Neutral Endopeptidase Promotes Phorbol Ester-induced Apoptosis in Prostate Cancer Cells by Inhibiting Neuropeptide-induced Protein Kinase C δ Degradation¹

TPA³exerts numerous effects on cells, including proliferation, malignant transformation, differentiation, and cell death (1, 2). These effects are mediated in part through modulation of PKC isoenzymes. Numerous investigators have shown that phorbol esters can induce apoptotic cell death in androgen-sensitive PC LNCaP cells but not in androgen-independent PC-3 or DU145 cells(3, 4, 5). Powell et al. (6) reported that TPA-induced cell death in LNCaP cells correlated with increased expression of PKCα mRNA, but not other PKC isoforms, and with translocation of PKCα to non-nuclear membranes. However, Fujii et al. (7) recently reported that PKCδmediates phorbol ester-mediated apoptosis in LNCaP cells, demonstrating that phorbol ester-induced cell death can be partially (∼50%)blocked by a PKCδ-inhibitor or a dominant-negative PKCδ mutant. The biological effect of PKCδ is cell-type specific, and overexpression can both inhibit cell growth (8) or enhance anchorage-independent growth and metastatic potential (9). PKCδ activity appears to be regulated in part through tyrosine phosphorylation by Src kinase, which results in degradation of PKCδprotein (10).

LNCaP cells express neutral endopeptidase 24.11 (NEP, CD10, CALLA, EC 3.4.24.11), a M_r90,000–110,000 zinc-dependent cell-surface metallopeptidase,whereas androgen-independent PC cell lines do not (11). NEP can regulate through its enzymatic function access of neuropeptides such as bombesin, neurotensin, and ET-1 to their cell-surface G-protein-coupled receptors. Neuropeptide signaling involves activation of cSrc kinase activity, which in turn leads to phosphorylation of several downstream substrates such as focal adhesion kinase and p130Cas(12, 13, 14). Mari et al. (15)reported previously that a catalytically active NEP protein is required for phorbol ester-induced growth arrest in Jurkatt T cells. In the present study, we considered whether NEP and its substrates were involved in regulating the expression of PKCδ in LNCaP cells and whether NEP and PKCδ expression were required for TPA-induced apoptosis. We report that ET-1 and bombesin induce PKCδdown-regulation caused by rapid PKCδ degradation in PC cells, which is mediated by cSrc kinase activation, and that PKCδ down-regulation is blocked by NEP in LNCaP cells, which facilitates TPA-induced apoptosis.

Prostate cancer cell lines were maintained in RPMI 1640 media supplemented with 2 mm glutamine, 1% nonessential amino acids, 100 units/ml streptomycin and penicillin, and 10% FCS. WT-5,TN-12, and M-22 were derived and maintained as described previously(14). TPA-resistant, LNCaP-derivative LN10H(6) was kindly provided by Dr. C. T. Powell (Cleveland Clinic Foundation) and maintained in the above media containing 10 nm TPA. Recombinant NEP was obtained from Arris Pharmaceutical Corp. CGS24592, a competitive inhibitor of NEP, was supplied by Novartis Pharmaceuticals. The specific Src inhibitor PP2 and the specific PKC inhibitors Rottlerin and Gö6976 were purchased from Calbiochem-Novabiochem, Ltd. (La Jolla, CA).

PC cells (1 × 10⁴/well) were plated in 12-well tissue culture plates (Falcon Division, Becton Dickinson, Cockeysville, MD). After overnight culture in regular media(LNCaP, TSU-Pr1, DU145, and PC-3) or culture for 48 h in media with or without tetracycline (WT-5, TN-12, and M-22), cells were treated with various reagents for 48 h. Cells were harvested and counted using a Coulter Counter ZM (Coulter Electronics, Hialeah, FL). Each data point represents the average cell number of triplicate samples from a single experiment. Statistical analyses were performed using an unpaired t test. Ps less than 0.005 are reported as <0.005. All growth assays were performed on three separate occasions with similar results.

Early apoptotic cells were detected using the Annexin V apoptosis detection kit (Santa Cruz Biotechnology, Inc., Santa Cruz, CA). Briefly, cells evenly distributed in Lab-Tek chamber slides (Nalge Nunc International, Naperville, IL) were treated with various reagents for 24 h. Cells were washed twice with cold PBS, washed once with 1 × Assay Buffer and with 1 μg of Annexin V-FITC with 500 μl of 1 × Assay Buffer added. Propidium iodide (0.5 μg ) was added to each well for nuclear counterstain. After incubation for 15 min at room temperature in the dark, positively stained cells were enumerated using a fluorescence microscope at ×100–400. Each data point represented the average cell number in six independent microscopic fields of a single experiment. The statistical analysis was performed using an unpaired ttest. Ps <0.05 were regarded as statistically significant. All assays were performed on three separate occasions with similar results.

For cell cycle analysis, cells were fixed in 70% ethanol and stained with 50 μg/ml propidium iodide. Cell cycle progression and apoptosis were analyzed by flow cytometry using a Becton Dickinson fluorescence-activated cell sorting system. Twenty thousand events were recorded for each treatment.

Cells were lysed in 1 ml of RIPA buffer (10 mm Tris-HCl (pH 7.4), 150 mm NaCl, 1% Triton X-100, 5 mm EDTA,1% sodium deoxycholate, 0.1% SDS, 1.2% aprotinin, 5 μmleupeptin, 4 μm antipain, 1 mmphenylmethylsulfonyl fluoride, and 0.1 mmNa₃VO₄), and lysates were separated on an 8% SDS-PAGE, transferred to nitrocellulose, and incubated in 1% BSA for 2 h. Membranes were immunoblotted with anti-NEP (NCL-CD10-270; Vector Laboratories, Inc., Burlingame, CA;1:100), anti-PKCα (1:2000), anti-PKCδ (Santa Cruz Biotechnology,Inc.; 1:1000), or anti-actin (Chemicon International, Inc., Temecula,CA; 1:3000) and detected using enhanced chemiluminescence (Amersham Pharmacia, Piscataway, NJ).

An equal number of cells were cultured in RPMI 1640 lacking methionine for 30 min, in the same media containing 300 μCi/ml[³⁵S]methionine for 1 h, and washed with PBS and then in RPMI supplemented with 10% FCS and 0.15 mg/ml nonradioactive methionine for specific time periods. Cells were lysed in RIPA buffer. For immunoprecipitation, 300-μg lysates were incubated for 1 h with 1 μg of anti-PKCδ antibody and then for 1 h with 40 μl of protein G-Sepharose beads (Amersham Pharmacia)at 4°C. Immunoprecipitates were collected by centrifugation at 12,000 × g for 1 min, washed with RIPA buffer, and resuspended in 2 × Laemmli sample buffer. Samples were resolved on an 8% SDS-PAGE and transferred to nitrocellulose. Autoradiography and immunoblotting were performed using the same membrane. The relative intensity of each band obtained by autoradiography was measured by NIH image.

Cell lysates (500 μg) were incubated with cSrc antibody (2 μg;Santa Cruz Biotechnology, Inc.) and immunoprecipitated as described above. cSrc kinase activity was measured using an Src kinase assay kit(Upstate Biotechnology, Inc.) as per the manufacturer’s recommendation. Briefly, 20 μl of kinase buffer (100 mmTris-HCl (pH 7.2), 125 mm MgCl₂, 25 mm MnCl₂, 2 mm EGTA, 0.25 mm Na₃VO₄, and 2 mm DTT), 0.5 mg/ml of a specific substrate peptide(KVEKIGEGTYGVVYK), and 10 μl of [γ-³²P]ATP diluted to 1 μCi/μl with Mn/ATP mixture (75 mmMnCl₂, 500 μm ATP) were added to 20μl of washed protein G-Sepharose beads. The reaction mixture was incubated for 10 min at 30°C, the reaction was stopped by adding 20μl of 40% trichloroacetic acid, and the phosphorylated substrate was separated from the residual [γ-³²P]ATP using P81 phosphocellulose paper and quantified with a scintillation counter. Each experiment was performed at least three times in triplicate. Results are expressed as fold increases compared with untreated controls.

Western blot analysis revealed high levels of NEP and PKCδ proteins in total cell lysates derived from LNCaP cells but not in lysates derived from androgen-independent TSU-Pr1, DU145, or PC-3 cells (Fig. 1,A, panel 1 and panel 2). In comparison,PKCα protein was expressed at lower levels in LNCaP cells relative to other PC cell lines (Fig. 1 A, panel 3), as reported previously (6). In contrast to parental TPA-sensitive LNCaP cells, NEP and PKCδ proteins could not be detected in a TPA-resistant subclone of LNCaP, LN10H cells(6), whereas PKCα protein was expressed at similar levels.

Incubation of PC cells in media containing FCS plus 10 nmTPA for 48 h induced ∼80% growth inhibition in LNCaP cells compared with untreated control (P < 0.005)but not in TSU-Pr1, DU145, or PC-3 cells (Fig. 1,B). Both PKCα and PKCδ have been implicated in TPA-induced growth inhibition of LNCaP cells (6, 7). We therefore assessed the effects on TPA-induced growth inhibition of the PKC-inhibitors Rottlerin(IC₅₀ = 3–6 μmfor PKCδ, IC₅₀ = 40μ m for PKCα, PKCβI, and PKCγ) at a concentration which selectively inhibits PKCδ, and Gö6976(IC₅₀ = 2–6 nm for PKCα and PKCβI; no inhibition at μm concentrations for PKCδ) as a PKCα inhibitor. As illustrated in Fig. 1 B,TPA-induced growth inhibition of LNCaP cells was reversed by pretreatment with 10 μm Rottlerin 2 h before TPA (P < 0.005) but not by 100 nm Gö6976. No significant effect of either inhibitor was observed in TSU-Pr1, DU145, and PC-3 cells. Taken together, these results show that TPA sensitivity correlates with NEP and PKCδ expression in PC cells and support previous studies which suggest that TPA-induced growth inhibition in LNCaP cells is mediated by PKCδ.

To assess whether NEP is needed for PKCδ expression and TPA sensitivity in PC cells, we cultured LNCaP cells in media containing FCS with the addition of the specific NEP enzyme inhibitor CGS24592 at a concentration of 100 nm, which completely inhibits NEP enzyme activity (14), and found that PKCδ protein levels were significantly less than control-treated LNCaP cells (Fig. 1,C). Next we examined PKCδ protein expression in TSU-Pr1 cells containing a tetracycline-repressed (tet-off) inducible wild-type NEP (WT-5 cells), catalytically inactive NEP (M-22 cells), which contain a point mutation in the zinc-binding domain required for NEP enzymatic function (14), and control (empty vector; TN-12 cells). Western blot analysis (Fig. 1,D, panel 1)and enzyme studies confirmed NEP protein expression in both WT-5 and M-22 cells, but not in control TN-12 cells, 48 h after withdrawal of tetracycline from the media, whereas high levels of NEP-specific activity could be detected only in total cell lysates from WT-5 cells (not shown; see Ref. 14). High levels of PKCδ protein expression were present in cells expressing wild-type NEP proteins (WT-5), whereas barely detectable PKCδ protein was detected in cells which did not express NEP (tet-repressed WT5, TN-12)or which expressed catalytically inactive NEP proteins (M-22; Fig. 1,D, panel 2). In contrast, PKCα expression was not affected by NEP expression or tetracycline in these cell lines(Fig. 1,D, panel 3). Cell growth assays showed that culturing in media containing FCS with 10 nmTPA for 48 h after expression of wild-type cell-surface NEP resulted in a >60% decrease in cell number in WT-5 cells(P < 0.005) but did not alter cell growth in TN-12 cells or in M-22 cells expressing catalytically inactive NEP(Fig. 1,E). Similar to experiments using LNCaP cells,pretreatment with 10 μm Rottlerin 2 h before 10 nm TPA treatment in NEP-expressing WT-5 cells reversed TPA-induced growth inhibition in TPA-treated WT-5 cells(P < 0.005; Fig. 1 E). Rottlerin alone had no significant effect on these cells (data not shown). Taken together, these results suggest that the expression of catalytically active NEP protein is required for increase in PKCδ protein expression in PC cells, and that PKCδ mediates susceptibility of PC cells to TPA-induced growth inhibition.

Previous studies show that TPA induces apoptosis in LNCaP cells(7). Cell cycle analysis showed that 10 nm TPA treatment of LNCaP cells for 24 h resulted in 35.1% of cells with sub-G₀-G₁ DNA content (Fig. 2,A, lower left), which was blocked by incubation with 100 nm CGS24592 16 h before TPA treatment (Fig. 2,A, lower right). As illustrated in Fig. 2,B, determination of Annexin V-FITC staining as a marker of apoptosis in LNCaP cells revealed that TPA treatment for 24 h resulted in 53.7% (Lane 2; range 46.7–62.7%)stained cells compared with untreated control (Lane 1;8.5%, range 6.2–10.8%; P < 0.005). Pretreatment for 16 h with 100 nm CGS24592 resulted in only 12% of cells staining positive with Annexin V-FITC(Lane 4; P < 0.005, compared with Lane 2). Pretreatment for 2 h with 10μ m Rottlerin also blocked TPA-induced apoptosis (Lane 6, 14.8%, P < 0.005, compared with Lane 2). Annexin V-FITC-positive staining cells did not increase after TPA treatment in TPA-resistant LN10H cells, which do not express NEP or PKCδ protein(Lane 7). Similar results to LNCaP cells were observed in WT-5 cells expressing NEP (following tetracycline removal; Fig. 2 C). These results suggest that NEP enzyme activity is necessary for TPA-induced apoptotic cell death in LNCaP cells.

PKCδ protein is highly expressed in PC cells expressing NEP. To assess the mechanisms by which NEP may regulate PKCδ expression, we first determined whether NEP increased protein production or prolonged turnover of PKCδ protein. As shown in Fig. 3,A (upper panel), the level of PKCδ protein production in TSU-Pr1 cells at time 0 was similar to that in LNCaP cells, but rapidly decreased within 2 h. Pulse-chase assays revealed that the half-life of PKCδ protein was longer in NEP-positive LNCaP cells (T_1/2 = 6.9 h) compared with NEP-negative TSU-Pr1 cells(T_1/2 = 1.3 h). Western blot analysis of immunoprecipitated PKCδ protein showed a marked decrease in total PKCδ protein expression in TSU-Pr1 cells compared with LNCaP cells (Fig. 3,A, lower panel). Similar results were obtained in WT-5 cells incubated with or without tetracycline for 48 h. Pulse-chase assays showed that the half-life of PKCδ protein was prolonged in NEP-expressing WT-5 cells(no tetracycline, T_1/2 = 4.5 h) compared with NEP-negative WT-5 cells (with tetracycline, T_1/2 =1.4 h; Fig. 3,B, upper panel). Western blot analysis of immunoprecipitated PKCδ protein showed a marked increase in PKCδ protein in NEP-expressing WT-5 cells compared with NEP-negative WT-5 cells (Fig. 3 B, lower panel).

The effect of NEP expression in WT-5 cells on PKCδ protein levels occurs within 2 h, suggesting that NEP stabilizes PKCδprotein in PC cells, possibly by blocking PKCδ degradation (Fig. 3,B, bottom panel). Furthermore, as shown above,NEP catalytic activity is critical to the interaction between NEP and PKCδ proteins. NEP neuropeptide substrates such as ET-1 and bombesin can contribute to paracrine/autocrine PC cell growth and cell survival by activating various signal transduction pathways (16, 17). We therefore assessed whether these neuropeptides affected PKCδ protein expression. Pulse-chase assays revealed that incubation of TSU-Pr1 cells with 10 nm ET-1 in serum-free media resulted in a decrease in the half-life of PKCδ from 5.7 h to 1.2 h (Fig. 3,C, upper panel). In addition, Western blot of immunoprecipitated PKCδ protein showed that PKCδ protein levels decreased by 6 h after ET-1 treatment (Fig. 3,C, lower panel). Furthermore, as shown in Fig. 3,D, Western blot analysis showed that incubation of TSU-Pr1 cells in serum-free media containing either 10% FCS (Lane 3), 10 nm ET-1 (Lane 5), or 10 nm bombesin (Lane 7) for 16 h resulted in significantly lower levels of PKCδ protein compared with control. Pretreatment with 50 μg/ml recombinant NEP for 2 h before the addition of FCS, ET-1, or bombesin inhibited the neuropeptide-induced down-regulation of PKCδ protein (Fig. 3,D, Lanes 4, 6, and 8). These incubations had no effect on PKCα protein expression (Fig. 3 D). Similarly, in LNCaP cells, pretreatment with 100 nm CGS24592 for 2 h before treatment with 10 nm ET-1 resulted in decreased PKCδ expression,whereas ET-1 alone did not alter PKCδ expression (data not shown). Taken together, these results suggest that NEP neuropeptide substrates such as ET-1 and bombesin stimulate PKCδ degradation, and that this effect is inhibited by NEP.

Recent studies show that cSrc kinase activation regulates PKCδtyrosine phosphorylation leading to a decrease in PKCδ protein expression (10, 18). Furthermore, neuropeptides can activate cSrc kinase activity, which contributes to neuropeptides-mediated action (12). We therefore considered that neuropeptides stimulate cSrc kinase activity, which in turn phosphorylates PKCδ, inducing rapid PKCδ protein degradation and inactivation. As shown in Fig. 4,A, cSrc kinase activity of TSU-Pr1 cells incubated with 10%FCS for 16 h was 6.1-fold higher than that of LNCaP cells. Incubation of TSU-Pr1 cells in serum-free media containing 10% FCS(Fig. 4,B, Lane 3), 10 nmET-1 (Fig. 4,B, Lane 5), or 10 nm bombesin (Fig. 4,B, Lane 7) for 20 min resulted in 4.4, 6.9, or 5.6-fold increase in cSrc kinase activity, respectively, compared with serum-free control (Fig. 4,B, Lane 1; P < 0.005). Pretreatment with 50 μg/ml recombinant NEP for 2 h partially blocked cSrc kinase activation (Lanes 4, 6, and 8). In addition, the expression of wild-type cell-surface NEP in WT-5 cells for 48 h resulted in a 75% decrease in the cSrc kinase activity compared with untreated control, but did not alter cSrc kinase activity in control TN-12 cells(data not shown). To confirm the requirement of cSrc activity in neuropeptide-mediated down-regulation of PKCδ expression, we incubated TSU-Pr1 cells with the specific Src kinase inhibitor PP2 (10μ m) for 2 h before treatment with ET-1 or 10% FCS for 16 h, and we showed that PP2 completely blocked ET-1-induced (Fig. 4,C, Lane 5, compared with Lane 2) or 10% FCS-induced (Fig. 4 C, Lane 6, compared with Lane 3) down-regulation of PKCδprotein expression.

Finally, to confirm that the inhibition of cSrc kinase activity is required for TPA-induced apoptotic cell death, we incubated TSU-Pr1 cells with TPA, PP2, plus Rottlerin and performed Annexin V detection. As illustrated in Fig. 4 D, whereas 10 nm TPA alone (Lane 2) did not induce apoptosis, pretreatment with 10 μm PP2 resulted in 64.3% (range 57.5–71.1%) positively stained cells (Lane 4, P < 0.005; compared with Lane 2). Moreover, pretreatment with the PKCδ inhibitor Rottlerin (10μ m) before TPA plus PP2 (Lane 6)partially reversed the TPA-plus-PP2 induction of positive-staining cells (20.3%, P < 0.005; compared with Lane 4). These data suggest that, in TSU-Pr1 cells,inhibition of cSrc kinase activity permits TPA-induced apoptotic cell death and that PKCδ activation is needed for apoptosis to occur.

The results presented here help clarify previous reports on phorbol ester-induced cell death in androgen-sensitive LNCaP cells but not in androgen-independent PC cells. Our data suggest that neuropeptides such as ET-1 and bombesin stimulate cSrc kinase activity,which in turn phosphorylates PKCδ and leads to rapid degradation of PKCδ protein. TPA-induced apoptosis is mediated through PKCδ. Thus, androgen-independent PC cells, which do not express NEP,are resistant to TPA treatment because they express low levels of PKCδ protein. In contrast, LNCaP cells constitutively express NEP. NEP inactivates through hydrolysis neuropeptides such as ET-1 and bombesin, leading to diminished cSrc kinase activity and stable expression of PKCδ protein. Consequently, PKCδ-expressing LNCaP cells are extremely sensitive to TPA-induced apoptosis. Although a previous study reported that NEP is required for TPA-induced growth arrest in Jurkatt T cells (15), the mechanism of NEP in allowing susceptibility to TPA had not been elucidated. Our results highlight the involvement of NEP in TPA-induced apoptosis mediated by PKCδ in PC cells.

Our studies indicate that TPA-induced apoptosis and growth inhibition in LNCaP cells is predominantly mediated by PKCδ rather than by PKCα. Henttu et al. (5) have reported that calcium-independent PKC isoenzymes such as PKCδ, rather than PKCα, are predominantly activated in TPA-treated LNCaP cells, which supports our studies. Recent reports also implicate PKCδ as a proapoptotic kinase (19, 20, 21). A proteolytic cleavage site for caspase-3 has been identified at the V3 (hinge) region of PKCδwith cleavage resulting in the release of an active M_r 40,000 fragment corresponding to the PKCδ COOH-terminal kinase domain (19). However,similar to a previous report (7), pretreatment with the selective caspase-3 inhibitor DEVD-CHO showed little inhibitory effect on TPA-induced cell death in LNCaP or WT-5 cells (data not shown). These results suggest that the TPA-induced apoptotic pathway mediated by PKCδ in PC cells is independent of caspase-3 activity or caspase-3-mediated PKCδ cleavage, and that PKCδ can act as a primary effector or is involved in other pathways for apoptosis via its allosteric activation (7). Recent studies show that translocation of PKCδ holoenzyme, and not its catalytic fragment,onto mitochondria induces cytochrome c release and apoptosis, and that this translocation precedes the activation of caspases (22, 23). This suggests that proteolytic cleavage may not be required for PKCδ kinase activation and apoptosis induction. Others have suggested that various cell cycle regulators(3) or ceramide (4) mediate TPA-induced apoptosis in LNCaP cells. We have found NEP expression in WT-5 cells up-regulates p21WAF/CIP1 expression and inactivates the retinoblastoma protein, which induces G₀-G₁arrest,⁴leading us to speculate on the possibility that PKCδ stabilized by NEP can affect these cell cycle regulators via its allosteric activation.

NEP neuropeptide substrates such as ET-1 and bombesin can act as survival and antiapoptotic factors (24, 25),transactivators of epidermal growth factor receptor (26),and activators of Akt/protein kinase B cell survival pathway (27, 28). As a regulator of these peptides to their cell surface receptors, NEP is involved in various critical signaling pathways. Our data for the first time define one mechanism of NEP function as an inducer of apoptosis through stabilization of PKCδ expression. PKCδactivity has been implicated in mediating apoptosis in response to various antitumor reagents such as etoposide (20) and cis-platinum (29) as well as TPA. Thus, through its ability to inhibit various cell survival pathways by inactivating mitogenic neuropeptides, NEP may be a potential therapeutic modality to use in combination with various agents to treat prostate cancer.

We thank Dr. Thomas Powell for useful discussions and for supplying the LN10H PC cell line, and Catherine Kearney and Lana Winter for secretarial assistance.