Abstract
We showed previously that angiotensin II activated the proliferation of prostate cancer cells and that angiotensin II receptor blockers (ARB) could inhibit it. Here, we investigated whether angiotensin II exerts mitogenic effects on the cross-talk between stromal and cancer cells and whether an ARB can inhibit tumor growth through actions on stromal cells. Cell proliferation and interleukin-6 secretion of prostate stromal PrSC cells stimulated with angiotensin II, tumor necrosis factor-α, or epidermal growth factor were examined in the absence and presence of ARB. We examined the effect of ARB on mitogen-activated protein kinase (MAPK) phosphorylation of PrSC and PC-3 cells treated with conditioned medium of PrSC cells and determined the effect of ARB on tumor growth induced by paracrine factors from PrSC cells. Angiotensin II activated the cell proliferation and interleukin-6 secretion of PrSC cells, and ARB inhibited it. Angiotensin II, tumor necrosis factor-α, or epidermal growth factor induced MAPK phosphorylation in PrSC cells, and this phosphorylation was inhibited by ARB. Conditioned medium of PrSC cells with angiotensin II activated MAPK phosphorylation in PC-3 cells, and ARB-treated conditioned medium of PrSC cells inhibited it. The tumor growth and angiogenesis of a mixture of PC-3 with PrSC were inhibited by ARB administration, whereas those of PC-3 xenografts were not inhibited. ARB exerted an antiproliferative effect on prostate cancer through paracrine factors from stromal cells. Because prostate stromal cells are thought to be involved in the initiation and development of prostate cancer, the present data suggest the possibility that ARBs are a novel therapeutic class of agents for prostate cancer.
Introduction
Prostate cancer is the most common disease in elderly men and the second most frequent cause of death from cancer among men in the United States. Despite the lower prevalence of prostate cancer in Japan than in the United States, the prevalence has recently been continuously increasing (1). Changes in diet or lifestyle in western countries are presumably the cause of the increase of patients with prostate cancer. In general, genetics also plays major roles in the development and progression of prostate cancer besides diet and lifestyle. To identify specific genes related to prostate cancer, we previously did differential display PCR and GeneChip analysis using prostate cancer cells and tissue. To date, by differential display PCR analysis, we have identified several genes, including liprin-α2 and nmt55; liprin-α2 gene expression was down-regulated by dihydrotestosterone in prostate cancer cells, and nmt55 gene expression was up-regulated in human prostate cancer tissue (2, 3). By GeneChip analysis, we have found the gene neuroserpin (a protease inhibitor-12), whose expression was higher in cancer than in normal tissue (4). Currently, numerous studies to seek key genes related to prostate cancer have been done in many laboratories.
Initial hormone therapy, consisting of androgen ablation and antiandrogen therapy, for advanced-stage prostate cancer provides good efficacy with a high response rate of ∼90%. However, the beneficial effects are temporary, and most cases develop resistance to hormone therapy within several years. Although the detailed mechanism of hormone-refractory prostate cancer remains unknown, it is conceivable that cross-talk between stromal and prostate cancer cells plays a critical role in the progression of cancer, especially through the secretion of growth factors and cytokines.
From the viewpoint of growth factors or cytokines in prostate cancer, we should understand the autocrine or paracrine mechanism surrounding cancer cells. For instance, many reports revealed an elevation of serum interleukin (IL)-6 level in patients with hormone-refractory prostate cancer, and this cytokine is therefore thought to be involved in the progression of prostate cancer. Lee et al. (5) reported that overexpression of IL-6 rendered androgen-sensitive prostate cancer cells more resistant to apoptosis induced by androgen deprivation. Although it has been suggested that prostate stromal cells may contribute to the initiation and progression of this disease, a recent report showed that lysophosphatidic acid regulated secretion of IL-6 from stromal cells, which induced mitogenic signaling and growth of prostate cancer cells (6). As other growth factors or cytokines involved in the progression of prostate cancer, epidermal growth factor (EGF), tumor necrosis factor-α (TNF-α), heparin-binding EGF, and insulin-like growth factor (7) have been shown to be expressed in stromal tissues.
Angiotensin II is well known to be associated with hypertension as a main peptide of the renin-angiotensin system, and the detailed molecular mechanisms have recently been elucidated. For instance, angiotensin II activates not only the mitogen-activated protein kinase (MAPK) but also the Janus tyrosine kinase-signal tranducers and activators of transcription (STAT) pathway directly through angiotensin II receptor type 1 (AT1 receptor) in smooth muscle cells and cardiac myocytes (8, 9). A recent report revealed that angiotensin II transactivated the EGF receptor via the AT1 receptor and induced angiogenesis by enhancement of the activity of vascular endothelial growth factor (VEGF; ref. 10). In vascular smooth muscle cells and cardiac myocytes, endothelin-1 or angiotensin II is a key factor to stimulate the secretion of IL-6 (11, 12).
Our previous report showed that angiotensin II as well as EGF could activate the cell proliferation of prostate cancer, and an angiotensin II receptor blocker (ARB) could inhibit it through the suppression of phosphorylation of MAPK and STAT3 (13). Administration of ARB to nude mice inhibited the growth of prostate cancer cell xenografts in a dose-dependent manner. In addition, microvessel density (MVD) was reduced in xenografts treated with an ARB, which indicated that the ARB inhibited the angiogenesis of xenografts. Interestingly, recent studies have shown that the AT1 receptor is expressed in human prostate stromal tissues, and the growth of stromal cells is markedly regulated by angiotensin II (14).
The aim of this study was to investigate the molecular mechanism of the paracrine loop involving growth factors or cytokines secreted from prostate stromal cells in the development of prostate cancer. In particular, we tested the hypothesis that angiotensin II exerts mitogenic effects on regulating the cross-talk between stromal cells and cancer cells, and ARB can inhibit tumor growth through stromal cells.
Materials and Methods
Cell Lines
PC-3 cells, a human prostate cancer cell line, and PrSC cells, a human prostate stromal cell line, were obtained from the American Type Culture Collection (Rockville, MD) and Clonetics (Walkersville, MD), respectively. PC-3 cells were cultured in F-12 medium and PrSC cells were cultured in RPMI supplemented with 10% FCS under 5% CO2. In the experiments, these cells were cultured in phenol red–free RPMI plus 0.1% bovine serum albumin (BSA) and stimulated with reagents.
Reagents
Angiotensin II and EGF were purchased from Sigma (Atlanta, GA). TNF-α was purchased from R&D Systems (Minneapolis, MN). Candesartan (CV11974 and TCV116) was provided by Takeda Pharmaceutical Co. (Osaka, Japan), and losartan was provided by Merck Co. (Whitehouse Station, NJ). CV11974 is the active metabolite of TCV116 and was used for in vitro experiments. TCV116 is the prodrug of CV11974 and was used for in vivo experiments. Anti-phosphorylated MAPK antibody, anti-MAPK antibody, anti-phosphorylated STAT3 antibody, and anti-STAT3 antibody were purchased from Cell Signaling Technology (Beverly, MA). Anti-IL-6 antibody was purchased from Calbiochem (San Diego, CA).
Cell Growth Analysis and Measurement of IL-6
Cell growth was estimated by counting the cell number using a microcell counter (Toha Co., Tokyo, Japan). Briefly, PC-3 and PrSC cells were seeded onto 12- or 24-well plates at a density of 104 to 105 per well. Cells were cultured in phenol red–free RPMI with 0.1% BSA for 18 to 24 hours before the experiments and then treated with angiotensin II, TNF-α, or EGF for 5 days. Simultaneously, the cells were pretreated with CV11974, losartan, or anti-IL-6 antibody for 30 minutes and cultured in phenol red–free RPMI plus 0.1% BSA in the presence of angiotensin II or EGF for 5 days. After incubation in 5% CO2 at 37°C, cells were harvested with trypsin and cell numbers were determined with a cell counter on day 5. To measure IL-6 level secreted from PrSC cells, the cells were pretreated with CV11974 or losartan for 30 minutes and cultured in phenol red–free RPMI with 0.1% BSA in the presence of angiotensin II, TNF-α, or EGF for 24 hours. Then, IL-6 level of the collected medium was measured using a chemiluminescent enzyme immunoassay kit (Fujirebio, Inc., Tokyo, Japan). The assay was done at the laboratory of SRL, Inc. (Tokyo, Japan).
Preparation of PrSC Conditioned Medium
PrSC cells were cultured in either phenol red–free RPMI with 0.1% BSA or with angiotensin II, TNF-α, or RPMI with 10% FCS for 18 to 24 hours. In addition, PrSC cells were treated with or without CV11974 or losartan (10 μmol/L) for 30 minutes. Conditioned medium from PrSC cells under these conditions was collected.
Western Blot Analysis
PC-3 and PrSC cells were cultured in phenol red–free RPMI plus 0.1% BSA for 18 to 24 hours before the experiments. Then, cells were harvested after angiotensin II, EGF, or TNF-α treatment as indicated in the figures. Cells were pretreated with CV11974 or losartan for 30 minutes, stimulated with the reagents, and harvested at the indicated times in the figures. Cells under the appropriate conditions were washed twice with ice-cold PBS, lysed in ice-cold buffer consisting of 20 mmol/L Tris (pH 8.0), 137 mmol/L NaCl, 10% glycerol, 0.1% SDS, 0.5% NP40, 100 mmol/L NaF, 200 mmol/L sodium orthovanadate, 1 mmol/L EGTA, 2 mmol/L phenylmethylsulfonyl fluoride, 1 mg/mL leupeptin, and 3 mg/mL aprotinin, and centrifuged (30 minutes, 4°C, 14,500 × g). Following quantitation, cell lysate (20 μg) was added to SDS gel-loading buffer (containing a reducing agent) and boiled for 5 minutes. The samples were subjected to SDS-PAGE on 10% gel and electrotransferred to Immobilon-P purchased from Millipore (Bedford, MA). After blocking the membrane with 5% albumin, Western blotting was done using the antibody of interest, and the product was detected with an enhanced chemiluminescence detection system (Amersham Biosciences, Buckinghamshire, United Kingdom).
Antitumor Activity of TCV116 in Nude Mice
The antitumor activity of TCV116 (an AT1 receptor antagonist) was determined in athymic nude mice bearing PC-3 and/or PrSC tumors. PrSC and/or PC-3 cells (4.5 × 106) were injected into the flank region of athymic male nude mice (4–6 weeks old), and treatment was started on day 8 when the tumor measured 5 mm in diameter. Each mouse received one of two different doses of TCV116 (5.0 or 10 mg/kg/d). Each group consisted of eight animals. The control group received only the diluent. Tumors were measured with a caliper every 7 days. The volume of the tumor was calculated using the formula: tumor volume (mm3) = length × (width)2 × 0.5. Each tumor volume on the first day when each mouse received treatment was expressed as a relative tumor volume of 1.0.
Microvessel Density
To assess the density of blood vessels, immunohistochemical staining for mouse CD31 was done according to our previous report (13). Briefly, frozen nude mouse xenografts embedded in OCT compound (Sakura Finetechnol Co. Ltd., Tokyo, Japan) were cut with a cryostat. Sections were fixed with ice-cold acetone for 5 minutes. After drying, the sections were immersed in 0.3% H2O2–containing methanol to inactivate intrinsic peroxidase followed by treatment with 10% normal goat serum. Then, the sections were treated with rat anti-mouse CD31 antibody (PharMingen, San Diego, CA; diluted to 1:100) at 4°C overnight. The labeled antigen was visualized by streptavidin-biotin complex method followed by diaminobenzidine reaction. Two investigators counted the microvessels independently in a blinded fashion. The tissues were examined at high power (×200), and the four fields with the highest MVD were identified for vessel count. The mean number of CD31-positive vessels in the four selected fields (high-power field, ×20 objective and ×10 ocular) was used to express the vascular density.
Immunohistochemistry and Terminal Deoxynucleotidyl Transferase–Mediated Nick End Labeling
The tumor tissue was fixed in 10% formalin and embedded in paraffin for H&E, immunohistochemical, and terminal deoxynucleotidyl transferase–mediated nick end labeling studies. Immunohistochemical staining for epithelial cancer cells was done using cytokeratin AE1/AE3 (Nichirei, Tokyo, Japan). Staining procedure was done using DAKO Envision Plus kit (DAKO Corp., Carpinteria, CA) following the manufacturer's recommendations. The antibody was added and the slides were incubated for 2 hours at room temperature. Then, the slides were rinsed and incubated for 30 minutes with biotinylated second antibody (Nichirei). After washes with PBS, they were incubated with horseradish peroxidase–conjugated streptavidin (Nichirei) for 10 minutes and treated with 3-amino-9-ethylcarbazole (Nichirei) in PBS containing 0.01% H2O2. The slides were counterstained with Meyer's hematoxylin. As negative controls, the primary antibody was replaced with normal rabbit or goat IgG at an appropriate dilution. To assess the incidence of apoptotic cell death, the sections were stained using an In situ Cell Death Detection Peroxidase kit (Boehringer Mannheim, Mannheim, Germany) according to the manufacturer's recommendations. The detailed methods were reported previously (15). The tissues were examined at high power (×200), and five fields were selected for the apoptotic cell count. The mean number of apoptotic cells in the five selected fields (high-power field, ×20 objective and ×10 ocular) was used to express the apoptotic change.
Cytokine Array
Forty-two cytokines were assayed using the Human Cytokine Array III (Ray Biotech, Norcross, GA). The membranes were exposed to blocking buffer and incubated with angiotensin II–induced conditioned medium or control medium for 2 hours. Membranes were processed according to the manufacturer's instructions.
Statistics
Values are given as mean ± SD. For the results of cell number, IL-6 concentration, and xenograft volume, group data were compared by unpaired Student's t test. P < 0.05 was considered statistically significant. Quantification of band densities on Western blot was done using the public domain NIH image software (version 1.61).
Results
ARB Inhibited Prostate Stromal Cell Growth
To investigate the effect of angiotensin II on human prostate stromal cells, we applied it with its cognate receptor blockers, CV11974 and losartan, selective blockers for the AT1 receptor (ARB). As shown in Fig. 1A, angiotensin II treatment increased the number of PrSC prostate stromal cells in a dose-dependent manner. Furthermore, 1.0 and 10 μmol/L CV11974 significantly suppressed the cell growth induced by angiotensin II treatment (P < 0.05). Next, we investigated the effect of CV11974 on PrSC cells, when these cells were stimulated with EGF or TNF-α, an important factor for the growth and development of stromal fibroblasts. Stimulation with 1 ng/mL EGF increased the cell number of PrSC cells (Fig. 1B). Interestingly, when 10 μmol/L CV11974 was added to cells treated with EGF, growth was suppressed by 18.6% (P < 0.05). Similarly, PrSC cell proliferation was induced by TNF-α treatment and suppressed by 26% by CV11974 (P < 0.02) as shown in Fig. 1C. On the other hand, another ARB, losartan, significantly suppressed PrSC cell proliferation in a dose-dependent manner under 10% FCS in RPMI (Fig. 1D). Losartan also suppressed cell proliferation induced by TNF-α treatment by 24.1% (P < 0.05; Fig. 1E).
As our previous report showed (13), PC-3 cells, a human metastatic prostate cancer cell line, have lower expression of the AT1 receptor compared with other prostate cancer cell lines, LNCaP or DU145. To examine whether cell proliferation of PC-3 is affected by angiotensin II, we treated PC-3 cells with angiotensin II and/or ARB. Unlike LNCaP or DU145 cells, cell proliferation of PC-3 was not influenced by angiotensin II stimulation or inhibited by CV11974 or losartan treatment as shown in Fig. 1F. Hence, we used PC-3 cells as a nonresponsive prostate cancer cell line to angiotensin II or ARB in the following experiments.
IL-6 is one of the cytokines involved in the progression of prostate cancer. A recent report showed that IL-6 was synthesized by fibroblasts in prostate stromal cells (6). We thus investigated whether IL-6 secretion of PrSC cells was influenced by stimulation with the above agents and ARB treatment. First, we confirmed whether IL-6 secretion of PrSC cells was induced when cells were stimulated with 10 μmol/L angiotensin II. In the presence of CV11974, IL-6 secretion was inhibited by 23.5% (P < 0.02) as shown in Fig. 2A. Furthermore, treatment of PrSC cells with TNF-α markedly augmented IL-6 secretion, and this was also inhibited by CV11974 (P < 0.05; Fig. 2B). A similar phenomenon was confirmed when PrSC cells were stimulated with EGF and/or CV11974 as shown in Fig. 2C. To further confirm the effect of IL-6 on PC-3 cell growth, we used conditioned medium collected from PrSC cells. Figure 2D showed that conditioned medium–mediated PC-3 cell growth was significantly attenuated in the presence of IL-6-neutralizing antibody compared with cells cultured in conditioned medium without anti-IL-6 antibody. This result suggests that IL-6 secreted from PrSC cells is a major mitogenic growth factor in PC-3 cells.
ARB Regulated Angiotensin II–Stimulated and Other Cytokine-Stimulated Signaling in PrSC and PC-3 Cells
Because we confirmed that angiotensin II was a mitogenic factor in PrSC cells, we investigated whether 10 μmol/L angiotensin II could induce the activation of MAPK and STAT3 in PrSC cells, which have important roles in signaling to mediate cell proliferation induced by growth factor or cytokine stimulation. MAPK was activated immediately after stimulation with 10 μmol/L angiotensin II as shown in Fig. 3A. Similarly, 10 μmol/L angiotensin II strongly induced phosphorylation of STAT3 in a time-dependent manner. Maximal activation of MAPK and STAT3 was achieved with 10 minutes of angiotensin II stimulation and was sustained for ∼15 minutes. Furthermore, we confirmed that 1.0 μmol/L angiotensin II activated MAPK phosphorylation as well as 10 μmol/L angiotensin II at 10 minutes after the start of treatment as shown in Fig. 3B. On the other hand, angiotensin II stimulation of androgen-independent human prostate cancer PC-3 cells did not promote phosphorylation of MAPK (data not shown). These results showed that angiotensin II stimulation activated MAPK and STAT3 in stromal PrSC cells.
To investigate whether an ARB, CV11974 or losartan, can suppress the activation of signal transduction pathways by angiotensin II, TNF-α, or EGF, we carried out Western blotting of phosphorylated MAPK. Figure 3C shows that CV11974 or losartan suppressed the phosphorylation induced by stimulation with angiotensin II or TNF-α in PrSC cells. Furthermore, as shown in Fig. 3D, CV11974 suppressed the phosphorylation of MAPK induced by EGF in PrSC cells. These data are consistent with the inhibition of PrSC cell proliferation by treatment with CV11974 or losartan.
Based on the above observations in PrSC and PC-3 cells, we then did Western blot analyses to investigate the possible interactions between epithelial (PC-3) and stromal (PrSC) prostate cells, especially when PrSC cells were treated with angiotensin II and/or ARB. As shown in Fig. 3E, even if only conditioned medium (nontreated conditioned medium) of PrSC cells was added to PC-3 cells, phosphorylated MAPK was recognized after 10 minutes of treatment (lane 2). When PrSC cells were treated with angiotensin II, conditioned medium induced phosphorylation of MAPK in PC-3 cells more strongly (lane 3) than nontreated conditioned medium (lane 2). Furthermore, to determine whether the conditioned medium of CV11974-treated PrSC cells can affect the phosphorylation of MAPK in PC-3 cells, Western blot was done. As a result, when PrSC cells were stimulated by angiotensin II, addition of CV11973 to the conditioned medium could inhibit the phosphorylation of MAPK in PC-3 cells (lane 4). This result implies that CV11974 may decrease soluble factors secreted by PrSC cells.
Next, we examined whether 0.1% BSA conditioned medium of PrSC cells treated with another ARB, losartan, could affect MAPK phosphorylation in PC-3. As shown in Fig. 3F, conditioned medium of angiotensin II–treated PrSC cells induced MAPK phosphorylation, and conditioned medium of PrSC cells treated with losartan as well as well as CV11974 could inhibit it in PC-3 (lane 3). Interestingly, similar inhibition of MAPK phosphorylation in PC-3 cells was recognized in 10% FCS conditioned medium of PrSC cells treated with ARBs (lanes 5 and 6). These results indicate that ARBs could inhibit paracrine factors secreted from PrSC cells in 0.1% BSA, including angiotensin II or 10% FCS; accordingly, they may cause inhibition of MAPK phosphorylation in PC-3 cells.
Identification of Growth Factors and Cytokines Secreted by PrSC Cells
To identify growth factors or cytokines secreted from prostate stromal PrSC cells, we used cytokine array analysis. Figure 4 shows the immunoreactivity of the cytokines and growth factors present in the conditioned medium of PrSC cells treated with or without angiotensin II. Intensified spots in the array membranes of angiotensin II–treated conditioned medium (Fig. 4B) were compared with respective spots in the conditioned medium of nontreated PrSC cells as a control (Fig. 4A). Levels of IL-1α, IL-6, IL-8, monocyte chemoattractant protein-1 (MCP-1), and macrophage colony-stimulating factor were significantly increased in the conditioned medium of angiotensin II–treated PrSC cells compared with that of nontreated PrSC cells as shown in Fig. 4B. We examined whether the above cytokines secreted from PrSC cells are the growth factors expressed in conditioned medium responsible for prostate cancer cell growth. The cytokines IL-8 and MCP-1 were added to human prostate cancer LNCaP cells. After 5 days, the number of LNCaP cells was significantly increased by stimulation with IL-8 or MCP-1 compared with that without treatment (data not shown). These results suggest that some growth factors or cytokines, including IL-6, may be present in the conditioned medium of PrSC treated with angiotensin II and contribute to cell proliferation of prostate cancer.
Antitumor Activity of ARB
To determine whether the in vitro antiproliferative activity of ARB could be translated to antitumor activity in vivo, TCV116, the prodrug of CV11974, was given to athymic male nude mice with tumor xenografts. More specifically, to investigate the possible physiologic interaction of stromal PrSC cells with epithelial PC-3 cancer cells, we tested the hypothesis that ARB could confer antitumor activity through stromal cells. Tumors were induced by inoculation of PC-3 cells alone (Fig. 5A) or coinoculation of PC-3 and PrSC cells (Fig. 5B) in athymic mice. When the tumors reached ∼5 mm in diameter, the animals were given TCV116 at 5.0 or 10 mg/kg/d. The control group received water containing sodium hypochlorite (10 ppm). As shown in Fig. 5A, at 5 weeks, control animals had developed large tumors of 5.6 ± 2.1 relative volume compared with those at week 0. Mice treated with TCV116 at 5.0 or 10 mg/kg/d showed no significant difference in tumor relative volume at 5 weeks compared with the control group (Fig. 5A).
On the other hand, to investigate whether growth inhibition by TCV116 is observed in tumors consisting of a mixture of PC-3 and PrSC cells, a mixture of both cells was established as xenografts in nude mice. Mice were treated with TCV116 at 5.0 or 10 mg/kg/d. At 5 weeks, whereas control mice had developed a large tumor of 6.5 ± 3.5, TCV116-treated mice showed inhibition of tumor relative volume, 4.3 ± 1.6. As shown in Fig. 5B, there was a significant difference in tumor growth between control and 10 mg/kg/d TCV116 groups (P < 0.02) as early as 1 week from the start of treatment. Thus, TCV116 could suppress tumor growth of a mixture of androgen-independent PC-3 cells with PrSC cells.
Our previous report showed that ARB reduced MVD in xenografts of DU145 cells (13). In the present study, we confirmed the antitumor effect of ARB; hence, we measured MVD of xenografts in mice treated with TCV116. Immunohistochemical staining for mouse CD31 revealed a marked difference in microvessel numbers of PC-3 plus PrSC xenografts at 5 weeks between control (nontreated) and TCV116-treated (10 mg/kg/d) mice as shown in Fig. 5C. MVD was quantitated in eight xenografts each in the control and treatment groups. As shown in Fig. 5C, the TCV116 treatment group had a reduced mean MVD of 45.4 ± 8.8 compared with 85.3 ± 11.4 in the control group (P < 0.01). In contrast, there was no significant difference in microvessel number of PC-3 xenografts between control (nontreated) and TCV116-treated (10 mg/kg/d) mice (data not shown).
Analysis of Immunohistochemical Staining and Apoptosis Induced by TCV116
To examine the difference in morphology between PC-3 alone and chimeric (PC-3 and PrSC) xenografts at later time points (5 weeks), we did histologic examination. Because H&E stain could not clearly differentiate epithelial cancer cells and stromal cells (Fig. 6A and B), the sections were stained immunohistochemically. The AE1/AE3 staining distinguished epithelial cancer cells from stromal cells originated from PrSC cells (unstained regions) in chimeric tumor (5 weeks) as shown in Fig. 6C, whereas PC-3 xenografts contained no stromal cells (Fig. 6D). Furthermore, to assess the incidence of apoptotic cell death induced by TCV116 treatment, we did terminal deoxynucleotidyl transferase–mediated nick end labeling staining. The data clearly showed that apoptotic cells in TCV116-treated xenograft of chimeric tumor increased significantly more than those in PC-3 xenografts (Fig. 6E and F, arrows) and untreated chimeric tumors (data not shown). Apoptotic cells were quantitated in the TCV116 treatment groups of PC-3 alone and chimeric tumors (5 weeks). The TCV116 treatment group of chimeric tumors had an increased mean apoptotic cell number of 43.6 ± 12.4 compared with 19.4 ± 2.7 in the TCV116 treatment group of PC-3 tumors.
Discussion
It is reported that there is an apparent low prevalence of cancer in hypertensive patients receiving angiotensin-converting enzyme inhibitors (16); however, the molecular mechanisms have never been elucidated. We presented previously strong evidence of the effects of angiotensin II and its receptor blocker (ARB) on prostate cancer cells and the tumor growth of prostate cancer (13). Those data clearly indicated that angiotensin II directly affected prostate cancer cells, which means angiotensin II enhanced the proliferation of prostate cancer cells through AT1 receptor–mediated activation of MAPK and STAT3 phosphorylation. In addition, cell proliferation induced by angiotensin II or EGF was inhibited by an ARB, indicating that ARB could indeed inhibit the cell proliferation of prostate cancer by interacting with signal transduction via the EGF receptor or by blocking angiotensin II binding to the AT1 receptor.
Several recent studies indicated that prostate stromal cells contain the AT1 receptor (14), and as shown in this study, the prostate stromal cell number was increased by angiotensin II treatment. It is well known that prostatic stromal cells, especially fibroblasts, are involved in the development of hormone-refractory prostate cancer accompanied by the secretion of several growth factors (17–20). We confirmed here that angiotensin II augmented the secretion of IL-6 and another cytokine from prostatic stromal cells. From the results of cytokine array, several kinds of cytokines, including IL-1α, IL-6, IL-8, and MCP-1, were identified to be induced by angiotensin II treatment (Fig. 5). Interestingly, these cytokines have been reported to be involved in tumor angiogenesis. IL-1α is required for in vivo angiogenesis and invasiveness of different tumor cells and contributed to the production of vascular endothelial cell growth factor (VEGF) and TNF in tumor cells cocultured with peritoneal macrophages (21). IL-8, a chemokine involved in the metastasis and angiogenesis of some tumors, has been reported to be overexpressed in prostate cancer (22). In particular, IL-8 confers androgen-independent growth and migration of LNCaP cells through activation of the androgen receptor without androgen stimulation (23). Therefore, IL-8 may play a role in the development of androgen-independent prostate cancer. Ohta et al. (24) reported that MCP-1 mRNA was expressed in gastric carcinoma and its expression was significantly correlated with VEGF level. These factors secreted from PrSC stimulated by angiotensin II treatment may contribute to the mechanisms underlying androgen independence through multiple pathways. It is conceivable that angiotensin II might induce neovascularization through activation of angiogenic factors via reactive prostate stroma, and specific ARBs might inhibit carcinogenesis through suppression of angiogenesis.
Although the detailed molecular mechanism by which TNF-α plus CV11974 caused a greater reduction in cell number than CV11974 alone is unknown, it is conceivable that activation of other angiotensin II receptor, AT2 receptor, signaling may occur. In brief, TNF-α stimulation may induce angiotensin II synthesis in PrSC cells, leading to activation of AT2 receptor signaling by endogenous angiotensin II binding to them. The functions of the AT2 receptor may oppose some actions of the AT1 receptor, which means AT2 receptor signaling exerts the opposite effect to the AT1 receptor, differentiation and/or apoptosis (25). Therefore, TNF-α-induced cell proliferation is inhibited by CV11987, and simultaneously, TNF-α-induced activation of AT2 receptor signaling occurs, which may augment the inhibition of PrSC cell proliferation. Similarly to our results, a recent report showed that CV11974 plus angiotensin II more strongly inhibited bromodeoxyuridine incorporation index than CV11974 alone in myofibroblasts and keratinocytes (26).
As reported previously, prostate cancer and its stromal cells have lysophosphatidic acid, α/β-adrenergic receptors, endothelin-1, and angiotensin II receptors, and these receptors are categorized as GMP-binding protein-coupled receptors (27). These receptors have been viewed as critical regulators of the interaction between epithelial and stromal cells (28). Hence, we consider that it is very important for overcoming prostate cancer to inhibit GMP-binding protein-coupled receptor signaling in cancer cells. Interestingly, if cells have AT1 receptors and ARB binds to them, ARB has the potential to suppress the MAPK or STAT pathway stimulated through tyrosine kinase receptors bound to some growth factors or cytokines. The major finding in this study was that activation of tyrosine kinase stimulated by TNF-α, EGF, or IL-6 in prostate stromal PrSC cells was inhibited by ARB as illustrated in Fig. 7. The secretion of some cytokines, including IL-6, from stromal cells is crucial for the development and progression of prostate cancer. Thus, ARB inhibits the intracellular signal pathways of those cytokines and growth factors, which means ARB is capable of negatively interacting with the paracrine loop in prostate cancer.
A previous report indicated that reactive stromal cells adjacent to prostate cancer cells or prostate intraepithelial neoplastic cells are involved in angiogenesis (29). Further, it was shown that coincubation of prostate stromal cells with LNCaP cells increased the incidence of tumor formation and accelerated tumor growth, a pathologic finding implying the promotion of angiogenesis (30). As a candidate factor for regulation of prostate stromal cells in cancer, transforming growth factor-β (TGF-β) has emerged because this growth factor is a key mediator of stromal response, especially in wound repair (31). TGF-β is overexpressed in cancer tissues, especially in reactive stroma and in breast and colon cancer (32, 33). In prostate cancer, TGF-β promotes tumor progression by stimulating angiogenesis and metastasis (34). In view of the relationship between TGF-β and angiotensin II, it is of interest that angiotensin II induces TGF-β and ARB reduces TGF-β expression in the development of renal fibrosis, which suggests that ARBs may have the potential to inhibit fibrotic changes. Indeed, ARBs are clinically used in the treatment of albuminuria in patients with chronic renal failure. In an animal model of renal cell carcinoma, ARB treatment significantly decreased the microvessel number of metastatic lung lesions accompanied by decreased TGF-β content and VEGF staining intensity (35). This result suggests that ARBs may inhibit angiogenesis in cancer tissue by inhibiting TGF-β secretion simultaneously.
It is reported that angiotensin II is present in human seminal fluid at a concentration 3- to 5-fold higher than that found in blood (36). In vitro study showed that angiotensin II physiologically affected sperm function through the AT1 receptor, for instance, increasing the oocyte-penetrating ability (37, 38). More surprisingly, there is considerable evidence that the male reproductive organs contain renin-angiotensin system components, such as angiotensinogen, renin, angiotensin I–converting enzyme, and AT1 and AT2 receptors. O'Mahony et al. suggested that the prostate gland may be the source of angiotensin II found in seminal fluid and that angiotensin II production may be regulated by androgen (39, 40). Based on this accumulated evidence and our results, the prostate itself seems to be a target for angiotensin, and the prostatic renin-angiotensin system may affect the development of prostate cancer. For more complete understanding of the regulation of renin-angiotensin system in prostate cancer, further analysis is required.
In summary, the present study provided the following new information: (a) angiotensin II activates the proliferation of prostate stromal cells accompanied by secretion of several cytokines, (b) angiotensin II activates the signal pathways of MAPK and STAT in prostate stromal cells, (c) conditioned medium of prostate stromal PrSC cells stimulated by angiotensin II activates the signal pathways of MAPK in prostate cancer PC-3 cells, and (d) ARB inhibits it through the interaction between prostate cancer and stromal cells both in vitro and in vivo. Although we have not thoroughly analyzed the interaction of angiotensin II and ARB between prostate cancer and stromal cells, we believe that ARBs have the novel ability to suppress the development or progression of prostate cancer. Furthermore, based on the idea that inhibition of GMP-binding protein-coupled receptor signaling in cancer and stromal cells could suppress prostate cancer growth by a negative interaction with tyrosine kinase–induced signal pathways, a novel treatment to overcome this devastating disease could be possible in the future.
Grant support: Yokohama Medical Foundation Umehara Grant, Ministry of Education, Culture, Science and Technology Grant-in-Aid for Scientific Research (C), and Japan Society for the Promotion of Science Center of Excellence Research.
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.
Acknowledgments
We thank Dr. Minoru Kihara (Department of Second Internal Medicine, Yokohama City University) for constructive discussions.