Bombesin Stimulates Nuclear Factor κB Activation and Expression of Proangiogenic Factors in Prostate Cancer Cells¹

Prostate cancer is the most commonly diagnosed form of cancer among men in the United States and is second only to lung cancer as a cause of cancer-related death. The American Cancer Society estimates that 220,900 new cases of prostate cancer will be diagnosed in 2003 and >28,000 men will die of the disease. Despite recent progress in prostate cancer detection and treatment, most of the deaths from prostate cancer occur in patients with androgen-insensitive metastatic disease (1).

Cancer metastasis is a complex multistep process involving tumor cell proliferation, angiogenesis, cell migration, invasion, adhesion, and growth in distant organs (2). An important early event in the development of the metastatic phenotype is the induction of genes regulating angiogenesis. Prostate cancers cells secrete a variety of proangiogenic molecules, including VEGF,³ IL-8, and extracellular matrix proteinases (3, 4, 5); however, the factors regulating the expression of proangiogenic genes remain largely undefined.

Coincident with progression from prostatic carcinoma in situ to metastatic disease is an increase in the number of cells exhibiting NE differentiation (6, 7). NE differentiation is characterized by the appearance of clusters of cells within the prostate tumor that: (a) express NE markers such as neuron-specific enolase; and (b) contain abundant secretory granules rich in neuropeptides such as calcitonin, calcitonin gene-related peptide (8), parathyroid hormone-related protein (9), and BBS-like peptides.

BBS is a 14-amino acid peptide originally isolated form the skin of the frog, Bombina bombina, and is a functional homologue to the mammalian peptide hormone, GRP. In humans, BBS and GRP bind with high affinities to the GRP-R, a member of the G protein-coupled receptor superfamily (10). Several clinical, histological, and experimental observations have indicated that BBS-like peptides and GRP-R play a role in the biology of prostate cancer. Logothetis and Hoosein (11) reported that 40% of patients with hormone-refractory androgen-independent prostate cancer had significantly elevated levels of BBS-like peptides in their serum. Several studies have shown expression of GRP and GRP-R by NE cells of human prostate cancer tissue and by prostate cancer-derived cell lines (12, 13). Additionally, BBS has been shown to stimulate the growth of both orthotopic and ectopic prostate cancer cell xenografts in athymic nude mice through a GRP-R-mediated mechanism (14, 15).

Although there is a strong positive correlation between the degree of NE differentiation and metastatic potential of prostate cancer, a mechanistic link between increased expression of neuropeptides, such as BBS-like peptides, and the proangiogenic gene expression associated with the development of a metastatic phenotype have not been established. Here, we report that BBS stimulation of GPR-R induces NFκB activation and up-regulation of IL-8 gene expression in DU-145 and PC-3 cells, and increased VEGF gene expression in PC-3 cells. We show that BBS treatment of PC-3 cells induced IκB degradation, NFκB translocation to the cell nucleus, increased NFκB binding to its DNA consensus sequence, increased IL-8 and VEGF mRNA expression, and increased protein secretion. BBS-stimulated IL-8 and VEGF expression and secretion were blocked by the proteasome inhibitor, MG-132. Finally, media collected from cultured PC-3 cells, after BBS treatment, stimulated an NFκB-dependent migration of HUVECs in vitro. These data suggest that BBS-like peptides and their cognate receptors may contribute to the increased metastatic potential of NE-differentiated androgen-insensitive prostate cancers through the stimulation of proangiogenic gene expression.

The human, androgen-independent PC-3 and DU145 prostate cancer cell lines were purchased from the American Type Culture Collection (Rockville, MD). PC-3 cells were maintained in either DMEM supplement with 10% FBS or in RPMI 1640 supplemented with l-glutamine (Mediatech, Inc., Herndon, VA), 10% heat-inactivated FBS (Hyclone Laboratories, Inc., Logan, UT), and 1 mm sodium pyruvate (Sigma-Aldrich Corp., St. Louis, MO), as described previously (16). DU-145 cells were cultured in MEM supplemented with 10% FBS, 1.5 g/liter sodium bicarbonate, 1 mm sodium pyruvate, 2 mm l-glutamine, and 0.1 mm nonessential amino acids.

PC-3 cells, growing in log-phase, were plated in 12-well plates at an initial density of 1 × 10⁴ cells/well in either DMEM or RPMI 1640 plus 10% FBS. One day after plating, the culture medium was removed and replaced with medium containing either no FBS or supplemented with 1, 2, 5, or 10% FBS. Two days after plating, the cells were treated with either BBS or vehicle (PBS and 0.1% BSA), and initial cell counts were performed (day 0). The number of PC-3 cells was determined at 24 h intervals for up to 5 days using both a Coulter Cell Counter and hemocytometer. Cell viability was assessed by trypan blue exclusion.

PC-3 cell s.c. xenografts were initiated by injecting 2 × 10⁶ cells into the right and left flanks of athymic nude mice. Two days after injection of the PC-3 cells, mice were randomized into two groups. One group received injections of BBS (20 μg/kg) in PBS containing 0.1% BSA at 8-h intervals for 5 weeks. The second group received injections of PBS plus 0.1% BSA (vehicle) on the same schedule. After 5 weeks, the mice were sacrificed, and tumors were excised and weighed.

Cells were plated at a density of 5 × 10⁴ cells/well in 24-well plates in 1 ml of medium supplemented with FBS and cultured for 48 h. For competition binding studies, cells were washed twice with binding buffer [medium plus 10 mm HEPES (pH 7.4) and 0.1% BSA] and then incubated (20 min at 37°C) in binding buffer containing [¹²⁵I]BBS (0.08 nm) and various concentrations of unlabeled competitor (1 pm to 10 μm). At the end of the incubation period, unbound radiolabeled BBS was removed by suction, and the cells were washed with ice-cold wash solution (PBS and 0.1% BSA) and lysed with 1 ml of 1 n NaOH. Nonspecific binding was defined as the amount of radioactivity detected in the presence of 10 μm unlabeled BBS. Best-fit curves were generated using nonlinear regression analyses of a one-site binding model with GraphPad Prism software (GraphPad, San Diego, CA).

Cells, grown on 25-mm glass coverslips, were washed with a physiological medium (KRH) containing NaCl (125 mm), KCl (5 mm), KH₂PO₄ (1.2 mm), MgSO₄ (1.2 mm), CaCl₂ (2 mm), glucose (6 mm), HEPES (25 mm; pH 7.4), and loaded with 2 μm Fura-2 AM (Molecular Probes, Eugene, OR) for 50 min at 25°C. The cells were treated with BBS (100 nm), and single cell changes in the concentration of free intracellular Ca²⁺ ([Ca²⁺]_i) were recorded using a Nikon Diaphot inverted microscope (Garden City, NY) and a CCD camera (Dage-MTI, Inc., Michigan City, IN). Data points were collected every 1–8 s from ∼35 cells/coverslip and processed using ImageMaster software. Data are presented as the mean change in [Ca²⁺]_i ± SE.

Total RNA was isolated from cells using Ultraspec reagent (Biotecx, Houston, TX). For Northern blot hybridization, 10 μg of total RNA was resolved on 1% agarose/formaldehyde gels by electrophoresis and transferred to Hybond-N+ membrane (Amersham Pharmacia Biotech, Buckinghamshire, England). Membranes were probed with either IL-8 or VEGF cDNA labeled with [α-³²P]dATP (Perkin-Elmer Life Sciences Inc., Boston, MA) using a random-priming DNA-labeling kit (Stratagene, Cedar Creek, TX). Prehybridization (30 min) and hybridization (3 h) were performed at 68°C in Express Hybridization Solution (Clontech, Palo Alto, CA). After hybridization, the blots were washed a final time at high stringency in a solution containing 0.1× SSC and 0.1% SDS for 30 min at 65°C. Specific hybridization was visualized by autoradiography using Bio-Max MR film (Kodak, Rochester, NY). To ensure RNA integrity and to confirm equal loading of lanes and transfer of RNA during blotting, the membranes were stripped and hybridized with a probe for 18S RNase.

PC-3 cell culture medium was collected from cells grown in 24-well plates after treatment with BBS for various lengths of time. After collection, the medium was spun at 800 × g for 3 min at 4°C to remove cell debris. The supernatant was either frozen at −20°C or assayed immediately for either VEGF or IL-8 levels using commercially available ELISA kits (R&D Systems, Inc., Minneapolis, MN).

PC-3 cells were cultured on glass coverslips. Before immunostaining for NFκB p65 subunit, the cells were treated with either the proteasome inhibitor, MG-132 (10 μm), or BBS (10 nm), or a combination of MG-132 and BBS for 30 min at 37°C, fixed with 4% paraformaldehyde (15 min), permeabilized with 0.3% Triton X-100 (10 min), and incubated in blocking solution (1% BSA in PBS; 20 min). After incubating the cells with anti-NFκB antiserum (1:100 dilution; San Cruz Biotechnology Inc., San Cruz, CA) for 1 h at room temperature, the cells were washed three times with PBS and incubated with a goat antirabbit IgG antibody labeled with Alexa 488 (Molecular Probes, Inc.; 1:2000; 30 min). Specific immunostaining was visualized with a Nikon Eclipse fluorescence microscope.

Nuclear extracts were prepared as described previously (17). An oligonucleotide (Stratagene, La Jolla, CA) of which the sequence corresponded to the NFκB binding site consensus sequence (5′-AGTTGAGGGGACTTTCCCAGGC-3′) was end-labeled with [γ-³²P]ATP using T4 polynucleotide kinase. EMSA reaction mixtures contained ∼50,000 cpm of ³²P-labeled oligonucleotide, 20 μg of nuclear protein extract, and 1.0 μg of synthetic polymer poly(dI-dc)·poly(dI-dc) (Amersham Pharmacia Biotech) in a final volume of 20 μl. Reaction mixtures were resolved on 4% nondenaturing PAGE at 200 V for 2 h. Gels were dried and visualized by autoradiography.

A 13-mm polycarbonate membrane filter (12-μm pore size) was coated with 100 μg of Matrigel (GFR) and placed in Boyden chemotaxis chambers (Neuroprobe, Cabin John, MD). The lower chamber was filled with RPMI 1640 collected from PC-3 cell cultures treated with BBS (10 nm), or MG-132, or a combination of BBS and MG-132 for 24 h. HUVECs (6 × 10⁴/well) were then added to the upper chamber. The cells were incubated in the chamber for 6 h at 37°C. At the end of the incubation period, the cells on the upper surface of the filter were mechanically removed by scraping, and the filters were fixed and stained with Diff-Quick (Baxter Scientific Products, McGaw Park, IL). HUVEC migration was quantified by counting cells on the bottom surface of the membrane using bright field illumination with a Nikon Eclipse 600 microscope equipped with a 16-square reticle.

Statistical analysis was performed using GraphPad InStat 3.0 (GraphPad Software, Inc.). Statistical significance was assumed if P ≤ 0.05.

Contrary to two widely cited reports (5, 18), we were unable to demonstrate that BBS affected PC-3 cell growth in vitro. Various cell culture conditions have been described for PC-3 cells, including growing them in either DMEM (5) or RPMI 1640 (16). We did not detect a significant BBS effect on PC-3 cell proliferation under any of the culture conditions tested (Fig. 1, A and B). We assessed the effects of BBS (100 nm) treatment on PC-3 cell growth using either DMEM supplemented with 2% FBS (5) or RPMI 1640 containing 0, 1, 5, or 10% FBS. When cells were cultured in DMEM supplemented with 2% FBS, BBS treatment tended to enhance cell growth, but the effect did not reach significance after multiple trials (Fig. 1,A). When cells were grown in RPMI 1640 in the absence of FBS, they grew slower than cells grown in medium supplemented with serum (Fig. 1,B); however, BBS treatment did not affect either serum-starved or serum-supplemented cultures. The rate of PC-3 cell proliferation was maximal in RPMI 1640 supplemented with 1% FBS (Fig. 1 B); increasing the serum concentration to 5 or 10% did not additionally increase the growth rate (data not shown).

Because our data contradicted growth studies reported previously, it was necessary to determine whether the PC-3 cells used in our laboratory expressed functional GRP receptors. We found that the lack of a growth response to BBS was not because of the absence of functional cell surface receptors. Competition binding studies revealed high-affinity binding sites for [¹²⁵I] BBS that were competed, in a dose-dependent manner, by both unlabeled BBS and the GRP-R-selective antagonist, BIM26226 [(d-F₅-Phe⁶, d-Ala¹¹)BBS (6, 7, 8, 9, 10, 11, 12, 13) OMe] (Ref. 19; Fig. 1,C). Agonist binding to GRP-R initiates the activation of intracellular signaling pathways involving heterotrimeric G proteins of the Gq subfamily (20), activation of phospholipase Cβ, generation of the second messenger inositol 1,4,5-triphosphate, and the release of Ca²⁺ from inositol 1,4,5-triphosphate-sensitive stores. To test whether BBS induced an increase in the concentration of [Ca²⁺]_i, we performed single cell florescence image analysis using the Ca²⁺-binding dye, Fura-2. BBS (10 nm) stimulated a time-dependent increase in [Ca²⁺]_I, which was blocked by pretreating the cells with BIM26226 (100 nm; Fig. 1 D), confirming the expression of functional GRP-R on PC-3 cells.

Consistent with previous studies (21, 22), we observed a dramatic stimulatory effect of BBS on PC-3 cell growth in vivo. PC-3 cell tumors were initiated by s.c. injection of 2 × 10⁶ cells into the right and left flanks of athymic nude mice. Two days after injection of the cells, the mice were randomized into two groups. The first group containing 9 mice received an injection of BBS (20 μg/kg) diluted in PBS supplemented with 0.1% BSA approximately every 8 h for 5 weeks. The second group of 7 mice received an injection of PBS containing 0.1% BSA on the same schedule. After 5 weeks, all of the mice in the BBS-treated group had developed at least 1 large tumor. Seven of the 9 mice in the BBS-treated group developed large tumors on both the right and left flanks (Fig. 2, A and B). By comparison, the average size and weight of the tumors collected from mice receiving injections of PBS plus 0.1% BSA were significantly smaller (Fig. 2, A and B), and only 1 of the 7 mice in the PBS-treated group developed two tumors. At the time of sacrifice, there was not a significant difference in the body weights of BBS-treated mice when compared with the control group.

The dichotomy between the effects of BBS treatment on PC-3 cell growth in culture versus its effects in vivo lead us to hypothesize that BBS promotes tumor growth in vivo by mechanisms other than the direct regulation of cell cycle progression. Because the growth of cancer cells in vivo requires neovascularization, which is regulated, in part, by proangiogenic factors such as IL-8 and VEGF, we reasoned that BBS might promote tumor growth through the regulation of these genes. To test this hypothesis, we assessed the effects of BBS treatment on steady-state levels of IL-8 and VEGF mRNA by Northern blotting.

Although IL-8 expression is regulated by various factors, including hypoxia, acidosis, nitric oxide, and cell density (23), a role for BBS-like peptides has not been established. We found that BBS treatment resulted in a time- and dose-dependent increase in the steady-state levels of IL-8 mRNA expression (Fig. 3). The levels of IL-8 mRNA increased at 1, 3, and 7 h after addition of BBS (100 nm final concentration) when compared with vehicle (PBS)-treated control cultures (Fig. 3,A). By 12 h after BBS stimulation, IL-8 mRNA levels returned to baseline. The increase in the level of IL-8 mRNA at 3 h was dependent on the concentration of BBS used to stimulate the cells. Increased IL-8 mRNA was detected in cells treated with 1 nm BBS, and reached a maximum at 10 and 100 nm BBS (Fig. 3 B).

VEGF is a member of a family of endothelial cell mitogens that regulate both physiological and pathological neovascularization (24). Similar to IL-8, we found that BBS stimulated a time- and dose-dependent increase in VEGF mRNA expression in PC-3 cells. An increase in the steady-state levels of VEGF mRNA was observed at 4 and 8 h after BBS stimulation when compared with vehicle (PBS)-treated control cultures (Fig. 3,C). By 12 h after BBS treatment, VEGF mRNA levels returned to control levels. After 3 h of stimulation with different concentrations of BBS, increased VEGF mRNA was detected in cells treated with 0.1, 1, 10, and 100 nm BBS, with a maximum increase between 10 and 100 nm (Fig. 3 D).

To assess whether the BBS-stimulated increases in IL-8 and VEGF mRNA were accompanied by an increase in the secretion of these proteins from PC-3 cells, we measured the levels of IL-8 and VEGF in culture medium over a time course after BBS treatment using ELISA. BBS stimulated a time-dependent increase in both IL-8 and VEGF release from PC-3 cells (Fig. 4, A and B, respectively).

It has been shown previously that inhibition of constitutively active NFκB in PC-3M cells, a metastatic derivative of the PC-3 cell line, caused a decrease in the levels of IL-8 and VEGF, suggesting that NFκB activity is required for the expression of these proangiogenic factors (4). NFκB is an inducible dimeric transcription factor that belongs to the Rel/NFκB family of proteins (25). Activation of NFκB involves its dissociation from the inhibitor protein, IκB, which is degraded by 26S proteasome complex, followed by its translocation to the nucleus where it binds to specific DNA sequences in the promoter regions of multiple genes (26, 27). To determine whether BBS stimulated an increase in NFκB activity, we first assessed whether BBS treatment of PC-3 cells induced the translocation of NFκB to the nucleus. Cells were treated with BBS for various lengths of time, fixed, and immunostained with an antibody to the p65 subunit of NFκB. By 30 min after BBS treatment, an increase in NFκB immunoreactivity was detected in the nuclei of PC-3 cells when compared with PBS-treated (control) cultures (Fig. 5,A). Similar to BBS-treated cells, cells treated with TNF-α, an established activator of NFκB, also showed an increase in p65 immunoreactivity in the nuclei (Fig. 5,A). Coincident with the translocation of NFκB to the cell nucleus was a BBS-dependent decrease in the level of IκB-α protein detected by immunoblot (Fig. 5 B).

To determine whether the translocation of NFκB immunoreactivity was accompanied by an increase in the level of NFκB binding activity, nuclear protein extracts were prepared from PC-3 cells after stimulation with BBS over a time course and analyzed by EMSA. BBS induced a time-dependent increase in NFκB binding activity by 30 min. The level of NFκB binding activity continued to increase up to 1 h, and then declined to near baseline levels at the 2, 4, and 6 h time points. Addition of an antibody specific for the p65 or p50 subunit of NFκB to the nuclear protein extracts of cells treated for 0.5 h with BBS (100 nm) retarded the mobility of the radiolabeled oligonucleotide, indicating the presence of both of these proteins in the DNA-binding complex. Furthermore, treatment of cells with the proteasome inhibitor, MG-132, which prevents the 26S proteasome-mediated destruction of IκB resulting in the inhibition of NFκB, blocked the BBS-stimulated increase in NFκB binding. Together, these data demonstrate that BBS activates NFκB in PC-3 cells, suggesting a possible mechanism for the up-regulation of NFκB activity associated with metastatic prostate cancer.

Consensus DNA sequences for NFκB binding have been identified in the 5′-untranslated region of the IL-8 promoter (23), and treatment of UV-irradiated skin cells with NFκB decoy oligodeoxynucleotides resulted in decreased expression of VEGF (28), suggesting a role for NFκB in the regulation of these genes. To determine whether NFκB mediated the BBS-stimulated increases in IL-8 and VEGF mRNA expression, PC-3 cells were pretreated with the proteasome inhibitor, MG-132, and stimulated over a time course with BBS. Inhibition of NFκB by MG-132 blocked the BBS-stimulated increase in NFκB binding (Fig. 5,B), the increase in steady-state levels of both IL-8 and VEGF mRNA expression (Fig. 6,A), and significantly reduced the amount of IL-8 and VEGF released into the culture medium (Fig. 6 B) 24 h after BBS treatment.

Angiogenesis involves proliferation and movement of endothelial cells toward the site of production of proangiogenic factors. VEGF and IL-8 are potent stimulators of cell migration. To determine whether media from PC-3 cells treated with BBS produced sufficient levels of chemoattractant to stimulate endothelial cell migration, culture medium was collected 24 h after treatment with BBS and used in an in vitro HUVEC migration assay (Fig. 7,A). After 6 h of exposure to medium from PC-3 cells treated with BBS, we observed an ∼43% increase in the number of HUVECs migrating through the membrane separating the upper and lower chambers of a Boyden chemotaxis chamber as compared with HUVECs exposed to medium from untreated cells (conditioned medium). The medium from cells treated with a combination of BBS and MG-132 for 24 h did not induce HUVEC migration above the level of untreated conditioned medium, and RPMI 1640 alone had little effect on migration (Fig. 7 A).

Previous studies have shown that activation of the receptors for BBS, IL-8, and VEGF results in an increase in [Ca²⁺]_i (29, 30, 31). To determine whether HUVECs were responsive to one or more of these peptides, we measured the change in [Ca²⁺]_i in HUVEC after treatment with each of the peptides (Fig. 7 B). Treatment with VEGF, but not BBS or IL-8 (data not shown), induced an increase in [Ca²⁺]_i in HUVEC, suggesting that VEGF, but not the other peptides, was the factor in the culture medium stimulating HUVEC migration. Together, these data demonstrate that BBS-treated PC-3 produce sufficient levels of chemoattractants to stimulate endothelial cell migration, and the production of the chemoattractant(s) is sensitive to the inhibition of proteasome activity by MG-132.

To determine whether GRP-R was coupled to the regulation the of proangiogenic gene expression in another androgen-insensitive prostate cancer cell line, we assessed the effect of BBS treatment on IL-8 and VEGF expression using DU-145 cells. Binding data confirmed results published previously showing expression of the GRP-R on DU-145 cells. In comparison with PC-3 cells, DU-145 cells express fewer binding sites for [¹²⁵I]BBS (Fig. 8,A; ∼2 fmol/mg protein versus 150 fmol/mg protein; Fig. 1,C). Like PC-3 cells, BBS induced an increase in [Ca²⁺]_i in DU-145 cells, which was blocked with the selective GRP-R antagonist BIM26226 (Fig. 8,B). BBS also stimulated a time- and dose-dependent increase in the steady-state levels of IL-8 mRNA (Fig. 8, C and D). However, we did not observe an effect of BBS treatment on the levels of VEGF mRNA. In contrast to PC-3 cells, DU-145 cells express a high baseline level of VEGF mRNA, which may affect our ability to detect a BBS-stimulated increase. Similar to PC-3 cells, EMSA revealed that after 30-min treatment with BBS (100 nm), NFκB binding activity increased in nuclear protein extracts from DU-145 cells. The stimulatory effect of BBS on NFκB binding was blocked with a peptide (SN50), which inhibits NFκB translocation to the nucleus (Fig. 8 E). Treatment of DU-145 cells with proteasome inhibitor MG-132 did not inhibit BBS-stimulated IL-8 expression (data not shown), suggesting that NFκB is not involved in the regulation of IL-8 expression in this cell line.

Our results demonstrate that BBS stimulates activation of NFκB and proangiogenic gene expression in two androgen-insensitive cell lines, PC-3 and DU-145. One androgen-sensitive cell line (LnCaP) was examined but was not responsive to treatment with BBS (data not shown). In PC-3 cells, BBS-stimulation of IL-8 and VEGF expression and secretion were regulated, in part, by a NFκB-dependent pathway.

Although BBS has been reported to stimulate PC-3 cell growth in vitro (5, 18), we were unable to reproduce the results of those previous studies. However, despite the lack of a growth effect in vitro, we show that BBS is a potent stimulator of PC-3 cell xenograft growth in nude mice. The dichotomy between our results in vitro versus those from in vivo experiments leads us to hypothesize that BBS promotes tumor growth through the regulation of proangiogenic gene expression.

Tumor growth and metastasis require the development of new blood vessels in and around the tumor. Formation of new blood vessels facilitates the delivery of nutrients and oxygen, as well as removal of waste products, thus enhancing tumor proliferation and sustenance. Angiogenesis is a dynamic process involving basement membrane degradation, endothelial cell proliferation and migration, and capillary tubule formation. It is regulated by a balance of both stimulatory and inhibitory factors released from tumor cells into the microenvironment. Prostate cancer cells produce a variety of proangiogenic factors, including VEGF, IL-8, basic fibroblast growth factor, plasminogen activator urokinase, and MMPs (32); however, little is known about the molecular mechanisms regulating the expression of these genes. The data presented in this report demonstrate that BBS and GRP-R play a role in the regulation of IL-8 and VEGF gene expression in the androgen-insensitive prostate cancer cell line, PC-3, and in the regulation of IL-8 expression in DU-145 cells.

Huang et al. (4) have reported a NFκB-dependent increase in the expression of IL-8 and VEGF in a highly metastatic derivative of the PC-3 cell line, called PC-3M. The PC-3M cell line was isolated from a nude mouse liver metastasis subsequent to intrasplenic injection of PC-3 cells (33). PC-3M cells constitutively overexpress IL-8 and VEGF. Overexpression of these proangiogenic genes was blocked in cells transfected with a dominant-negative IκB expression plasmid (34). Associated with the down-regulation of IL-8 and VEGF gene expression was a decrease in neoplastic angiogenesis in orthotopic tumors of PC-3M cells in vivo as well as decreased invasion of PC-3M cells through a Matrigel matrix in vitro. Huang et al. (4) clearly demonstrated that increased NFκB activity resulted in an increase in proangiogenic gene expression and subsequent metastatic potential of PC-3M cells (4); however, the molecular mechanisms leading to NFκB up-regulation in PC-3M cells were not identified. Our studies extend the observations of Huang et al. (4) by providing a potential molecular mechanism for the up-regulation of NFκB expression in metastatic prostate cancers, namely BBS activation of GRP-R. We have shown that BBS stimulates a rapid activation of NFκB in PC-3 cells, which was blocked by treating the cells with the proteasome inhibitor, MG-132. Associated with the inhibition of NFκB was an inhibition of both IL-8 and VEGF mRNA expression and secretion, suggesting that NFκB mediates the BBS regulation of these genes. Although BBS has been shown to regulate the expression of various transcription factors, including members of the AP-1 family (c-fos and jun; Ref. 35, 36) and c-myc (37), this is the first report demonstrating GRP-R coupling to NFκB-dependent signal transduction and regulation of IL-8 and VEGF expression.

Although IL-8 was originally discovered as a chemotactic factor for leukocytes, it has been shown recently to contribute to human cancer progression (23). Experiments with melanoma cells have shown that overexpression of recombinant IL-8 resulted in an increase in MMP-2 activity, increased growth rate, and increased metastasis (38). In patients with prostate cancer, serum concentrations of IL-8 were elevated when compared with patients with either normal prostates or BTH, and there was a positive correlation between IL-8 serum concentrations and increasing prostate cancer stage (39). Serum levels of IL-8 were found to be twice as high in patients with stages A-C prostate cancer and four times as high in patients with stage D prostate cancer, when compared with healthy individuals. Ferrer et al. (40), using immunohistochemical analysis, found that prostate cancer specimens stained positively for IL-8, whereas BTH and normal tissue exhibited little staining. Like IL-8, the levels of VEGF are also elevated in malignant prostate tissue when compared with either BTH or normal tissue (40, 41), and patients with metastatic prostate cancer have elevated levels of serum VEGF when compared with healthy subjects (42). Additionally, 40% of patients with hormone-refractory androgen-independent prostate cancer had significantly elevated levels of BBS-like peptides in their serum (11), and the levels of functional GRP receptors were elevated in prostatic intraepithelial neoplasia, invasive prostate carcinomas, and bone metastases (43). The data presented in this report suggest that the overexpression of GRP-R, IL-8, and VEGF are not a coincidence, but rather are likely mechanistically linked in the subset of GRP-R-positive prostate cancers.

BBS-like peptides regulate the expression of multiple genes and intracellular signaling pathways, which together may enhance prostate cancer growth and metastasis. BBS has been shown to regulate the expression of MMP, which plays a central role in the regulation of cell migration and invasion into and through the extracellular matrix. In recent studies using various prostate cancer-derived cell lines, including PC-3, BBS has been shown to induce the expression and activation of the extracellular matrix proteinases, plasminogen activator urokinase and MMP-9 (5). BBS-stimulated expression of these proteases is associated with enhanced cell motility and invasion into reconstituted (Matrigel) basement membranes (44). We have shown that BBS-treated PC-3 culture medium contains chemoattractants that stimulate HUVEC migration in vitro. HUVEC showed an increase in [Ca²⁺]_i in response to VEGF, but not BBS and IL-8, suggesting that, of the 3 peptides, VEGF is the major stimulator of cell migration in that assay. Our findings, together with results of others, indicate a role for BBS-like peptides and GRP-R in the regulation of angiogenesis and, thus, prostate cancer proliferation and, perhaps, metastasis. In this regard, the regulation of proangiogenic genes by BBS-like peptides and their cognate receptors begin to provide a mechanistic link between increasing NE differentiation and the increasing metastatic potential of androgen-insensitive prostate cancers.

We thank Dr. Javier Navarro for the plasmids containing the IL-8 cDNA and Dr. Fang Mei for developing the VEGF probe. We also thank Kirk L. Ives and Elzbeita Goluszko for their technical support, and Eileen Figueroa, Karen Martin, and Steve Schuenke for their assistance with the preparation of this manuscript.