| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLE |
Department of Public Health and Cellular Biology, Unit of Histology (D.F., A.P., G.S.), and Department of Internal Medicine, Unit of Endocrinology (C.Ma., C.Mo.), University of Rome "Tor Vergata," 00133 Rome, Italy
Address all correspondence and requests for reprints to: Costanzo Moretti, M.D., Department of Internal Medicine, Chair of Endocrinology, Faculty of Medicine, University of Rome "TorVergata," Via di TorVergata 135, 00133 Rome, Italy. E-mail: moretti{at}med.uniroma2.it.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The effects of cAMP on cell cycle progression depends on the cell system studied. Indeed, both induction of prolifera-tion (thyroid cells, granulosa, and pituitary cells) or growth inhibition (fibroblasts, smooth muscle cells, adipocytes) have been described (15, 16, 17, 18). It has been proposed that the inhibi-tory effects of cAMP on cell growth reflect its ability to suppress the activation of ERK1/2 by growth factors (19). In other cell types, PC12 for example, cAMP activates ERK1/2 and potentiates the effects of growth factor on differentiation and gene expression (20).
Recently, increasing interest has been developed on the role of cAMP in prostate cancer. It has been shown that in androgen-dependent LNCaP and androgen-independent PC-3 and DU-145 prostate cancer cell lines PACAP elevates cAMP, leading to c-fos gene expression and proliferation (5). However, conflicting data are present in the literature concerning the role of cAMP in the control of growth and neuroendocrine differentiation of prostate cancer cell lines. Some neuropeptides that increase the levels of cAMP induce the LNCaP cells growth (5, 21). Moreover, a pretreatment with cAMP-elevating agents synergize with activation of MAPK cascade by epidermal growth factor (EGF; Ref. 22). Paradoxically, Cox et al. (23, 24) have shown that cAMP elevating factors or transfected protein kinase A (PKA) induce mitotic arrest and the acquisition of a reversible NE phenotype. NE differentiation is accompanied by an increased synthesis of neurosecretory factors, some of which are characterized by mitogenic activity (23). LNCaP is a cell line displaying many of the hallmarks of primary prostate cancer cells (25), and it can undergo NE differentiation in response to increased intracellular cAMP levels (23), long-term androgen deprivation (26), or stimulation with IL-6 (27). For this reason, these cells have emerged as a useful model to study the molecular mechanisms that control NE differentiation in prostatic carcinoma.
In the present study, we show that transient or sustained cAMP accumulation triggered by PACAP in LNCaP cells respectively stimulates either proliferation, through PKA-dependent activation of ERK1/2, or NE differentiation, by an ERK1/2-independent mechanism. Moreover, we demonstrate that PACAP induces activation of cAMP response element (CRE) binding protein (CREB) in a PKA-dependent manner, and that induction of CRE-mediated transcription requires ERKs activation. In summary, our study indicates that the timing of signaling through accumulation of intracellular cAMP may hold the key to the different outcomes of PACAP stimulation of LNCaP cells and suggest that the regulation of CREB function may be important in mediating the complex PACAP-cAMP-PKA actions at the gene expression level.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Reagents and antibodies
PACAP-27, PACAP-38, PACAP 6–27, VIP, Fsk, 1-methyl-2-isobutylxanthine (IBMX), H89, 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide (MTT), Hoechst 33258, and direct-cAMP enzyme immunoassay kit were purchased from Sigma (Milano, Italy). U0126 and Taq polymerase were obtained from Promega Corp. (Milano, Italy). 3[H]-tritiated-thymidine (TdR) was from NEN Life Science Products (Boston, MA). Rabbit polyclonal antibodies to ERK1/2, phospho-ERKs, CREB, and phospho-CREB were from New England Biolabs, Inc. (Beverly, MA). Rabbit anti-PACAP (ovine) serum was from Peninsula Laboratories, Inc. (San Carlos, CA). Affinity purified rabbit antiactin antibody was purchased from Sigma and used as indicated by the manufacturer. Cy3 conjugate goat antirabbit IgG was also from Sigma. Horseradish peroxidase-linked donkey antirabbit antibody was from Amersham Pharmacia Biotech (Milano, Italy) and SuperSignal chemiluminescent substrate was from Pierce Chemical Co. (Rockford, IL).
RNA preparation and amplification of human (h) PACAP and hPAC1R isoforms cDNA
Whole cell RNA was extracted from LNCaP cells with TRIZOL rea-gent (Roche Molecular Biochemicals, Milano, Italy) in accordance to manufacturer’s specifications. First-strand cDNA synthesis was performed as follows: 1 µg total RNA was reverse transcribed by 200 U of Maloney-murine leukemia virus reverse transcriptase (Life Technologies, Inc.) using 2.5 µM random hexamers, in the presence of 250 µM deoxynucleotide triphosphates in a final volume of 20 µl. DNA contamination or PCR carryover controls were performed omitting Maloney-murine leukemia virus reverse transcriptase during reverse transcription. The reaction mixture was incubated for 1 h at 42 C, then heat-denatured for 5 min at 75 C. Five microliters of the obtained cDNA were used to amplify hPAC1-R or hPACAP. hPAC1-R was PCR- and nested-amplified as follows: the first round of PCR was carried out using 15 pmol of primers PAC1R-1/PAC1R-2 (see Table 1
), Taq polymerase (2 U/tube), and 1.5 mM magnesium chloride in a final volume of 50 µl. The amplification consisted of 35 cycles at 94 C for 30 sec, 60 C for 30 sec, and 72 C for 30 sec, with a 5-min final extension. The PCR product was nested-amplified using primers PAC1-R-3/PAC1-R-4 and 35 cycles were carried out at 94 C for 30 sec, 48 C for 30 sec, 72 C for 30 sec, with 5 min final extension. hPACAP PCR was performed using primers listed in Table 1 (28). Then, 30 cycles (94 C for 60 sec, 60 C for 60 sec, 72 C for 60 sec, with 10 min final extension) were applied. The amplified products were analyzed on 2% (wt/vol) agarose gel by loading 10-µl aliquots and followed by ethidium bromide staining. The specificity of the PCR products was evaluated by restriction analysis and sequencing.
|
(29). cDNA was obtained as indicated before. The PCR conditions were optimized for NT or for actin gene used as control. For NT, 28 cycles (94 C for 1 min, 60 C for 1 min, and 72 C for 1 min, with a 10-min final extension) were applied. For actin, 20 cycles (94 C for 1 min, 55 C for 1 min, and 72 C for 1 min, with 10-min final extension). The levels of amplified PCR products were quantified by densitometry and expressed in comparison to actin.
Immunohistochemistry
Immunohistochemistry for hPACAP was performed on LNCaP cells cultured on glass coverslips in standard conditions. After two washes with PBS, cells were fixed in 4% neutral buffered formaldehyde for 20 min at room temperature, permeabilized with 0.1% Triton X-100 in PBS for 2 min on ice and incubated with rabbit anti-PACAP diluted 1:200 in 1% BSA/PBS overnight at 4 C. In control samples, the primary antibody was replaced by either rabbit preimmune serum or primary antibody preadsorbed with a 1000-fold excess of synthetic human PACAP. Cy3-conjugated goat antirabbit IgG was used at 1:500 dilution for 1 h at room temperature. The samples were observed under a fluorescence microscope equipped with phase contrast optics (Axioplan, Carl Zeiss, Jena, Germany).
DNA synthesis
After treatment for 24 h at 37 C, LNCaP cells were incubated with 3[H]-TdR (0.5 µCi/ml) for 3 h at 37 C. The medium was subsequently discarded and cells washed twice with ice-cold PBS. The incorporated 3[H]-TdR, after precipitation with 10% trichloracetic acid and solubilization with 0.1 N NaOH, was quantified by a liquid scintillation spectrometer (LS 301, Beckman Instruments, Inc., Irvine, CA).
MTT assay
The effect of PACAP on cell growth was assessed by MTT assay. The MTT-cell proliferation assay is based on the ability of viable cells to reduce the yellow tetrazolium salt, MTT, in a water-insoluble, dark blue formazan crystals that can be measured spectrophotometrically. The assay was carried out plating 3000 cells in 96-well microtiter plates and exposing LNCaP cells to the different treatments for 3 d. At the end of incubation, MTT was added to each well to a final concentration of 1 mg/ml and incubated 4 h at 37 C. The reaction was stopped by adding 100 µl of lysis buffer (20% wt/vol of sodium dodecyl sulfate in 50% N,N-dimethylformamide at pH 4.7) and then analyzed at 595 nm on Bio-Rad Laboratories, Inc. (Hercules, CA) multiscan plate reader. Usually six replicate wells were used for each group. Controls included untreated cells, whereas medium alone was used as a blank. A standard curve with increasing amounts of cells was performed to normalize the results.
cAMP assay
LNCaP cells were cultured in DMEM (Life Technologies, Italia) because RPMI 1640 medium contains high concentration of endogenous cAMP. For intracellular cAMP assay, cells were seeded in 24-well plates at 100,000 cells/well and treated with PACAP-27 (100 nM) or Fsk (10 µM) for various time lengths. Cultures were then extracted for 1 h at 37 C in 0.1 N HCl. For extracellular cAMP accumulation assay, cells were exposed to PACAP-27 (100 nM) with or without 0.1 mM IBMX or to Fsk (10 µM) for 3 d. At the end of the treatment, medium was removed, centrifuged at 3000 x g and stored at -80 C. Intra-extra-cellular cAMP was assayed by direct enzyme immunoassay as specified by the manufacturer.
Measurement of MAPK and CREB phosphorylation
For the measurement of ERK1/2 and CREB phosphorylation LNCaP cells were directly lysed in 1x Laemmli sample buffer. Twenty-five micrograms of protein extract per lane were resolved on 12% sodium dodecyl sulfate-polyacrylamide gels and transferred onto nitrocellulose membranes. Phospho-ERK1/2 and phospho-CREB were detected by using a 1:1000 dilution of polyclonal rabbit anti-phospho-p44/42 MAPK (Thr202/Tyr204) antibody or anti-phospho-CREB (Ser133) antibody. Membranes were routinely stripped and reprobed with anti-ERK1/2 antibody or anti-CREB antibody to check for the specificity of the signal and with polyclonal rabbit antiactin antibody for the total amount of proteins. The immunoreactive bands were visualized by enhanced chemiluminescence system. Subsequent quantitation of phospho-ERK1/2 and phospho-CREB immunoblots by scanning densitometry was normalized to the total amount of actin present or to phosphorylation state-independent CREB, respectively.
Plasmids, transfections, and luciferase reporter gene assay
Eighty percent confluent LNCaP cells in 24-well plates were cotransfected with the indicated cDNAs using a TransFast Transfection kit (Promega Corp.) according to the manufacturer’s instructions. The vector pcDNA3 (Invitrogen, Milano, Italy) was added to each set of transfections to normalize the total amount of DNA. Following transfection, cells were allowed to recover for 18 h before being maintained in low serum medium and treated, as described. The following plasmids were used: 5x CRE-luciferase (500 ng) provided by Dr. P. J. Storks (Vollum Institute, Oregon Health Science University, Portland, OR), pRSV-ßgal (250 ng); A-CREB (500 ng) for a dominant-negative CREB mutant provided by Dr. D. Ginty (Howard Hughes Medical Institute, Johns Hopkins University, Baltimore, MD; Ref. 30). Cells were treated with the appropriate stimuli for 5 h and then lysed with a Reporter Lysis Buffer (Promega Corp.). Equal protein amounts of lysate were assayed for luciferase activity with a luciferase assay kit (Stratagene, La Jolla, CA). To normalize for differences in transfection efficiency, the ß-gal enzyme assay was performed using a commercial kit (Promega Corp.).
Statistical analysis
All experiments were performed at least three times and one set of representative results was shown. Results are reported as means ± SD. Treatment groups were compared using one-way ANOVA followed by post hoc comparison of Tukey. Differences between two means with P < 0.05 were considered as significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
), and a specific signal for PACAP-38 protein was detected in the cytoplasm of these cells by immunofluorescence (Fig. 1B). LNCaP cells also express PAC1-R mRNA (Fig. 1A), and amplification by nested RT-PCR of cDNA from these cells identified the isoform without SV-1/SV-2 cassettes (PAC1-R null) as the predominant product. The expected product corresponding to PAC1-R containing SV-1/SV-2 cassettes (274 bp) was also identified with variable intensity in some experiments (not shown). These data suggest a possible autocrine action of the peptide on these tumoral cells.
|
) or PACAP-38 (not shown) was able to induce a significant and dose-dependent increase in 3[H]-TdR uptake. PACAP-27 was effective in stimulating DNA synthesis from a concentration of 10 nM, and the maximal increase, obtained with 1 µM (plus 85%), was similar to that elicited, under the same experimental conditions, by EGF (10 ng/ml), a well-known mitogenic factor for LNCaP cells (31). Moreover, 100 nM PACAP-27 added only once at the beginning of the culture caused doubling of the cell number, as monitored after 3 d of culture (Fig. 2B). On the contrary, 100 nM PACAP-27 added every 24 h (PACAP-27) was ineffective to influence LNCaP cell proliferation (data not shown). To investigate whether PACAP-27 acts through PAC1-R, we measured the mitogenic response induced by the neuropeptide in the absence or presence of PACAP 6–27, a selective PAC1-R antagonist. We found that the response was only partially inhibited by the PAC1-R antagonist (6041 ± 208 cells vs. 4228 ± 305 cells), a result that could be explained by the presence of VPAC-Rs in LNCaP cells (13, 14). Accordingly, we found that VIP (100 nM) was also able to increase LNCaP cell growth to the same extent as PACAP-27, likely through binding to VPAC-Rs because VIP action was not affected by PACAP 6–27 (Fig. 2B). Next, we investigated if cAMP-mediated activation of PKA is involved in PACAP mitogenic signaling by using the specific PKA inhibitor H89. Indeed, pretreatment with 5 µM H89 almost completely inhibited PACAP-stimulated LNCaP cell proliferation, suggesting that activation of PKA is required for such event. Interestingly, under the same conditions, EGF stimulation of LNCaP cell growth was not significantly affected (Fig. 2C), indicating that H89 did not have a nonspecific toxic effect.
|
, both Fsk and PACAP-27 rapidly elevated intracellular cAMP, with a maximal increase obtained after 10 min of the agonist addition. After 3 h of culture, we observed that in PACAP-27-treated cells cAMP levels were lowered to approximately 10% of the maximal value, whereas in Fsk-treated cells, cAMP levels remained high (
50% of the maximal value). Because both PACAP-27 and Fsk caused intracellular cAMP elevation but exerted different cellular events, proliferation in the case of PACAP (Fig. 2A) and differentiation in the case of Fsk (23), we hypothesized that sustained high levels of cAMP were necessary for the mitotic arrest. To confirm this hypothesis, we added PACAP-27 to the proliferating cells every 24 h for 3 d or we treated the cells with PACAP-27 added only at the beginning of the culture in the presence of IBMX, a phosphodiesterase inhibitor. When measured after 3 d, cAMP levels were still elevated in cells that had been treated with PACAP-27 (100 nM) every 24 h compared with untreated cells (55 pmol/well ± 3 vs. 5 pmol/well ± 0.3), whereas a single dose of PACAP-27 at the beginning of the culture was insufficient to maintain high cAMP levels (Fig. 4A). An even more pronounced effect was observed when PACAP-27 was added only at the beginning of the culture in the presence of IBMX (110 pmol/well ± 20), resembling the effect obtained with a single dose of Fsk (10 µM; Fig. 4A). Furthermore, we found that the continuous PACAP-27 treatment for 3 d caused a 40% cell growth inhibition compared with cells treated with 5% FBS (Fig. 4B, lane 3). Similar results were also observed in PACAP-27/IBMX culture conditions, whereas PACAP-27 added only at the beginning of the culture didn’t affect LNCaP cell proliferation (data not shown). As previously reported, Fsk, which we have shown to induce a sustained accumulation of cAMP, was able to inhibit cell growth of 60% (Fig. 4B, lane 5). Both agonists exerted their effect through a PKA-depending pathway, because the pretreatment with H89 was able to revert the inhibitory response of both compounds (Fig. 4B, lanes 4–6). The action of H89 was also evident when it was added after 24 h or 48 h from the beginning of PACAP-27 or Fsk stimulation (Fig. 4C). These data suggest that treatments causing a prolonged increase of intracellular cAMP and PKA activation elicit inhibition of LNCaP cell growth.
|
|
). This effect was partially prevented by a pretreatment with PACAP 6–27, indicating that PAC1-R is involved in this phenomenon (Fig. 5B). On the contrary, 10 µm Fsk induced NE morphology as rapidly as PACAP-27, but the effect was still evident after 3 d (data not shown; and see Ref. 23). Chronic PACAP-27 treatment or cotreatment with PACAP-27 and IBMX, conditions that maintain a sustained cAMP accumulation, exerted the same effect as Fsk, and the NE morphology of LNCaP cells was still observed after 3 d (see Fig. 7C). These data suggest that the inhibitory effect of the peptide on the cell growth correlates with the maintenance of NE differentiation.
|
|
, lane 5). Cells treated with PACAP-27 only at the beginning of the 3 d of culture expressed slightly higher levels of NT compared with proliferating cells (Fig. 5B, lane 2). However, NT mRNA expression in LNCaP cells was strongly stimulated following either a chronic treatment with PACAP-27 or a cotreatment with PACAP-27 and IBMX (Fig. 5B, lanes 3 and 4). These results are in agreement with the requirement of a sustained accumulation of cAMP to induce NE differentiation in LNCaP cells.
PACAP induces ERK1/2 activation through a PKA-MAPK kinase (MEK)-dependent pathway
Activation of the ERK1/2 MAPK cascade is a common feature of many mitogenic signals (32). To determine if MAPK’s activation is involved in PACAP-induced LNCaP cell growth, we assessed ERK1/2 activity after treatment with the neuropeptide. ERK activity was judged by the levels of ERK1/2 phosphorylation as determined by Western blot analysis of cell lysates using a phosphospecific antibody. Stimulation with different concentrations of PACAP-27 for 10 min resulted in a dose-dependent increase of ERK1/2 phosphorylation: when the peptide was used at 100 nM, the standard concentration used to stimulate cell proliferation, a 3-fold increase of ERK1/2 phosphorylation over basal levels was observed (Fig. 6A). PACAP-27 (100 nM) induced a rapid and transient ERK1/2 activation, peaking at 15 min after agonist addition (Fig. 6B). In comparison, activation of ERK1/2 induced by EGF was sustained, as protein phosphorylation was still over basal level 2 h after application of the growth factor (data not shown; and see Ref. 31). Pretreatment with PACAP 6–27 significantly reduced ERK1/2 phosphorylation (Fig. 6C), indicating that PAC1-R is functionally involved in ERKs activation.
|
). Finally, pretreatment of LNCaP cells for 20 min with 10 µM U0126, an MEK-specific inhibitor, attenuated both EGF- and PACAP-induced stimulation of ERK1/2 phosphorylation (Fig. 6D).
ERK1/2 phosphorylation is differently involved in PACAP regulation of cell proliferation and NE differentiation
To verify whether ERK1/2 activation was required for PACAP-induced cell proliferation, the effect of U0126 on PACAP- and EGF-induced cell proliferation was investigated. The growth of LNCaP cells induced by a short exposure to 100 nM PACAP-27 was partially inhibited in the presence of 10 µM U0126 for 3 d. Indeed, under these conditions, the percentage of viable cells was decreased from 208.3 ± 15.1% (PACAP-27 only) to 128.8 ± 21% (PACAP-27 plus U0126). In similar experiments, U0126 was more effective in reducing EGF action (from 270.5 ± 37% to 138.2 ± 45%; Fig. 7A). We ruled out the possibility that the inhibition of proliferation by U0126 could be due to the induction of apoptosis because throughout the culture, the fraction of apoptotic nuclei after staining with Hoechst was negligible both in cells stimulated with agonist or with agonist plus inhibitor (data not shown).
In many cell types, cAMP inhibits cell proliferation negatively altering ERKs activation by growth factors (33). We evaluated if this mechanism could be related to the negative PACAP control of LNCaP cells growth studying the phosphorylation state of ERKs in the inhibitory growth conditions. LNCaP cells were cultured in 5% FBS and stimulated with Fsk (10 µM) or with PACAP-27 (100 nM) added every day. ERK1/2 activation was evident in proliferating cells during the 3 d of stimulation, but neither Fsk nor PACAP-27 influenced the phosphorylation state at any time analyzed (Fig. 7B). The total ERKs activity was also not modified by the two agonists during the culture. These results indicated that, in this cell type, cAMP exerts growth inhibitory action without altering ERK activation. Moreover, ERK1/2 activation by PKA seemed not to be involved in the acquisition of neuroendocrine morphology because the pretreatment with U0126 failed to prevent neurite appearance in LNCaP cells induced by Fsk or chronic PACAP-27 treatments (Fig. 7C). The MEK inhibitor was ineffective also when added to the culture medium at various time points after PACAP-27 or Fsk addition (not shown).
The requirement of ERK activation for NT expression was evaluated by semiquantitative PCR performed with RNA obtained from LNCaP cells cultured in the absence or presence of U0126. As shown in Fig. 7D, the stimulation of NT expression induced by a chronic PACAP-27 or Fsk treatment was not affected by the MEK inhibitor, thus indicating that ERKs were not involved in cAMP-induced NE differentiation.
PACAP stimulates CREB activation via an ERK-independent pathway and CRE-dependent transcription
A well-established event triggered by activation of PKA is the phosphorylation of the transcription factor CREB on Ser133, which causes activation of many CRE-responsive genes (34). We found that both PACAP-27 and Fsk induced a rapid phosphorylation of CREB as assayed by Western blot using a phospho-Ser133 antibody. After 5 min of stimulation, there was a 5-fold PACAP-27-induced increase in the levels of phosphorylated CREB in comparison to unstimulated cells (Fig. 8A), returning to baseline after 4 h of treatment, whereas Fsk-induced CREB phosphorylation was maintained at the maximal level for 24 h in line with the capacity of this AC activator to induce a prolonged cAMP accumulation.
|
). In contrast, pretreatment with U0126 failed to affect the PACAP-induced phosphorylation of CREB, suggesting that phosphorylation of the transcription factor on Ser133 was ERK independent (Fig. 8B). Furthermore, we investigated the ability of PACAP-27 to regulate CREB transcriptional activity using a CRE reporter system consisting of five reiterated CREs controlling the expression of luciferase. As shown in Fig. 8C, 100 nM PACAP-27 led to a 4-fold increase in CRE-dependent transcription; this effect requires CREB activation because it was blocked by cotransfection with a dominant-negative mutant of CREB (A-CREB). As expected from the results on CREB phosphorylation, transcriptional activity induced by PACAP-27 or Fsk was dependent on PKA activity because it was completely inhibited by pretreatment with H89 (Fig. 8C). Interestingly, we found that, in addition to CREB phosphorylation, an ERK-induced event, was also required for the stimulation of CRE-luciferase transcription because U0126 blocked both PACAP- and Fsk-induced transcriptional activation of the reporter gene (Fig. 8D). |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The present study was addressed to evaluate the role of PACAP on LNCaP cell proliferation and NE differentiation. Our results demonstrate that LNCaP cells express the PACAP gene and synthesize its peptide product. As also reported by others (14), we show that these cancer cells express PACAP receptor type 1 (PAC1-R) variants. Alternative splicing of two exons in the human PAC1 gene generates protein variants with variable tissue/cell-specific expression. Electrophoretic analysis of PCR products provided evidence that in LNCaP cells the null isoform is the most represented.
The second messenger cAMP has been shown to exert diverse functions in different cell systems. Furthermore, a different outcome can be reached in the same cell depending on whether a transient vs. a sustained increase in cAMP is elicited (39). Here we show that PACAP stimulates a rapid but transient cAMP accumulation in LNCaP cells, and that this is associated to increased cell proliferation. The mitogenic effect of PACAP is likely to depend on cAMP and PKA activation because H89 almost completely inhibits PACAP action. LnCAP cells express high levels of VPAC1-R (14) and lower levels of PAC1-R (this study and Ref. 14). Although both these receptors are coupled to AC, PAC1-R stimulation also elicits activation of phospholipase C (9). Because we show that the selective PAC1-R antagonist PACAP 6–27 exerts only a limited inhibition of mitogenesis, and that the VPAC1-R agonist VIP can replace PACAP as mitogenic factor, we propose that PAC1-R is only partially involved in PACAP-induced LNCaP cell proliferation.
We have observed that PACAP triggers a rapid and dose-dependent increase of ERK1/2 phosphorylation in LNCaP cells. Experiments using the selective inhibitors H89 and U0126 demonstrate that ERKs activation requires PKA and MEK activity. In both LNCaP and PC12 neuroendocrine cells, cAMP activates MEK by an Rap1/B-Raf-dependent pathway (22, 40). Because in LNCaP cells cAMP elevation induced by Fsk or other agonists by itself only weakly activates MAPK, it has been proposed that Rap1/B-RAF mediators are unable to fully activate the pathway without the participation of an additional receptor-initiated signal (22). Indeed, it has been shown that cAMP-elevating agents potentiate the action of growth factors such as EGF on MAPK activation in LnCAP cells (22). We found that PACAP-induced ERKs activation is transient, compared with the sustained EGF-mediated phosphorylation (Farini, D., personal observations; and Ref. 31). Whether PACAP alone is able to stimulate LNCaP cell proliferation through the ERKs signal, or rather it requires the additional presence of other growth factors or trans-activation of the EGF receptor, involved in the growth promoting activity of G protein-coupled receptor (41), is still uncertain. The possible interaction of PACAP with growth factor signaling is further suggested by the observation that the blockade of the EGF receptor in prostate cancer cells attenuates not only the actions of EGF, but also that of IGF-I on the MAPK and PKA pathways (42). Our data indicate that PACAP resembles the action of other cAMP-raising compounds (21, 22) in effectively increasing the mitogenicity of the ERK pathway, at least in the androgen-sensitive LNCaP cells.
The positive control of LNCaP cell growth by PACAP is apparently in contrast, however, with data concerning the acquisition of a NE phenotype by these cells when exposed to different agonists inducing cAMP accumulation. Indeed, Fsk or isoproterenol rapidly induce the development of neuritic processes and increase the expression of NE markers whereas LNCaP progressively stop proliferating (23). Furthermore, ectopic expression of a constitutively activated PKA catalytic subunits in LNCaP cells is able per se to induce NE differentiation (24). The NE phenotype of these cells is reverted upon withdrawal of differentiation-inducing agents, suggesting that trans-differentiation is a dynamic process, in part due to the balance of different cues in the local environment. Because the persistent presence of the agonist is required for differentiation, we reasoned that the lifetime of a given intracellular transducer (i.e. cAMP and PKA activation) could be a critical factor to address the cell life toward proliferation or differentiation. Indeed, we found that a single dose of PACAP elicits a transient rise in cAMP levels and is unable to induce differentiation. However, repetitive addition of PACAP or cotreatment with the cAMP-phosphodiesterase inhibitor IBMX, led to a sustained accumulation of cAMP and were sufficient to maintain the NE phenotype. We also show that PKA activity is required for inhibition of proliferation when PACAP is continuously present in the culture medium. It remains unknown what are the effectors of PKA action in LNCaP cells.
Our data indicated that ERK1/2 are not involved in PKA-mediated induction of NE differentiation. In fact, the ERKs activation during the culture conditions in which is evident the NE differentiation is unchanged and the MEK inhibitor U0126 is unable to influence NE morphology and NT expression. Interestingly, in a different way from other cell types, the inhibitory action on LNCaP cell growth exerted by sustained cAMP accumulation and PKA activation might be also independent by the ERKs activation pathway.
It is known that translocation of PKA to the nucleus results in phosphorylation of transcription factors and this may lead to induction of specific gene expression in parallel with the stimulation or the inhibition of LNCaP cell proliferation. PKA is known to stimulate the activity of CREB by phosphorylating Ser-133 (43). CREB is actively involved in the cell cycle control in a variety of cell types (44, 45). In this study, we report for the first time that PACAP is able to stimulate CRE-mediated gene expression through CREB activation in LNCaP cells. We show that CREB phosphorylation stimulated by PACAP occurs via a PKA-dependent pathway, without the involvement of ERK activation. Multiple PKA-regulated events are required for transcriptional activity of CREB (34). In particular, MAPK-mediated phosphorylation of coactivators like p300/CREB-binding protein is required for their interaction with CREB (46). Our experiments suggest that also in LNCaP cells CRE-mediated transcription by PACAP require ERK1/2 activation, independently from their known action on CREB phosphorylation. Therefore, a coordinate cross-talk between cAMP-PKA and MAPK cascade might be necessary for the complex regulation of CREB-mediated gene expression. An involvement of CREB in PACAP-PKA-mediated stimulation of cell proliferation cannot be excluded, because in certain cell systems growth factors induce CREB phosphorylation and transcription of c-fos, fundamental events for their mitogenic action (47).
In conclusion, we support the hypothesis that a different availability of cAMP in the neoplastic cell microenvironment may change the fate of the prostate tumor. A transient neuropeptide-induced cAMP rise, during the androgen-dependent state, may contribute to the neoplastic growth. A large and sustained cAMP production might be involved in the loss of androgen sensitivity (48), helping in the process by which tumoral cells undergo NE differentiation thus increasing tumor malignancy. Further studies will be necessary in particular to explore the role of CREB in the molecular pathways underlying prostate cancer progression.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: AC, Adenylate cyclase; CRE, cAMP response element; CREB, CRE binding protein; EGF, epidermal growth factor; FBS, fetal bovine serum; Fsk, forskolin; gal, galactosidase; h, human; IBMX, 1-methyl-2-isobutylxanthine; MEK, MAPK kinase; MTT, 3-[4,5,dimethylthiazol-2-yl]-2,5,diphenyl tetrazolium bromide; NE, neuroendocrine; NT, neurotensin; PACAP, pituitary adenylate cyclase-activating peptide; PAC1-R, PACAP type 1 receptor; PKA, protein kinase A; TdR, tritiated-thymidine; VIP, vasoactive intestinal peptide; VPAC2-R, VIP/PACAP subtype 2 receptor; VPAC1-R, VIP/PACAP subtype 1 receptor.
Received September 26, 2002.
Accepted for publication December 17, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2-macroglobulin: role in mitogenesis and cell proliferation. J Biol Chem 277:4069–4078This article has been cited by other articles:
 |
|
 |
|
  T.-C. Yuan, S. Veeramani, and M.-F. Lin Neuroendocrine-like prostate cancer cells: neuroendocrine transdifferentiation of prostate adenocarcinoma cells Endocr. Relat. Cancer, September 1, 2007; 14(3): 531 - 547. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A M Lobascio, F G Klinger, M L Scaldaferri, D Farini, and M De Felici Analysis of programmed cell death in mouse fetal oocytes Reproduction, August 1, 2007; 134(2): 241 - 252. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. D. Deeble, M. E. Cox, H. F. Frierson Jr., R. A. Sikes, J. B. Palmer, R. J. Davidson, E. V. Casarez, G. P. Amorino, and S. J. Parsons Androgen-Independent Growth and Tumorigenesis of Prostate Cancer Cells Are Enhanced by the Presence of PKA-Differentiated Neuroendocrine Cells Cancer Res., April 15, 2007; 67(8): 3663 - 3672. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
, P < 0.01 compared with samples without treatments; *, P < 0.05 compared with PACAP-27 treatment. C, Inhibition of PKA activity by H89 reduces PACAP-stimulated proliferation of LNCaP cells line. Cells were pretreated with H89 (5 µM) for 1 h and then stimulated with PACAP-27 (100 nM) or EGF (10 ng/ml) for 3 d at 37 C. Proliferation was measured by MTT assay as described in Materials and Methods. Bars represent the averages of six replicate measurements ± SD. *, P < 0.01 respect to samples without treatment; 
, P < 0.01 with respect to PACAP treatments.