Background
Prostate cancer (PCA) is the second leading cause of cancer-related death in men with a death rate reaching 26.7% for 2001-2005 in United States[
1]. Chronic inflammation has been suggested to play a major role in prostate oncogenesis[
2]. Furthermore, local and systemic immunosuppression have also been reported in patients bearing PCA[
3‐
6]. The clarification of molecular mechanisms underlying these phenomena might provide novel insights into PCA induction and progression, of potentially high clinical relevance.
Cytokines represent key mediators of inflammation and play pivotal roles in the interaction between immune system and cancer. A number of them have been suggested to be associated with advanced stage PCA[
7]. IL-4 and, in particular, IL-6 have been shown to exert antiapoptotic effects on PCA cells, whereas proangiogenic effects of IL-8 have been demonstrated[
7].
A group of lymphokines including, among others, IL-2, IL-7, IL-15 and IL-21 share a receptor γ-chain[
8]. Common receptor γ-chain (c- γ) cytokines regulate lymphocyte development[
9‐
12] and support CD4+ and CD8+ T cell homeostatic proliferation and functions[
13,
14]. IL-2, IL-7, IL-15 and IL-21 are of particular interest in cancer immunotherapy[
15]. Importantly, IL-2 is FDA approved for the treatment of melanoma and renal cell carcinoma, whereas IL-7 has been used with promising results in phase I/II trials in clinical oncology[
16,
17]. IL-15 has not yet been used in clinical trials. However, experimental models and "in vitro" evidence suggest that it may represent a cancer treatment with high potential clinical relevance[
18‐
20]. Finally, IL-21 has been shown in murine models to prevent T lymphocyte exhaustion induced by chronic stimulation[
21‐
23]. In phase I/II clinical trials, IL-21 has also shown reproducible antitumor effects[
24,
25].
Notably, however, elevated IL-7 serum levels have been detected in Hodgkin disease and in ovarian cancers [
26‐
29]. Furthermore, IL-7 has been shown to be produced by breast and colorectal cancer cells[
30,
31]. IL-15 has also been shown to be produced by colon cancer cells and specific gene expression has been shown to be associated with distant metastases [
32]. Moreover, elevated IL-15 serum levels were detected in multiple myeloma[
33].
Little is known about IL-2, IL-7, IL-15 and IL-21 gene expression and protein secretion in patients bearing PCA or BPH. Indeed, IL-2 gene expression and IL-15 protein have been detected in BPH tissues[
34,
35]. Furthermore, IL-7 gene has been found to be expressed in PCA tissues[
36,
37] and the corresponding protein has been detected in patients' sera. However, no comparative analysis in patients with BPH and PCA has been performed so far and no data are currently available for IL-21 expression in PCA and BPH.
In order to evaluate the potential role of these IL in early phases of prostate oncogenesis, in this study we have evaluated the expression of specific genes in prostatic tissues from patients bearing either localized pT1-2c PCA or BPH and we have measured the titres of circulating proteins in their sera.
Our results indicate that in prostatic tissues from patients bearing early stage PCA, the expression of IL-7 and IL-15 genes is significantly enhanced, as compared to BPH, while no significant differences could be observed in IL-2 or IL-21 gene expression. Accordingly, titres of circulating IL-7 and IL-15 are increased in patients with localized PCA, as compared to patients with BPH. Most interestingly, IL-7 titres might help to discriminate sera from patients with PCA or BPH.
Patients and methods
Patients
We investigated a consecutive series of 159 specimens from men diagnosed for BPH or PCA at the Department of Urology of the University Hospital of Basel (Switzerland) from 2007 to 2010. Patients with BPH underwent conventional transurethral resection (TUR-P), while patients with PCA underwent either palliative TUR-P or endoscopic extraperitoneal radical prostatectomy (EERP). Relevant clinical data were collected by reviewing patients' files.
Written informed consent was obtained from patients in accordance with the requirements of the Ethical Committee of Basel (EKBB, Ref.Nr. EK: 176/07).
Quantification of gene expression in prostatic tissues by quantitative Real-Time PCR
Prostatic tissues were screened for the presence of tumors by an experienced pathologist (L. Bubendorf). Total RNA was extracted by using RNeasy® MiniKit (Qiagen, Basel, Switzerland), deoxyribonuclease I (Invitrogen, Carlsbad, California) treated and reverse transcribed by using M-MLV Reverse Transcriptase (Invitrogen, Carlsbad, California). Quantitative RT-PCR was performed using the TaqMan® Universal PCR Master Mix (Applied Biosystems, Forster City, CA), and the following primers and probes:
Fwd ATGGGGAAGGTGAAGGTCG
Rev TAAAAGCAGCCCTGGTGACC
Probe FAM-CGCCCAATACGACCAAATCCGTTGAC-TAMRA
Fwd AACTCACCAGGATGCTCACATTTA
Rev TCCCTGGGTCTTAAGTGAAAGTTT
Probe FAM-TTTTACATGCCCAAGAAGGCCACAGAACT-TAMRA
Fwd CAGCCCTGAGAAAGGAGACATG
Rev GGTTCAGGTTGTTTTCTGCCA
Probe FAM-AGTAACATGTGTGAAAGCAGCAAAGAGGCAC-TAMRA
IL-21, IL-7 and IL-15 primers and probes were provided by Assays-on-Demand, Gene Expression Products (Applied Biosystems, Forster City, CA).
Specific gene expression was quantified by using the 2
-ΔΔCT method[
41]. Normalization of gene expression was performed using GAPDH as reference gene.
ELISA assays
Commercial ELISA kits for IL-6, IL-7 and IL-15 were obtained from Pharmingen (San Diego, CA). Detection limit was 1pg/ml for all assays. HMGB1 ELISA kit was obtained from Shino-Test Corporation (Kanagawa, Japan). SAA and CRP ELISA kits were obtained from USB (Swampscott, MA) and Roche Diagnostics Cobas® (Mannheim, Germany), respectively. Detection limits were 1 ng/ml, 0.6 ng/ml, and 0.6 μg/ml, for HMGB1, SAA and CRP, respectively. All assays were performed in accordance to the instructions from the producers.
Statistical Analysis
Mann-Whitney tests were used to compare mean expression of cytokine genes and cytokine serum titers for independent samples while Wilcoxon test was used for paired samples. When n ≥ 30, Student's T-Tests were used to compare means. Frequencies of specific gene expression in two groups were assessed by χ
2 tests. Differences with p < 0.05 were considered significant. Receiver operating characteristic (ROC) curves were created to verify whether cytokine titers could be used to distinguish patients with PCA from patients with BPH [
27].
Discussion
A large body of literature based on histopathological, epidemiological and molecular pathology data suggests that chronic inflammation might play an important role in prostate oncogenesis [
2]. IL represent essential mediators of inflammation and titres of a number of them have been shown to be increased in patients bearing PCA[
7]. Most of these studies, however, have addressed IL detection in patients with advanced stage cancers. In these conditions, high tumor burdens and mechanisms inherent with metastatic spread, e,g, osteolysis, might be involved in the induction of IL production. Furthermore, sterile inflammation primed by intratumoral ischemia and associated necrosis could also promote IL release [
51]. Therefore, data from patients with advanced tumors are unlikely to provide useful information on the role eventually played by inflammation or immunosuppression in prostate oncogenesis. In order to gain insights into events occurring in early stages of prostate cancerogenesis, in this study we evaluated IL gene expression and protein production at local and systemic levels, respectively, in localized PCA, as compared with BPH.
Our data clearly indicate that the expression of a number of genes encoding pro-inflammatory and homeostatic IL, including IL-6, IL-7 and IL-15 is detectable more frequently and to a higher extent, in early stage PCA as compared to BPH tissues. Notably, the genes encoding these IL were also found to be expressed to different levels by established PCA cell lines.
To investigate the possibly systemic nature of the pro-inflammatory state detectable in PCA[
37,
52], we measured IL titres in sera from patients with BPH or early stage PCA. We report here for the first time that IL-7 and IL-15 serum levels are significantly higher in patients with localized PCA than in patients with BPH. IL-6 titres were also significantly higher in patients with PCA than in patients with BPH despite the early stage of their disease [
53].
IL-6 is known to be produced by multiple cell types, including activated macrophages, smooth muscle cells and PCA cells [
53]. Instead, IL-7 is produced by multiple stromal cell types and epithelial cells from different districts including thymus and gut[
10]. Most recently IL-7 production has been detected in normal, but not in transformed prostate epithelial cells[
36]. IL-15 is typically produced by activated antigen presenting cells of the myeloid lineage, including monocytes, macrophages and dendritic cells. However, IL-15 gene has also been found to be expressed by a multiplicity of other cell types, including stromal and epithelial cells[
9]. Importantly, epithelial BPH cells have been shown to express IL-15[
34]. In our patients, IL serum titres were not significantly correlated with specific gene expression in the autologous prostate tissues, thereby suggesting that cells other than tumor cells might be at least in part responsible for IL production. Thus, while we here show that IL-6, IL-7 and IL-15 genes are expressed in established PCA cell lines, future studies are warranted to clarify the cellular sources of these IL within the prostate and systemically.
The nature of the mechanisms promoting IL-6, IL-7 and IL-15 gene expression and protein production in PCA is also unclear. Different microorganisms, including virus and bacteria[
54,
55] have been suspected to contribute to oncogenesis in the prostate. Indeed, their presence could result in chronic inflammation. Alternatively, it has been suggested that direct damage of epithelial cells might be caused by toxic compounds contained in urines or by dietary factors[
2]. Cell death might, in turn, result in the activation of the innate immune system by DAMPs [
49]. Soluble factors produced by DAMP-triggered innate immune system cells, including IL-6, have repeatedly been suggested to promote prostate cancerogenesis[
2]. However, here we show that at least circulating levels of HMGB1, SAA and CRP acute phase proteins are similar in patients with BPH or localized, early stage PCA.
Analysis of ROC curves indicates that IL-7, but neither IL-6 nor IL-15 titres do distinguish sera from patients with PCA from sera from patients with BPH. These findings might set the stage for larger studies also addressing the clinical relevance of IL-7 serum titres in the monitoring of PCA response to treatment and recurrences.
On the other hand, our data raise the issue of potential functional consequences of c- γ cytokine gene expression and secretion in PCA. Preliminary experiments indicate that none of the IL under investigation enhances the proliferation of cells from established PCA cell lines (data not shown). However, studies performed in breast cancer[
56] indicate that IL-7 promotes the production of pro-angiogenic factors by tumor cells and proliferation of endothelial cells. IL-15 has also been shown to stimulate angiogenesis "in vivo"[
57]. Furthermore, although IL-7 and IL-15 are known to promote survival and functional maturation of T cells, they have also been demonstrated to induce the expression of PD-1 "exhaustion" marker in these cells[
58]. Indeed, PCA tissues have been shown to be infiltrated by T cells largely expressing PD-1, and, less frequently, PD-L1 and PD-L2[
59,
60]. These data are consistent with previous reports on the phenotypic characteristics of circulating and tissue infiltrating lymphocytes in chronic viral infections[
61‐
64]. In this context, it is remarkable that we found that the expression of the IL-21 gene, that is of the gene encoding the cytokine preventing PD-1 mediated T cell exhaustion[
21‐
23] is rarely detectable in PCA.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
All authors read and approved the final manuscript. CM, CLM, ET and KY participated in the design, acquisition and analysis of data. GCS and MP and SW conceived the study; GCS, MH and AB provided funding support and revised the manuscript critically for important intellectual content. LB made substantial contributions to the analysis and interpretation of data.