Regulation of Erk1/2 activation by osteopontin in PC3 human prostate cancer cells

We measured the phosphorylation state of the three most widely known members of the mitogen-activated kinase (MAPK) family proteins including Erk1/2, JNK, or p38 MAPK in PC3 cells over expressing OPN (PC3/OPN). Stable PC3/OPN cells were generated as described previously [5]. PC3/OPN stable cell lines display an increased expression of OPN compared with stable PC3 cell lines expressing empty vector (Figure 1A, lane 2). Previous studies have shown that metastatic PC3 and DU145 prostate cancer cells have relatively low levels of active Erk1/2 [23]. Western blot analysis with indicated phosphor-specific antibody was performed. Consistent with those findings, we show here that PC3 cells expressing pCEP4 vector (henceforth represented as PC3) displayed either minimal or barely detectable levels of phosphorylation of Erk 1/2 (Figure 1B, lane 1). The phosphorylation is increased to a greater extent in PC3/OPN cells (Figure 1C, lane 2). An increase in the phosphorylation at Thr 202/204 represents the activation of Erk1/2 in PC3/OPN cells (Figure 1B, lane 2). Confocal analysis of PC3 and PC3/OPN cells stained for phospho-Erk1/2 also revealed a robust and diffuse staining of activated Erk1/2 in PC3/OPN cells (Figure 1E, bottom middle panel). An increased staining substantiates the activation of Erk1/2 in PC3/OPN cells since staining was performed with phosphor-Erk1/2 antibody. PC3 cells show sparse staining of phospho- Erk1/2 (Figure 1E, Top middle panel, green). This is consistent with the immunoblotting analysis shown in Figure 1B which demonstrates a decrease in the phosphorylation and activation of Erk1/2 in PC3 cells. Actin staining was used to demonstrate the cell periphery.

Immunoblotting analyses demonstrated a small increase in the phosphorylation of JNK at Threonine 183 and Tyrosine 185 in PC3/OPN cells (Figure 1C, lane 2). Furthermore, OPN had a very negligible effect on the phosphorylation of p38 MAPK at Thr180/Tyr182 (Figure 1D, lane 2). GAPDH was used as a loading control when probing total OPN expression levels (Figure 1A; lower panel). There were no observed differences in the protein levels of non-phosphorylated MAPK family members in either PC3 or PC3/OPN cell lines (Figure 1B-D; lower panels).

Raf and MEK have been shown to be the upstream regulators of Erk1/2 [17]. In order to determine the role of Raf and MEK1/2 in OPN-mediated activation of Erk1/2, western blot analysis was employed. Structures of the Raf proteins (A-Raf, B-Raf, and c-Raf) have been shown to be similar, but the proteins maintain differences in how they are activated and how they activate downstream targets such as MEK1/2 [17]. Activation of A-Raf and B-Raf is represented by the phosphorylation at Ser 299 and 245, respectively. Activation of c-Raf is measured by phosphorylation at Ser 338 [24]. Phosphorylation of A-Raf was almost not detected in PC3 and PC3/OPN cells (Figure 2A). Conversely, PC3 cells exhibited a higher basal level phosphorylation of B-Raf at Ser445 in PC3 cells (lane 1) and OPN expression had no effect in increasing the phosphorylation state of B-Raf (Figure 2B, lane 2). However, activation of c-Raf appears to highly dependent on OPN over-expression (Figure 2C, lane 2). An increase in the phosphorylation of c-Raf at Ser338 suggests that activation of c-Raf may have a role in the OPN-dependent Raf/MEK/ERK pathway and control apoptosis. Therefore we next proceed to investigate the activation of MEK1/2 in response to OPN over-expression. MEK1/2 activation is characterized by phosphorylation at two activation loop residues, Ser 217 and Ser 221. We found an increase in the activation of MEK1/2 in PC3/OPN cells as compared to PC3 control cells (Figure 2D).

Recent observations have demonstrated an increase in the activation of Akt in PC3/OPN cells [6, 25]. Little is known about the role of Akt in the Erk pathway in PC3 cells. Therefore, we have investigated the effects of Akt inhibitor on the phosphorylation of c-Raf and ERK1/2 on Thr202/204. OPN expression in PC3 cells increased Akt activation, as measured the phosphorylation of ser473 (Figure 3A, lane 2). Serine 259 of c-Raf has been shown to be regulated by Akt. Its phosphorylation provides a docking site for the cytosolic protein 14-3-3 and the subsequent inhibition of c-Raf activation [17]. OPN, presumably through Akt induces the phosphorylation of c-Raf at ser259 (Figure 3B). PC3 cells treated with Akt inhibitor showed an almost undetectable amount of c-Raf phosphorylation at ser259 (Figure 3C, lane 2) when compared with vehicle treated PC3 cells (Figure 3C, lane 1). In order to more fully understand the role of OPN in c-Raf activation and its association with Akt, the activation of Erk1/2 and c-Raf was studied in the presence of Akt inhibitor (Figure 3D and 3E). In the presence of an Akt inhibitor, PC3/OPN cells displayed a further increase in phosphorylation of c-Raf at Ser338 (lane 4 in Figure 1D) and Erk1/2 at Thr202/204 (Figure 3E, lane 4) as measured by immunoblotting analyses with respective phospho-specific antibody. These results indicate that while OPN ultimately activates c-Raf and Erk1/2, its activation of Akt plays an inhibitory role through the increased phosphorylation of c-Raf Serine 259, a known docking site for 14-3-3 protein.

Integrin αvβ3 and CD44 are receptors of osteopontin and CD44 is frequently over expressed in cancer cells [13]. To assess whether both the CD44 and αVβ3 receptors have a role in OPN- mediated Akt activation, we used a specific inhibitor (cyclo RGD; Figure 4A and 4D) to the αVβ3 integrin and siRNA to CD44 (Figure 4B and 4C). PC3 cells over expressing OPN with a mutation in the integrin binding domain RGDΔRGA (referred to as PC3/RGA) and thus no longer able to activate integrins were used to further define the individual roles of αVβ3 integrin and CD44 in the activation of Akt. The expression levels OPN and OPN (RGA) in these cell lines were shown previously. We do not see any differences in the molecular mass of cellular or secreted OPN in PC3, PC3/OPN or PC3/OPN (RGA) cells. The molecular mass of native OPN protein is approximately 30-36 kDa. These cells express ~60-68 kDa OPN protein which indicates that OPN is glycosylated [5]. PC3/OPN and PC3/RGA cells increase Akt activation (Figure 4A and 4C; lanes 2 and 3) when compared with PC3 cells, suggesting that OPN can induce activation of Akt in the absence of integrin signaling (Figure 4A and 4C, lane 1). In the presence of the αV inhibitor, PC3/OPN cells no longer have the ability to induce activation of Akt (Figure 4A, lane 6), while expression of mutant OPN in PC3 cells (PC3/RGA) did not affect the phosphorylation of Akt (Figure 4A, lane 5). The ability of PC3/RGA cells to activate Akt in the presence of the αV inhibitor suggests a role for an additional receptor.

CD44 is another receptor for OPN [13] and previous work from our laboratory showed that CD44 has an important role in the activation of MMP-9 and migration of PC3 cells [5]. Therefore, we sought to determine the role of CD44 in the activation of Akt using CD44 knock-down strategy with SiRNA to standard (s)CD44 (Figure 4B-E). We arrived at about 75-85% knockdown of sCD44 when using SiRNA to sCD44 (Figure 4B, lane 2). Scrambled RNAi was used as a control (lane 1). Mutation in OPN (RGA) abolishes Akt activation only in the cells depleted of CD44 (Figure 4C, lane 5) while PC3/OPN cells retain the ability to induce Akt activation, presumably through the interaction of αVβ3 and OPN via RGD-sequence (lane 6). However, cells treated with SiRNA to CD44 and an inhibitor to αv demonstrated a considerable decrease of both CD44 and αVβ3 integrin-mediated Akt activation (Figure 4E, lane 6). A graphical representation of changes in AKT phosphorylation (i.e. activation) is provided (Figure 4E) for the Western blot shown in Figure 4D. Cells treated with both αv-inhibitor and SiRNA to CD44 was normalized to the corresponding control cells untreated with αv-inhibitor but treated with scrambled RNAi (Figure 4E). These experiments illustrate that the interaction between OPN and either CD44 or integrin is sufficient to induce phosphorylation of Akt, which is largely responsible for the anti-apoptotic mechanisms vital to cancer cell survival and progression.

Monoclonal rabbit anti-phospho-p44/42MAPK (Erk1/2) (Thr202/Tyr204), anti-phospho-SAPK/JNK (Thr183/Tyr185), anti-phospho-c-Raf (Ser338), anti-p44/42MAPK (Erk1/2), anti-B-Raf, polyclonal rabbit anti-phospho-p38MAPK (Thr180/Tyr182), anti-phospho-c-Raf (Ser259), anti-phospho-c-Raf (Ser289/296/301), anti-phospho-A-Raf (Ser299), anti-phospho-B-Raf (Ser445), anti-p38MAPK, anti-SAPK/JNK, anti-A-Raf, and anti-c-Raf were purchased from Cell Signaling Technology (Danvers, MA). GAPDH and CD44 antibodies were purchased from Santa Cruz Biotechnology Inc (Santa Cruz, CA). OPN antibody was purchased from Rockland Immunochemicals (Gilbertsville, PA). Roswell Park Memorial Institute-1640 (RPMI-1640) media, fetal bovine serum (FBS), penicillin-streptomycin (10,000 units/ml penicillin and 10,000 μg/ml streptomycin), 0.25% Trypsin-EDTA, and phosphate buffered saline (PBS) pH 7.4 were purchased from Invitrogen (GIBCO; Auckland, NZ). Akt (5 μM) inhibitor, rhodamine phalloidin, and other chemicals were purchased from Sigma-Aldrich (St. Louis, MO). Protein assay reagent kit, reagents for polyacrylamide gel electrophoresis (PAGE), and molecular weight standards were purchased from Bio-Rad (Hercules, CA). Polyvinyldifluoride (PVDF) membrane for immunoblotting analysis was obtained from Millipore Corp. (Bedford, MA).

Stable prostate cancer cell (PC3) lines that either over express unmutated OPN (PC3/OPN) or a mutant OPN in the integrin-binding site (PC3/RGDΔRGA) was generated as described previously [5]. PC3 cells transfected with empty pCEP4 vector were used as control. Cells were cultured at 37°C in RPMI-1640 media containing 10% Fetal Bovine Serum (FBS) and 1% Penicillin-Streptomycin. Upon reaching 100% confluency, cells were passaged with two brief phosphate buffered saline (PBS) washes, removed from tissue culture plates using 0.25% Trypsin-EDTA, and transferred to larger dishes.

Cells were washed two times with cold PBS and lysed in ice-cold RIPA lysis buffer (10 mM Tris-HCl, pH 7.2, 150 mM NaCl, 1% deoxycholate, 1% Triton X-100, 0.1% SDS, 1% aprotinin, 2 mM phenylmethysulfonyl fluoride, and 0.6 mM sodium orthovanadate). After incubating on ice for 10 min, lysates were centrifuged for 5 min at 6,000 rpm at 4°C. The supernatants were saved and protein concentrations were measured using the Bio-Rad protein assay reagent kit.

PC3 cell lines (PC3, PC3/OPN, and PC3/OPN (RGA) were cultured in a 6-well culture dish and then treated with one of the following inhibitor in the presence of RPMI-1640 media containing 10% FBS at 37°C: Akt inhibitor for 16 h (1:500 dilution of 5 μM stock), αV integrin inhibitor (Cyclo-RGD, 10 μM for 16 h), siRNA to CD44 (200 nM for 48 h). CD44 siRNA (product number sc-29342) and scrambled siRNA (product number sc-37007) nucleotides were purchased from Santa Cruz Biotechnology Inc (Santa Cruz, CA). siRNA transfection reagent, RNAiFect, was purchased from Qiagen (Valencia, CA). Protein lysates were subjected to 12% SDS-PAGE and Western blot analysis as described below.

Cell lysates were denatured by boiling for 5 minutes in Novagen 1× SDS sample buffer (EMD Biosciences, Inc. Madison, WI). Proteins were resolved by SDS-polyacrylamide gel electrophoresis on 8% or 12% gels and then transferred to PVDF membranes. The membranes were initially blocked with PBS containing 0.05% TWEEN 20 (PBS-T) and 5% BSA for 1 h at room temperature (RT) and were then probed overnight at 4°C using a dilution of 1:1000 with the following primary antibodies in PBS-T and 5% BSA: anti-phospho-p44/42MAPK (Erk1/2) (Thr202/Tyr204), anti-phospho-SAPK/JNK (Thr183/Tyr185), anti-phospho-c-Raf (Ser338), anti-phospho-p38MAPK (Thr180/Tyr182), anti-phospho-c-Raf (Ser259), anti-phospho-c-Raf (Ser289/296/301), anti-phospho-A-Raf (Ser299), and anti-phospho-b-Raf (Ser445). After three washes with PBS-T for 5 minutes each, the membranes were incubated with a 1:1000 dilution of species-specific horseradish peroxidase (HRP)- linked secondary antibody (Little Chalfont, UK) in PBS-T and 5% blotting grade blocker non-fat dry milk (Bio-Rad, Hercules, CA) for 2 h at RT. Blots were washed three times with PBS-T for 15 min. each. Protein bands were visualized by chemiluminescence using a SuperSignal West Pico Chemiluminescent Substrate Kit (Thermoscientific, Rockford, IL). PVDF membranes were stored in PBS-T at 4°C until being stripped and re-probed with the corresponding control antibodies to determine the loading in each lane as described below.

The PVDF membranes were incubated in stripping buffer (2% sodium dodecyl sulfate (SDS), 62.5 mM Tris HCl pH 7.2, and 100 mM β-mercaptoethanol) at 55°C for 15 min. After three washes with PBS-T for 15 minutes each, the membranes were blocked with PBS-T and 5% blotting grade blocker non-fat dry milk (Bio-Rad, Hercules, CA) for 1 h at room temperature (RT) and were then probed overnight at 4°C using a dilution of 1:1000 of the primary antibody of interest (e.g. anti-GAPDH at 1:1000 dilution) in PBS-T and 5% blotting grade blocker non-fat dry milk. The membranes were washed three times with PBS-T for 5 min each and were then incubated with a 1:1000 dilution of species-specific horse-radish peroxidase (HRP-) linked secondary antibody in PBS-T and 5% blotting grade blocker non-fat dry milk (Bio-Rad, Hercules, CA) for 3 h at RT. Membranes were washed and proteins bands were visualized as described above.

PC3 and PC3/OPN cells were cultured onto cover slips in a 12-well dish for 14-16 h at 37°C. Cells were washed three times with room temperature PBS (RT-PBS) and fixed in 4% formaldehyde-PBS for 10 min. After washing three times with RT-PBS, cells were permeabilized with 0.5% Triton X-PBS for 10 min. Cells were washed three times with RT-PBS, followed by incubation in 5% boiled goat serum for 1 h at RT. After washing three times with RT-PBS, cells were incubated with a 1:100 dilution of anti-phospho-p-44/42(ERK1/2) (Thr202/Tyr204) in 5% boiled goat serum overnight at 4°C. Cells were washed three times with RT-PBS. Subsequently, cells were incubated for 3 h at RT in the dark with the following: 1:1000 dilution of FITC-conjugated species specific secondary antibody and 1:500 dilution of rhodamine phalloidin for actin staining. Cells were washed three times with RT-PBS for 15 minutes each and the cover slips were transferred cell side down onto glass slides containing perma fluor mounting medium (Thermo Scientific, Pittsburgh, PA) and sealed with clear nail polish around the edge of the cover slips. The immunostained cells were viewed and photomicrographed on a Bio-Rad 6000 (Hercules, CA) confocal microscope. Images were stored in TIF image format and processed by the Adobe Photoshop software program (Adobe Systems, Inc., Mountain View, CA).