- Email: [email protected]
Protein kinase A (PKA) pathway is functionally linked to androgen receptor (AR) in the progression of prostate cancer
Urologic Oncology: Seminars and Original Investigations 32 (2014) 25.e1–25.e12
Original article
Protein kinase A (PKA) pathway is functionally linked to androgen receptor (AR) in the progression of prostate cancer Martuza Sarwara, Sabina Sandberga, Per-Anders Abrahamsson, M.D., Ph.D.b, Jenny L. Persson, Ph.D.a,* a
Division of Experimental Cancer Research, Department of Laboratory Medicine, Malm¨o, CRC, Lund University, Malm¨o, Sweden b Department of Clinical Sciences, Division of Urologic Research, Sk˚ane University Hospital, Lund University, Malm¨o, Sweden Received 14 October 2011; received in revised form 10 August 2012; accepted 20 August 2012
Abstract Objectives: In the present study, we investigated whether the cyclic adenosine monophosphate (cAMP)-activated protein kinase A (PKA) pathway may regulate the expression of AR and prostate-specific antigen (PSA) and whether there is a correlation between the expression of cAMP/PKA-associated genes and androgen receptor (AR) in patients with prostate cancer (CaP). Materials and methods: The functional studies were performed in LNCaP and PC3 cell lines. Data on the mRNA expression of sets of genes in human clinical samples, including prostate tissues from organ donors, prostate primary cancer, and metastatic cancer, were extracted from the National Center for Biotechnology Informations Gene Expression Omnibus (GEO) database. Statistical tests were applied. Results: We showed that elevated levels of cAMP/PKA pathways induced an increased expression of AR and PSA proteins in LNCaP cells in the absence of androgen. A cAMP-associated phosphodiesterase-4 (PDE4) inhibitor, rolipram induced an up-regulation in AR expression, whereas a cAMP enhancer, forskolin increased PSA level without affecting AR expression. Forskolin treatment increased the level of PKA R1a in LNCaP cells, but remarkably inhibited R1a expression in aggressive PC3 cells. In patients with CaP, we found that the expression of genes encoding R1a and phosphodiesterase-4B was statistically significantly lower in the metastatic specimens than that in the primary CaP specimens or in the normal prostate tissues (P o 0.01) and was reversely correlated with AR expression. Conversely, AR and PRKAR2B mRNA expressions were significantly higher in metastatic lesions than those in the primary CaP specimens or in the normal prostate tissues (P o 0.01). Conclusion: Our study revealed a novel mechanism to precisely define the functional and clinical interrelationship between the cAMP/ PKA pathway and AR signaling in the development of androgen-independent growth of CaPs and metastasis progression. r 2014 Elsevier Inc. All rights reserved. Keywords: Rolipram; Forskolin; AR; cAMP/PKA pathways; CaP metastasis
1. Introduction Prostate cancer (CaP) is one of the most common malignancies in men. The growth of cancer cells is highly dependent on androgens in the early stages. Androgen receptors (ARs) mediate the effects of hormones and play a central role in the formation and progression of tumors. Abbreviations: cAMP, cyclic adenosine monophosphate; PKA, protein kinase A; CaP, prostate cancer; CRPC, castration-resistant prostate cancer; PDE, phosphodiesterase; GEO, Gene Expression Omnibus. * Corresponding author. Tel.: þ46-40-391106; fax: þ46-40-391222. E-mail address: [email protected] (J.L. Persson). 1078-1439/$ – see front matter r 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.urolonc.2012.08.019
Most primary CaPs exhibit elevated levels of androgen and are initially sensitive to androgen deprivation therapy (ADT). Despite therapy, the disease often progresses into a castration-resistant prostate cancer (CRPC) [1–3]. Although CRPC is no longer dependent on androgen stimulation, AR is still expressed in tumor cells and the disease becomes highly aggressive [3]. Thus, alterations in AR expression and its associated pathways may lead to the development of CRPC and CaP metastasis [4]. AR-mediated growth-promoting pathways are therefore important therapeutic targets in CRPC. A number of cellular pathways that crosstalk with AR signalings have been studied. It has been shown that the epidermal growth factor (EGF) pathway can trigger AR
25.e2
M. Sarwar et al. / Urologic Oncology: Seminars and Original Investigations 32 (2014) 25.e1–25.e12
transcriptional activity in CaP cell lines [5,6]. The cytokine interleukin-6 (IL-6) and a family of neuropeptides, including bombesin- and gastrin-releasing peptides, have been shown to promote androgen-independent growth in several CaP cell lines [6,7]. One of the major intracellular signal transduction pathways – the cyclic adenosine monophosphate (cAMP)activated protein kinase A (PKA) pathway – is implicated in CaP progression [8]. An elevated cAMP level leads to an increased expression of prostate-specific antigen (PSA) mRNA, and the rise in the PSA gene expression by PKA requires a functional AR, suggesting that cAMP and its downstream PKA pathways may regulate the expression and activity of AR [9]. However, the precise role of cAMP/PKA in the development of CRPC and the relationship between cAMP/PKA and AR signaling in CRPC are less explored. The PKA pathway is critical for the fundamental cellular processes, including proliferation, differentiation, and apoptosis [10]. The synthesis of cAMP is mediated by adenylyl cyclase from adenosine triphosphate (ATP). The degradation of cAMP is mediated by a large family of cAMP-specific phosphodiesterases (PDEs) enzymes. Inhibition of PDEs, in particular by rolipram, a PDE4 inhibitor, leads to elevated levels of cAMP [8]. It is known that cAMP triggers the activation of downstream PKA, which facilitates the major cellular effects of cAMP [8]. PKA exists as a tetrameric holoenzyme consisting of 2 regulatory (R) and 2 catalytic (C) subunits forming a holoenzyme R2C2 [11]. The R subunits are the major intracellular receptors of cAMP. Binding of cAMP to the R subunits leads to the release of C subunits from the R-C complexes and allows the C subunits to phosphorylate downstream substrates [10,11]. The R subunits are further distinguished as R1a and R1b, and R2a and R2b, each of which are encoded by different genes like PRKAR1A, PRKAR1B, PRKAR2A, and PRKAR2B, respectively [11]. The C subunits, Ca, Cb, and Cg, are coded by the genes PRKACA, PRKACB, and PRKACG, respectively [10,11]. PKA exerts the effects of cAMP on several key transcriptional factors, and among them the cAMP response element (CRE)-binding protein 1 (CREB/CREB1) is the principle mediator [10,12]. The effects of cAMP/PKA on the cells are partly mediated by the transcriptional factor CREB1, which modulates the expression of a large number of genes involved in growth, survival, metabolism, reproduction, transport, and immune regulation [13,14]. Thus, cAMP/ PKA pathways are recognized as the central signaling transduction pathways that are required for multiple cellular functions, including metabolism, cellular growth, differentiation, gene expression, and apoptosis [10]. Increasing evidence has shown that cAMP/PKA pathways may play a role in CaP progression. It has been shown that ARs can be activated by the PKA activator, forskolin in an androgen-independent manner in PC3 cells cotransfected with AR responsive reporter vectors and AR expression vectors [15]. Furthermore, activation of PKA by forskolin leads to the increased mRNA level of PSA in CaP cell lines
[9]. Further, forskolin-induced increase in PSA mRNA expression was reversed by the addition of PKA inhibitor or by an AR antagonist, bicalutamide, in LNCaP cells [9]. Moreover, double knockdown of the AR and PKA subunit R1a induced a growth arrest of CaP cells with a better effect in comparison with what was achieved by a single knockdown of AR [16]. These data indicate that cAMP/ PKA pathways may be the upstream mediators of ARs in CaP cells. However, the precise role of cAMP/PKA in the development of CRPC and the relationship between cAMP/ PKA and AR signaling are less explored. The aim of our study is therefore to investigate whether cAMP/PKA may be responsible for the activation of AR/ PSA signaling during the progression of CRPC in the absence of androgen. We also assessed whether elevated levels of cAMP/PKA may sensitize CaP cells to low levels of androgen and thus promote progression of tumor cells from the androgen-dependent state to CRPC status. To further validate the correlation between the cAMP/PKAassociated regulators and AR expression, we utilized National Center for Biotechnology Information’s Gene Expression Omnibus (GEO) database and analyzed the mRNA expression profiles of the sets of genes of interest from the prostate tissues of organ donors, and the tumor tissues and metastatic lesions from patients with CaP. Our findings for the first time showed that cAMP-dependent pathways regulate AR/PSA expression. In the aggressive CaP cells, cAMP-dependent pathways mediate the matrix metalloproteinase (MMP)-9 expression, suggesting that cAMP-dependent pathways may promote adaptation of CaP cells to a more aggressive phenotype in the absence of hormones. Our data from patient studies support our findings in cell line-based studies and suggest that cAMP/ PKA may in part contribute to the progression of CRPC by increasing the levels of AR/PSA expression. 2. Materials and method 2.1. Cell lines The PNT1A cell line, androgen-dependent cell line LNCaP, and the androgen-independent cell lines PC3, DU-145, and PC3M were purchased from the American Type Culture Collection (ATCC, Manassas, VA). The cell lines were maintained in RPMI 1640 medium supplemented with 10% fetal bovine serum, 1% penicillin-streptomycin, and 1 mM L-glutamine (Life Technologies, Paisley, UK) at 37 1C in a humidified atmosphere with 5% CO2. 2.2. Treatment For the treatment, rolipram, the cAMP-specific PDE4 inhibitor, and forskolin, the cell permeable cAMP enhancer (Calbiochem, Darmstadt, Germany) dissolved in 100% dimethyl sulfoxide (DMSO), were used. LNCaP or PC3
M. Sarwar et al. / Urologic Oncology: Seminars and Original Investigations 32 (2014) 25.e1–25.e12
cells seeded at a density of 0.25106 or 0.15106 cells/ml in RPMI 1640 medium supplemented with 10% fetal bovine serum were cultured for 24 hours. Cells were treated for the final concentration of 1 mM rolipram or 1 mM forskolin or 5 nM dihydrotestosterone (DHT) or a combination of them for 24 hours. For the establishment of serumfree condition, LNCaP cells were cultured for 24 hours, washed with 1 phosphate buffered saline, and subsequently cultured in serum-free phenol red-free RPMI 1640 medium for a further 24 hours prior to the addition of different agents as mentioned previously. 2.3. Antibodies Primary antibodies against PKA RIa/b at 1:500 dilutions (Cell signaling technology, Danvers, MA), PKA RIIb at 1:1,000 (BD Biosciences Transduction Laboratories, San Jose, CA), AR at 1:250, PKAa at 1:500, p-CREB1 at Ser 133 sites at 1:500 (Santa Cruz Biotechnology Inc., Santa Cruz, CA), PSA at 1:500 (DAKO, Glostrup, Denmark), PDE4 (Abcam, Cambridge, UK) at 1:250, MMP-9 (Abcam, Cambridge, UK), and anti-b-actin at 1:10,000 (MP biochemicals, Illkirch, France) were used. 2.4. Immunoblot analysis The cells were harvested and lysed in ice-cold radioimmunoprecipitation assay buffer (120 mM NaCl, 50 mM Tris–HCl, pH 7.6, 50 mM NaF, 0.1 mM Na3VO4, 1% NP40, 1 mM phenylmethylsulfonyl fluoride) (Sigma, St. Louis, MD) and 15% protease inhibitor cocktail Complete Mini (Roche, Basel, Switzerland). Then, 20 mg of protein was separated on a 12% sodium dodecyl sulfate polyacrylamide gel electrophoresis gel and transferred onto nitrocellulose membranes. Signals were visualized using an Enhanced ChemiLuminescence detection system (Millipore Corp Sweden, Solna, Sweden) and documented with an AlphaImager CCD system. Densitometric quantification of immunoblots was carried out by the ImageJ Image Analysis Software (NIH, Baltimore, USA), represented as a fold change relative to control, and was normalized with an actin band.
25.e3
(Invitrogen, Stockholm, Sweden) were used. The cells were counterstained with 40 ,6-diamidino-2-phenylindole, and microphotographs were taken under 20 magnifications with an Olympus AX70 microscope equipped with appropriate filters from Semrock and fitted with a digital color camera (Nikon DS-U1, Stockholm, Sweden). The software ACT2U was used (ACT2U version 1.5, Stockholm, Sweden). For morphologic analysis, cells were prepared as described previously, and the phase-contrast images were monitored using an Olympus AX70 microscope. Alternatively, the images were accessed, and the photomicrographs were taken at 10 magnification using a HoloMonitorTM M3 (Phase Holographic Imaging AB, Lund, Sweden). 2.6. Patient samples and data collections For the immunoblot analysis, tumor and adjacent normal specimens that had been surgically removed from patients were obtained from Departments of Surgery and Urology, and Department of Clinical Pathology and Cytology, Lund University, University hospital in Malm¨o, Sweden. The above-mentioned tissues were paraffin-embedded and sectioned for histologic analysis. The tumor samples did not contain a significant amount of normal tissues and viceversa. This study was approved by the ethics committee of Lund University, Sweden, and the Helsinki Declaration of Human Rights was strictly observed. Gene expression data from the data set GDS2545 in the Gene Expression Omnibus database at the National Center for Biotechnology Information website was used. The data set was obtained by performing an Affymetrix HG-U95Av2 oligonucleotide array platform as described [17,18]. The mean mRNA values of genes of interest from a total of 171 human samples in the data set GDS2545 were used in this study. The samples included organ donor prostate tissues free of any pathologic diseases (n = 18), normal prostate tissues adjacent to tumor (n = 63), primary tumor (n = 65), and metastatic lesions (n = 25) from the liver, para-aortic lymph node, paratracheal lymph node, retroperitoneal lymph node, lung, and adrenal gland of 4 patients with CRPC. 2.7. Statistical analysis of gene expression profiles
2.5. Immunofluorescence and morphologic analysis LNCaP cells were grown on glass coverslips, starved for serum and androgen by growing in serum-free, phenol redfree RPMI for 24 hours, and treated with 1 mM rolipram, 5 nM DHT, and 1 mM forskolin alone or in combination, for 24 hours. Cells were fixed with 4% paraformaldehyde in phosphate buffered saline. For blocking background staining from nonspecific interactions, an Image-iTTM FX signal enhancer (Molecular Probes, Inc.) was used. Primary antibodies including anti-AR at 1:100 dilutions and anti-PSA at 1:250 dilutions were used. The secondary antibodies including rabbit anti-mouse Rhodamine (Chemicon/Millipore) at 1:200 and goat anti-rabbit alexa fluor 488 at 1:500
We analyzed the original data of DNA microarray gene expression profiles (data sets GDS2545) using the BRBArray Tool (Version: 4.1.0), a software developed by R. Simon and A. Pang Lam at the Biometric Research Branch of the National Cancer Institute [19]. The raw signal intensities of the arrays were preprocessed by setting the intensity at the minimum value, if the intensity is below the minimum value of 10, and normalized by subtracting the median log ratio of an array by all the log ratios on that array. The genes were excluded if less than 20% of expression data showed at least a 1.1-fold change in either direction from the median value of the gene. Two-sample t-tests were used to identify the genes that are differentially
25.e4
M. Sarwar et al. / Urologic Oncology: Seminars and Original Investigations 32 (2014) 25.e1–25.e12
Table 1 Names of the genes included in the present study Gene symbol Description AR PRKAR1A PRKAR2B PDE4B
Androgen receptor Protein kinase, cAMP dependent, regulatory, type I, alpha Protein kinase, cAMP dependent, regulatory, type II, beta Phosphodiesterase 4B, cAMP-specific
expressed between metastasis and primary tumor with a significance level of P o 0.05. False discovery rates were calculated based on 1,000 sample permutations. Tukey test, ANOVA, Kruskal-Wallis test, Mann-Whitney test, and Spearman rank correlation test were performed with the Statistical Package for the Social Sciences software (SPSS, version 17; Chicago). P-value o0.05 was considered as significant. The expression of genes and their correlations are listed in Tables 1 and 2.
3. Results 3.1. Expression of several key components in cAMP/ PKA-associated pathways in patients with CaP and prostate cancer cell lines To understand the role of cAMP signaling cascades in CaP progression, we first examined protein levels of PDE4, phosphorylated CREB1, and phosphorylated ATF1 that are cAMP/PKA downstream effectors in various types of CaP cell lines. PDE4, phosphor-CREB1, and phosphor-ATF1 were detected in all of the 5 cell lines, including the nonmalignant PNT1A cells, malignant androgen-sensitive LNCaP cells, and androgen-insensitive DU-145, PC3, and Table 2 Comparison of the expression of genes among clinical subgroups Genes
P values between groups
AR Normal vs. primary tumor Normal vs. metastasis Primary tumor vs. metastasis
0.14 0.00** 0.000**
PRKAR1A Normal vs. primary tumor Normal vs. metastasis Primary tumor vs. metastasis
0.000** 0.000** 0.877
PRKAR2B Normal vs. primary tumor Normal vs. metastasis Primary tumor vs. metastasis
0.615 0.025* 0.006**
PDE4B Normal vs primary tumor Normal vs. metastasis Primary tumor vs. metastasis
0.161 0.000** 0.000**
*
P o 0.05 level (2-tailed). P o 0.01 level (2-tailed).
**
PC3M cells, which represent different stages of CaPs. The lowest levels of PDE4 and phosphor-CREB1 were observed in PNT1A and PC3M cells (Fig. 1A). We next examined the expression of PDE4 and PKA subunits in clinical samples including normal and cancer specimens from patients with CaP, breast cancer, or testicular cancer. PDE4D was observed in all normal and paired cancer specimens examined (Fig. 1B), whereas PKA catalytic subunit Ca (PKACa was barely detectable in breast specimens, but was observed in normal testes and testicular cancers. In prostate specimens, PKACa protein was expressed in normal prostate and in the PC3 cell line, but was absent in 2 CaP specimens (Fig. 1B).
3.2. Regulation of AR and PSA by cAMP/PKA-associated regulatory pathways It is known that PDE enzymes facilitate the degradation of cAMP to keep the balance of cAMP levels in the cell. Treatment of different types of cells with rolipram, a selective PDE4 inhibitor, leads to the elevated levels of cAMP [8]. Similarly, treatment of CaP cells with forskolin leads to the increased activity of cAMP [8]. Thus, cAMP/PKA levels in CaP cells can be induced by rolipram and forskolin treatment. To this end, we investigated the interrelationship between cAMP/PKA pathways and AR signaling. We examined the effects of rolipram or forskolin on the protein levels of AR and PSA in LNCaP cells using immunoblot analysis. Rolipram treatment led to an increased level of AR protein expression and a moderate induction of PSA (Fig. 2A and C). In parallel, forskolin treatment increased the PSA level (Fig. 2A and C), but had no effect on AR expression (Fig. 2A). Taken together, these data suggest that the cAMP mediators, rolipram and forskolin influence the AR signaling via distinct mechanisms. Rolipram and forskolin combination treatment had no effect on AR and PSA expressions (Fig. 2A and C). This result suggests that rolipram antagonizes the effect of forskolin on AR and PSA expressions via an unknown mechanism. Thus, rolipram and forskolin in combination had no additive effect on AR signaling compared with rolipram or forskolin alone. Because the above-mentioned data showed that cAMP/ PKA pathways participated in the regulation of AR/PSA expression in the absence of androgen, we aimed to investigate whether cAMP/PKA and androgen together might have an additive effect on AR/PSA expression. We treated LNCaP cells with DHT alone or in combination with rolipram and forskolin. As expected, DHT treatment induced a robust increase in AR and PSA expressions in LNCaP cells, and this is statistically significant (P ¼ 0.04). Interestingly, the effect of rolipram on AR was comparable to the effect of DHT (Fig. 2A). Combined treatment of DHT together with forskolin or rolipram also significantly increased PSA expression (P o 0.01) (Fig. 2A and C). This indicates that cAMP/PKA pathways participate in the regulation of AR/PSA expression in the presence of androgen.
M. Sarwar et al. / Urologic Oncology: Seminars and Original Investigations 32 (2014) 25.e1–25.e12
25.e5
Fig. 1. Expression of several key components of the cAMP- or PKA-associated pathways in prostate cancer cell lines and in prostate cancer, along with breast cancer and testicular cancer samples. (A) Immunoblot analysis of the expression of PDE4 and phosphorylation of CREB1/ATF1 in human immortalized prostatic cell line PNT1A, androgen-sensitive LNCaP, and the hormone-insensitive DU145, PC3, and PC3M cells. (B) Expression of PDE4D and PKA Ca in normal indicated as “N” and prostate cancer, along with breast cancer and testicular cancer specimens indicated as “C”.
3.3. Regulation of PKA subunits by DHT in LNCaP cells We next examined the effects of rolipram, forskolin, and DHT, alone or in combinations, on the levels of PKA regulatory subunits R1a and R2b and PKA catalytic subunit Ca. Forskolin alone induced the increased levels of R2b (P ¼ 0.024) and Ca proteins, but had no pronounced effect on R1a (Fig. 2B). Interestingly, DHT treatment alone or in combination with rolipram or forskolin significantly increased the expression of all 3 PKA subunits, including R1a and R2b and Ca (Fig. 2B and C). These intriguing results provided the direct evidence that androgen may be participated in the regulation of protein expression of the key components of cAMP/PKA pathways. 3.4. Regulation of AR and PSA by cAMP pathways under serum-free conditions Because AR and PSA are expressed in most of the castration-resistant CaPs after androgen deprivation therapy, we therefore aimed to investigate whether cAMP/PKA pathways may replace androgens to activate AR/PSA signalings. We examined the expression of AR and PSA in LNCaP cells treated with rolipram or forskolin alone or in combination under serum-free condition. Rolipram treatment increased AR and PSA protein expressions, whereas forskolin treatment only induced an increase in PSA expression (Fig. 3A). However, this did not achieve statistical significance due to the small numbers of experiments (3 repeats) (Fig. 3C). The combination of rolipram and forskolin only slightly decreased PSA level through an unknown mechanism, but did not influence the AR level (Fig. 3A). These data showed that activation of cAMP/PKA pathways by rolipram or forskolin alone increased the levels of AR/PSA signaling in the absence of mitogenic growth factors. DHT alone or in
combination with rolipram or forskolin increased PSA or AR expression under serum-free condition, and this was statistically significant (P o 0.05) (Fig. 3A and C). In contrast to what was observed in LNCaP cells cultured in serum-containing medium, in the absence of serum, rolipram, forskolin, or DHT, alone or in combination, did not have a pronounced effect on R2b and Ca (Fig. 3B and C). Expression of R1a was slightly increased in LNCaP cells treated with rolipram or rolipram in combination with DHT (Fig. 3B), suggesting that R1a may be the downstream effector of rolipram under serum-free conditions. 3.5. The effects of rolipram, forskolin, and DHT on the subcellular localization of AR and PSA Immunofluorescence analysis was also carried out to verify the effect of rolipram, forskolin, and DHT on the expression and subcellular localization of AR and PSA in LNCaP cells under serum-free condition. Similar to what was observed in the immunoblot analysis, rolipram, forskolin, and DHT, alone or in combination, enhanced AR staining intensity and induced a pronounced increase in PSA signals on cell membranes under serum-free condition (Fig. 4). LNCaP cells treated with DHT in combination with rolipram and forskolin showed altered morphologies compared with the control LNCaP cells (Fig. 4). 3.6. Enhanced cAMP signaling led to the morphologic changes in LNCaP cells Because we observed that rolipram and forskolin influenced the expression of AR/PSA, we next aimed to examine the consequences of rolipram and forskolin treatment on the growth of LNCaP cells. Treatment of LNCaP cells with
25.e6
M. Sarwar et al. / Urologic Oncology: Seminars and Original Investigations 32 (2014) 25.e1–25.e12
Fig. 2. cAMP-mediated modulation of AR and PSA. (A) Immunoblot analysis of expression of AR and PSA in LNCaP cells treated with different agents, including DMSO as a vehicle control, the PDE4-specific inhibitor rolipram (1 mM), the cAMP enhancer forskolin (1 mM), 5a-dihydrotestosterone (5 nM), and combinations of these agents in complete medium for 24 hours. Then, 20 mg of total protein lysate was immunoblotted with anti-AR and anti-PSA antibodies. (B) Immunoblot analysis of the expression of PKA regulatory subunits R1a and R2b and Ca in the samples mentioned in (A). (C) Densitometric quantification of PSA and R2b expressions from 3 independent experiments. Bars represent standard errors of mean, and asterisks denote statistical significance between control and other treatments evaluated by the 2-tailed Student’s t- tests (**P o 0.01 and *P o 0.05).
rolipram or forskolin had no significant effect on proliferation of LNCaP cells (data not shown). As mentioned earlier, the effect of forskolin on cell morphology was readily detected after 24 hours of treatment. The majority of cells displayed the reduced cell bodies with elongated membranes and structures after treatment with forskolin alone or in combinations with rolipram and DHT (Fig. 5). These morphologic characteristics are similar to what were known for neuroendocrine-like features. This suggests that the enhanced activity of cAMP induced by forskolin may lead to neuroendocrine differentiation in LNCaP cells under serum-free condition. 3.7. The effect of cAMP/PKA on androgen-insensitive PC3 cells lacking AR and PSA expressions Because PC3 cells lack the expression of AR and PSA and are androgen insensitive, we therefore aimed to closely examine the activity of cAMP/PKA pathways in PC3 cells.
To this end, we treated PC3 cells with rolipram and forskolin alone or in combination and examined PKA subunits by immunoblot analysis. Surprisingly, forskolin treatment completely diminished R1a and reduced Ca level as well (Fig. 6A). In contrast, R2b was slightly increased by rolipram or forskolin alone (Fig. 6A). These data suggest that forskolin treatment inhibited the expression of PKA subunits R1a and Ca but increased the expression of R2b in PC3 cells. However, rolipram treatment did not show markedly effect on the expression of R1a and Ca (Fig. 6A). It is known that extracellular MMPs can facilitate the invasion of CaP cells to secondary organs [20]. Alterations in MMP-9 expression, a key component of the MMP family of proteins, may result in the progression of CaP and subsequently tumor metastasis [20,21]. We therefore decided to examine whether elevated level of cAMP/ PKA pathways may be linked to the increased level of MMP-9 in PC3 cells. Immunoblot analysis showed that
M. Sarwar et al. / Urologic Oncology: Seminars and Original Investigations 32 (2014) 25.e1–25.e12
25.e7
Fig. 3. Modulation of AR and PSA in response to cAMP in the absence of serum. (A) Immunoblot analysis shows the expression of AR and PSA in LNCaP cells treated with different agents under serum-free medium for 24 hours. (B) Expression of anti-PKA regulatory subunits R1a and R2b and the PKA catalytic subunit Ca in LNCaP cells treated with different agents under serum-free condition. (C) Densitometric quantification of PSA and R2b expressions from 3 independent experiments. Bars represent standard errors of mean, and asterisks denote statistical significance between control and other treatments by the unpaired 2-tailed Student’s t-tests (**P o 0.01 and *P o 0.05).
rolipram and forskolin treatment alone or together induced an increase in MMP-9 level in PC3 cells in the presence of serum (Fig. 6B). We next depleted serum from the culturing medium and examined the effect of rolipram and forskolin on the PKA subunits and MMP-9 expression under serum-free condition. Similar to what was observed under serum-containing condition, forskolin treatment alone or in combination with rolipram abolished R1a expression (Fig. 6C). Unlike what was observed under serum-enriched condition, rolipram and forskolin treatment had no detectable effect on the levels of R2b and Ca proteins under serum-free condition (Fig. 6C). However, rolipram and forskolin treatment led to a slightly increased level of MMP-9 (Fig. 6D). Taken together, the effect of forskolin on R2b and Ca proteins differed between androgen-sensitive LNCaP cells and androgen-insensitive PC3 cells.
3.8. mRNA expression of AR and cAMP subunits in normal prostate tissues, primary CaPs, and metastatic lesions from patients with CaP To investigate the clinical relevance of cAMP/PKA pathways in the progression and metastasis of CaPs, we evaluated mRNA expression of AR and the major components of cAMP/PKA pathways in 171 clinical specimens obtained from men at various age groups, including organ donor prostate tissues free of any pathologic condition (n ¼ 18), normal prostate tissues adjacent to the tumors (n ¼ 63), primary CaP specimens (n ¼ 65), and specimens from CaP metastatic sites (n ¼ 25). Metastatic specimens had a significantly higher level of AR expression than the normal prostate tissues or primary tumor specimens (P o 0.01) (Fig. 7). Conversely, expression of cAMP-specific PDE4B was significantly lower in metastatic tissues (P o 0.01)
25.e8
M. Sarwar et al. / Urologic Oncology: Seminars and Original Investigations 32 (2014) 25.e1–25.e12
Fig. 4. Immunofluorescence analysis of the effect of rolipram, forskolin, and DHT on AR and PSA cellular expressions and localization in LNCaP cells. LNCaP cells were starved in serum-free medium for 24 hours and were treated with different agents as indicated for an additional 16 hours under serum-free condition.
compared with that of normal prostate tissues or primary CaP specimens (Fig. 7, Table 2). The Spearman rank correlation statistical analysis showed that PDE4B is inversely correlated with AR in cancer specimens (P o 0.01) (Fig. 7, Table 3).
Fig. 5. Enhanced cAMP signaling led to the morphologic changes with the characteristics of neuroendocrine cells. LNCaP cells were starved in serumfree medium for 24 hours and were treated with different agents as indicated for an additional 24 hours under serum-free condition.
metastatic tissues compared with the primary tumor or normal prostate tissues (P o 0.05) and was positively correlated with AR in CaP cancer specimens (Fig. 7, Tables 2 and 3). Expression of PRKACA encoding for Ca did not show any significant differences between metastases and primary tumor (data not shown).
3.9. mRNA expression of PKA subunits and its correlation with AR in samples of patients with CaP
4. Discussion
mRNA expression of PRKAR1A encoding for the R1a protein was significantly lower in primary CaP specimens or in metastatic specimens compared with the normal prostate tissues (P o 0.01) (Fig. 6, Table 2). However, there was no significant correlation between PRKAR1A and AR expressions in cancer specimens. In contrast, PRKAR2B encoding for the R2b protein was significantly higher in
Multiple cellular pathways cooperatively contribute to the development of CaPs from a hormone-dependent to an incurable hormone-refractory state. Alterations in AR signaling contribute to the growth and survival of CaP cells and render the tumor cells resistant to androgen deprivation therapy, suggesting a key role for AR in CRPC. Additional pathways, in particular, the intracellular cAMP/PKA
M. Sarwar et al. / Urologic Oncology: Seminars and Original Investigations 32 (2014) 25.e1–25.e12
25.e9
Fig. 6. Effect of PDE4 inhibition or cAMP enhancement in androgen-independent PC3 cells that lack the expression of AR. (A) Immunoblot analysis of expression of PKA subunits R1a, R2b, and Ca in PC3 cells treated with DMSO (control vehicle), rolipram (1 mM), and forskolin (1 mM) in complete medium for 24 hours. (B) Expression of MMP-9 in PC3 cells treated with rolipram, forskolin, alone or in combination. (C) PC3 cells were serum starved for 24 hours and were treated with different agents for an additional 24 hours under serum-free condition. Expression of R1a, R2b, and Ca in PC3 cells treated with different agents under serum-free condition was assessed by immunoblot. (D) MMP-9 expression in PC3 cells treated with different agents under serum-free condition.
signalings, may be essential for CaP tumor cells to adapt invasive phenotypes in the absence of androgen. Several previous studies have provided direct evidence that cAMP/ PKA crosstalks with AR, such that it can activate AR and increase PSA expression in an androgen-independent fashion [8,9,15,16]. In the present study, we aimed to gain deeper understanding of the precise role of cAMP/PKA pathways and their relationship with AR signalings during the progression of CaP from androgen-dependent to hormone-refractory high-risk CaPs. We also aimed to define the key
components of cAMP/PKA pathways that may be responsible for transducing the cAMP and androgenic effects in cooperation with AR. We attempted to further explore the cellular mechanisms and clinical relevance underlying the progression of the androgen-sensitive to the androgeninsensitive state, and for these purposes, we utilized CaP cell lines and patient samples. In this study, we presented several novel findings on the interrelationship between cAMP/PKA and AR signalings and their clinical importance in CaPs, which have not been reported before. We for the first time showed that inhibition
25.e10
M. Sarwar et al. / Urologic Oncology: Seminars and Original Investigations 32 (2014) 25.e1–25.e12
Fig. 7. Evaluation of mRNA expression of the androgen receptor (AR), PDE4B, R1a (PRKAR1A), and R2b (PRKAR2B) in normal prostate tissues, primary CaP tissues, and metastatic lesions. Normal donor (n ¼ 18), normal adjacent (n ¼ 63), primary tumor (n ¼ 65), and metastases (n ¼ 25). **P o 0.01 level (2-tailed). *P o 0.05 level (2-tailed).
of PDE4 using rolipram resulted in the up-regulation of AR and PSA expressions in LNCaP cells. The effect of rolipram on AR was equivalent to what was achieved by using androgen. Given the positive role of the cAMP/PKA pathway in CaP growth, it is plausible that lower expression of its negative regulators (PDE enzymes) would favor the constitutive activation of the AR pathway through PKA. Table 3 Correlation of AR and PKA pathway–related genes between metastases and primary tumor
Spearman’s r and Pearson’s r
AR
Correlation coefficient Sig. (2-tailed) n
PRKAR2B
PDE4B
0.554**a 0.000 90
–0.523**b 0.000 90
Correlation is significant at the P o 0.01 level (2-tailed). Pearson’s correlation coefficient. b Spearman’s r. ** a
Forskolin treatment of LNCaP cells led to the activation of cAMP/PKA and increased PSA levels, but showed no pronounced effect on AR expression. Our observed differences in the effects between rolipram and forskolin treatments reflect the different modes of action of these 2 different agents. Because PDE is responsible for limiting the cAMP production and fine-tuning the PKA pathway [22], rolipram releases the inhibitory effect of PDE4 on cAMP/PKA, thereby increasing the level of cAMP. Forskolin is a cAMP enhancer and can directly increase the activity of cAMP [8]. We found that a combination of rolipram and forskolin did not have more pronounced effects than either drug alone, which may arise as a result of chemical interaction between these 2. Our findings suggest that rolipram may target pathways upstream of AR, whereas froskolin may crosstalk with AR to regulate PSA. It has been shown that cAMP/PKA could crosstalk with AR to enhance the transcriptional activity of AR on
M. Sarwar et al. / Urologic Oncology: Seminars and Original Investigations 32 (2014) 25.e1–25.e12
Fig. 8. Hypothetical model by which the PKA pathway can activate AR and AR-responsive genes. Various hormones and neurotransmitters activate the b-adrenergic receptor, which in turn causes the activation of adenylyl cyclase (AC) to produce cAMP from ATP. Up-regulation of the R2b subunit or down-regulation of the R1a subunit may lead to stabilization of the Ca subunit and activation of CREB1 by phosphorylation. Activated CREB1 can bind to the upstream promoter element of AR. The activation of CREB1 may lead to recruitment of coactivator protein CBP to the AR responsive PSA promoter and thereby can increase the expression of the PSA gene. Up-regulated genes are indicated with an upward arrow, and the down-regulated ones with a downward arrow.
PSA gene expression in LNCaP cells [8,9,15,16]. Our data that forskolin increased PSA protein levels supported previous reported studies. As illustrated in Fig. 8, up-regulation of the R2b subunit or the down-regulation of the R1a subunit may lead to the stabilization of the Ca subunit and activation of CREB1 by phosphorylation. Activated CREB1 can bind to the upstream promoter element of AR. The activation of CREB1 may lead to recruitment of the coactivator protein CREB binding protein (CBP) to the AR responsive PSA promoter and thereby can increase the expression of the PSA gene. We demonstrate that cAMP/PKA pathways not only crosstalk with AR and enhance the expression of PSA protein, but also directly regulate protein expression of AR. Further, the effects of cAMP/PKA on AR and PSA were independent of androgen and other hormones in LNCaP cells. Interestingly, the activity of cAMP/PKA and its intracellular signaling transduction appeared to be different between androgen-sensitive LNCaP cells and androgeninsensitive PC3 cells. As forskolin treatment led to the increased expression of R1a and R2b and Ca in LNCaP cells, in contrast, forskolin treatment completely abolished the expression of R1a in PC3 cells. This suggests that cAMP/PKA activation and its downstream signaling effectors are functionally associated with AR and androgen status. Further, in androgen-insensitive PC3 cells, activation of cAMP/PKA by forskolin or rolipram increased the level of
25.e11
MMP-9 expression, a marker for metastasis of cancer cells. It is likely that under low levels of hormones, the constitutive activation of cAMP/PKA pathways may replace the androgen effect, increase the level of AR, and enhance the transcriptional activity of AR to enable the high level of PSA production in the tumor cells. We showed that constitutive activation of the cAMP/ PKA pathway by forskolin induced large proportion of cells with features of neuroendocrine differentiation [23]. It is known that CaP cells can differentiate into neuroendocrine cells. Thus, the appearance of neuroendocrine cells in prostate tumors has been considered as a potential indicator for disease progression toward an androgen-independent state [8,23,24]. We showed that metastatic lesions from patients with CaP had significantly higher levels of AR and R2b (PRKAR2B) mRNA expression compared with the primary CaP specimens or normal prostate tissues. In contrast, expression of R1a (PRKAR1A) and PDE4B was significantly lower in metastatic specimens and was reversely correlated with AR. Our results suggested the clinical importance of PDE4B in CaP. More recently, Kashiwagi et al. [25] have shown that the expression of PDE4B is down-regulated in advanced CaP, and their knockdown promotes proliferation of LNCaP cells in androgendeprived conditions. Chen et al. [26] have shown that increase in AR mRNA is sufficient for the conversion of the phenotype of CaP from androgen-dependent to androgenindependent in a xenograft model. Our finding is in agreement with previous studies, which suggest the prognostic values of PKA pathways in patients with CaP [27,28]. Further, our novel findings by using clinical data strongly supported our results in cell line studies and provided evidence that cAMP/PKA pathways play an important role in the progression from androgendependent to androgen-refractory CaP. In conclusion, our finding suggests that deregulation of multiple signal transduction pathways and their correlation with AR in CaP cells may reflect the complexity of prostate progression and metastasis. Our study provides novel information and raised the possibility for future studies to investigate the different aspects of signaling pathways. Furthermore, the possible post-translational modification and protein-protein interaction of PKA and its negative regulators may also play a vital role in CaP metastasis. Acknowledgments This work was supported by the Swedish National Research Council, the Swedish Cancer Society, the Medical Faculty Grant, the Government Health Grant, Malm¨o Cancer Foundation, and Malm¨o Hospital Foundation (JLP). We thank Dr. Federico A. Monzon from the Department of Pathology and Laboratory Medicine, The Methodist Hospital, 6565 Fannin St., MS205, Houston, TX 77030, USA, for providing and sharing the clinical data sets.
25.e12
M. Sarwar et al. / Urologic Oncology: Seminars and Original Investigations 32 (2014) 25.e1–25.e12
Appendix A. Supplementary materials Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j. urolonc.2012.
References [1] Wegiel B, Evans S, Hellsten R, Otterbein LE, Bjartell A, Persson JL. Molecular pathways in the progression of hormone-independent and metastatic prostate cancer. Curr Cancer Drug Targets 2010;10:392–401. [2] Nieto M, Finn S, Loda M, Hahn WC. Prostate cancer: re-focusing on androgen receptor signaling. Int J Biochem Cell Biol 2007;39:1562–8. [3] Labrie F. Blockade of testicular and adrenal androgens in prostate cancer treatment. Nat Rev Urol 2011;8:73–85. [4] Taplin ME, Balk SP. Androgen receptor: a key molecule in the progression of prostate cancer to hormone independence. J Cell Biochem 2004;91:483–90. [5] Culig Z, Hobisch A, Cronauer MV, et al. Androgen receptor activation in prostatic tumor cell lines by insulin-like growth factor-I, keratinocyte growth factor, and epidermal growth factor. Cancer Res 1994;54:5474–8. [6] Grossmann ME, Huang H, Tindall DJ. Androgen receptor signaling in androgen-refractory prostate cancer. J Natl Cancer Inst 2001;93: 1687–97. [7] Ueda T, Mawji NR, Bruchovsky N, Sadar MD. Ligand-independent activation of the androgen receptor by interleukin-6 and the role of steroid receptor coactivator-1 in prostate cancer cells. J Biol Chem 2002;277:38087–94. [8] Merkle D, Hoffmann R. Roles of cAMP and cAMP-dependent protein kinase in the progression of prostate cancer: cross-talk with the androgen receptor. Cell Signal 2011;23:507–15. [9] Sadar MD. Androgen-independent induction of prostate-specific antigen gene expression via cross-talk between the androgen receptor and protein kinase a signal transduction pathways. J Biol Chem 1999;274:7777–83. [10] Chin KV, Yang WL, Ravatn R, et al. Reinventing the wheel of cyclic AMP: novel mechanisms of cAMP signaling. Ann NY Acad Sci 2002;968:49–64. [11] Taske´n K, Sk˚alhegg BS, Solberg R, et al. Novel isozymes of cAMPdependent protein kinase exist in human cells due to formation of RI alpha-RI beta heterodimeric complexes. J Biol Chem 1993;268: 21276–83. [12] Sands WA, Palmer TM. Regulating gene transcription in response to cyclic AMP elevation. Cell Signal 2008;20:460–6.
[13] Mayr B, Montminy M. Transcriptional regulation by the phosphorylation-dependent factor CREB. Nat Rev Mol Cell Biol 2001;2:599–609. [14] Sakamoto KM, Frank DA. CREB in the pathophysiology of cancer: implications for targeting transcription factors for cancer therapy. Clin Cancer Res 2009;15:2583–7. [15] Nazareth LV, Weigel NL. Activation of the human androgen receptor through a protein kinase a signaling pathway. J Biol Chem 1996;271:19900–7. [16] Desiniotis A, Schafer G, Klocker H, Eder IE. Enhanced antiproliferative and proapoptotic effects on prostate cancer cells by simultaneously inhibiting androgen receptor and cAMP-dependent protein kinase A. Int J Cancer 2010;126:775–89. [17] Chandran UR, Ma C, Dhir R, et al. Gene expression profiles of prostate cancer reveal involvement of multiple molecular pathways in the metastatic process. BMC Cancer 2007;7:64. [18] Yu YP, Landsittel D, Jing L, et al. Gene expression alterations in prostate cancer predicting tumor aggression and preceding development of malignancy. J Clin Oncol 2004;22:2790–9. [19] Simon R, Lam A, Li M-C, Ngan M, Menenzes S, Zhao Y. Analysis of gene expression data using BRB-array tools. Cancer Inf 2007;3:11–7. [20] Egeblad M, Werb Z. New functions for the matrix metalloproteinases in cancer progression. Nat Rev Cancer 2002;2:161–74. [21] Wegiel B, Bjartell A, Tuomela J, et al. Multiple cellular mechanisms related to cyclin A1 in prostate cancer invasion and metastasis. J Natl Cancer Inst 2008;100:1022–36. [22] Conti M, Beavo J. Biochemistry and physiology of cyclic nucleotide phosphodiesterases: essential components in cyclic nucleotide signaling. Annu Rev Biochem 2007;76:481–511. [23] Cox ME, Deeble PD, Bissonette EA, Parsons SJ. Activated 30 , 50 -cyclic amp-dependent protein kinase is sufficient to induce neuroendocrine-like differentiation of the LNCaP prostate tumor cell line. J Biol Chem 2000;275:13812–8. [24] Cox ME, Deeble PD, Lakhani S, Parsons SJ. Acquisition of neuroendocrine characteristics by prostate tumor cells is reversible. Cancer Res 1999;59:3821–30. [25] Kashiwagi E, Shiota M, Yokomizo A, et al. Downregulation of phosphodiesterase 4B (PDE4B) activates protein kinase A and contributes to the progression of prostate cancer. The Prostate 2012;72:741–51. [26] Chen CD, Welsbie DS, Tran C, et al. Molecular determinants of resistance to antiandrogen therapy. Nat Med 2004;10:33–9. [27] Khor L-Y, Bae K, Al-Saleem T, et al. Protein kinase A RI-a predicts for prostate cancer outcome: analysis of radiation therapy oncology group trial 86-10. Int J Radiat Oncol Biol Phys 2008;71:1309–15. [28] Pollack A, Bae K, Khor L-Y, et al. The importance of protein kinase A in prostate cancer: relationship to patient outcome in radiation therapy oncology group trial 92-02. Clin Cancer Res 2009;15:5478–84.
Original article
Protein kinase A (PKA) pathway is functionally linked to androgen receptor (AR) in the progression of prostate cancer Martuza Sarwara, Sabina Sandberga, Per-Anders Abrahamsson, M.D., Ph.D.b, Jenny L. Persson, Ph.D.a,* a
Division of Experimental Cancer Research, Department of Laboratory Medicine, Malm¨o, CRC, Lund University, Malm¨o, Sweden b Department of Clinical Sciences, Division of Urologic Research, Sk˚ane University Hospital, Lund University, Malm¨o, Sweden Received 14 October 2011; received in revised form 10 August 2012; accepted 20 August 2012
Abstract Objectives: In the present study, we investigated whether the cyclic adenosine monophosphate (cAMP)-activated protein kinase A (PKA) pathway may regulate the expression of AR and prostate-specific antigen (PSA) and whether there is a correlation between the expression of cAMP/PKA-associated genes and androgen receptor (AR) in patients with prostate cancer (CaP). Materials and methods: The functional studies were performed in LNCaP and PC3 cell lines. Data on the mRNA expression of sets of genes in human clinical samples, including prostate tissues from organ donors, prostate primary cancer, and metastatic cancer, were extracted from the National Center for Biotechnology Informations Gene Expression Omnibus (GEO) database. Statistical tests were applied. Results: We showed that elevated levels of cAMP/PKA pathways induced an increased expression of AR and PSA proteins in LNCaP cells in the absence of androgen. A cAMP-associated phosphodiesterase-4 (PDE4) inhibitor, rolipram induced an up-regulation in AR expression, whereas a cAMP enhancer, forskolin increased PSA level without affecting AR expression. Forskolin treatment increased the level of PKA R1a in LNCaP cells, but remarkably inhibited R1a expression in aggressive PC3 cells. In patients with CaP, we found that the expression of genes encoding R1a and phosphodiesterase-4B was statistically significantly lower in the metastatic specimens than that in the primary CaP specimens or in the normal prostate tissues (P o 0.01) and was reversely correlated with AR expression. Conversely, AR and PRKAR2B mRNA expressions were significantly higher in metastatic lesions than those in the primary CaP specimens or in the normal prostate tissues (P o 0.01). Conclusion: Our study revealed a novel mechanism to precisely define the functional and clinical interrelationship between the cAMP/ PKA pathway and AR signaling in the development of androgen-independent growth of CaPs and metastasis progression. r 2014 Elsevier Inc. All rights reserved. Keywords: Rolipram; Forskolin; AR; cAMP/PKA pathways; CaP metastasis
1. Introduction Prostate cancer (CaP) is one of the most common malignancies in men. The growth of cancer cells is highly dependent on androgens in the early stages. Androgen receptors (ARs) mediate the effects of hormones and play a central role in the formation and progression of tumors. Abbreviations: cAMP, cyclic adenosine monophosphate; PKA, protein kinase A; CaP, prostate cancer; CRPC, castration-resistant prostate cancer; PDE, phosphodiesterase; GEO, Gene Expression Omnibus. * Corresponding author. Tel.: þ46-40-391106; fax: þ46-40-391222. E-mail address: [email protected] (J.L. Persson). 1078-1439/$ – see front matter r 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.urolonc.2012.08.019
Most primary CaPs exhibit elevated levels of androgen and are initially sensitive to androgen deprivation therapy (ADT). Despite therapy, the disease often progresses into a castration-resistant prostate cancer (CRPC) [1–3]. Although CRPC is no longer dependent on androgen stimulation, AR is still expressed in tumor cells and the disease becomes highly aggressive [3]. Thus, alterations in AR expression and its associated pathways may lead to the development of CRPC and CaP metastasis [4]. AR-mediated growth-promoting pathways are therefore important therapeutic targets in CRPC. A number of cellular pathways that crosstalk with AR signalings have been studied. It has been shown that the epidermal growth factor (EGF) pathway can trigger AR
25.e2
M. Sarwar et al. / Urologic Oncology: Seminars and Original Investigations 32 (2014) 25.e1–25.e12
transcriptional activity in CaP cell lines [5,6]. The cytokine interleukin-6 (IL-6) and a family of neuropeptides, including bombesin- and gastrin-releasing peptides, have been shown to promote androgen-independent growth in several CaP cell lines [6,7]. One of the major intracellular signal transduction pathways – the cyclic adenosine monophosphate (cAMP)activated protein kinase A (PKA) pathway – is implicated in CaP progression [8]. An elevated cAMP level leads to an increased expression of prostate-specific antigen (PSA) mRNA, and the rise in the PSA gene expression by PKA requires a functional AR, suggesting that cAMP and its downstream PKA pathways may regulate the expression and activity of AR [9]. However, the precise role of cAMP/PKA in the development of CRPC and the relationship between cAMP/PKA and AR signaling in CRPC are less explored. The PKA pathway is critical for the fundamental cellular processes, including proliferation, differentiation, and apoptosis [10]. The synthesis of cAMP is mediated by adenylyl cyclase from adenosine triphosphate (ATP). The degradation of cAMP is mediated by a large family of cAMP-specific phosphodiesterases (PDEs) enzymes. Inhibition of PDEs, in particular by rolipram, a PDE4 inhibitor, leads to elevated levels of cAMP [8]. It is known that cAMP triggers the activation of downstream PKA, which facilitates the major cellular effects of cAMP [8]. PKA exists as a tetrameric holoenzyme consisting of 2 regulatory (R) and 2 catalytic (C) subunits forming a holoenzyme R2C2 [11]. The R subunits are the major intracellular receptors of cAMP. Binding of cAMP to the R subunits leads to the release of C subunits from the R-C complexes and allows the C subunits to phosphorylate downstream substrates [10,11]. The R subunits are further distinguished as R1a and R1b, and R2a and R2b, each of which are encoded by different genes like PRKAR1A, PRKAR1B, PRKAR2A, and PRKAR2B, respectively [11]. The C subunits, Ca, Cb, and Cg, are coded by the genes PRKACA, PRKACB, and PRKACG, respectively [10,11]. PKA exerts the effects of cAMP on several key transcriptional factors, and among them the cAMP response element (CRE)-binding protein 1 (CREB/CREB1) is the principle mediator [10,12]. The effects of cAMP/PKA on the cells are partly mediated by the transcriptional factor CREB1, which modulates the expression of a large number of genes involved in growth, survival, metabolism, reproduction, transport, and immune regulation [13,14]. Thus, cAMP/ PKA pathways are recognized as the central signaling transduction pathways that are required for multiple cellular functions, including metabolism, cellular growth, differentiation, gene expression, and apoptosis [10]. Increasing evidence has shown that cAMP/PKA pathways may play a role in CaP progression. It has been shown that ARs can be activated by the PKA activator, forskolin in an androgen-independent manner in PC3 cells cotransfected with AR responsive reporter vectors and AR expression vectors [15]. Furthermore, activation of PKA by forskolin leads to the increased mRNA level of PSA in CaP cell lines
[9]. Further, forskolin-induced increase in PSA mRNA expression was reversed by the addition of PKA inhibitor or by an AR antagonist, bicalutamide, in LNCaP cells [9]. Moreover, double knockdown of the AR and PKA subunit R1a induced a growth arrest of CaP cells with a better effect in comparison with what was achieved by a single knockdown of AR [16]. These data indicate that cAMP/ PKA pathways may be the upstream mediators of ARs in CaP cells. However, the precise role of cAMP/PKA in the development of CRPC and the relationship between cAMP/ PKA and AR signaling are less explored. The aim of our study is therefore to investigate whether cAMP/PKA may be responsible for the activation of AR/ PSA signaling during the progression of CRPC in the absence of androgen. We also assessed whether elevated levels of cAMP/PKA may sensitize CaP cells to low levels of androgen and thus promote progression of tumor cells from the androgen-dependent state to CRPC status. To further validate the correlation between the cAMP/PKAassociated regulators and AR expression, we utilized National Center for Biotechnology Information’s Gene Expression Omnibus (GEO) database and analyzed the mRNA expression profiles of the sets of genes of interest from the prostate tissues of organ donors, and the tumor tissues and metastatic lesions from patients with CaP. Our findings for the first time showed that cAMP-dependent pathways regulate AR/PSA expression. In the aggressive CaP cells, cAMP-dependent pathways mediate the matrix metalloproteinase (MMP)-9 expression, suggesting that cAMP-dependent pathways may promote adaptation of CaP cells to a more aggressive phenotype in the absence of hormones. Our data from patient studies support our findings in cell line-based studies and suggest that cAMP/ PKA may in part contribute to the progression of CRPC by increasing the levels of AR/PSA expression. 2. Materials and method 2.1. Cell lines The PNT1A cell line, androgen-dependent cell line LNCaP, and the androgen-independent cell lines PC3, DU-145, and PC3M were purchased from the American Type Culture Collection (ATCC, Manassas, VA). The cell lines were maintained in RPMI 1640 medium supplemented with 10% fetal bovine serum, 1% penicillin-streptomycin, and 1 mM L-glutamine (Life Technologies, Paisley, UK) at 37 1C in a humidified atmosphere with 5% CO2. 2.2. Treatment For the treatment, rolipram, the cAMP-specific PDE4 inhibitor, and forskolin, the cell permeable cAMP enhancer (Calbiochem, Darmstadt, Germany) dissolved in 100% dimethyl sulfoxide (DMSO), were used. LNCaP or PC3
M. Sarwar et al. / Urologic Oncology: Seminars and Original Investigations 32 (2014) 25.e1–25.e12
cells seeded at a density of 0.25106 or 0.15106 cells/ml in RPMI 1640 medium supplemented with 10% fetal bovine serum were cultured for 24 hours. Cells were treated for the final concentration of 1 mM rolipram or 1 mM forskolin or 5 nM dihydrotestosterone (DHT) or a combination of them for 24 hours. For the establishment of serumfree condition, LNCaP cells were cultured for 24 hours, washed with 1 phosphate buffered saline, and subsequently cultured in serum-free phenol red-free RPMI 1640 medium for a further 24 hours prior to the addition of different agents as mentioned previously. 2.3. Antibodies Primary antibodies against PKA RIa/b at 1:500 dilutions (Cell signaling technology, Danvers, MA), PKA RIIb at 1:1,000 (BD Biosciences Transduction Laboratories, San Jose, CA), AR at 1:250, PKAa at 1:500, p-CREB1 at Ser 133 sites at 1:500 (Santa Cruz Biotechnology Inc., Santa Cruz, CA), PSA at 1:500 (DAKO, Glostrup, Denmark), PDE4 (Abcam, Cambridge, UK) at 1:250, MMP-9 (Abcam, Cambridge, UK), and anti-b-actin at 1:10,000 (MP biochemicals, Illkirch, France) were used. 2.4. Immunoblot analysis The cells were harvested and lysed in ice-cold radioimmunoprecipitation assay buffer (120 mM NaCl, 50 mM Tris–HCl, pH 7.6, 50 mM NaF, 0.1 mM Na3VO4, 1% NP40, 1 mM phenylmethylsulfonyl fluoride) (Sigma, St. Louis, MD) and 15% protease inhibitor cocktail Complete Mini (Roche, Basel, Switzerland). Then, 20 mg of protein was separated on a 12% sodium dodecyl sulfate polyacrylamide gel electrophoresis gel and transferred onto nitrocellulose membranes. Signals were visualized using an Enhanced ChemiLuminescence detection system (Millipore Corp Sweden, Solna, Sweden) and documented with an AlphaImager CCD system. Densitometric quantification of immunoblots was carried out by the ImageJ Image Analysis Software (NIH, Baltimore, USA), represented as a fold change relative to control, and was normalized with an actin band.
25.e3
(Invitrogen, Stockholm, Sweden) were used. The cells were counterstained with 40 ,6-diamidino-2-phenylindole, and microphotographs were taken under 20 magnifications with an Olympus AX70 microscope equipped with appropriate filters from Semrock and fitted with a digital color camera (Nikon DS-U1, Stockholm, Sweden). The software ACT2U was used (ACT2U version 1.5, Stockholm, Sweden). For morphologic analysis, cells were prepared as described previously, and the phase-contrast images were monitored using an Olympus AX70 microscope. Alternatively, the images were accessed, and the photomicrographs were taken at 10 magnification using a HoloMonitorTM M3 (Phase Holographic Imaging AB, Lund, Sweden). 2.6. Patient samples and data collections For the immunoblot analysis, tumor and adjacent normal specimens that had been surgically removed from patients were obtained from Departments of Surgery and Urology, and Department of Clinical Pathology and Cytology, Lund University, University hospital in Malm¨o, Sweden. The above-mentioned tissues were paraffin-embedded and sectioned for histologic analysis. The tumor samples did not contain a significant amount of normal tissues and viceversa. This study was approved by the ethics committee of Lund University, Sweden, and the Helsinki Declaration of Human Rights was strictly observed. Gene expression data from the data set GDS2545 in the Gene Expression Omnibus database at the National Center for Biotechnology Information website was used. The data set was obtained by performing an Affymetrix HG-U95Av2 oligonucleotide array platform as described [17,18]. The mean mRNA values of genes of interest from a total of 171 human samples in the data set GDS2545 were used in this study. The samples included organ donor prostate tissues free of any pathologic diseases (n = 18), normal prostate tissues adjacent to tumor (n = 63), primary tumor (n = 65), and metastatic lesions (n = 25) from the liver, para-aortic lymph node, paratracheal lymph node, retroperitoneal lymph node, lung, and adrenal gland of 4 patients with CRPC. 2.7. Statistical analysis of gene expression profiles
2.5. Immunofluorescence and morphologic analysis LNCaP cells were grown on glass coverslips, starved for serum and androgen by growing in serum-free, phenol redfree RPMI for 24 hours, and treated with 1 mM rolipram, 5 nM DHT, and 1 mM forskolin alone or in combination, for 24 hours. Cells were fixed with 4% paraformaldehyde in phosphate buffered saline. For blocking background staining from nonspecific interactions, an Image-iTTM FX signal enhancer (Molecular Probes, Inc.) was used. Primary antibodies including anti-AR at 1:100 dilutions and anti-PSA at 1:250 dilutions were used. The secondary antibodies including rabbit anti-mouse Rhodamine (Chemicon/Millipore) at 1:200 and goat anti-rabbit alexa fluor 488 at 1:500
We analyzed the original data of DNA microarray gene expression profiles (data sets GDS2545) using the BRBArray Tool (Version: 4.1.0), a software developed by R. Simon and A. Pang Lam at the Biometric Research Branch of the National Cancer Institute [19]. The raw signal intensities of the arrays were preprocessed by setting the intensity at the minimum value, if the intensity is below the minimum value of 10, and normalized by subtracting the median log ratio of an array by all the log ratios on that array. The genes were excluded if less than 20% of expression data showed at least a 1.1-fold change in either direction from the median value of the gene. Two-sample t-tests were used to identify the genes that are differentially
25.e4
M. Sarwar et al. / Urologic Oncology: Seminars and Original Investigations 32 (2014) 25.e1–25.e12
Table 1 Names of the genes included in the present study Gene symbol Description AR PRKAR1A PRKAR2B PDE4B
Androgen receptor Protein kinase, cAMP dependent, regulatory, type I, alpha Protein kinase, cAMP dependent, regulatory, type II, beta Phosphodiesterase 4B, cAMP-specific
expressed between metastasis and primary tumor with a significance level of P o 0.05. False discovery rates were calculated based on 1,000 sample permutations. Tukey test, ANOVA, Kruskal-Wallis test, Mann-Whitney test, and Spearman rank correlation test were performed with the Statistical Package for the Social Sciences software (SPSS, version 17; Chicago). P-value o0.05 was considered as significant. The expression of genes and their correlations are listed in Tables 1 and 2.
3. Results 3.1. Expression of several key components in cAMP/ PKA-associated pathways in patients with CaP and prostate cancer cell lines To understand the role of cAMP signaling cascades in CaP progression, we first examined protein levels of PDE4, phosphorylated CREB1, and phosphorylated ATF1 that are cAMP/PKA downstream effectors in various types of CaP cell lines. PDE4, phosphor-CREB1, and phosphor-ATF1 were detected in all of the 5 cell lines, including the nonmalignant PNT1A cells, malignant androgen-sensitive LNCaP cells, and androgen-insensitive DU-145, PC3, and Table 2 Comparison of the expression of genes among clinical subgroups Genes
P values between groups
AR Normal vs. primary tumor Normal vs. metastasis Primary tumor vs. metastasis
0.14 0.00** 0.000**
PRKAR1A Normal vs. primary tumor Normal vs. metastasis Primary tumor vs. metastasis
0.000** 0.000** 0.877
PRKAR2B Normal vs. primary tumor Normal vs. metastasis Primary tumor vs. metastasis
0.615 0.025* 0.006**
PDE4B Normal vs primary tumor Normal vs. metastasis Primary tumor vs. metastasis
0.161 0.000** 0.000**
*
P o 0.05 level (2-tailed). P o 0.01 level (2-tailed).
**
PC3M cells, which represent different stages of CaPs. The lowest levels of PDE4 and phosphor-CREB1 were observed in PNT1A and PC3M cells (Fig. 1A). We next examined the expression of PDE4 and PKA subunits in clinical samples including normal and cancer specimens from patients with CaP, breast cancer, or testicular cancer. PDE4D was observed in all normal and paired cancer specimens examined (Fig. 1B), whereas PKA catalytic subunit Ca (PKACa was barely detectable in breast specimens, but was observed in normal testes and testicular cancers. In prostate specimens, PKACa protein was expressed in normal prostate and in the PC3 cell line, but was absent in 2 CaP specimens (Fig. 1B).
3.2. Regulation of AR and PSA by cAMP/PKA-associated regulatory pathways It is known that PDE enzymes facilitate the degradation of cAMP to keep the balance of cAMP levels in the cell. Treatment of different types of cells with rolipram, a selective PDE4 inhibitor, leads to the elevated levels of cAMP [8]. Similarly, treatment of CaP cells with forskolin leads to the increased activity of cAMP [8]. Thus, cAMP/PKA levels in CaP cells can be induced by rolipram and forskolin treatment. To this end, we investigated the interrelationship between cAMP/PKA pathways and AR signaling. We examined the effects of rolipram or forskolin on the protein levels of AR and PSA in LNCaP cells using immunoblot analysis. Rolipram treatment led to an increased level of AR protein expression and a moderate induction of PSA (Fig. 2A and C). In parallel, forskolin treatment increased the PSA level (Fig. 2A and C), but had no effect on AR expression (Fig. 2A). Taken together, these data suggest that the cAMP mediators, rolipram and forskolin influence the AR signaling via distinct mechanisms. Rolipram and forskolin combination treatment had no effect on AR and PSA expressions (Fig. 2A and C). This result suggests that rolipram antagonizes the effect of forskolin on AR and PSA expressions via an unknown mechanism. Thus, rolipram and forskolin in combination had no additive effect on AR signaling compared with rolipram or forskolin alone. Because the above-mentioned data showed that cAMP/ PKA pathways participated in the regulation of AR/PSA expression in the absence of androgen, we aimed to investigate whether cAMP/PKA and androgen together might have an additive effect on AR/PSA expression. We treated LNCaP cells with DHT alone or in combination with rolipram and forskolin. As expected, DHT treatment induced a robust increase in AR and PSA expressions in LNCaP cells, and this is statistically significant (P ¼ 0.04). Interestingly, the effect of rolipram on AR was comparable to the effect of DHT (Fig. 2A). Combined treatment of DHT together with forskolin or rolipram also significantly increased PSA expression (P o 0.01) (Fig. 2A and C). This indicates that cAMP/PKA pathways participate in the regulation of AR/PSA expression in the presence of androgen.
M. Sarwar et al. / Urologic Oncology: Seminars and Original Investigations 32 (2014) 25.e1–25.e12
25.e5
Fig. 1. Expression of several key components of the cAMP- or PKA-associated pathways in prostate cancer cell lines and in prostate cancer, along with breast cancer and testicular cancer samples. (A) Immunoblot analysis of the expression of PDE4 and phosphorylation of CREB1/ATF1 in human immortalized prostatic cell line PNT1A, androgen-sensitive LNCaP, and the hormone-insensitive DU145, PC3, and PC3M cells. (B) Expression of PDE4D and PKA Ca in normal indicated as “N” and prostate cancer, along with breast cancer and testicular cancer specimens indicated as “C”.
3.3. Regulation of PKA subunits by DHT in LNCaP cells We next examined the effects of rolipram, forskolin, and DHT, alone or in combinations, on the levels of PKA regulatory subunits R1a and R2b and PKA catalytic subunit Ca. Forskolin alone induced the increased levels of R2b (P ¼ 0.024) and Ca proteins, but had no pronounced effect on R1a (Fig. 2B). Interestingly, DHT treatment alone or in combination with rolipram or forskolin significantly increased the expression of all 3 PKA subunits, including R1a and R2b and Ca (Fig. 2B and C). These intriguing results provided the direct evidence that androgen may be participated in the regulation of protein expression of the key components of cAMP/PKA pathways. 3.4. Regulation of AR and PSA by cAMP pathways under serum-free conditions Because AR and PSA are expressed in most of the castration-resistant CaPs after androgen deprivation therapy, we therefore aimed to investigate whether cAMP/PKA pathways may replace androgens to activate AR/PSA signalings. We examined the expression of AR and PSA in LNCaP cells treated with rolipram or forskolin alone or in combination under serum-free condition. Rolipram treatment increased AR and PSA protein expressions, whereas forskolin treatment only induced an increase in PSA expression (Fig. 3A). However, this did not achieve statistical significance due to the small numbers of experiments (3 repeats) (Fig. 3C). The combination of rolipram and forskolin only slightly decreased PSA level through an unknown mechanism, but did not influence the AR level (Fig. 3A). These data showed that activation of cAMP/PKA pathways by rolipram or forskolin alone increased the levels of AR/PSA signaling in the absence of mitogenic growth factors. DHT alone or in
combination with rolipram or forskolin increased PSA or AR expression under serum-free condition, and this was statistically significant (P o 0.05) (Fig. 3A and C). In contrast to what was observed in LNCaP cells cultured in serum-containing medium, in the absence of serum, rolipram, forskolin, or DHT, alone or in combination, did not have a pronounced effect on R2b and Ca (Fig. 3B and C). Expression of R1a was slightly increased in LNCaP cells treated with rolipram or rolipram in combination with DHT (Fig. 3B), suggesting that R1a may be the downstream effector of rolipram under serum-free conditions. 3.5. The effects of rolipram, forskolin, and DHT on the subcellular localization of AR and PSA Immunofluorescence analysis was also carried out to verify the effect of rolipram, forskolin, and DHT on the expression and subcellular localization of AR and PSA in LNCaP cells under serum-free condition. Similar to what was observed in the immunoblot analysis, rolipram, forskolin, and DHT, alone or in combination, enhanced AR staining intensity and induced a pronounced increase in PSA signals on cell membranes under serum-free condition (Fig. 4). LNCaP cells treated with DHT in combination with rolipram and forskolin showed altered morphologies compared with the control LNCaP cells (Fig. 4). 3.6. Enhanced cAMP signaling led to the morphologic changes in LNCaP cells Because we observed that rolipram and forskolin influenced the expression of AR/PSA, we next aimed to examine the consequences of rolipram and forskolin treatment on the growth of LNCaP cells. Treatment of LNCaP cells with
25.e6
M. Sarwar et al. / Urologic Oncology: Seminars and Original Investigations 32 (2014) 25.e1–25.e12
Fig. 2. cAMP-mediated modulation of AR and PSA. (A) Immunoblot analysis of expression of AR and PSA in LNCaP cells treated with different agents, including DMSO as a vehicle control, the PDE4-specific inhibitor rolipram (1 mM), the cAMP enhancer forskolin (1 mM), 5a-dihydrotestosterone (5 nM), and combinations of these agents in complete medium for 24 hours. Then, 20 mg of total protein lysate was immunoblotted with anti-AR and anti-PSA antibodies. (B) Immunoblot analysis of the expression of PKA regulatory subunits R1a and R2b and Ca in the samples mentioned in (A). (C) Densitometric quantification of PSA and R2b expressions from 3 independent experiments. Bars represent standard errors of mean, and asterisks denote statistical significance between control and other treatments evaluated by the 2-tailed Student’s t- tests (**P o 0.01 and *P o 0.05).
rolipram or forskolin had no significant effect on proliferation of LNCaP cells (data not shown). As mentioned earlier, the effect of forskolin on cell morphology was readily detected after 24 hours of treatment. The majority of cells displayed the reduced cell bodies with elongated membranes and structures after treatment with forskolin alone or in combinations with rolipram and DHT (Fig. 5). These morphologic characteristics are similar to what were known for neuroendocrine-like features. This suggests that the enhanced activity of cAMP induced by forskolin may lead to neuroendocrine differentiation in LNCaP cells under serum-free condition. 3.7. The effect of cAMP/PKA on androgen-insensitive PC3 cells lacking AR and PSA expressions Because PC3 cells lack the expression of AR and PSA and are androgen insensitive, we therefore aimed to closely examine the activity of cAMP/PKA pathways in PC3 cells.
To this end, we treated PC3 cells with rolipram and forskolin alone or in combination and examined PKA subunits by immunoblot analysis. Surprisingly, forskolin treatment completely diminished R1a and reduced Ca level as well (Fig. 6A). In contrast, R2b was slightly increased by rolipram or forskolin alone (Fig. 6A). These data suggest that forskolin treatment inhibited the expression of PKA subunits R1a and Ca but increased the expression of R2b in PC3 cells. However, rolipram treatment did not show markedly effect on the expression of R1a and Ca (Fig. 6A). It is known that extracellular MMPs can facilitate the invasion of CaP cells to secondary organs [20]. Alterations in MMP-9 expression, a key component of the MMP family of proteins, may result in the progression of CaP and subsequently tumor metastasis [20,21]. We therefore decided to examine whether elevated level of cAMP/ PKA pathways may be linked to the increased level of MMP-9 in PC3 cells. Immunoblot analysis showed that
M. Sarwar et al. / Urologic Oncology: Seminars and Original Investigations 32 (2014) 25.e1–25.e12
25.e7
Fig. 3. Modulation of AR and PSA in response to cAMP in the absence of serum. (A) Immunoblot analysis shows the expression of AR and PSA in LNCaP cells treated with different agents under serum-free medium for 24 hours. (B) Expression of anti-PKA regulatory subunits R1a and R2b and the PKA catalytic subunit Ca in LNCaP cells treated with different agents under serum-free condition. (C) Densitometric quantification of PSA and R2b expressions from 3 independent experiments. Bars represent standard errors of mean, and asterisks denote statistical significance between control and other treatments by the unpaired 2-tailed Student’s t-tests (**P o 0.01 and *P o 0.05).
rolipram and forskolin treatment alone or together induced an increase in MMP-9 level in PC3 cells in the presence of serum (Fig. 6B). We next depleted serum from the culturing medium and examined the effect of rolipram and forskolin on the PKA subunits and MMP-9 expression under serum-free condition. Similar to what was observed under serum-containing condition, forskolin treatment alone or in combination with rolipram abolished R1a expression (Fig. 6C). Unlike what was observed under serum-enriched condition, rolipram and forskolin treatment had no detectable effect on the levels of R2b and Ca proteins under serum-free condition (Fig. 6C). However, rolipram and forskolin treatment led to a slightly increased level of MMP-9 (Fig. 6D). Taken together, the effect of forskolin on R2b and Ca proteins differed between androgen-sensitive LNCaP cells and androgen-insensitive PC3 cells.
3.8. mRNA expression of AR and cAMP subunits in normal prostate tissues, primary CaPs, and metastatic lesions from patients with CaP To investigate the clinical relevance of cAMP/PKA pathways in the progression and metastasis of CaPs, we evaluated mRNA expression of AR and the major components of cAMP/PKA pathways in 171 clinical specimens obtained from men at various age groups, including organ donor prostate tissues free of any pathologic condition (n ¼ 18), normal prostate tissues adjacent to the tumors (n ¼ 63), primary CaP specimens (n ¼ 65), and specimens from CaP metastatic sites (n ¼ 25). Metastatic specimens had a significantly higher level of AR expression than the normal prostate tissues or primary tumor specimens (P o 0.01) (Fig. 7). Conversely, expression of cAMP-specific PDE4B was significantly lower in metastatic tissues (P o 0.01)
25.e8
M. Sarwar et al. / Urologic Oncology: Seminars and Original Investigations 32 (2014) 25.e1–25.e12
Fig. 4. Immunofluorescence analysis of the effect of rolipram, forskolin, and DHT on AR and PSA cellular expressions and localization in LNCaP cells. LNCaP cells were starved in serum-free medium for 24 hours and were treated with different agents as indicated for an additional 16 hours under serum-free condition.
compared with that of normal prostate tissues or primary CaP specimens (Fig. 7, Table 2). The Spearman rank correlation statistical analysis showed that PDE4B is inversely correlated with AR in cancer specimens (P o 0.01) (Fig. 7, Table 3).
Fig. 5. Enhanced cAMP signaling led to the morphologic changes with the characteristics of neuroendocrine cells. LNCaP cells were starved in serumfree medium for 24 hours and were treated with different agents as indicated for an additional 24 hours under serum-free condition.
metastatic tissues compared with the primary tumor or normal prostate tissues (P o 0.05) and was positively correlated with AR in CaP cancer specimens (Fig. 7, Tables 2 and 3). Expression of PRKACA encoding for Ca did not show any significant differences between metastases and primary tumor (data not shown).
3.9. mRNA expression of PKA subunits and its correlation with AR in samples of patients with CaP
4. Discussion
mRNA expression of PRKAR1A encoding for the R1a protein was significantly lower in primary CaP specimens or in metastatic specimens compared with the normal prostate tissues (P o 0.01) (Fig. 6, Table 2). However, there was no significant correlation between PRKAR1A and AR expressions in cancer specimens. In contrast, PRKAR2B encoding for the R2b protein was significantly higher in
Multiple cellular pathways cooperatively contribute to the development of CaPs from a hormone-dependent to an incurable hormone-refractory state. Alterations in AR signaling contribute to the growth and survival of CaP cells and render the tumor cells resistant to androgen deprivation therapy, suggesting a key role for AR in CRPC. Additional pathways, in particular, the intracellular cAMP/PKA
M. Sarwar et al. / Urologic Oncology: Seminars and Original Investigations 32 (2014) 25.e1–25.e12
25.e9
Fig. 6. Effect of PDE4 inhibition or cAMP enhancement in androgen-independent PC3 cells that lack the expression of AR. (A) Immunoblot analysis of expression of PKA subunits R1a, R2b, and Ca in PC3 cells treated with DMSO (control vehicle), rolipram (1 mM), and forskolin (1 mM) in complete medium for 24 hours. (B) Expression of MMP-9 in PC3 cells treated with rolipram, forskolin, alone or in combination. (C) PC3 cells were serum starved for 24 hours and were treated with different agents for an additional 24 hours under serum-free condition. Expression of R1a, R2b, and Ca in PC3 cells treated with different agents under serum-free condition was assessed by immunoblot. (D) MMP-9 expression in PC3 cells treated with different agents under serum-free condition.
signalings, may be essential for CaP tumor cells to adapt invasive phenotypes in the absence of androgen. Several previous studies have provided direct evidence that cAMP/ PKA crosstalks with AR, such that it can activate AR and increase PSA expression in an androgen-independent fashion [8,9,15,16]. In the present study, we aimed to gain deeper understanding of the precise role of cAMP/PKA pathways and their relationship with AR signalings during the progression of CaP from androgen-dependent to hormone-refractory high-risk CaPs. We also aimed to define the key
components of cAMP/PKA pathways that may be responsible for transducing the cAMP and androgenic effects in cooperation with AR. We attempted to further explore the cellular mechanisms and clinical relevance underlying the progression of the androgen-sensitive to the androgeninsensitive state, and for these purposes, we utilized CaP cell lines and patient samples. In this study, we presented several novel findings on the interrelationship between cAMP/PKA and AR signalings and their clinical importance in CaPs, which have not been reported before. We for the first time showed that inhibition
25.e10
M. Sarwar et al. / Urologic Oncology: Seminars and Original Investigations 32 (2014) 25.e1–25.e12
Fig. 7. Evaluation of mRNA expression of the androgen receptor (AR), PDE4B, R1a (PRKAR1A), and R2b (PRKAR2B) in normal prostate tissues, primary CaP tissues, and metastatic lesions. Normal donor (n ¼ 18), normal adjacent (n ¼ 63), primary tumor (n ¼ 65), and metastases (n ¼ 25). **P o 0.01 level (2-tailed). *P o 0.05 level (2-tailed).
of PDE4 using rolipram resulted in the up-regulation of AR and PSA expressions in LNCaP cells. The effect of rolipram on AR was equivalent to what was achieved by using androgen. Given the positive role of the cAMP/PKA pathway in CaP growth, it is plausible that lower expression of its negative regulators (PDE enzymes) would favor the constitutive activation of the AR pathway through PKA. Table 3 Correlation of AR and PKA pathway–related genes between metastases and primary tumor
Spearman’s r and Pearson’s r
AR
Correlation coefficient Sig. (2-tailed) n
PRKAR2B
PDE4B
0.554**a 0.000 90
–0.523**b 0.000 90
Correlation is significant at the P o 0.01 level (2-tailed). Pearson’s correlation coefficient. b Spearman’s r. ** a
Forskolin treatment of LNCaP cells led to the activation of cAMP/PKA and increased PSA levels, but showed no pronounced effect on AR expression. Our observed differences in the effects between rolipram and forskolin treatments reflect the different modes of action of these 2 different agents. Because PDE is responsible for limiting the cAMP production and fine-tuning the PKA pathway [22], rolipram releases the inhibitory effect of PDE4 on cAMP/PKA, thereby increasing the level of cAMP. Forskolin is a cAMP enhancer and can directly increase the activity of cAMP [8]. We found that a combination of rolipram and forskolin did not have more pronounced effects than either drug alone, which may arise as a result of chemical interaction between these 2. Our findings suggest that rolipram may target pathways upstream of AR, whereas froskolin may crosstalk with AR to regulate PSA. It has been shown that cAMP/PKA could crosstalk with AR to enhance the transcriptional activity of AR on
M. Sarwar et al. / Urologic Oncology: Seminars and Original Investigations 32 (2014) 25.e1–25.e12
Fig. 8. Hypothetical model by which the PKA pathway can activate AR and AR-responsive genes. Various hormones and neurotransmitters activate the b-adrenergic receptor, which in turn causes the activation of adenylyl cyclase (AC) to produce cAMP from ATP. Up-regulation of the R2b subunit or down-regulation of the R1a subunit may lead to stabilization of the Ca subunit and activation of CREB1 by phosphorylation. Activated CREB1 can bind to the upstream promoter element of AR. The activation of CREB1 may lead to recruitment of coactivator protein CBP to the AR responsive PSA promoter and thereby can increase the expression of the PSA gene. Up-regulated genes are indicated with an upward arrow, and the down-regulated ones with a downward arrow.
PSA gene expression in LNCaP cells [8,9,15,16]. Our data that forskolin increased PSA protein levels supported previous reported studies. As illustrated in Fig. 8, up-regulation of the R2b subunit or the down-regulation of the R1a subunit may lead to the stabilization of the Ca subunit and activation of CREB1 by phosphorylation. Activated CREB1 can bind to the upstream promoter element of AR. The activation of CREB1 may lead to recruitment of the coactivator protein CREB binding protein (CBP) to the AR responsive PSA promoter and thereby can increase the expression of the PSA gene. We demonstrate that cAMP/PKA pathways not only crosstalk with AR and enhance the expression of PSA protein, but also directly regulate protein expression of AR. Further, the effects of cAMP/PKA on AR and PSA were independent of androgen and other hormones in LNCaP cells. Interestingly, the activity of cAMP/PKA and its intracellular signaling transduction appeared to be different between androgen-sensitive LNCaP cells and androgeninsensitive PC3 cells. As forskolin treatment led to the increased expression of R1a and R2b and Ca in LNCaP cells, in contrast, forskolin treatment completely abolished the expression of R1a in PC3 cells. This suggests that cAMP/PKA activation and its downstream signaling effectors are functionally associated with AR and androgen status. Further, in androgen-insensitive PC3 cells, activation of cAMP/PKA by forskolin or rolipram increased the level of
25.e11
MMP-9 expression, a marker for metastasis of cancer cells. It is likely that under low levels of hormones, the constitutive activation of cAMP/PKA pathways may replace the androgen effect, increase the level of AR, and enhance the transcriptional activity of AR to enable the high level of PSA production in the tumor cells. We showed that constitutive activation of the cAMP/ PKA pathway by forskolin induced large proportion of cells with features of neuroendocrine differentiation [23]. It is known that CaP cells can differentiate into neuroendocrine cells. Thus, the appearance of neuroendocrine cells in prostate tumors has been considered as a potential indicator for disease progression toward an androgen-independent state [8,23,24]. We showed that metastatic lesions from patients with CaP had significantly higher levels of AR and R2b (PRKAR2B) mRNA expression compared with the primary CaP specimens or normal prostate tissues. In contrast, expression of R1a (PRKAR1A) and PDE4B was significantly lower in metastatic specimens and was reversely correlated with AR. Our results suggested the clinical importance of PDE4B in CaP. More recently, Kashiwagi et al. [25] have shown that the expression of PDE4B is down-regulated in advanced CaP, and their knockdown promotes proliferation of LNCaP cells in androgendeprived conditions. Chen et al. [26] have shown that increase in AR mRNA is sufficient for the conversion of the phenotype of CaP from androgen-dependent to androgenindependent in a xenograft model. Our finding is in agreement with previous studies, which suggest the prognostic values of PKA pathways in patients with CaP [27,28]. Further, our novel findings by using clinical data strongly supported our results in cell line studies and provided evidence that cAMP/PKA pathways play an important role in the progression from androgendependent to androgen-refractory CaP. In conclusion, our finding suggests that deregulation of multiple signal transduction pathways and their correlation with AR in CaP cells may reflect the complexity of prostate progression and metastasis. Our study provides novel information and raised the possibility for future studies to investigate the different aspects of signaling pathways. Furthermore, the possible post-translational modification and protein-protein interaction of PKA and its negative regulators may also play a vital role in CaP metastasis. Acknowledgments This work was supported by the Swedish National Research Council, the Swedish Cancer Society, the Medical Faculty Grant, the Government Health Grant, Malm¨o Cancer Foundation, and Malm¨o Hospital Foundation (JLP). We thank Dr. Federico A. Monzon from the Department of Pathology and Laboratory Medicine, The Methodist Hospital, 6565 Fannin St., MS205, Houston, TX 77030, USA, for providing and sharing the clinical data sets.
25.e12
M. Sarwar et al. / Urologic Oncology: Seminars and Original Investigations 32 (2014) 25.e1–25.e12
Appendix A. Supplementary materials Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j. urolonc.2012.
References [1] Wegiel B, Evans S, Hellsten R, Otterbein LE, Bjartell A, Persson JL. Molecular pathways in the progression of hormone-independent and metastatic prostate cancer. Curr Cancer Drug Targets 2010;10:392–401. [2] Nieto M, Finn S, Loda M, Hahn WC. Prostate cancer: re-focusing on androgen receptor signaling. Int J Biochem Cell Biol 2007;39:1562–8. [3] Labrie F. Blockade of testicular and adrenal androgens in prostate cancer treatment. Nat Rev Urol 2011;8:73–85. [4] Taplin ME, Balk SP. Androgen receptor: a key molecule in the progression of prostate cancer to hormone independence. J Cell Biochem 2004;91:483–90. [5] Culig Z, Hobisch A, Cronauer MV, et al. Androgen receptor activation in prostatic tumor cell lines by insulin-like growth factor-I, keratinocyte growth factor, and epidermal growth factor. Cancer Res 1994;54:5474–8. [6] Grossmann ME, Huang H, Tindall DJ. Androgen receptor signaling in androgen-refractory prostate cancer. J Natl Cancer Inst 2001;93: 1687–97. [7] Ueda T, Mawji NR, Bruchovsky N, Sadar MD. Ligand-independent activation of the androgen receptor by interleukin-6 and the role of steroid receptor coactivator-1 in prostate cancer cells. J Biol Chem 2002;277:38087–94. [8] Merkle D, Hoffmann R. Roles of cAMP and cAMP-dependent protein kinase in the progression of prostate cancer: cross-talk with the androgen receptor. Cell Signal 2011;23:507–15. [9] Sadar MD. Androgen-independent induction of prostate-specific antigen gene expression via cross-talk between the androgen receptor and protein kinase a signal transduction pathways. J Biol Chem 1999;274:7777–83. [10] Chin KV, Yang WL, Ravatn R, et al. Reinventing the wheel of cyclic AMP: novel mechanisms of cAMP signaling. Ann NY Acad Sci 2002;968:49–64. [11] Taske´n K, Sk˚alhegg BS, Solberg R, et al. Novel isozymes of cAMPdependent protein kinase exist in human cells due to formation of RI alpha-RI beta heterodimeric complexes. J Biol Chem 1993;268: 21276–83. [12] Sands WA, Palmer TM. Regulating gene transcription in response to cyclic AMP elevation. Cell Signal 2008;20:460–6.
[13] Mayr B, Montminy M. Transcriptional regulation by the phosphorylation-dependent factor CREB. Nat Rev Mol Cell Biol 2001;2:599–609. [14] Sakamoto KM, Frank DA. CREB in the pathophysiology of cancer: implications for targeting transcription factors for cancer therapy. Clin Cancer Res 2009;15:2583–7. [15] Nazareth LV, Weigel NL. Activation of the human androgen receptor through a protein kinase a signaling pathway. J Biol Chem 1996;271:19900–7. [16] Desiniotis A, Schafer G, Klocker H, Eder IE. Enhanced antiproliferative and proapoptotic effects on prostate cancer cells by simultaneously inhibiting androgen receptor and cAMP-dependent protein kinase A. Int J Cancer 2010;126:775–89. [17] Chandran UR, Ma C, Dhir R, et al. Gene expression profiles of prostate cancer reveal involvement of multiple molecular pathways in the metastatic process. BMC Cancer 2007;7:64. [18] Yu YP, Landsittel D, Jing L, et al. Gene expression alterations in prostate cancer predicting tumor aggression and preceding development of malignancy. J Clin Oncol 2004;22:2790–9. [19] Simon R, Lam A, Li M-C, Ngan M, Menenzes S, Zhao Y. Analysis of gene expression data using BRB-array tools. Cancer Inf 2007;3:11–7. [20] Egeblad M, Werb Z. New functions for the matrix metalloproteinases in cancer progression. Nat Rev Cancer 2002;2:161–74. [21] Wegiel B, Bjartell A, Tuomela J, et al. Multiple cellular mechanisms related to cyclin A1 in prostate cancer invasion and metastasis. J Natl Cancer Inst 2008;100:1022–36. [22] Conti M, Beavo J. Biochemistry and physiology of cyclic nucleotide phosphodiesterases: essential components in cyclic nucleotide signaling. Annu Rev Biochem 2007;76:481–511. [23] Cox ME, Deeble PD, Bissonette EA, Parsons SJ. Activated 30 , 50 -cyclic amp-dependent protein kinase is sufficient to induce neuroendocrine-like differentiation of the LNCaP prostate tumor cell line. J Biol Chem 2000;275:13812–8. [24] Cox ME, Deeble PD, Lakhani S, Parsons SJ. Acquisition of neuroendocrine characteristics by prostate tumor cells is reversible. Cancer Res 1999;59:3821–30. [25] Kashiwagi E, Shiota M, Yokomizo A, et al. Downregulation of phosphodiesterase 4B (PDE4B) activates protein kinase A and contributes to the progression of prostate cancer. The Prostate 2012;72:741–51. [26] Chen CD, Welsbie DS, Tran C, et al. Molecular determinants of resistance to antiandrogen therapy. Nat Med 2004;10:33–9. [27] Khor L-Y, Bae K, Al-Saleem T, et al. Protein kinase A RI-a predicts for prostate cancer outcome: analysis of radiation therapy oncology group trial 86-10. Int J Radiat Oncol Biol Phys 2008;71:1309–15. [28] Pollack A, Bae K, Khor L-Y, et al. The importance of protein kinase A in prostate cancer: relationship to patient outcome in radiation therapy oncology group trial 92-02. Clin Cancer Res 2009;15:5478–84.