Introduction
Prostate cancer (PCa) is the third leading cause of death among men in America [
1,
2]. The mortality from PCa results from metastases to bones and lymph nodes and progression from androgen-dependent to androgen-independent disease. While androgen deprivation has been effective in treating androgen-dependent PCa, it is ineffective in treating advanced PCas, the primary cause of mortality. Epidemiological and histopathological studies have implicated inflammation in the pathogenesis of PCa [
3‐
5]. Studies have consistently shown a decreased risk of PCa among men who regularly take aspirin or other nonsteroidal anti-inflammatory drugs (NSAIDs) [
6‐
8]. Despite beneficial effects, the side effects from using high doses of COX-2 inhibitors for cancer prevention are a major concern. These observations emphasize the need for development of new effective treatments for advanced PCa.
The family of natriuretic peptide hormones has broad physiologic effects. In addition to vasodilation, cardiovascular homeostasis, sodium excretion and inhibition of aldosterone secretion, they have been implicated in immunity and inflammation [
9‐
18]. The effects of atrial natriuretic peptide (ANP) are mediated by its interaction with the cell surface natriuretic peptide receptor A (NPRA; high affinity) and natriuretic peptide receptor C (NPRC; low affinity). In patients with prostate tumors, the immune response plays a large part in the progression of the disease and it is likely that the NPRA system is involved; but the role of NPRA in human cancers remains unknown. A novel peptide, NP
73-102, has been identified [
14] whose sequence is immediately N-terminal to the ANP peptide and which is an inhibitor of NPRA (iNPRA). NP
73-102 does not bind to NPRA but blocks its expression, and we have shown that it possesses bronchodilatory, anti-inflammatory [
14,
16,
19,
20] and antitumor activity [
19].
We previously reported that mice deficient in NPRA (NPRA-knockout, KO) exhibit significantly decreased inflammation [
16,
19‐
21]. Furthermore, we found that NPRA-KO mice do not permit growth of implanted human lung cancer, melanoma and ovarian cancer cells [
19], suggesting that NPRA may be a novel therapeutic candidate. Given the evolutionary conservation of ANP in many species, we reasoned that NPRA expression may be relevant in human cancers. In this study, we examined the expression of NPRA in PCa cell lines and human tissue samples and determined whether NPRA can be used as a target for PCa therapy. The results show that increased NPRA expression is strongly associated with progression of human PCa and that NPRA deficiency prevents growth of transplanted PCa cells and inhibits tumor burden in part by downregulating macrophage migration inhibitory factor (MIF) in PCa cells.
Discussion
There remain several overarching challenges in PCa research: the lack of specific clinical markers for early diagnosis and prognosis of PCa and the need to identify drugs that target androgen-independent PCa tumor cells directly without damaging healthy cells. In this study we show that NPRA is a potential biomarker for PCa and candidate for PCa therapy.
One important finding of our study is the demonstration that NPRA is significantly over-expressed in mouse and human PCa cells compared to normal cells. Screening of a human PCa tissue microarray containing 240 tissue samples shows that NPRA is also over-expressed in human tissues including high grade PIN (prostatic intraepithelial neoplasm) and prostatic adenocarcinoma. The benign hyperplastic glands exhibited significantly lower NPRA expression than localized PCas. These data are consistent with our previous report and with the data in this study, showing that NPRA is highly expressed in both human and mouse PCa cell lines and in advanced PCa tissues, but not in a normal prostate epithelial cell line or in a benign prostate hyperplasia epithelial cell line [
19,
34,
35]. It is to be noted that NPRA was expressed in the androgen-dependent cell line LNCaP but not in the stromal cell line, WPMY. However, expression of ANP was detected in culture supernatants of PC3 and DU145 PCa cells and WPMY stromal cells but not in supernatants from normal prostate epithelial cells or LNCaP cells. These results suggest that ANP produced by stromal cells can signal via NPRA on androgen-dependent cells (paracrine), whereas androgen-independent cells produce both ANP and NPRA and can signal in an autocrine manner. Thus, ANP-NPRA signaling may play a key role in engaging PCa cells with stroma during PCa pathogenesis. Hence, PCa may be better managed by inhibiting ANP-NPRA signaling.
Further, we found a significant association between NPRA expression and Gleason score and pathological stage. Results from the tissue array studies show that NPRA is an independent predictor of advanced PCa, and may therefore be useful as a clinical marker. Although, a number of marker antigens have been reported for PCa, none of them is specific enough to pass the clinical test for use in PCa prognosis. Given the strong positive correlation (r = 0.64, p < 0.005) between NPRA expression and the severity of the clinical stage, particularly in androgen-independent PCa, NPRA may prove to be an effective clinical prognostic marker.
Our study also suggests that NPRA may be a drug target for treating PCa. Using the TRAMP-C1 spontaneous PCa model, we demonstrated that NPRA-KO mice, which have normal heart, kidney and vascular function, have no detectable increase in postnatal mortality, do not permit growth of implanted PCa cells and have a normal lifespan of over 24 months. Tumor growth is observed in NPRA-het mice but at a significantly reduced level compared to that in WT C57BL/6 mice, which indicates that host NPRA gene dosage is a determining factor for the growth of tumor cells in mice. This finding is consistent with the reports that atrial natriuretic factor peptides (ANP and VD) inhibit the proliferation of PCa cells
in vitro and in mice [
34]. This is presumably due to the feedback inhibition of NPRA expression caused by high doses of ANP or other natriuretic peptides, such as NP
73-102 (our data) [
18,
34,
36,
37]. Thus, while low doses of these peptides stimulate NPRA signaling, high doses inhibit NPRA signaling and show anticancer effects. In sum, NPRA provides a heretofore undescribed target for PCa. This hypothesis is also supported by the observation that NPRA is an upstream regulator of IL-6, which has been reported as a target for PCa therapy [
38,
39].
The finding that pNP
73-102 inhibits NPRA expression prompted us to examine its role in treating PCa. TRAMP-C1 cells injected into C57BL/6 mice induced tumors in the control mice but not in pNP
73-102-treated mice. These findings demonstrate the potential utility of pNP
73-102 for the treatment of PCa. Although the mechanism of tumor inhibition by pNP
73-102 is unknown, the evidence that pNP
73-102 significantly decreases the expression of NPRA suggests that this may be the explanation for its antitumor effect. A perceived limitation in iNPRA therapy for PCa is the normal physiological role of NPRA in blood pressure regulation. To address this issue we compared blood pressure of NPRA-KO mice with that of TRAMP mice and found no relationship between NPRA expression, blood pressure levels and PCa incidence (Additional file
6, Fig. S4), which is consistent with studies in humans that showed no relationship between blood pressure and PCa [
40,
41].
Another major finding of our report is that the antitumor effects of limiting NPRA expression may be due to a reduction in inflammation in the tumor environment. Our evidence shows that a number of molecules may be regulated by NPRA signaling including MIF and IL-6, both of which have been implicated in PCa development. Increased MIF mRNA expression and serum MIF levels have been associated with progression of PCa when tumor and benign tissue from matched samples were compared [
31,
33,
42]. Elevated IL-6 levels are found in patients with metastatic PCa and are associated with a poor prognosis [
43]. Furthermore, aberrant expression of the IL-6 gene and increased production of IL-6 are associated with advanced bone metastasis and increased morbidity [
43‐
46], as well as resistance to chemotherapy [
47]. There are three lines of evidence supporting the idea that NPRA is an upstream regulator of MIF in PCa cells: (i) a 2.5-fold reduction in MIF mRNA was found after LPS treatment of NPRA-KO mice compared to WT mice. (ii) MIF expression was detectable in the prostate tissues of TRAMP mice, but not in WT mice, and (iii) NPRA downregulation reduced MIF expression in cultured TRAMP-C1 cells and xenografts. Consistent with these observations, a PCa tissue array stained for NPRA showed expression of MIF (
data not shown).
Since intratumoral expression of MIF was correlated with serum IL-6 in patients with non-small cell lung cancer [
48] and IL-6 was shown to be one of the potential MIF-regulated genes in DU145 cells [
32], we speculate that NPRA signaling may regulate IL-6 in PCa cells via MIF. In support of this hypothesis, we found elevated IL-6 in the serum of TRAMP mice during PCa development (
unpublished observation). These data support our previously reported studies, where lung tissues of NPRA-KO mice failed to induce IL-6 during OVA-induced inflammatory challenge and showed reduced expression of activated p65- and p50-NF-kB [
19,
20]. Together, these studies show that NPRA may affect PCa progression by regulating in part MIF and IL-6 expression, both of which have been linked to PCa.
In summary, we demonstrate that increased NPRA expression is strongly associated with progression of human PCa and that NPRA deficiency prevents growth of transplanted PCa cells and inhibits tumor burden in TRAMP mice in part by downregulating MIF in PCa cells.
Materials and methods
Materials
Normal prostate epithelial cell line (PrEC) was purchased from Lonza (Allendale). RWPE, WPMY, Tramp-C1 and PC3 cells were purchased from the American Type Culture Collection (Manassas, VA, USA). DU145, PC3, benign prostatic hyperplasia (BPH), LNCaP, C4-2, the rat gastric mucosa cell line (RGM1), P2 and CaP2 were described before [
22,
24,
32,
49]. The beta-actin antibody was obtained from Sigma, the PARP antibody from Santa Cruz Biotechnology, and the MIF antibody from Abcam. Lipofectamine 2000 reagent was obtained from Invitrogen.
NPRA antibody production and purification
Antibody against NPRA was generated by injecting rabbits (New Zealand white) with 400 μg of synthetic NPRA peptide (amino acid 1010-1031 of mouse NPRA protein, which is homologous to rat and human NPRA) conjugated to keyhole limpet hemocyanin (BioSynthesis, Inc., Lewisville TX). Antibody was purified by applying serum to a column of protein A/G agarose (Invitrogen, Carlsbad, CA) equilibrated with 20 mM Tris, pH 7.5, 150 mM NaCl, and eluting with 100 mM citrate, pH 3.0. The eluate was neutralized with 5 M NaOH, glycerol was added to 50% and the purified aliquots were stored at -20°C.
NPRA antibody competition assay
For determining NPRA antibody titer, a 96 well plate was coated with the non-KLH-conjugated NPRA-specific peptide (amino acids 1010-1031 of mouse NPRA protein) that was used to raise the antibody or an unrelated peptide. Rabbit sera from 6 animals were pooled and purified using a protein A/G sepharose column (Pierce). A serial dilution of the antibody was added to each well of a microtiter plate coated with peptides overnight. For the competition assay, purified antibody was incubated with NPRA-specific peptide on ice for 1 hr and then added to the plate. The plate was washed and developed using HRP-conjugated anti-rabbit IgG (Cell Signaling) and HRP-substrate (R & D Systems). The plate was read at 450 nm using a Synergy H4 plate reader (Biotek). The values presented are means of four wells.
Cell counting and colony assay
At the indicated times, cells were harvested by trypsinization and viable cell numbers enumerated by trypan blue dye-exclusion. To test colonization potential, TR-C1or TR-C3 cells were plated in 100 mm dishes at 1000 cells/plate. After 3 weeks, the resulting colonies were stained with 0.25% crystal violet, photographed and counted.
Luciferase reporter assays
PC3 cells were co-transfected with hNP73-102, mNP73-102 or vector alone (pVAX), reporter plasmid (pNPRA-Luc), and a transfection normalization vector (pRenilla-luc). DNA (0.5-1 μg/106 cells) was transfected into 60% confluent PC3 cells using lipofectamine (Life Technologies). Forty-eight hours after transfection, the reporter activity was measured with the Dual-Luciferase Reporter assay system (Promega) according to the manufacturer's instructions. Luminescent signals were quantified with the Synergy H4 (Biotek). Reporter assay results were based on data averaged from three replicates.
Tissue microarray (TMA) staining
A human prostate cancer TMA containing 240 samples, prepared in the histology laboratory of the Moffitt Cancer Center Tissue Core Facility was used to test for expression of NPRA and MIF. The TMA slide was stained using a Ventana Discovery XT automated system (Ventana Medical Systems, Tucson, AZ), according to the manufacturer's protocol. Briefly, slides were deparaffinized on the automated system with EZ Prep solution (Ventana). Following heat-induced antigen retrieval, the slide was incubated with NPRA antibody (1:300) for 32 min and Ventana anti-rabbit or anti-goat secondary antibody for 20 min. The detection system used was the Ventana OmniMap kit, and the slide was then counterstained with hematoxylin and dehydrated.
TMA data analysis
The TMA slide was scored for intensity and cellularity by an expert pathologist. Positive staining for NPRA was scored into four grades, according to the intensity: 0, 1+, 2+ and 3+. The percentage of NPRA-positive cells was scored into three categories: 1 (0-33%), 2 (34-64%) and 3 (65-100%). The product of the intensity and percentage scores was used as the final score. The final score was classified as: 0, negative; 1-3, weak; 4-6, moderate; and 7-9, strong. A median analysis of NPRA scores and the frequency in each disease group of having a score at or below the median was performed. Also, the chi-squared test, the Kruskal-Wallis test and the Wilcoxon-Mann-Whitney test were used to compare the scores by groups. Comparisons were done for (1) PIN-L vs. BPH; (2) PIN-H vs. BPH; (3) Gleason-6 vs. BPH; (4) Gleason-7 vs. BPH; (5) Gleason-8 to 10 vs. BPH and (6) AI vs. BPH.
Animals
Male C57BL/6 mice were purchased from the National Cancer Institute. Male C57BL/6 NPRA-KO or NPRA-het were described before [
19]. All mice were maintained in a pathogen-free environment and all procedures were reviewed and approved by the University of South Florida Institutional Animal Care and Use Committee.
Preparation of plasmid nanoparticles and administration to mice
Plasmids encoding NP
73-102, hNP
73-102 and VD were constructed as described previously [
14,
19]. Plasmids encoding siRNAs against NPRA were described previously [
19]. Plasmids encoding shNPRAs were purchased from Origene. For transfection, epithelial cells at 60% confluence (log phase) were incubated in complete medium at 37°C with plasmid DNA (1 μg/10
6 cells) complexed with lipofectamine (GibcoBRL Life Technologies, Carlsbad, CA). For tumor cell inoculation, TRAMP-C1 cells were trypsinized, washed and resuspended in PBS at 5 × 10
7 cells per ml. Mice were injected s.c. in the flank with 100 μL of resuspended cancer cells. For evaluating the effects of iNPRA in modulating tumor progression, plasmids encapsulated in chitosan nanoparticles (25 μg of plasmid plus 125 μg of chitosan) were administered i.p. twice a week until sacrificed. Tumor sizes were measured externally by calipers, and at the end of experiment (day 62), the mice were euthanized and the tumors were removed and weighed. Proteins from tumors were extracted and examined for NPRA and MIF expression by western blotting.
Blood pressure measurement
Diastolic and systolic pressures of age-matched mice were measured using the CODA noninvasive blood pressure system (Kent Scientific). Briefly, mice were placed in a restrainer on a hot water blanket and the restrainer was covered with a warm water glove. The occlusion cuff was fitted to the base of the tail and the VPR cuff slid down until it reached the occlusion cuff. Maximum occlusion pressure was set to 250 μL with a deflation time of 20 seconds and a minimum volume of blood flow in the tail of 10 μL. The occlusion cuff was inflated to impede the blood flow to the tail. As the occlusion cuff is deflated, a second tail cuff with the VPR sensors records the pressure at the point where blood flow returns. The systolic is measured at the first appearance of tail swelling and the diastolic is calculated when the increasing rate of swelling ceases in the tail.
Western blot analysis
Western blot assay was performed as previously described [
49]. Cells were lysed, total cellular protein (120 μg) was separated by SDS-PAGE, blotted to nitrocellulose, and incubated with antibodies to specific proteins. Bands were visualized by enhanced chemiluminescence (Amersham Life Sciences, Piscataway, NJ) on Kodak X-OMAT-AR film.
Real-time PCR analysis
Total RNA was isolated using the RNeasy mini kit (Qiagen). One tube cDNA synthesis followed by real-time PCR was performed in a 25 μl reaction mixture using Taqman RNA-to-CT™
1-Step Kit (Applied Biosystems). Quantitative real-time PCR was carried out on the CFX96 real-time System (Bio-Rad). Taqman gene expression assays (Applied Biosystems) Hs00418568, Hs00236988 and 4333762, respectively are used for amplification of NPR1, MIF and β-actin. The conditions for the real-time PCR assay were 15 min at 48°C, 10 min at 95°C, 40 cycles of 15 sec at 95°C and 60 sec at 60°C. Expression of each target mRNA relative to β-actin was calculated under experimental and control conditions based on threshold cycle (
Ct) as
, where Δ
Ct =
Ct target -
Ct β-actin and Δ (Δ
Ct) = Δ
Ct experimental -
Ct control.
ANP ELISA
Duplicate aliquots of 50 μl of culture supernatants were assayed for ANP concentration using a fluorescent immunoassay kit (Phoenix Pharmaceuticals, Burlingame CA). ANP standards were run to generate a standard curve that was used to calculate the average ANP concentration.
SuperArray analysis of prostate tissues
NPRA-KO and WT C57BL/6 mice (n = 4) were injected i.p. with LPS (1 mg/kg body weight) for 3 hrs, prior to prostate harvesting. Total RNA was isolated using an RNAeasy kit (QIAGEN, Valencia, CA) and a pool of total RNA by group hybridized to the mouse autoimmune and inflammatory response Oligo GEarray (SuperArray Frederick, MD), according to the manufacturer's instructions. The X-ray films were scanned, and the spots were analyzed using SuperArray Software. The relative expression level was determined by comparing the signal intensity of each gene in the array after normalization to the signal of a set of housekeeping genes.
Statistics
The number of mice used in each test group was a minimum of four. Experiments were repeated at least once, and measurements were expressed as means ± SD. Pairs of groups were compared through the use of Student's t tests. Differences between groups were considered significant at p ≤ 0.05.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
XW designed the experiments, interpreted the results and prepared the manuscript under the supervision of SM. PR, RP, GH and XK provided technical assistance for experiments. GH also contributed to the manuscript editing. DC provided expertise in scoring TMA slides. KLM and PLD provided expertise in MIF related studies. SSM provided key reagents for the study. SSM, KLM and PLD critically evaluated the manuscript. All authors read and approved the final manuscript.