Abstract
The mechanisms by which Ca2+-independent phospholipase A2 (iPLA2) mediates cell growth in p53-positive LNCaP and p53-negative PC-3 prostate cancer cell lines were studied. Exposure of cells to the iPLA2 selective inhibitor bromoenol lactone (BEL; 0–20 μM) induced concentration- and time-dependent decreases in cell growth based on 3-(4, dimethylthiazolyl-2)-2,5-diphenyltetrazolium bromide staining and cell number. Decreased cell growth was not caused by cell death as BEL exposure did not alter nuclear morphology or increase annexin V (apoptotic cell marker) or propidium iodide (necrotic cell marker) staining after 48 h. Decreased growth correlated to a G1/G0 arrest in LNCaP cells and aG2/M arrest in PC-3 cells. In LNCaP cells, G1 arrest was preceded by time- (0–48 h) and concentration-dependent (0–10 μM) increases in the expression of the tumor suppresser protein p53 and the cyclin-dependent kinase inhibitor p21. Increases in p53 expression preceded increases in p21 expression by 8 h. In LNCaP cells, BEL treatment decreased the expression of the p53 antagonist Mdm2, while increasing Akt phosphorylation. BEL treatment also increased Akt phosphorylation in PC-3 cells, but Mdm2 was not detected. The ability of BEL to increase Akt phosphorylation was inhibited by the phosphoinositide 3-kinase inhibitor LY294002 [2-(4-morpholinyl)-8-phenyl-4H-1-benzopyran-4-one]. BEL treatment also decreased agonist-induced activation of the epidermal growth factor receptor. These data suggest that inhibition of iPLA2 decreases prostate cancer cell growth by p53-dependent and independent mechanisms. Furthermore, alterations in Mdm2 and epidermal growth factor receptor activation following BEL exposure suggest novel roles for iPLA2 in prostate cancer cell signaling.
Prostate cancer is one of the most prevalent cancers in men in the United States, accounting for 9% of cancer cases reported in 2005 (http://www.cancer.org/docroot/stt/stt_0.asp). This resulted in 218,890 new prostate cancer cases and approximately 27,000 deaths (http://www.cancer.org/docroot/stt/stt_0.asp). To date, androgen ablation is the most widely used treatment for prostate cancer; however, most advanced prostate cancers will develop into an androgen refractory phenotype (Murillo et al., 2006). Thus, it is critical to identify novel targets for the treatment of prostate cancers.
Phospholipase A2 (PLA2) are esterases that cleave glycero-phospholipids at the sn-2 ester bond to release a fatty acid and a lysophospholipid (Cummings, 2007). PLA2 are classified into three major categories: the secretory PLA2 (sPLA2), the calcium-dependent cytosolic PLA2 (cPLA2), and the Ca2+-independent phospholipase A2 (iPLA2). sPLA2 and cPLA2 are relatively more studied than iPLA2 and participate in numerous physiological functions, including inflammation, phospholipid remodeling, and possibly cancer cell growth (Yamashita et al., 1995).
In comparison to cPLA2 and sPLA2, iPLA2 is relatively novel. However, it recently has drawn attention as a cell growth regulator because of its possible role in insulin secretion (Ramanadham et al., 1999), eicosanoid metabolism (Akiba et al., 1998), phospholipid remodeling (Balsinde and Dennis, 1997), apoptosis (Atsumi et al., 2000; Cummings et al., 2004), vascular relaxation (Seegers et al., 2002), and adipogenesis (Su et al., 2004). Recent studies also suggest that iPLA2 mediates cell proliferation in HEK293 (Saavedra et al., 2006), human insulinoma, colon cancer (Bao et al., 2006; Zhang et al., 2006), and ovarian cancer cells (Song et al., 2007). It is not known whether iPLA2 mediates prostate cancer cell growth.
Multiple types of iPLA2 exist (Cummings, 2007), but the most studied isoform is a cytosolic localized 85-kDa protein called iPLA2β or Group VIA PLA2 (Balsinde et al., 1997; Ramanadham et al., 1999; Bao et al., 2006). This isoform is suggested to mediate phospholipid remodeling and cell growth in a variety of cell types, including insulinoma (INS-1) and ovarian cancer cells (Bao et al., 2006; Song et al., 2007). However, a membrane-localized iPLA2 called iPLA2γ (Group VIB PLA2) is also expressed in several cells types (Jenkins et al., 2002; Kinsey et al., 2005; Murakami et al., 2005) and may mediate these same functions (Cummings et al., 2002, 2004; Murakami et al., 2005). The differential expression of iPLA2β and γ in prostate cancer cells, or their role in prostate cancer cell growth, is not known.
The exact mechanisms by which iPLA2 mediates cell growth are not known. Studies in INS-1 and human HCT116 colon cancer cells demonstrated that iPLA2 inhibition correlated to decreased cell growth and G1 arrest (Zhang et al., 2006). G1 arrest correlated to increased expression of the tumor suppressor protein p53 and the cyclin-dependent kinase inhibitor p21 by unknown mechanisms. In contrast, iPLA2 inhibition using the selective inhibitor bromoenol lactone (BEL) decreased cell growth in ovarian cancer cells by inducing apoptosis and G2/M arrest (Song et al., 2007). The G2/M arrest and apoptosis in ovarian cancer cells were p53-independent.
The possibility that iPLA2 mediates cell growth by p53-dependent and independent pathways is of special concern in prostate cancer. Alterations in p53 expression correlate to prostate cancer progression, metastases, and androgen-independent growth (Burchardt et al., 2001; Song et al., 2007). Analysis of human prostate tumors suggests that untreated and localized tumors express unmutated p53, whereas androgen-resistant and metastatic prostate cancers express mutant p53 or no p53 at all (Thompson et al., 1995; Burchardt et al., 2001). These studies suggest that identifying signaling pathways that alter p53 expression in prostate cancer cells may lead to novel treatments for this disease. Toward this goal, we tested the effect of iPLA2 inhibition on cell growth, death, cell cycle, and cell signaling in p53-positive prostate cancer LNCaP and p53-negative PC-3 cell lines.
Materials and Methods
Materials. PC-3 (human prostate cancer), LNCaP (human prostate cancer) cells, Ham's F-12K medium, RPMI-1640, and fetal bovine serum were purchased from American Type Culture Collection (Manassas, VA). Penicillin and streptomycin were obtained from Invitrogen (Carlsbad, CA). Annexin-V FITC was obtained from R&D Systems (San Diego, CA). Mouse anti-human p53, p21, phospho-Akt, and GAPDH monoclonal antibodies, the goat anti-human phospho-EGFR monoclonal antibody, and the goat anti-human regular Akt antibody were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). The rabbit anti-human EGFR antibody was purchased from Cell Signaling Technology (Danvers, MA). 4′,6-Diamidino-2-phenylindole (DAPI), LY294002, propidium iodide (PI), racemic bromoenol lactone (BEL), R-BEL, and S-BEL were obtained from Cayman Chemical Co. (Ann Arbor, MI). All other chemicals were obtained from Sigma-Aldrich (St. Louis, MO).
Immunoblot Analysis. After treatment, cells were washed three times and removed using a rubber policeman in immunoblot buffer [0.25 M Tris-HCl, pH 6.8, 4% SDS, 10% glycerol, 1 mg/ml bromphenol blue, and 0.5% (v/v) 2-mercaptoethanol]. Protein levels were determined in duplicate wells using lysis buffer that contained 1% (v/v) Triton X-100. After isolation, 20 μg of total protein was mixed with SDS-sample buffer, heated to 70°C for 10 min, separated under reducing conditions on a 12% SDS-polyacrylamide gel, and transferred to a nitrocellulose membrane. Nonspecific binding was blocked by incubating the membrane with 3% (w/v) bovine serum albumin in TBS buffer (10 mM Tris-HCl, pH 7.5, 150 mM NaCl) overnight at room temperature. Membranes were then incubated with the indicated primary antibody for 2 h followed by a 2-h incubation in the appropriate secondary antibody. Bands were detected by enhanced chemiluminescence (GE Healthcare, Chalfont St. Giles, UK).
Measurement of iPLA2 Activity. After treatment, cells were collected, and subcellular fractions were isolated using differential centrifugation. Arachidonoyl thio-phosphatidylcholine was used as a synthetic substrate to detect PLA2 activity. Hydrolysis of the arachidonoyl thioester bond at the sn-2 bond by PLA2 releases a free thiol that can be detected by 5,5′-dithiobis(2-nitrobenzoic acid). PLA2 activity was measured in cytosolic and microsomal fractions in the absence and presence of racemic, R-or S-BEL, and 4 mM EGTA. Samples were treated for 30 min at room temperature before the assay. Activity was calculated by measuring the absorbance of 5,5′-dithiobis(2-nitrobenzoic acid) (ϵ = 10.66) at 404 nm and normalized with the protein content of each sample.
Analysis of Cell Growth and Proliferation. Cell growth and proliferation were determined using MTT staining and cell counting. Cells were seeded at the indicated concentration and allowed to attach for 24 h before treatment. At the desired time points, 20 μlof 5 mg/ml MTT was added to each well; the cells were incubated for 2 h, after which the medium was removed; 400 μl of dimethyl sulfoxide was added to dissolve the resulting purple formazan; and absorbance was read at 544 nm with a FLUOstar OPTIMA plate reader (BMG Labtechnologies, Inc., Durham, NC). Cell number was determined using trypan blue and a hemacytometer.
Assessment of Nuclear Morphology. For assessment of nuclear morphology, cells were grown on glass coverslips for the indicated time. After treatment, medium was removed, and cells were washed twice with PBS, fixed for 20 min using 10% buffered formalin/4% formaldehyde, and washed with PBS. After washing, cells were incubated at 25°C with DAPI (16.6 μM final concentration) for 10 min and washed three times, mounting medium was applied, and the coverslips were inverted onto glass slides. Visualization of DAPI staining was performed using a Nikon AZ100 fluorescence microscope (Nikon, Melville, NY) with excitation and emission filters of 350 and 486 nm, respectively.
Immunocytochemistry. Cells were grown on coverslips before exposure to either solvent control or BEL for 16 h. After treatment, cells were washed with PBS, fixed using 10% buffered formalin/4% formaldehyde, and washed with PBS (Cummings and Schnellmann, 2002). Cells were then permeabilized and washed, and nonspecific binding was blocked by incubation with PBS/8% bovine serum albumin for 30 min. After washing, cells were incubated at 4°C overnight with either the primary antibody against p53 (10 μg/ml) or an IgG control, washed three times, and incubated with a secondary antibody conjugated to FITC and DAPI (16.6 μM) for 2 h. Samples were washed three times and covered with mounting medium, and coverslips were inverted onto glass slides. Visualization of staining was done using a Nikon AZ100 fluorescence microscope with excitation and emission filters of 488 and 520, respectively, for p53 and 350 and 486 nm, respectively, for DAPI.
Measurement of Cell Death. Cell death was assessed using annexin V (apoptotic cell marker) and propidium iodide (PI, necrotic cell marker) staining and flow cytometry as previously described with modifications. Cisplatin (50 μM) and tert-butylhydroperoxide (500 μM) were used as positive controls for apoptosis and necrosis, respectively. In brief, after treatment, medium was removed, and cells were washed twice with PBS and incubated in binding buffer (10 mM HEPES, 140 mM NaCl, 5 mM KCl, 1 mM MgCl2, 1.8 mM CaCl2, pH 7.4) containing annexin V-FITC (25 μg/ml) and PI (25 μg/ml) for 10 min. Cells were washed three times using binding buffer and released from the monolayers using a rubber policeman, and staining was quantified using a FACSCalibur flow cytometer (BD Immunocytometry Systems, San Jose, CA). For each measurement, 10,000 events were counted.
Measurement of Cell Cycle. Cell cycle and DNA hypoploidy were assessed using methods described previously. In brief, cells were washed twice with sample buffer [PBS plus glucose (1 g/liter)], dislodged using Cellstripper (Mediatech, Herndon, VA), centrifuged at 400g for 10 min, and suspended in sample buffer. Cells were fixed in ice-cold ethanol (70% v/v) and stained with PI (50 μg/ml) in sample buffer containing RNase A (100 U/ml) for 30 min at room temperature with gentle shaking. Samples were analyzed within 24 h by flow cytometry with a FACSCalibur flow cytometer.
Protein Determination. Protein concentrations were determined using the bicinchonic acid assay method as described by Sigma-Aldrich.
Statistical Analysis. Cells isolated from a distinct passage represented one experiment (n = 1). Data are represented as the average ± S.E.M. of at least three separate experiments (n = 3). The appropriate analysis of variance was performed for each data set using SAS software (SAS Institute, Cary, NC). Individual means were compared using Fisher's protected least significant difference test with P < 0.05 considered as indicative of a statistically significant difference between mean values.
Results
Expression of iPLA2 Isoforms in LNCaP and PC-3 Cells. The major iPLA2 isoforms are a cytosolic 80 to 85-kDa isoform, referred to as iPLA2β (Group VIA), and a 66 to 85-kDa membrane-associated isoform, referred to as iPLA2γ (Group VIB) (Balsinde et al., 2002). The expression of these proteins in human prostate cancer cells has never been determined. Immunoblot analysis using a polyclonal antibody against cytosolic iPLA2β resulted in the detection of bands at approximately 85 kDa in both PC-3 and LNCaP cells (Fig. 1A). Immunoblot analysis using a peptide antibody against iPLA2γ resulted in the detection of two bands at approximately 63 and 78 kDa in both PC-3 and LNCaP cells (Fig. 1B).
PC-3 cells contained both cytosolic and microsomal iPLA2 activity as indicated by the cleavage of the synthetic substrate arachidonoyl thio-phosphatidylcholine (Fig. 1C). Treatment of cells with racemic 2.5 μM BEL, a selective iPLA2 inhibitor (Cummings et al., 2002; Kinsey et al., 2005), decreased activity in both cytosolic and microsomal fractions, compared to control cells. Treatment of cells with R-BEL, a selective inhibitor of microsomal iPLA2 (Jenkins et al., 2002; Kinsey et al., 2005), did not decrease cytosolic iPLA2 activity but did decrease activity in microsomes. In contrast, treatment of cells with S-BEL (2.5 μM), a selective inhibitor of cytosolic iPLA2 (Jenkins et al., 2002; Kinsey et al., 2005), decreased cytosolic but not microsomal iPLA2 activity. Collectively, these data demonstrate that PC-3 and LNCaP cells express both iPLA2β and γ and that this activity is inhibited by BEL and its enantiomers.
Effect of Inhibition of iPLA2 on LNCaP and PC-3 Cell Growth. The effect of iPLA2 inhibition on prostate cancer cell growth was assessed by measurement of MTT and cell number in LNCaP and PC-3 cells after exposure to racemic R- and S-BEL. Both cell lines were seeded in 48-well plates at 64,000 cells/ml, allowed to attach for 24 h, and then exposed to 0–10 μM BEL for 24 or 48 h. Treatment with racemic BEL significantly decreased MTT staining in both cell lines in a concentration- and time-dependent manner (Fig. 2, A and C). MTT staining was decreased 50% in both cell lines after a 48-h exposure to 10 μM racemic BEL. Significant decreases in MTT staining were detected at earlier time points in LNCaP cells, compared with PC-3 cells. Treatment of cells with Rand S-BEL also decreased MTT staining, but only after 48 h (Fig. 2, B and D). Racemic-BEL was more effective than R-or S-BEL at decreasing MTT staining in both cell lines. Thus, it was used exclusively throughout the rest of the experiments.
The effect of BEL on nuclear morphology was studied to confirm the effect of BEL on cell growth and to rule out the presence of cell death (Fig. 3, A–D). Nuclear morphology in untreated LNCaP (Fig. 3, A and C) and PC-3 cells (Fig. 3, B and D) was normal with a minimal number of nuclei displaying chromatin condensation or nuclear fragmentation. Treatment of cells with BEL for 24 h did not significantly increase either type of apoptotic morphology (Fig. 3, C and D). However, a decrease in the number of nuclei was observed. This decrease correlated to a decrease in cell number after 48 h of exposure (Fig. 3, E and F).
To further test the hypothesis that BEL did not induce cell death at the concentrations used, annexin V and PI staining were determined using flow cytometry (Fig. 4, A and B). Treatment of cells with concentrations of BEL as high as 20 μM did not increase annexin V or PI staining above control levels after 48 h (Fig. 4, A and B). Collectively, these data suggest that BEL treatment decreases prostate cancer cell growth in the absence of cell death.
Previous studies in U937, THP-1, and MonoMac (human phagocyte), RAW264.7 Jurkat (human T lymphocyte), and GH3 (human pituitary) demonstrated that concentrations of BEL below 15 μM do not induce apoptosis, whereas concentrations above do (Fuentes et al., 2003). Concentrations of BEL as low as 2 μM also induced apoptosis in ovarian cancer cells (Song et al., 2007). In contrast, studies in insulinoma and colon carcinoma cells demonstrated that BEL does not induce cell death at concentrations as high as 12.5 μM (Zhang et al., 2006). Based on data from a previous study (Song et al., 2007), we hypothesized that these differences are caused by different serum or medium conditions. We tested this hypothesis by exposing serum-starved LNCaP and PC-3 cells to BEL (0 to 20 μM) for 24 and 48 h. It is noteworthy that removal of serum increased annexin V binding only in LNCaP cells after exposure to concentrations of BEL higher than 10 μM BEL (Fig. 4). Furthermore, PI staining was increased in both cell types, but only after exposure to 20 μM BEL for 48 h (Fig. 4C). These data demonstrate that BEL can induce cell death in prostate cancer cells, but only in the absence of serum and after exposure to high concentrations.
Effect of Inhibition of iPLA2 on Cell Cycle. Treatment of LNCaP cells with 0 to 10 μM racemic BEL for 24 h modestly, but significantly, increased the percentage of cells in the G1 phase of the cell cycle compared with control cells (Fig. 5A). In contrast, treatment of PC-3 cells with similar doses of BEL caused S- and G2/M-phase arrests (Fig. 5B) compared with control cells.
Effect of Inhibition of iPLA2 on p53 and p21 Expression. Cell cycle arrest and decreased cell growth can be mediated by alterations in cyclin-dependent kinase complexes, which are mediated by p21 and p53 (Zhang et al., 2006). Not unexpectedly, treatment of PC-3 cells with BEL did not induce p53 or p21 expression at any concentration or time point tested (data not shown). In contrast, treatment of LNCaP cells with 5 μM racemic BEL increased p53 expression compared with controls as early as 2 h after exposure (Fig. 6, A and B). These increases were more rapid than those induced by cisplatin. The expression of p53 began to decrease after 21 h of exposure to BEL (Fig. 6A) and was reduced to control levels by 48 h (Fig. 6C).
Treatment of LNCaP cells with BEL increased p21 expression after 16 h (Fig. 6A). Like p53, the ability of BEL to induce p21 expression was concentration-dependent (Fig. 6B). Increases in p21 expression occurred after increase in p53 and also after treatment with cisplatin. Increases in p53 and p21 were not a result of differences in loading as demonstrated by the expression of GAPDH.
To study the expression of p53 at longer time points cells were exposed to 5 μM BEL for periods up to 48 h (Fig. 6C). As expected, treatment of cells with BEL increased expression of p53 at 6 and 12 h. By 24 h p53 expression decreased, and by 48 h p53 expression almost returned to control levels.
To verify the expression of p53, LNCaP cells were treated with either solvent control or BEL and p53 expression was assessed using fluorescence microscopy (Fig. 6, D and E). Little, if no, p53 staining was detected in control cells (Fig. 6D) or cells incubated with the IgG control (data not shown). In contrast, treatment of cells with 5.0 μM BEL for 16 h resulted in clear punctate staining, which was mainly perinuclear as determined by the comparison with the nuclear dye DAPI (Fig. 6E). Collectively, these data demonstrate that inhibition of iPLA2 using BEL increases the expression of p53 and p21 in LNCaP cells.
Effect of BEL on Mdm2 and Akt Kinase Phosphorylation in LNCaP and PC-3 Cells. How iPLA2 inhibition increases p53 expression is not known. p53 can be regulated by Mdm2, which in turn can be regulated by Akt (Ogawara et al., 2002). Treatment of LNCaP cells with BEL decreased Mdm2 expression as early as 30 min after exposure (Fig. 7A). Mdm2 expression was decreased in LNCaP cells at 0.5 and 1 h after exposure to BEL but rebounded after 2 h and continued to increase at later time points. Mdm2 expression was not detected in PC-3 cells at any time point or concentration of BEL tested (data not shown).
Treatment of LNCaP cells with BEL did not decrease Akt levels at any time point measured (Fig. 7B). However, Akt phosphorylation increased after just 30 min of exposure to BEL (Fig. 7B). Akt phosphorylation remained elevated above controls after 4 h of treatment. Treatment of PC-3 cells with BEL also increased Akt phosphorylation (Fig. 7C). Similar to LNCaP cells, increases were detected as early as 30 min after exposure and remained elevated up to 4 h. In both cell types, Akt phosphorylation returned to control levels after 24 h (data not shown).
Effect of BEL on Epidermal Growth Factor-Activated Signaling Pathways in LNCaP and PC-3 Cells. Alterations in p53 and Mdm2 expression can only explain alterations in cell growth in LNCaP cells. Therefore, we sought to identify a signaling pathway common to both LNCaP and PC-3 cells that was altered by BEL. Recent studies suggest that cPLA2 regulates Akt activity by activating epidermal growth factor receptors (EFGR) (Hassan and Carraway, 2006). Furthermore, EGFR is expressed in both LNCaP and PC-3 cells (El Sheikh et al., 2004; Hassan and Carraway, 2006). Thus, we investigated the hypothesis that BEL alters Akt activation through EGFR. To directly study the effect of BEL on EGFR, studies were performed in the absence of serum. EGFR levels and activity in LNCaP cells were lower than those in PC-3 cells under basal conditions (Fig. 8). Exposure of cells to epidermal growth factor (EGF) for 5 min induced time- and concentration-dependent increases in the phosphorylation of EGF, compared with control cells (data not shown). Decreases in the level of EGFR itself are due to the inability of the antibody to recognize phosphorylated EGFR. Maximal activation of EGF was accomplished with doses of EGF of 100 ng/ml (Fig. 8). Treatment with BEL (5 and 10 μM) before exposure to EGF reduced EGFR phosphorylation in both cell types compared with cell exposed to EGF alone.
Treatment of cells with EGF for 5 min increased the phosphorylation of Akt compared with control cells (Fig. 8). Increases in Akt phosphorylation were not accompanied by changes in the level of Akt itself. Similar to EGFR phosphorylation, treatment of cells with BEL (5 and 10 μM) decreased Akt phosphorylation compared with cells treated with EGF alone. Changes in either EGFR or Akt phosphorylation were not due to differences in loading, as demonstrated by the level of GAPDH expression. Collectively, these data suggest that inhibition of iPLA2 decreases EGFR activation, which may decrease cell growth.
To test the role of PI3-kinase in BEL-induced Akt phosphorylation, we pretreated cells with LY294002, a PI3-kinase inhibitor demonstrated to decrease Akt phosphorylation in response to numerous stimuli (Ogawara et al., 2002; Pommery and Henichart, 2005; Liu et al., 2006). Exposure of serum-exposed cells to LY294002 for 30 min before exposure to BEL decreased Akt phosphorylation in both LNCaP and PC-3 cells (Fig. 8, C and D). Similar results were seen when cells were exposed to EGF alone (data not shown). These data suggest that BEL induces Akt phosphorylation via mechanisms that include PI3-kinase.
Discussion
Inhibition of iPLA2 decreases proliferation in several types of cancer and noncancerous cells (Saavedra et al., 2006; Zhang et al., 2006). Decreased INS-1 cell growth correlated to p53-mediated G1 arrest (Zhang et al., 2006). In contrast, decreased ovarian cancer cell growth correlated to G2/M arrest and is p53-independent. Although these studies suggest that the effect of iPLA2 inhibition on cell growth is cell-dependent, they do not identify the roles of individual iPLA2 isoforms in cell growth or identify the mechanisms involved in p53 induction.
We studied the effect of iPLA2 inhibition in PC-3 and LNCaP cells lines, which represent two of the most common used cell models for studying prostate cancer. Furthermore, LNCaP cells are androgen-sensitive and express wild-type p53, whereas PC-3 cells are androgen-insensitive and p53-null. We demonstrated that LNCaP and PC-3 cells expressed iPLA2β and γ based on immunoblot analysis and activity assays. The expression of iPLA2β was higher than iPLA2γ in both cell lines, although direct comparisons are difficult due to differences in antibody specificities. Regardless, these data demonstrate the novel finding that iPLA2β and γ are both expressed in prostate cancer cells.
We previously demonstrated that BEL inhibits iPLA2 activity with an IC50 of 1 to 2.5 μM (Kinsey et al., 2005; Saavedra et al., 2006). Furthermore, we and others (Jenkins et al., 2002; Kinsey et al., 2005; Saavedra et al., 2006) demonstrated the selectivity of R-BEL for iPLA2γ and S-BEL for iPLA2β. We report similar findings in LNCaP and PC-3 cells. Its unlikely that the effects of BEL on activity are due to inhibition of phosphatidic acid phosphohydrolase-1, which is only inhibited 50% by 25 μM BEL (Balsinde and Dennis, 1996), a 10-fold higher concentration than the one used in this study. Furthermore, its unlikely that decreased activities are a result of inhibition of cPLA2γ (Group IVC), as the concentration of BEL used does not significantly decrease cPLA2γ activity (Stewart et al., 2002). Therefore, data presented above support the conclusion that PC-3 cells contain both cytosolic and microsomal iPLA2 activities, which are inhibited by BEL.
Racemic-BEL decreased MTT staining and cell number to a greater extent than S-BEL or R-BEL. These results differ from our previous studies that showed that S-BEL and racemic BEL equally inhibited HEK293 cell growth (Saavedra et al., 2006). These differences are not due to differences in specificity of these compounds for iPLA2 as demonstrated by data in Fig. 1 and by our previous studies (Kinsey et al., 2005). These differences may result from higher levels of iPLA2 activity in PC-3 cells compared with HEK293 cells. Regardless, these data suggest that both iPLA2β and γ have roles in prostate cancer cell growth. This conclusion resulted in our choice to not focus on either iPLA2β or γ using molecular techniques. Thus, using racemic BEL allowed for assessment of total iPLA2 activity on prostate cancer cell growth.
The ability of BEL to induce G1-phase arrest in LNCaP cells agrees with data from INS-1 cells (Zhang et al., 2006) where 10 to 15% increases in cells in G1 were reported. G1 arrest and p53 induction in LNCaP cells occurred in the absence of DNA hypoploidy (data not shown), alterations in nuclear morphology, or increases in annexin V or PI staining. Thus, we do not believe that G1 arrest is a result of DNA damage or cell death. Most likely, BEL-induced G1 arrest is mediated by p53, which is activating p21. Checkpoints exist within the G1 phase of the cell cycle to ensure that phospholipid levels and metabolism are adequate to ensure cell entry into S phase (Manguikian and Barbour, 2004; Zhang et al., 2006). It is possible that inhibition of iPLA2 decreases phospholipid levels and activates these checkpoints.
Similar to ovarian cancer cells (Song et al., 2007) iPLA2 inhibition in PC-3 cells induced G2/M- and S-phase arrests independently of p53. In contrast to ovarian cancer cells, iPLA2 inhibition did not induce death at low concentrations, even in the absence of serum. Increases in cell death in the absence of serum suggest that BEL-induced cell death is serum-dependent. This may explain conflicting data concerning cell death in several studies (Fuentes et al., 2003; Zhang et al., 2006; Song et al., 2007). It is also possible that differences in iPLA2 activities account for such differences.
Data reported above and in other studies (Bao et al., 2006; Zhang et al., 2006) suggest that at least two different pathways mediate decreased prostate cancer cell growth during iPLA2 inhibition. One pathway is p53-dependent and results in G1 arrest. The other pathway is p53-independent and results in G2/M- and S-phase arrest. The former pathway may be mediated by Mdm2. Another pathway exists that involves decreased EGFR activation, which appears to be p53-independent as BEL inhibited EGFR activation in both LNCaP and PC-3 cells. To our knowledge, this is the first report that iPLA2 inhibitors decrease Mdm2 expression and EGFR activation in prostate cancer cells.
The kinetics of Mdm2 expression may appear a bit surprising as Mdm2 expression appears to increase in tandem with p53; however, Mdm2 levels actually decrease after 30 min before increases in p53. Mdm2 expression does increase at later time points but never reach control levels. This increase may be mediated by p53 itself, which can induce Mdm2 (Hamstra et al., 2006; Lu et al., 2007), or by BEL-induced Akt phosphorylation. Increases in Akt and Mdm2 at later time points may mediate decreases in p53 at 24 and 48 h.
Exactly how inhibitors of iPLA2 decrease Mdm2 expression is not known. It is unlikely that Mdm2 expression is mediated by Akt because BEL treatment actually increased Akt activity. Mdm2 expression is decreased during DNA damage (Lozano and Montes de Oca Luna, 1998), but the absence of DNA damage in either cell type suggests that this mechanism is not involved. Mdm2 is also regulated by the phosphatase and tensin homolog (Zhou et al., 2003), glycogen synthase kinase (Rudelius et al., 2006), and ADP-ribosylation factor (Zhang et al., 1998). Future studies are needed to determine whether iPLA2 inhibition alters Mdm2 by these proteins.
Decreases in Mdm2 and increases in p53 only explain alterations in growth and cell cycle in LNCaP cells. Another mechanism must exist in PC-3 cells. This mechanism results in G2/M- and S-phase arrests, as opposed to G1 arrest. G2/M- and S-phase arrests can be induced by cyclin G1, B1, and E (Kimura et al., 2001; Liu et al., 2007; Song et al., 2007). Song et al. (2007) demonstrated that inhibition of iPLA2 increases the expression of cyclin B and E1. Thus, future studies will determine the effect of iPLA2 inhibition on these cyclins.
These data suggests the novel finding that iPLA2 inhibition decreases agonist-stimulated EGFR activation. This event occurred in both cell types, suggesting that the effect of BEL on EGFR activation is p53-independent. Roles for EGFR and cPLA2 in cell growth have been suggested previously (Hassan and Carraway, 2006; Xu et al., 2006), and arachidonic acid mediates EGFR activation in human cholangiocarcinoma cells (Xu et al., 2006). However, the addition of arachidonic acid to both LNCaP and PC-3 cells induced cell death at levels as low as 10 μM (data not shown), making it difficult to study this possibility. EGFR activation and cell growth are also mediated by G-protein-coupled receptors (Hernández et al., 2000; El Sheikh et al., 2004), and recent reports (McHowat et al., 2001) suggest that iPLA2 activity is mediated by G-protein-coupled receptors. This suggests that cross-talk may occur between iPLA2, G-protein-coupled receptors, and EGFR, and that this cross-talk may mediate cell growth.
The ability of BEL to alter Akt activity was EGF-dependent. In the absence of EGF, BEL induced Akt activation; however, in the presence of EGF, BEL decreased Akt activation. The ability of LY294002 to inhibit BEL-induced Akt phosphorylation suggests a role for PI3-kinase in these processes. EGF induces Akt activation by activating EGFR, which activates PI3-kinase (Pommery and Henichart, 2005). In EGF-treated cells, we believe that BEL reduces Akt phosphorylation by directly inhibiting EGFR transactivation and not by interacting with PI3-kinase activity. In contrast, in the absence of EGF, BEL may induce a survival response, independent of EGFR, which activates PI3-kinase. The mechanism by which PI3-kinase is activated by BEL in the absence of EGF is under study.
In conclusion, data in this article demonstrated the novel finding that inhibition of iPLA2 decreased prostate cancer cell growth. There are at least two mechanism involved. One is p53-dependent and correlates to G1-cycle arrest, and the other is p53-independent and correlates to G2/M-phase arrest. The expression of p53 in LNCaP cells correlated to decreases in the expression of Mdm2, suggesting that iPLA2 inhibitors regulate this protein. These data also suggest roles for iPLA2 in Akt and EGFR signaling, which appear to be mediated, in part, by PI3-kinase. Collectively, these data support the hypothesis that iPLA2 has novel roles in prostate cancer cell signaling and may represent a novel chemotherapeutic target for treatment of this disease.
Acknowledgments
We thank Drs. Mchowat and Rick Schnellmann for the gift of the antipeptide iPLA2γ antibody. We also thank Dr. Robert D. Arnold for use of Nikon AZ100 fluorescence microscope. Finally, we thank Dr. Jim Franklin for prereview and excellent suggestions.
Footnotes
-
This work was supported by a Georgia Cancer Coalition Distinguished Scholar Grant and a University of Georgia Junior Faculty Grant (to B.S.C.).
-
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
-
doi:10.1124/jpet.108.138958.
-
ABBREVIATIONS: PLA2, phospholipase A2; sPLA2, secretory PLA2; iPLA2, Ca2+-independent phospholipase A2; iPLA2β, Group VIA Ca2+-independent phospholipase A2; iPLA2γ, Group VIB Ca2+-independent phospholipase A2; BEL, bromoenol lactone, (E)-6-(bromoethylene)-3-(1-naphthaleny)-2H-tetrahydropyran-2-one; EGF, epidermal growth factor; EGFR, epidermal growth factor receptor; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; PBS, phosphate-buffered saline; FITC, fluorescein isothiocyanate; PI3-kinase, phosphoinositide 3-kinase; MTT, 3-(4, dimethylthiazolyl-2)-2, 5-diphenyltetrazolium bromide; PI, propidium iodide; LY294002, 2-(4-morpholinyl)-8-phenyl-4H-1-benzopyran-4-one; DAPI, 4′,6-diamidino-2-phenylindole; INS, insulinoma; cPLA2, calcium-dependent cytosolic PLA2.
- Received March 10, 2008.
- Revision received April 24, 2008.
- The American Society for Pharmacology and Experimental Therapeutics