, Volume 36, Issue 3, pp 643–650

The Proinflammatory Cytokine, IL-6, and its Interference with bFGF Signaling and PSMA in Prostate Cancer Cells


  • Awatef Ben Jemaa
    • Unit of Immunology and Microbiology Environmental and Carcinogenesis (IMEC), Faculty of Sciences of BizerteUniversity of Carthage
  • Sataa Sallami
    • Department of UrologyHospital of La Rabta Tunis
  • Dunia Ramarli
    • Clinical ImmunologyGiovanni Battista Rossi Hospital
  • Marco Colombatti
    • Clinical ImmunologyGiovanni Battista Rossi Hospital
    • Unit of Immunology and Microbiology Environmental and Carcinogenesis (IMEC), Faculty of Sciences of BizerteUniversity of Carthage

DOI: 10.1007/s10753-012-9586-7

Cite this article as:
Ben Jemaa, A., Sallami, S., Ramarli, D. et al. Inflammation (2013) 36: 643. doi:10.1007/s10753-012-9586-7


The aim of the present work was to study the expression of the proinflammatory cytokine, interleukin-6 (IL-6), mediated by bFGF signaling and its possible crosstalk with prostate-specific membrane antigen (PSMA) in LNCaP and PC3-PSMA prostate cancer cell lines. PC3 cells stably transfected with PSMA gene were used for restoring PSMA expression. LNCaP and PC3-PSMA cells were exposed to 10 ng/mL of basic fibroblast growth factor (bFGF). IL-6 production was measured by ELISA assay, and levels of PSMA expression were assessed by flow cytometry. AKT, ERK1/2, and p38 phosphorylation were detected by Western blot. bFGF enhances IL-6 production in LNCaP and PC3-PSMA prostate cancer cells. The effect of bFGF on stimulating IL-6 secretion was greater in LNCaP than in PC3-PSMA cells. In the presence of bFGF, PSMA expression was activated after 4 days of treatment in LNCaP and PC3-PSMA cells. This activation was not maintained after long term of treatment in both metastatic cell lines. Solely MAPKs pathways (ERK1/2 and p38) were activated after bFGF stimulation in both metastatic cell lines, whereas AKT did not show any activation. The interference of the proinflammatory cytokine, IL-6, with bFGF signaling and PSMA, should be of high clinical relevance in the treatment of metastatic prostate cancer. In developing novel therapeutic modalities targeting IL-6, significant attention should be given to PSMA and its inactivation to fight against prostate cancer.


proinflammatory cytokineIL-6bFGFPSMAMAPKsAKTangiogenesisprostate cancer


There is growing evidence that inflammation is a causal factor in the initiation and progression of various cancers, including prostate cancer [1]. The molecular mechanisms that underlie the pathogenesis of inflammation-associated cancer are complex and involve various proinflammatory cytokines. Among these is interleukin 6 (IL-6), which, while traditionally described as key mediator in the inflammatory response, has been also proven to be an integral part of prostate cancer biology [24]. Prostate cancer is an androgen-sensitive cancer, which soon progresses to an androgen-independent state, characterized by androgen ablation and chemotherapy refractoriness [5, 6]. Prostate cancer metastasizes primarily in the bones and also in lymph nodes. Metastasis microenvironment-related growth factors, such as IL-6 and basic fibroblast growth factor (bFGF), contribute to the development of androgen ablation and chemotherapy refractoriness of prostate cancer cells [7]. The biological activities of IL-6 are mediated by a membrane receptor complex composed by IL-6 receptor a (IL-6Ra) and glycoprotein 130 (gp130) resulting in high-affinity binding [8].

bFGF is a member of the fibroblast growth factor (FGF) superfamily, which is comprised of at least 23 different members [9]. bFGF is a pleiotropic growth factor that signals through four cognate tyrosine kinase receptors, designed FGFR-1 to FGFR-4 with different affinity promote prostate tumorigenesis [10]. Human prostate cancer cellular models have confirmed that aberrant bFGF signaling can promote tumor development by directly driving prostate cancer cell proliferation and survival and by supporting tumor angiogenesis [11]. bFGF-induced signal transduction involves the activation of multiple mitogen-activated protein kinases (MAPKs) such as ERK, p38, and c-Jun N-terminal kinase (JNK) [1214]. bFGF can also activate the phosphatidylinositol 3-kinase/AKT (PI3K/AKT) pathway [15]. These pathways are important for many fundamental cellular processes, including proliferation, differentiation, and survival [16, 17]. Another key regulator of prostate cancer is prostate-specific membrane antigen (PSMA) that functions as folate hydrolase and neuropeptidase [18]. Expression levels of PSMA increase several folds in primary, metastatic, and in hormone refractory prostate carcinoma as compared to normal and benign prostatic hyperplasia [19, 20]. This indicates that PSMA has a role not only as an enzyme but also as a protein in cell survival, cell proliferation, and cell migration [21]. Surprisingly, upon prostate cancer progression, the expression of PSMA is lost from androgen-dependent (like LNCaP) to androgen-independent state like in case of metastatic cell lines PC3 [22]. Stably transfection of these dedifferentiated cell line with the gene PSMA is able to restore PSMA expression in PC3 cells.

At present, little is known about the role of the proinflammatory cytokine, IL-6, and its relationship with PSMA and bFGF signaling in advanced prostate cancer cells. In this work, we studied the effect of exogenous bFGF on IL-6 production and its crosstalk with PSMA expression in androgen-dependent (LNCaP) and androgen-independent prostate cancer cell lines (PC3-PSMA). In parallel, the activation of MAPKs (ERK1/2 and p38) and AKT pathways in bFGF-treated LNCaP and PC3-PSMA cells were investigated.


Chemicals, Cell Culture Reagents and Antibodies

Poly-d-lysine, trypsin, and protease inhibitors were purchased from Sigma (Sigma, Milan, Italy). The Hybond-P PVDF membranes and the ECL plus Western Blotting Detection System were from Amersham Bioscience Inc. (Freiburg, Germany). RPMI 1640 medium and glutamine were from Società Prodotti Antibiotici (Milan, Italy). The fetal bovine serum (FBS) was from Celbio (Pero, Milan,Italy). Basic FGF was a product of R&D Systems (Milan, Italy). Fluorescein isothiocyanate (FITC)-labeled (Fab) 2 goat antimouse Ig was from Becton-Dickinson (San Jose, CA, USA), mouse mAbs anti-phospho AKT (Ser473), anti-phospho ERK1/2 and anti-phospho p38 were all from Cell Signaling Technology (Danvers, MA, USA). Rabbit mAb anti-actin was from Sigma (Sigma, Milan, Italy). The anti-PSMA mAb, used for this work, was produced in Pr.Colombatti’s laboratory and was used in a purified form following an affinity chromatography, as described elsewhere [23].

Cell Lines and Treatments

The PSMA-positive LNCaP human prostate carcinoma cell lines were obtained from the American Type Culture Collection and PC3-PSMA cell lines, which were stably transfected with PSMA gene, were kindly provided by Dr. Warren Heston (Cleveland Clinic Foundation, Cleveland, OH, USA). Cells were maintained in vitro by serial passages in RPMI-10 % FBS at 37 °C in a humidified atmosphere of 5 % CO2. In the case of the LNCaP cell line, plating was performed on poly-d-lysine-coated tissue culture plasticware (10 mg/mL). Cells were plated into 6- or 24-well plates and grown overnight in a regular culture medium. Before treatment with bFGF (10 ng/mL), cells were serum-starved by changing media to serum-free RMPI. Experiments were terminated with separate media from each sample of untreated or treated cells.


LNCaP and PC3-PSMA cells were seeded into 24-well plates in regular culture media and were let to attach and reach the subconfluence. The cells were treated with bFGF (10 ng/mL) for 48 h or 72 h. After the incubation period, supernatants were collected and used for further investigation. Cells were harvested by trypsinization and counted. The results of IL-6 measurements were then expressed as picograms of IL-6 per milliliter per 104 cells. Human IL-6 ELISA (Endogen, Woburn, MA, USA) assay was performed according to the manufacturer’s instructions. The results of the assay were normalized according to the cell number.

Flow Cytometry Analysis

LNCaP and PC3-PSMA cells, initially seeded into six-well plates, were maintained at 37 °C in a humidified atmosphere of 5 % CO2 without or with added agents: bFGF (10 ng/mL, R&D Systems) for 4, 5, or 7 days. Media were replaced every 48 h. For 1 h, cells were incubated with saturating amounts of the indicated antibody on ice. Then, they were washed three times with cold RPMI 1640 medium containing 5 % FBS. Cells were then incubated with 100 μL of goat anti-mouse FITC-labeled antiserum (Becton-Dickinson) on ice and in the dark for 1 h. After washings, cells were analyzed by a FACS Canto instrument (Becton-Dickinson). Conjugated isotype-matched, nonreactive monoclonal antibody was used as control.

Western Blot Analysis

Cell extracts were prepared from LNCaP and PC3-PSMA cells, which were cultured overnight into six-well plates with regular medium then with 2.5 % FBS overnight. Finally, they were starved from serum overnight before stimulation with 10 ng/mL bFGF. Aliquots of cell lysates were used to evaluate protein concentration by colorimetric assay. Briefly, untreated and treated cells were lysed for 30 min on ice in a buffer containing 25 mM Hepes pH 7.5, 150 mM NaCl, 10 % glycerol, 1 mM Na orthovanadate, 50 nM Na pyrophosphate, 25 mM Na fluoride, 10 mM MgCl2, 1 % NP40, phosphatase, and protease inhibitors. Equal protein concentrations from each of the cell lysates were analyzed by 10 % sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). After electroblotting the gels into polyvinylidene difluoride (PVDF) membranes, these were probed with different primary antibodies. Enhanced chemiluminescence (ECL) (GE Healthcare Bio-Sciences AB, Uppsala, Sweden) reagents were used to visualize the protein bands on the blots. Film densitometry was carried out with a laser densitometer and analyzed using a built-in software (LKB, Bromma, Sweden).

Statistical Analysis

Statistical analysis of results was performed by applying the parametric paired t test. Significance was accepted when p < 0.05 (GraphPad PRISMA 5.0 computer program).


To study the influence of bFGF on IL-6 secretion, LNCaP and PC3-PSMA cells were stimulated with bFGF (10 ng/mL) as confluent monolayer and harvested 48 or 72 h later. To determine the basal and inducible release of IL-6 in culture, both cell lines and the corresponding cell-free supernatants were separately recovered from each treated or untreated groups of cells. Levels of IL-6 expression in cells supernatant were measured using the ELISA technique. We found that LNCaP cells actively produced more IL-6 after bFGF treatment for 48 and 72 h. IL-6 levels increased 2.5 times more in LNCaP cells (p < 0.001). Similar data were seen in PC3-PSMA cells. In fact, bFGF treatment increased IL-6 production after 48 h of stimulation (1.6-fold; p < 0.001). This increase became moderate after 72 h of treatment in this androgen-independent cell line (Fig. 1).
Fig. 1

Increased IL-6 secretion in LNCaP and PC3-PSMA cells after bFGF treatment. Treatment with bFGF for 48 h or 72 h leads to an increase in IL-6 in both cell lines. LNCaP cells secrete 2.5-fold more of IL-6 in their supernatant after stimulation with bFGF for 48 or 72 h. PC3-PSMA cells exhibit moderate increase in IL-6 levels after bFGF treatment. IL-6 concentration in cell-free supernatants was measured with an ELISA kit used based on the manufacturer’s instructions (Endogen, Woburn, MA, USA). Optical density (OD) values were plotted onto a standard curve and expressed as pg/mL. Total IL-6 of each sample was then normalized according to the cell number. The experiments were performed in duplicate. Values represent the mean ± SD. Stars on top of the bars indicate statistical significance as follows: ***p < 0.001; **p < 0.01; *p < 0.05.

To look at the effect of bFGF on PSMA expression in LNCaP and PC3-PSMA cells, we treated them with bFGF for 4, 5, or 7 days. Levels of PSMA expression were determined by flow cytometry technique (Figs. 2 and 3). We observed a significant increase of PSMA expression (up to 5-fold vs. control) after 4 days of bFGF treatment in both cell lines. After 5 or 7 days of treatment, this effect was lost in PC3-PSMA cells. Surprisingly, it slightly increased in LNCaP cells (Figs. 2 and 3).
Fig. 2

Representative flow cytometry of LNCaP cells (a) and PC3-PSMA (b) treated or not with bFGF and stained with anti-PSMA. LNCaP and PC3-PSMA cells were treated with 10 ng/mL bFGF during a period of 4 or 5 or 7 days. LNCaP and PC3-PSMA cells stained with anti-PSMA mAb. Controls are represented by cells incubated with the second step reagent (goat anti-mouse-FITC IgG) only.
Fig. 3

PSMA expression is regulated by bFGF treatment in LNCaP and PC3-PSMA cells. Consecutive treatment with bFGF during a period of 4 days causes a significant increase of PSMA expression in both cell lines (p < 0.001 vs. control). This effect sustained during a period of 5 or 7 days for LNCaP cells but it is reversed for PC3-PSMA cell line. LNCaP and PC3-PSMA cells stained with anti-PSMA mAb. Controls are represented by cells incubated with the second step reagent (goat anti-mouse-FITC IgG) only. RFU relative fluorescence unit. The experiments were performed in triplicates and with three independent experiments. Values represent the mean ± SD. Stars on top of the bars indicate statistical significance as follows: ***p < 0.001; **p < 0.01; *p < 0.05.

In order to determine whether AKT, MAPK/ERK, and MAPK/p38 pathways are mediated by bFGF effect on IL-6 secretion and PSMA expression, an immunoblotting analysis of phosphorylated target proteins was performed. After 10 min of bFGF addition (10 ng/mL), p-AKT levels did not show any variation in PC3-PSMA cells, but did slightly decrease in LNCaP cells (Fig. 4).
Fig. 4

Specific activation of the MAPK pathway (ERK1/2 and p38) after bFGF stimulation. LNCaP or PC3-PSMA cells were untreated or subjected to bFGF (10 ng/mL) for the indicated times. Akt, ERK1/2, and p38 activation was assessed in crude lysates. Equal amounts (30 μg) of total proteins were boiled in sample buffer and separated by SDS-PAGE. Cell lysates were analyzed by immunoblotting with an anti-phospho-AKT or anti-phospho-ERK1/2 or anti-phospho-p38 mAb, followed by densitometric analysis, and then the ratio was calculated with β-actin as loading control. The experiments were performed in triplicates and with three independent experiments.

bFGF induced a high level of ERK1/2 phosphorylation in both cell lines. Upon bFGF treatment, ERK1/2 became considerably phosphorylated at 10 min. Sustained levels of phosphorylation lasted up to 20 min and then returned to basic levels (Fig. 4). Both cells lines expressed phosphorylated p38 at basic levels. After adding bFGF, a slight increase of p38 activation was seen in LNCaP cells. This increase was more predominant in PC3-PSMA cells (Fig. 4).


In the present study, we demonstrated that bFGF increased significantly IL-6 expression in LNCaP and PC3-PSMA advanced prostate cancer cell lines. bFGF is known to be a fundamental proangiogenic factor and a potent stimulator of vascular endothelial growth factor (VEGF) expression [24, 25]. Moreover, it activates IL-6 expression. We therefore suggest that IL-6 may enhance the angiogenic effect of bFGF in advanced prostate cancer. It was reported, previously, that IL-6 has an important role in the regulation of angiogenesis in several forms of tumors [26, 27]. IL-6 also induces expression of VEGF messenger RNA (mRNA) in various carcinoma cell lines [28]. IL-6 and bFGF play a fundamental role in the regulation of proliferation, apoptosis, development, progression, and angiogenesis in prostate carcinomas [11, 29]. In fact, we had previously shown that expression of IL-6 and its receptors are consistently established in human prostate carcinoma and benign prostate hyperplasia tissues [4, 30]. Therefore, we consider that observed increase in IL-6 expression in prostate tumor tissues [4, 30] could be associated to the paracrine expression of bFGF and its effect on cancer cell proliferation, motility, and angiogenesis. In line with this view, a paracrine interaction of bFGF and IL-6 has been shown in multiple myeloma [31]. Nevertheless, IL-6 could in return induce bFGF-dependent angiogenesis. In fact, Jee et al. [32] demonstrated that bFGF is a downstream effector of IL-6-induced angiogenesis in the skin cancer cell model.

The distinction between androgen-sensitive and androgen-insensitive prostate cancer cell lines is reflected in their responses to IL-6. Several experimental studies have shown that overexpression of IL-6 in androgen-responsive LNCaP cells promotes their androgen-independent growth in vitro and in vivo [33]. Nevertheless, we found that LNCaP responded more efficiently to bFGF than PC3-PSMA prostatic cell lines in the context of IL-6 production. Indeed, bFGF induced greater IL-6 production in androgen-dependent than androgen-insensitive metastatic prostate cancer cell lines. This phenomenon is not quite understood. It could be due to the difference in protein profiles between PSMA-transfected PC3 and PC3-mock model cell lines.

Clinically, the levels of IL-6 and PSMA expression are significantly elevated in many patients with localized, advanced, and hormone-refractory prostate cancer [19, 20, 34]. These clinical observations suggest a possible correlation between high levels of IL-6 production and PSMA expression in high-grade prostate cancer. Therefore, we wanted to investigate whether a functional relationship may exist between the presence of IL-6 bFGF-induced and the level of PSMA expression in LNCaP and PC3-PSMA cell lines.

The present study indicated that after long-term treatment with bFGF (up to 4 days), levels of PSMA expression significantly increased in both prostate metastatic cell lines (LNCaP and PC3-PSMA). In order to elucidate the importance of this observation, we extended our treatment time up to 7 days. We found that this increase was maintained but was not significant in androgen-dependent LNCaP cells. In contrast, levels of PSMA expression declined after 5 and 7 days of stimulation with 10 ng/mL bFGF in PC3-PSMA cells. This finding is not concordant with the data of Laidler et al. [22]. According to their study, the same dose of bFGF was able to significantly stimulate PSMA expression in androgen-dependent LNCaP cells. Noticeably, bFGF was able to restore expression of PSMA gene after 5 days of treatments in both metastatic cell lines PC3 and DU145 [22]. We could explain this disparity by the fact that we used a PSMA-transfected PC3 cell line, whereas the laboratory of Laidler et al. had used the PC3-mock model cell line. Moreover, these authors limited their investigation on PSMA expression at the gene level using real-time PCR analysis only [22], while we focused on PSMA expression at the protein levels using flow cytometry analysis. The mechanism by which the stimulation of PSMA expression by bFGF is suppressed after 5 or 7 days is not well understood. Perhaps, the androgen-independent PC3-PSMA cells are able to become insensitive to bFGF. bFGF could inhibit PSMA expression through biological mechanisms of attenuation and negative feedback control..

bFGF, as a fundamental agent of angiogenesis [24], activates PSMA expression as well. As a result, we suggested that PSMA may enhance the angiogenic effect of bFGF in metastatic prostate cancer. In this setting, we have previously demonstrated the association of PSMA expression with high intratumoral angiogenesis activity in prostate cancer [19]. In the same context, Liu et al. recently reported the presence of PSMA expression in newly formed vessels, which suggests its involvement in angiogenesis and in developing tumors. Indeed, PSMA has been detected in human umbilical vein endothelial cells in response to a breast-tumor-conditioned medium [35]. Several studies revealed the presence of PSMA in neovasculature of several nonprostatic tumors [21, 36], which indicates more its important potent role in angiogenesis. Our present results may imply that IL-6, PSMA, and bFGF are mitogenic and represent angiogenic molecules, which act in synergy in advanced prostate cancer cells. As a result, we suggested a crosstalk between IL-6, PSMA, and bFGF signaling in androgen-sensitive and androgen-insensitive prostate cancer cells. To investigate signaling pathways mediating the effect of bFGF on IL-6 and PSMA transcriptions, the phosphorylation status of AKT, ERK1/2, and p38 were determined by Western blot analysis. The treatment of LNCaP and PC3-PSMA cells with bFGF did not induce PI3K/AKT activation, whereas it activated MAPK p44 and p42. These data are in accordance with others previously published, like Puhr et al. [37], Wesley et al. [11], and Hatziapostolou et al. [38]. Similar to ERK1/2, p38 was activated after adding bFGF in both advanced prostatic adenocarcinoma models. Recently, and for the first time, it has been shown that PSMA recruitment at the surface of LNCaP cells induced IL-6 upregulation through eliciting a signaling wave involving MAPKs p38 and ERK1/2 activation in LNCaP cells [23]. We suggest that PSMA is also able to induce IL-6 upregulation through bFGF signaling (p38 MAPK and ERK1/2) in LNCaP and PC3-PSMA cells. We think that the same mechanism is happening in LNCaP and PC3-PSMA cells.

p38 MAPK pathway is mostly known for its involvement in the production and action of inflammatory mediators, such as IL-6, which plays a significant role in the recruitment of leukocytes to inflammatory sites and in chronic inflammation of prostate cancer progression [4, 39]. Thus, the capacity of metastatic prostate carcinoma cells to produce high amount of this proinflammatory cytokine in response to bFGF-induced PSMA makes these cells important components of the immune and inflammatory responses. Our results support the emergence of IL-6 as well PSMA as promising targets of immunotherapy trials for prostate cancer [29, 40].

The sensibility of the proinflammatory cytokine, IL-6, and PSMA to bFGF through activation of p44/p42 and p38 MAPKs pathways confirms the relationship between these two molecules and their involvement in chronic inflammation of prostate cancer progression. These results suggest that interference of PSMA and bFGF signaling with IL-6 function or expression should be of high clinical relevance in the treatment of prostate cancer. In developing novel therapeutic modalities targeting IL-6, significant attention should be given to PSMA and its inactivation to fight against prostate cancer.


We thank the University of Carthage (Tunisia). Awatef Ben Jemaa had a predoctoral fellowship from the University of Carthage (Tunisia) during the course of this work.

Conflict of interest

The authors declare no conflict of interest.

Copyright information

© Springer Science+Business Media New York 2012