Abstract
The 38-amino-acid isoform of pituitary adenylate cyclase-activating polypeptide (PACAP38) elicits a robust outgrowth of neurites in cultured PC12 cells. Initiation of neurite outgrowth occurs within 4–8 hr after the addition of PACAP38. Treatment with PACAP38 does not elicit collateral activation of p140trk nerve growth factor receptor tyrosine kinase activity, nor is it associated with tyrosine phosphorylation of suc1-associated neurotrophic factor target, a selective target of neurotrophin tyrosine kinase receptors. Coadministration of epidermal growth factor with PACAP38 elicits an enhanced response. Induction of neurites is also observed on the addition of PACAP38 to dominant negative Src and Ras PC12 cell variants. PACAP38 stimulates extracellular signal-regulated kinase (Erk) activity >10-fold within 5 min, and the effect is augmented by cotreatment with epidermal growth factor. Pretreatment with the cAMP-dependent protein kinase-selective inhibitor, H-89, is ineffective as an antagonist of PACAP38-induced neurite outgrowth, whereas down-regulation of protein kinase C (PKC) by phorbol ester or incubation with PKC-selective inhibitors GF109203X and calphostin C effectively blocks PACAP38-stimulated neurite formation. Stimulation of Erk activity is inhibited by incubation with PD90859, a pharmacological antagonist of the threonine/tyrosine dual-specificity Erk. Inhibition of ligand-stimulated Erk activation prevents PACAP38-induced neurite outgrowth. Collectively, these findings indicate that PACAP38-stimulated neuritogenesis requires PKC and Erk activation but is independent of cAMP-dependent protein kinase, nerve growth factor receptor tyrosine kinase, p21ras G protein, and pp60c-src cytoplasmic tyrosine kinase.
The PC12 cell line, derived from a rat adrenal medullary pheochromocytoma tumor, has been used extensively as a model for investigating biomolecular events involved in neuronal differentiation. In the presence of nanomolar concentrations of the neurotrophin NGF, replicating PC12 cells cease to divide and extend neuritic processes indicative of phenotypic transformation to a neuronal phenotype characteristic of sympathetic neurons (Fujita et al., 1989). The neurodifferentiative effects of NGF in PC12 cells are elicited by interaction of the neurotrophin with two distinct cell surface receptors: p75NGFR, which binds all members of the neurotrophin family, and p140trk, a prototypic receptor tyrosine kinase that selectively binds NGF and is activated by interaction with its cognate ligand (Kaplan and Miller, 1997). Many of the biological activities elicited by NGF are mediated by ligand-dependent activation p140trk tyrosine kinase activity. Binding of NGF to p140trkstimulates rapid tyrosine autophosphorylation of the receptor, as well as activation of several signal transduction cascades, including the Ras-Erk pathway, phospholipase-Cγ1, phosphatidylinositol-3 kinase, and SNT (Kaplan and Miller, 1997). Additionally, a role for pp60c-src in the transduction of NGF biological signals has been proposed (Keegan and Halegoua, 1993).
PC12 cells also express cell surface receptors for various other peptides, including EGF (Lazarovici et al., 1987), a mild mitogen for these cells (Huff et al., 1981), as well as type I and type II receptors for the PACAP38 and PACAP27 (Deutsch and Sun, 1992; Cavallaro et al., 1995). These two isoforms of PACAP represent alternatively processed forms of a precursor protein that share in common 27 amino-terminal amino acids. Both forms exhibit substantial amino acid sequence homology with vasoactive intestinal peptide (Miyata et al., 1990). PACAP type I and II receptors are seven membrane-spanning, G protein-coupled moieties that mediate activation of adenylate cyclase and phospholipase C (Spengler et al., 1993). In PC12 cells, both PACAP38 and PACAP27 activate these signaling pathways but with different efficiencies. PACAP38 is 200-fold more potent than PACAP27 in promoting phospholipase C-mediated inositol phospholipid metabolism (Deutsch and Sun, 1992), whereas both peptides are equally potent with respect to activation of adenylate cyclase and elevation of cAMP (Deutsch and Sun, 1992). The neurite-inducing activity of PACAP38 in PC12 cells is markedly more robust than that observed for PACAP27 (Deutsch and Sun, 1992). The vigorous neuritogenic activity of PACAP38 has been observed in a variety of systems, including numerous clones of PC12 (Deutsch and Sun, 1992; Hernandezet al., 1995; Colbert et al., 1994; Barrieet al., 1997; Lazarovici et al., 1997b), rat chromaffin cells (Wolf and Krieglstein, 1995), and human neuroblastoma cells (Deutsch et al., 1993). The signal transduction pathway or pathways responsible for mediating PACAP-induced neurite outgrowth remain to be definitively identified, and their elucidation could reveal fundamentally novel mechanisms responsible for regulating processes associated with morphological neuronal development. The coexistence of receptors for NGF, EGF, and PACAP on the membrane surface of PC12 cells, coupled with recent progress in detailing the intracellular signaling pathways activated by these receptors in a ligand-dependent manner, demonstrates that the PC12 cell model is well suited for investigating cooperative, synergistic interactions between distinct families of peptide growth factors. Specifically, the issue of signal transduction cascade cross-talk between G protein-linked PACAP receptors and the tyrosine kinase activity of EGF-stimulated growth factor receptors resulting in neurotrophic activity may be investigated.
In the current study, a number of variant PC12 cell lines expressing different levels of p140trk (Chao 1992), dominant-negative Ras (Green et al., 1986), or Src (Hempstead et al., 1992) PC12 transfectant cells and pharmacological inhibitors were used to elucidate the roles of Trk, Ras, Src, Erk, PKA, and PKC in promoting PACAP-stimulated neurite outgrowth. Results obtained indicate that PACAP-induced neuritogenesis in PC12 cells is Trk-, Ras-, Src-, and PKA-independent but PKC- and Erk-dependent.
Materials and Methods
Mouse NGF, EGF, and rat collagen type II were purchased from Collaborative Biomedical (Bedford, MA). Dexamethasone, poly-l-lysine, myelin basic protein, and PMA were obtained from Sigma Chemical (St. Louis, MO). PD98059 was acquired from Research Biochemicals International (Natick, MA). U73122, GF109203X, calphostin C, and staurosporine were purchased from Calbiochem (La Jolla, CA). H-89 [N-[2-((p-bromocinnamyl)amino)ethyl]-5-isoquinolinesulfonamide, dihydrochloride] was obtained from Alexis (San Diego, CA). Rat PACAP38 and the antagonist PACAP(6–38) were acquired from Peninsula Laboratories (Belmont, CA). Monoclonal anti-phosphotyrosine antibody (clone 4G10) was procured from Upstate Biotechnology (Lake Placid, NY). Polyclonal antibodies directed against TrkA (C14), Src, Ras, and Erk were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Monoclonal anti-EGF receptor antibody was obtained from Medical and Biological Laboratories (Nagoya, Japan). Anti-phosphorylated Erk antibody was generously provided by Dr. Eric Schaeffer (Promega, Madison, WI). [3H]Thymidine and [γ-32P]ATP were acquired from Amersham (Arlington Heights, IL). K-252a [8R*,9S*,11S*-(−)-9-hydroxy-9-methoxycarbonyl-8-methyl-2,3,9,10-tetrahydro-8,11-epoxy-1H,8H,11H-2,7β,11α-triazadibenzo(a,g)cycloocta(c,d,e)trindene-1-one] and staurosporine were kindly given by Dr. Y. Matsuda (Kyowa Hakko Koygo, Tokyo, Japan). Sp-cAMP was purchased from Boehringer-Mannheim (Mannheim, Germany).
Cell culture.
PC12 cells were grown in Dulbecco’s modified Eagle’s medium supplemented with 7% fetal bovine serum, 7% horse serum, 100 μg/ml streptomycin, and 100 units/ml penicillin (Life Technologies, Grand Island, NY) (Lazarovici et al., 1987). In experiments involving extended treatment with growth factors, the medium was changed, and test agents replenished factors every 48 hr. PC12nnr cells were grown in collagen-coated tissue culture dishes in RPMI 1640 medium (Life Technologies) supplemented with 10% horse serum and 5% fetal bovine serum (Lazarovici et al., 1997b). PC12–6.24 cells, a clone of PC12 manipulated to overexpress human p140trk, were grown under culture conditions comparable to those for wild-type PC12 cells except for the inclusion of 200 μg/ml G418 (Life Technologies) (Lazarovici et al., 1997b). During treatment with test article, G418 was removed from the medium. PC12 cell variant cell line, GSrasDN6, which expresses a dominant-negative ras gene under the transcriptional control of the mouse mammary tumor virus promoter (Lazarovici et al., 1997b); M-M17–26, expressing the Ha-ras Asn17 gene under the transcriptional control of the mouse metallothionein-I promoter (Lazarovici et al., 1997b); and srcDN2, modified to express the K295R mutant (kinase inactive) form of chicken Src under transcriptional control of the cytomegalovirus promoter (Lazaroviciet al., 1997b), were maintained in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum and 5% horse serum. In experiments involving dexamethasone treatment, cells were switched to media supplemented with charcoal-stripped serum 24 hr before stimulation with 500 nm dexamethasone.
Immunoprecipitation and immunoblotting.
Cells were plated in 10-cm tissue culture dishes 1 day before initiation of experiments. After treatment with selected growth factors, cultures were washed twice with Tris-buffered saline (20 mm Tris·HCl, pH 8.0, 137 mm NaCl) and subjected to lysis in 1 ml of 1% Nonidet P-40 lysis buffer (20 mm Tris·HCl, 137 mmNaCl, 0.5 mm EDTA, 10% glycerol, 1 mmphenylmethylsulfonyl fluoride, 0.15 unit/ml aprotonin, 20 mm leupeptin, 1 mm sodium vanadate) at 4° for 20 min. Insoluble material was removed by centrifugation for 10 min at 12,000 × g. Lysate aliquots were diluted in SDS sample buffer (0.06 mm Tris·HCl, pH 6.8, 12.5% glycerol, 1.25% SDS, 5% β-mercaptoethanol, 0.002% bromphenol blue), boiled for 5 min, and then subsequently electrophoresed through 7.5% polyacrylamide gels or lysates were subjected to immunoprecipitation with anti-TrkA or anti-Erk antibodies for 2 hr at 4° with continuous agitation, followed by an additional 2-hr incubation with Protein A-Sepharose. Precipitates were washed three times with lysis buffer and once with water and then boiled for 5 min in SDS sample buffer and subjected to SDS-PAGE on 7.5% polyacrylamide gels. After SDS-PAGE, resolved proteins were transferred to nitrocellulose. Blots were probed overnight at 4° with the primary antibodies and analyzed using enhanced chemiluminescence systems with horseradish peroxidase-coupled secondary antibodies. After visualization on film, quantitation was performed by densitometry.
Erk kinase assay.
Washed Protein A resin containing immunoprecipitated Erk was suspended in 30 μl of kinase assay buffer [7.5 mm HEPES, pH 7.5, 10 mmMgCl2, 1 mm dithiothreitol, 2.5 μm protein kinase A inhibitor peptide PKI(6–22)-amide, 225 μm cold ATP, 25 μCi of [γ-32P]ATP, 500 μg/ml MBP as the kinase substrate] and incubated for 30 min at room temperature. The reaction was terminated by adding 30 μl of 2× SDS sample buffer. Samples were heated for 5 min and then electrophoresed through a 4–20% Tris-glycine gradient gel (NOVEX). After electrophoretic resolution, radiolabeled proteins were transferred to Immobilon PVDF membranes. Phosphorylated MBP was visualized by exposure of the membranes to XAR film (Eastman Kodak). MBP-associated radioactivity was quantified by liquid scintillation counting of bands excised from the membrane. Dual threonine/tyrosine phosphorylation of activated Erk was demonstrated by Western blotting conducted as specified in a protocol developed by Promega.
PKA assay.
Growth factor-stimulated activation of cAMP-dependent PKA was assessed in PC12 cells using a protein kinase assay system obtained from Life Technologies. The assay was conducted in accordance with the manufacturer’s specifications. In brief, PC12 cells treated with specified test article were washed with ice-cold PBS containing 1 mmNa3VO4 and sonicated in extraction buffer (5 mm EDTA, 59 mm Tris, pH 7.5), and cellular debris was removed by centrifugation in a microfuge at 4°. Aliquots of cellular extract were incubated for 5 min at 30° in an assay buffer composed of 50 mm Tris, pH 7.5, 10 mm MgCl2, 100 μm[γ-32P]ATP, 0.25 mg/ml bovine serum albumin, and 50 μm Kemptide as a substrate for PKA. Reactions were terminated by spotting an aliquot of the reaction mixture onto phosphocellulose discs. The spotted discs were washed twice with 1% (v/v) phosphoric acid followed by two washes with H2O. Washed discs were placed into scintillation vials, and the peptide-incorporated 32P was quantified by liquid scintillation counting.
Neurite outgrowth assay.
To evaluate neurite outgrowth, cells were plated at low density (1000–5000 cells/dish) on collagen- and poly-l-lysine-coated tissue culture dishes in the appropriate medium. Cultures were treated for different periods of time with the indicated reagents, examined by light microscopy using an inverted Nikon-Diaphot microscope, and photographed at 320× magnification.
Results
PACAP38-induced neurite outgrowth in PC12 cells is p140trk independent and is facilitated by coadministration with EGF.
To demonstrate that PACAP38-elicited neuritogenesis does not involve collateral activation of high affinity, p140trk NGF receptors, PC12 cell variants expressing variable levels of neurotrophin receptors were used (Fig.1). Wild-type PC12 cells, which express both low affinity, p75 pan-neurotrophin binding receptors, and p140trk, exhibit neurite outgrowth on treatment with NGF (Chao, 1992; see references). PC12nnr5 cells, a PC12 variant expressing levels of p75 comparable to wild-type cells but undetectable levels of p140trk, do not elaborate neurites in response to NGF administration (Green et al., 1986). Overexpression of p140trk in wild-type PC12 cells results in accelerated neurite outgrowth when NGF is present (Hempsteadet al., 1992). Each of these PC12 cell lines expresses EGF receptors (Lazarovici et al., 1997b). Uniformly, treatment of each PC12 cell line variant with EGF (10 ng/ml) fails to elicit neurite outgrowth under routine culture conditions (data not shown), indicating that stimulation of EGF receptor tyrosine kinase activity in and of itself is insufficient to induce a neuronal morphological phenotype in any of these cell lines.
Treatment of PC12 variant cultures for 48 hr with 5 nmPACAP38 was observed to promote neurite outgrowth irrespective of the level of p140trk receptors expressed (Fig. 1, Table 1). Inhibition of PACAP38-stimulated neurite outgrowth in the variant PC12 cell lines by the antagonist PACAP 6–38 (data not shown) indicates that the effects on PC12 cell morphology are mediated by PACAP-selective receptors. To determine whether coactivation of a non-neurotrophin, tyrosine kinase-initiated signal transduction cascade might amplify PACAP38-induced neurite outgrowth, each of the variant PC12 cell lines was treated with a combination of PACAP38 (5 nm) and 10 ng/ml EGF. The consequence of this coadministration in each of the cell lines tested was a more robust neuritogenic response than noted for PACAP38 alone (Fig. 1). Collateral activation of p140trk was not elicited by PACAP38, EGF, or the combination of PACAP38 and EGF (Fig. 2). Similarly, these non-neurotrophin peptides did not elicit tyrosine phosphorylation of the nuclear protein SNT (Fig. 2), an event reported to be essential for the elaboration of neurites induced by NGF (Rabinet al., 1993). Moreover, PACAP38-induced neurite outgrowth was not affected in any of the PC12 variant cell lines by pretreatment with 250 nm K-252a (data not shown), a highly selective inhibitor of NGF-induced, p140trk tyrosine kinase activity (Koizumi et al., 1988; Berg et al., 1992) and subsequent induction of neurites in PC12 cells. Collectively, these findings demonstrate that PACAP38 does not elicit neurite outgrowth in PC12 cell variants through coactivation of p140trk NGF receptors.
PACAP38-induced neurite outgrowth in PC12 cells is Src and Ras independent
Variant PC12 cell lines were used to determine whether Src tyrosine kinase activity or Ras activation is involved in PACAP38-stimulated neurite outgrowth. Incubation of SrcDN2 variant PC12 cells with dexamethasone results in glucocorticoid-induced overexpression of a kinase-inactive dominant negative Src (Fig.3). Similarly, the addition of dexamethasone to cultures of GSrasDN6 PC12 variants results in transient overexpression of a dominant negative mutant Ras (Fig. 3). The PC12 variant line M-M17–26, in which the Ras mutation is stably expressed without requiring glucocorticoid treatment of the cultures (not shown), was used as well. All of the cell lines demonstrate detectable levels of EGF receptor (Fig. 3). Each of these variant cell lines was treated with PACAP38 or PACAP38 plus EGF, and the extent of neurite outgrowth was evaluated (Fig. 4). In SrcDN2 PC12 cells, PACAP38 unambiguously induces the initiation of neuritic process outgrowth with coadministration of EGF amplifying the response (Fig. 4). Addition of PACAP38 or PACAP38/EGF to the Ras dominant negative-expressing PC12 variants M-M17–26 or GSrasDN6 cells pretreated 24 hr with dexamethasone also elicited the outgrowth of neurites. In each instance, the extent of neuritic outgrowth induced by the addition of PACAP38 was significant compared with untreated, control cultures (Table 1). Again, amplification of the PACAP38-induced neurite outgrowth response by simultaneous addition of EGF was observed, consistent with the noted expression of EGF receptor in each cell line. Collectively, these observations indicate that Src and Ras are unlikely to be involved in mediating PACAP38-stimulated neuritogenesis. In contrast, NGF failed to elicit a neurite outgrowth response in these variant, dominant negative Src/Ras cell lines (data not shown) as reported previously (Lazarovici et al., 1997a). The decreased percentage of cells exhibiting neurite outgrowth in response to PACAP38 in nnr5, SrcDN, and RasDN variant lines compared with wild-type PC12 cells may be attributable, in part, to the clonal selection of colonies with lower levels of PACAP receptor expression. Additional subcloning for PACAP receptors has not been performed.
Time-dependent, PACAP38-induced stimulation of Erk kinase activity: potentiation by coadministration with EGF.
Addition of PACAP38 (5 nm) to wild-type PC12 cells elicits a rapid increase in Erk kinase activity, reaching a maximum of 12.1 ± 4.2-fold above untreated control cultures within 5 min of treatment (Fig.5). In the continuous presence of PACAP38, Erk activity declines to a level that remains ∼2-fold above basal activity to 2 hr after the addition (Fig. 5). Increasing the concentration of PACAP38 failed to further elevate the level of Erk activation observed at 5 min. Considerably higher peaks in stimulated Erk activity were achieved by the addition of NGF (50 ng/ml) or EGF (10 ng/ml) reaching levels that were 22.5 ± 6.7- and 19.5 ± 6.7-fold above control values, respectively, at 5 min after treatment (Fig. 5). Simultaneous addition of PACAP38 (5 nm) and EGF (10 ng/ml) resulted in the maximal (24.5 ± 9.9-fold) increase in Erk activity observed 5 min after treatment. The decline in PACAP38/EGF-stimulated Erk activity that occurs from the peak levels observed at 5 min is comparable to results obtained for NGF (Fig. 5). At 120 min after the addition, Erk phosphorylating activity remains at ∼4-fold above basal levels for both NGF and PACAP38/EGF-treated cultures. Continued culturing of PC12 cells in the presence of PACAP38 alone, beyond 2 hr, and to 72 hr reveals a secondary stimulation of Erk activity occurring at 24 hr after PACAP38 addition (Fig. 5,inset).
PACAP38-stimulated Erk kinase activity is partially mediated by PKA.
PACAP type II receptor activates adenylate cyclase through coupling to the guanine nucleotide binding protein Gs (Miyata et al., 1990; Deutsch and Sun, 1992). To determine whether PKA is involved in PACAP38-induced Erk kinase activation, experiments were conducted using an analog of cAMP to activate PKA. Treatment of PC12 cells with the stable, cAMP analog sp-cAMP (25 μm) for up to 72 hr was without effect on the level of Erk activity (Fig. 5, inset) as reported previously (Richter-Landsberg and Jastorff, 1986; Barrie et al., 1997), although a transient neurite outgrowth response has been observed in PC12 cells as a consequence of elevating intracellular cAMP levels (Richter-Landsberg and Jastorff, 1986). As an alternative approach for testing the hypothesis that activation of PKA is required for PACAP38-mediated induction of Erk activity, H89, a potent, relatively selective inhibitor of PKA, was used (Chijiwa et al., 1990). At a concentration of 20 μm, the amount used in experiments reported here, H89, an isoquinoline-sulfonamide, inhibits PKA but not calcium-calmodulin kinases, protein kinase C, casein kinase II, or cGMP-dependent protein kinases (Chijiwa et al., 1990). By comparison, K-252a, a microbial alkaloid, when added to PC12 cells at a concentration of 200 nm, inhibits not only NGF-induced p140trk tyrosine kinase activity but also a number of the previously mentioned protein kinases (Lazaroviciet al., 1997a). Pretreatment of PC12 cells for 30 min with 0.2% methanol (control), 20 μm H89, or 200 nm K-252a was followed by the addition of NGF (50 ng/ml), EGF (10 ng/ml), PACAP38 (5 μm), PACAP38 and EGF, or no peptides. In control cells, NGF, EGF, PACAP38, and PACAP38+EGF stimulated Erk kinase activity by 12.3-, 16.5-, 5.2-, and 20.0-fold respectively (Fig. 6). As was noted previously (Fig. 5), the combination of PACAP38 and EGF exhibited an additive effect with respect to Erk activation. Pretreatment with H-89 (30 min) inhibited NGF, EGF, PACAP38, or PACAP38/EGF stimulation of Erk kinase activity by 76%, 74%, 65%, and 75%, respectively. As anticipated (Lazarovici et al., 1997a), prior treatment with K-252a completely blocked NGF-stimulated Erk activity (>99% inhibition). In addition, Erk stimulation by EGF, PACAP, or PACAP38/EGF was inhibited by K-252a 69%, 48%, and 67%, respectively (Fig. 6). It should be noted that with the exception of the K-252a and NGF combination, the effect of inhibitor pretreatment on peptide-stimulated Erk activation was due, in large measure, to an inhibitor-mediated increase in basal Erk activity. Maximal peak stimulations of Erk elicited by acute (5-min) treatment with EGF, PACAP38, or PACAP38 plus EGF were not diminished by pretreatment with either K-252a or H-89. The peak level of Erk MBP phosphorylating activity induced by NGF was reduced to the base-line value observed in K-252a- but not H-89-pretreated control cultures.
Results for the activation of PKA by the same peptide additions as described above are presented in Table 2. Again, maximal stimulation was obtained by combined administration of PACAP38 and EGF. Pretreatment with 20 μm H-89 resulted in an 85–95% inhibition of PKA activity for all treatment groups evaluated.
PACAP38-induced neurite outgrowth and activation of Erk kinase activity is primarily PKC and MEK dependent.
To affirm the involvement of PKA, PKC, and MEK in PACAP38-elicited neurite outgrowth in PC12 cells, cultures were incubated for 1 hr with various selective pharmacological inhibitors before the addition of PACAP38 (Figs.7 and 8). Preincubation of wild-type and variant PC12 cells with 20 μm H-89 (PKA antagonist) failed to significantly diminish PACAP38-induced neurite outgrowth (Figs. 7 and 8), suggesting that the contribution of PKA to promotion of PACAP38-stimulated neuritogenesis is minimal.
In addition to adenylate cyclase, PACAP stimulates PLC phospholipid hydrolytic activity. In PC12 cells, activation of PLC by PACAP results in the generation of diacylglycerol and stimulation of PKC. To assess the role of PKC in PACAP38 signal transduction, PC12 cells were incubated with 1 μm concentration of the PKC inhibitors GF10923X (Figs. 7 and 8) and calphostin C (data not shown) for 1 hr before the addition of PACAP38. PACAP38-induced neurite outgrowth in wild-type and dominant negative Ras PC12 cells was inhibited 70% and 88%, respectively, by pretreatment with GF10923X. Confirming the PKC-dependent character of this inhibition, down-regulation of phorbol-sensitive PKC isoforms by treatment with 1 μm PMA (Zheng et al., 1996) inhibited PACAP38-induced neurite outgrowth by 84% and 71% in wild-type and dominant negative Ras PC12 cells, respectively. Representative photomicrographs (Fig. 7) also illustrate the inhibitory effects of MEK inhibition on PACAP38-stimulated neurite formation. In wild-type and dominant negative Ras PC12 cells, pretreatment with 20 μm PD98059, an inhibitor of the Erk kinase MEK, attenuated PACAP38-elicited neuritogenesis by 45 ± 16% and 87 ± 12%, respectively. Increasing the concentration of inhibitor to 50 μmcompletely blocks PACAP38-induced neurite outgrowth; however, at this concentration, considerable toxicity was observed during the 3-day course of treatment.
To further establish that PACAP38-elicited stimulation of PKC is involved in mediating Erk activation associated with induction of neurite outgrowth, the effect of the PKC-selective antagonist GF10923X and down-regulation of phorbol-sensitive PKC by PMA on PACAP38-stimulated Erk activity was evaluated. Cultures of wild-type PC12 cells were incubated for 1 hr with 1 μm GF10923X or overnight with 1 μm PMA to down-regulate PKC before acute treatment (5 min) with PACAP38 (5 nm). Levels of activated Erk were assessed using an anti-phospho Erk antibody that detects the dual threonine/tyrosine phosphorylated form of the kinase (Fig.9). Treatment of PC12 cells with NGF (100 ng/ml) or PACAP38 triggered robust, MEK-dependent phosphorylation of Erk. Both pharmacological antagonism and down-regulation of PKC attenuated PACAP38-elicited Erk activation as indicated by decreases in the level of antibody-detected, dual threonine/tyrosine phosphorylated Erk (Fig. 9). These findings, when coupled with the noted effects of PKC inhibition on PACAP38-elicited neuritogenesis, implicate a signaling cascade that is PKC and Erk dependent.
Discussion
The neurotrophin family of polypeptides consists of nerve growth factor, brain-derived neurotrophic factor, and neurotrophins 3, 4, and 6. Through ligand-specific interaction with neurotrophin-selective Trk A, B, and C receptors, as well as the nonselective p75 neurotrophin receptor, neurotrophins assume significant physiological roles in the nervous system. They are essential for neuronal development, survival, maintenance of phenotype, and structural/functional repair after injury (Chao, 1992). Recent characterization of the PACAPs and the cloning of their cognate receptors (Hashimoto et al., 1993) have lead to the suggestion that PACAP may serve as a neurotrophic factor for sympathetic neurons (Chang and Korolev, 1997).
Studies from numerous laboratories using PC12 cells as a model for nerve growth factor-induced neuronal morphological differentiation have served to elucidate signal transduction pathways used by this neurotrophin that are initiated by binding of NGF to the tyrosine kinase receptor p140trkA (Kaplan and Miller, 1997). Mutagenesis studies involving TrkA indicate that the Ras/Raf/MEK/Erk cascade, acting in conjunction with PLC-γ and tyrosine phosphorylation of the nuclear protein SNT, is involved in the initiation, elongation, and maintenance of neuritic processes (Kaplan and Miller, 1997). Nerve growth factor-evoked neuronal differentiation in PC12 cells is associated with prolonged Ras activity and, consequently, extended activation of Erk (Qui and Green, 1992), which has been demonstrated as necessary for the outgrowth of neurites (Fukuda et al., 1995). PACAP also promotes neurite outgrowth in PC12 cells and has been reported to activate Erk; however, the molecular mechanisms involved in transducing signals from PACAP receptors to downstream targets such as Erk are less well defined than those for NGF. In the current study, a predominant role for PKC is proposed in mediating PACAP-elicited Erk activation. Stimulation of Erk activity by PACAP38 in PC12 cells occurs in a Ras-independent manner and promotes neurite outgrowth.
Pharmacological studies (Deutsch and Sun, 1992; Hernandez et al., 1995) have indicated that PACAP38 stimulates both adenylate cyclase and phospholipase C enzymatic activity in PC12 cells (Fig.10). Elevation of intracellular cAMP activates PKA, which, in turn, activates Erk through B-Raf (Fig. 10) involving a Ras-independent pathway (Vossler et al., 1997). Indeed, results from experiments reported here using the PKA antagonist H89 reveal partial inhibition of PACAP38-stimulated Erk activity in wild-type and dominant negative PC12 cells; however, induction of neurite outgrowth is not prevented. The nature of this apparent inconsistency between inhibition of Erk activity and undiminished neurite outgrowth responsiveness can be most readily explained by fact that the majority of the inhibitory effect elicited by H-89 is due to an inhibitor-mediated increase in basal Erk activity and not a suppression of kinase activity level stimulated by PACAP38. The fact that chronic treatment with H-89 alone for up to 2 days fails to elicit a neurite outgrowth response suggests that this elevation in basal Erk activity observed on an acute basis as a consequence of H-89 treatment either is transient and short lived or the level of Erk activation is insufficient to support the induction of neurite formation. Similar findings have been obtained by others wherein H89 inhibits PACAP-stimulated neuropeptide Y gene expression in PC12 cells but not neurite outgrowth (Barrie et al., 1997). Collectively, these data indicate that any direct contribution of PKA to PACAP38-induced neurite formation in PC12 cells is minimal (Lazarovici et al., 1997b). In contrast, the cessation of cellular proliferation inferred from experiments demonstrating decreased thymidine incorporation on treatment with PACAP38 (data not shown) is consistent with the suggestion that PKA facilitates antiproliferative processes in PC12 cells (Mark and Storm, 1997). Moreover, the antiproliferative effect of PACAP38 was effectively abrogated by H-89 (data not shown), an observation that underscores further the featured role PKA subserves in regulation of PC12 cell proliferation. A hallmark feature of NGF-induced neuronal differentiation in PC12 cells is suppression of cellular replication; hence, inhibition of proliferation by PACAP38 may indicate activation of similar signal transduction cascades that involve PKA and are used by the neurotrophin NGF.
PACAP38 stimulation of phospholipase C in PC12 cells (Deutsch and Sun, 1992; Colbert et al., 1994) produces an increase in intracellular calcium levels (Barrie et al., 1997) and promotes diacylglycerol formation (Fig. 10). Together, these second messengers stimulate the activity of PKC isozymes, which, consequently, enhances the activity of Raf isoforms that, independent of Ras (Uedaet al., 1996), lead to activation of Erk (Fig. 10). In the current study, depletion of PKCα, PKCδ, and PKCε by prolonged phorbol treatment (Zheng et al., 1996), inhibition of PKCα, PKCβ, and PKCγ using the inhibitor GF109203X (Toullecet al., 1991), or both, profoundly inhibited PACAP38-induced Erk activation (Fig. 9) and induction of neurite outgrowth (Fig. 7). The identity of PKC isoforms stimulated by PACAP38 in PC12 cells is under investigation in our laboratories. The role of calcium in promoting PACAP38-stimulated Erk activity and the commensurate elaboration of neurites was not addressed in this study due to the fact that Ras is required for calcium-mediated increases in Erk activity (Rosen et al., 1994). In our hands, PACAP38 was capable of inducing neuritogenesis independent of the activities of either Ras or Src.
Consistent with features that characterize these signal transduction pathways (Fig. 10), we have shown that the MEK inhibitor PD98059, reported to block NGF-induced neurite outgrowth in PC12 cells (Panget al., 1995), significantly inhibited both PACAP38-stimulated Erk activity and neurite outgrowth. This finding supports the concept that stimulation of Erk activity in PC12 cells is sufficient for induction of neurite outgrowth, which, in turn, is extended by our current results suggesting PACAP38 effects are signaled through a PKC-Erk pathway.
In contrast to the long, robust, stable neurites elicited by NGF on p140trkA-mediated activation of the Ras-Erk pathway, PACAP38-stimulated neurites resulting from activation of a PKC-Erk cascade are abbreviated in length, less ramified, finer in diameter, and seem to be stable for only 3–4 days. However, on the concomitant administration of both PACAP38 and EGF, the latter at concentrations that do not elicit neurite outgrowth in PC12 cells, there is a pronounced amplification of both magnitude and duration of Erk activation accompanied by induction of neurites that more closely resemble in appearance those elicited by NGF with respect to length, complexity, and stability. These observations are reminiscent of the synergistic induction of neurite outgrowth reported for the combined treatment of PC12 cells with EGF and interleukin-6 (Wu and Bradshaw, 1996) or the addition of EGF in combination with depolarization or cAMP (Hilborn et al., 1997). This amplification is correlated with a more robust and persistent stimulation of Erk activity, which is consistent with current interpretations regarding molecular mechanisms that mediate the actions of neurotrophins such as NGF. Consequently, the combination of PACAP38 with EGF in PC12 cells may promote synergistic interactions between PKC/Erk and Ras/Erk pathways that support morphological neuronal differentiation, whereas the role of PKA in this scheme is proposed to be one that involves inhibition of DNA synthesis and cellular replication. In circumstances of inhibited cell division, signals generated by the EGF receptor in PC12 cells are switched from growth promoting to supporting neuronal phenotypic differentiation (Mark and Storm, 1997).
Additional interactions occurring between neurotrophins and electrical activity under physiological conditions that serve to control neuronal development and plasticity are only now being systematically addressed at the pharmacological level (Hilborn et al., 1997; Lu and Figurov, 1997). The principal outcomes of these events include neurite outgrowth, synapse formation, and the acquisition of electrical properties. It is tempting to suggest that depolarizing events, in combination with signals generated by mitogens (EGF) and neuropeptides (vasoactive intestinal peptide, PACAP) could promote morphological changes without altering physiological differentiation. The elaboration of neurites, synapse formation, and resulting development of neuron/neuron contact could evoke release of neurotrophins such as NGF, which, in turn, will be responsible for driving final differentiation and neuronal maturation.
Acknowledgments
We express our appreciation to Dr. M. Hayes and Dr. E. Sutkowski for their critical reading of the manuscript and thoughtful suggestions. The insights, support, and encouragement of Dr. G. Guroff are appreciated.
Footnotes
- Received March 9, 1998.
- Accepted June 3, 1998.
-
Send reprint requests to: Donald W. Fink, Jr., Ph.D., FDA/CBER-Division of Cytokine Biology, 1401 Rockville Pike, Suite 200 N/HFM-505, Rockville, MD 20852-1448. E-mail:finkd{at}al.cber.fda.gov
Abbreviations
- NGF
- nerve growth factor
- PACAP38
- pituitary adenylate cyclase-activating polypeptide with 38 residues and an amidated carboxyl terminus
- EGF
- epidermal growth factor
- Src
- pp60c-src cytoplasmic tyrosine kinase
- Trk
- p140trk nerve growth factor receptor tyrosine kinase
- Ras
- p21ras G protein
- Erk
- extracellular signal-regulated kinase
- MEK
- threonine/tyrosine dual-specificity extracellular signal-regulated kinase
- PVDF
- polyvinylidene difluoride
- SNT
- suc1-associated neurotrophic factor target
- PMA
- phorbol-12-myristate-13-acetate
- PKA
- cAMP-dependent protein kinase, PKC, calcium- and phospholipid-dependent protein kinase C
- MBP
- myelin basic protein
- PC12nnr5
- a pheochromocytoma PC12 clone nonresponsive to NGF
- PC12–6.24
- a PC12 clone overexpressing human p140trk
- GsRasDN6
- dexamethasone-inducible, dominant-negative Ras PC12 clone
- M-M17–26
- stable, dominant-negative Ras PC12 clone
- SDS
- sodium dodecyl sulfate
- PAGE
- polyacrylamide gel electrophoresis
- The American Society for Pharmacology and Experimental Therapeutics