine serum albumin. Immunohistochemical analysis was performed according to the manufacturer's instructions using DAKO LSAB kit. Images were taken using Olympus BX51 microscope.
Enzyme-linked immunosorbent assay (ELISA)
The effect of UA on CXCR4 levels in TRAMP prostate tumor tissue extracts was determined using CXCR4 ELISA kit (USCN Life Science, USA) according to manufacturer instructions. Briefly, tumor samples were thoroughly rinsed in ice-cold PBS to remove excess blood, weighed, cut the tissue into small pieces and kept on ice. Add 500 μL of PBS containing protease inhibitor to tissue fragments and homogenized using a tissue homogenizer on ice. The resulting tissue extract was sonicated for 30 sec, 10 sec pulse and subjected to two freeze thaw cycle to ensure complete breakdown of cells. The homogenate was centrifuged for 10 min at 10,000g. The supernatant was taken, protein concentration was determined using Bradford reagent and assayed immediately. A total of 150 μg of the protein was taken for ELISA assay and the sample CXCR4 concentration was determined by comparing to standard.
The experiments were carried out in triplicates and repeated twice. The significance of differences between groups was evaluated by Student's t-test and a p value of less than 0.05 was considered statistically significant.
Our study was designed to investigate the effect of UA (with structure shown in Fig. 1a) on both CXCR4 expression and invasion and metastasis in prostate cancer cells and TRAMP mice model.
UA suppresses the expression of CXCR4 protein in various prostate cancer cells
Several lines of evidence implicate the role of CXCR4 in prostate cancer metastasis.7, 12, 13 Hence, we first investigated the expression level of CXCR4 and HER2 in three different prostate tumor cell lines, namely PC-3, DU145 and LNCaP. As shown in Figure 1b, it is clearly evident that expression of CXCR4 in LNCaP cells is almost tenfold less as compared to that in PC-3 and DU145 cells. Because, HER2 has been shown to induce the expression of CXCR4 in tumor cells,31 we also analyzed the expression levels of HER2 in these three prostate cancer cells. However, we found that there is no obvious relationship between HER2 and CXCR4 expression levels in two prostate cancer cell lines, as HER2 expression level was relatively lower in PC-3 cells that showed higher expression of CXCR4 and vice-versa for LNCaP cells. Interestingly, only, in DU145 cells, the expression level of CXCR4 correlated with endogenous expression of HER2 in these cells (Fig. 1b). Hence, we first decided to investigate the effect of UA on CXCR4 expression in detail in DU145 cells. When DU145 cells were incubated either with different concentrations of UA for 12 hr or with 50 μM of UA for different times, UA suppressed the expression of CXCR4 in a dose- (Fig. 1c) and time- (Fig. 1d) dependent manner. The exposure of cells to 50 μM UA for 12 hr significantly inhibited the CXCR4 expression as evident by time kinetics study in Figure 1d. This downregulation was not due to decrease in cell viability as approximately 90% of cells were viable under these conditions (data not shown). Since HER2 has been reported to regulate the expression of CXCR4 by stimulating CXCR4 translation and attenuating CXCR4 degradation,6 we next examined whether UA downregulates CXCR4 expression through modulation of HER2 expression in prostate cancer cells. For this, HER2-overexpressing DU145 cells were incubated with different concentrations of UA for 12 hr and then examined for HER2 expression by Western blot analysis using specific antibodies. We found that HER2 expression was partially affected after UA treatment in DU145 cells (Fig. 1e), thus suggesting that downregulation of CXCR4 expression by UA is not completely due to modulation of HER2 expression. To determine whether UA is preferentially more toxic to tumor cells than normal cells, we treated DU145 and normal MEFs with increasing concentrations of UA for 24 hr. We observed that UA significantly decreased DU145 cell viability, although no significant toxicity was observed in normal fibroblast cells under identical conditions (Fig. 1f).
Downregulation of CXCR4 expression by UA is not mediated through its degradation
Previous studies have shown that CXCR4 undergoes ubiquitination at its lysine residue followed by degradation,32, 33 hence we next investigated the possibility that UA may enhance the rate of CXCR4 degradation via the activation of proteasomes. To determine this, we examined the ability of ALLN, a proteasome inhibitor, to block UA-induced degradation of CXCR4. DU145 cells were pretreated with ALLN for 1 hr before being exposed to UA. As shown in Figure 2a, ALLN had no effect on UA-induced degradation of CXCR4, suggesting that this is an unlikely basis for the suppression of CXCR4 expression by UA.
We also examined the ability of chloroquine, a lysosomal inhibitor, to block UA -induced degradation of CXCR4, as CXCR4 has been shown to undergo ligand-dependent lysosomal degradation.33 The cells were pretreated with chloroquine for 1 hr before exposure to UA. Our results showed that chloroquine at 200 μM only slightly prevented the degradation of CXCR4 (Fig. 2b), suggesting that this was arguably not the primary pathway for suppression of expression of CXCR4.
Downregulation of CXCR4 by UA occurs at the transcriptional level
As UA did not downregulate CXCR4 expression by enhancing its degradation, we investigated whether suppression occurs at the transcriptional level instead. Cells were treated with UA for different times and then extracted the mRNA for PCR analysis. As shown in Fig. 2c, UA induced downregulation of CXCR4 mRNA expression in a time dependent manner, with significant reduction observed as early as 6 hr after exposure.
UA suppresses constitutive activation of NF-κB in DU145 cells
The promoter of CXCR4 is known to contain several NF-κB binding sites.34 Thus, it is possible that UA manifests its effect on CXCR4 by suppressing NF-κB activation. The effect of UA on constitutive NF-κB activation in DU145 cells was determined by DNA binding assay. We found that treatment with UA suppressed NF-κB activation in a time-dependent manner (Fig. 3a). This result suggests that UA may downregulate CXCR4 expression through inhibition of NF-κB activation. However, DNA binding alone is not always associated with NF-κB-dependent gene transcription,35 suggesting that additional regulatory steps are involved. Subsequent results also indicated that UA inhibited NF-κB reporter activity in a time-dependent manner in DU145 cells (Fig. 3b).
UA inhibits binding of NF-κB to the CXCR4 promoter
Whether the downregulation of CXCR4 by UA in DU145 cells was due to suppression of NF-κB activation in vivo was examined by a ChIP assay targeting NF-κB binding in the CXCR4 promoter. We found that UA suppressed the NF-κB binding to the CXCR4 promoter (Fig. 3c), thereby indicating that UA inhibits CXCR4 expression by suppressing NF-κB binding to the CXCR4 promoter.
UA suppresses CXCL12-induced prostate cancer cells migration
Whether downregulation of CXCR4 by UA correlates with prostate tumor cells migration was examined using an in vitro wound healing assay. We found that both DU145 and PC-3 prostate cancer cells migrated faster under the influence of CXCL12 and this effect was abolished on treatment with UA (Figs. 4a and 4b).
UA inhibits CXCL12-induced prostate cancer cell invasion
We further elucidated the effect on UA on CXCL12-induced cell invasion and found using an in vitro invasion assay that treatment of UA suppressed CXCL12-induced invasion of both DU145 and PC-3 prostate cancer cells (Figs. 5a and 5b). We also observed that UA downregulated the expression of both mRNA (Fig. 5c) and protein (Fig. 5d) for CXCR4 in a time dependent manner in PC-3 cells.
UA inhibits metastasis of prostate tumor to lung, liver and GI tract in TRAMP mice model
CXCR4 over expression has been linked to prostate tumor progression and metastasis in mice tissues and clinical samples.16, 36, 37 To determine this, 24 week old TRAMP mice was fed with control or UA (1% w/w) containing diet for 12 weeks and then sacrificed at the end of 36 weeks period. As shown in Fig. 6a, mice fed with UA diet, showed significant reduction in number of metastatic colonies found in lung, liver and GI tract as compared to control mice, which showed higher number of metastatic colonies in these organs.
UA inhibits CXCR4 expression in TRAMP mice model
We next investigated the effect of UA on CXCR4 expression in UA treated TRAMP mice tissues. As shown in Fig. 6b, immunohistochemical analysis showed strong CXCR4 expression in TRAMP tumor tissues but not in nontransgenic C57Bl/6 prostate tissue. Immunohistochemical analysis revealed control prostate tissues were strongly reactive for CXCR4 staining and overexpression of CXCR4 correlated with increased malignancy as evidenced by increase in metastatic colonies in the control group animals (Fig. 6a). Interestingly, UA fed mice showed reduced staining thereby, indicating that UA can suppress the expression of CXCR4 in prostate tissues (Fig. 6b). Furthermore, we examined whether UA can inhibit CXCR4 expression in tumor tissues by Western blot analysis. Tissue extracts were prepared and probed with anti-CXCR4 antibody. Results obtained from immunoblot analysis clearly showed that UA significantly suppressed the expression of CXCR4 in prostate tumor tissues (Fig. 6c).
Additionally, the expression level of CXCR4 in tumor tissues was also determined by ELISA assay. As shown in Fig. 6d, a significant decrease in the levels of CXCR4 was observed in UA treated tissue homogenates. Collectively, all these data indicate that UA enriched diet suppressed the expression of CXCR4 in TRAMP tumor tissues which correlate well with its ability to reduce distal organ metastasis.
The goal of our study was to determine whether anticancer agent UA can suppress the expression and function of CXCR4, a chemokine receptor that has been closely linked with tumor cell proliferation, invasion and metastasis. Our results indicate for the first time that UA downregulated the expression of CXCR4 in prostate cancer cells, irrespective of their HER2 status. For example, UA was found to suppress CXCR4 expression in both high HER2 overexpressing, DU145 and low HER2 overexpressing PC-3 prostate cancer cells. Our results also showed that downregulation of CXCR4 did not occur through proteolytic degradation of the receptor but rather through downregulation of the transcript. Furthermore, suppression of receptor expression led to downregulation of migration and invasion induced by the ligand CXCL12 in prostate cancer cells and incidence of distant metastasis in transgenic TRAMP model.
The CXCR4 chemokine receptor has been found to be overexpressed in different tumors, including prostate cancer, breast cancer, ovarian cancer, glioma, pancreatic cancer, melanoma, cervical cancer, colorectal cancer, small-cell lung carcinoma, acute myeloid leukemia, chronic lymphoblastic leukemia (CLL), B-CLL and non-Hodgkin's lymphoma, as compared to normal cells which show little or no CXCR4 expression.6, 38, 39 Although it remains unclear, that what leads to the overexpression of CXCR4 in different tumor cells, prior studies point to genetic and microenvironmental factors.40 PAX3- and PAX7-FKHR gene fusion,41 mutations in the von Hippel Lindau tumor suppressor gene,42 hypoxia in the tumor microenvironment,43 NF-κB34 and inflammatory cytokine such as tumor necrosis factor alpha, and angiogenic molecule vascular endothelial growth factor44 have all been implicated in CXCR4 overexpression. Recently, the epidermal growth factor receptor, c-erbB2, and its encoding gene, HER2/neu, have also been implicated in the positive regulation of CXCR4 expression at the posttranscriptional level.45, 46 Given that CXCR4 has been linked with the metastasis of prostate cancer and CXCR4 expression has been correlated with poor prognosis and poor overall patient survival,16 CXCR4 appears an ideal therapeutic target for the investigation of novel therapeutic interventions for the prevention and treatment of metastatic prostate cancer.
Our results clearly indicate that UA suppressed CXCR4 expression in both high and low HER2 expressing prostate cancer cells, but had minimal effect on HER2 expression in DU145 cells. The ligand-dependent downregulation of the CXCR4 receptor by lysosomal degradation is well documented.6 Recent reports suggest that degradation involves atrophin-interacting protein (AIP)-4 mediated ubiquitination and degradation.33 However, our data indicate that UA does not downregulate the CXCR4 through this mechanism. As such, with downregulation of CXCR4 by UA arguably not occurring at the posttranslational level, we postulated that the inhibition of CXCR4 expression by this triterpene could occur at the transcriptional level. Indeed, we found that UA downregulated the expression of CXCR4 mRNA in prostate cancer cells as observed by PCR analysis.
UA has been previously reported to downregulate NF-κB activation in various tumor cells.21 Therefore, it is possible that downregulation of CXCR4 by UA occurs via the suppression of NF-κB, as binding site for this transcription factor has been identified in the proximal region of the CXCR4 promoter.34 Indeed, we found that inhibition of constitutive NF-κB activation by UA leads to downregulation of CXCR4 in DU145 cells. Whether mechanism(s) other than suppression of NF-κB activation are involved in downregulation of CXCR4 by UA, cannot currently be confirmed or ruled out. Furthermore, besides CXCR4, the activation of NF-κB also induces the expression of various molecules including cyclooxygenase-2, matrix metallopeptidase-9 and adhesion molecules such as intracellular adhesion molecule 1, vascular cell adhesion molecule 1 and endothelial-leukocyte adhesion molecule 1, all of which have been linked with cancer cell migration, invasion, and metastasis.47 Because UA can inhibit both DNA binding ability and transcriptional activation of NF-κB, as shown in our study, it is possible that UA can suppress the expression of other NF-κB regulated molecules as well in prostate cancer cells. We further investigated the effect of UA on CXCL12-induced migration and invasion of prostate cancer cells. We found that preincubation of cells with UA can also inhibit CXCL12-induced migration and invasion of both DU145 and PC-3 prostate cancer cells.
Using transgenic model, we further demonstrate that UA treatment can significantly downregulate expression of CXCR4, which in turn reduces the metastasis of prostate tumor to lung, liver and GI tract in TRAMP mice. In contrast, the control mice had a significantly higher incidence of metastasis, and also had metastatic lesions disseminating to multiple organs. The results that the CXCR4 expression in the mice tissues was suppressed by the administration of UA is consistent with the key role of CXCR4 in initiating the progression of clinical prostate cancer16, 36 and prostate cancer in the TRAMP mouse model.22, 37 Taken together, our findings indicate for the first time that UA after 12 weeks treatment period, might be able limit the process of metastasis in TRAMP model, and thus can form the basis of novel treatment regimen for patients with prostate cancer.
In summary, our data show for the first time that UA not only downregulates the expression of CXCR4, a key receptor involved in the cross-talk between tumor cells and its microenvironment, but can also limit malignant progression of prostate cancer in transgenic mouse model. Prior studies have already established that UA is potentially a nontoxic pharmacological agent and topical cosmetic preparations containing UA are already patented in Japan for the prevention of skin cancer.19, 20 Moreover, to the best of our knowledge, UA as such has never been tested in cancer patients before and hence its clinically relevant doses are not clear as yet. Thus, carefully designed double-blind, placebo-controlled clinical trials are needed to analyze the relevance of these observations to prostate cancer treatment in patients.
This work was supported by Grants from National Medical Research Council of Singapore [Grant R-184-000-168-275] to G.S., Cancer Science Institute of Singapore, Experimental Therapeutics I Program [Grant R-713-001-011-271] to A.K.; and Biomedical Research Council of Singapore [Grant R185-000-163-305] to L.L. Prof. Kam M. Hui was supported by grants from the Biomedical Research Council of Singapore, National Medical Research Council of Singapore, and the Singapore Millennium Foundation.