INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

The pheochromocytoma cell line PC12 is a well-utilized model for study of neurotrophic factors such as nerve growth factor (NGF). Treatment of these cells with NGF induces differentiation and survival. The NGF signaling cascade begins with the sequential action of a Src-Ras cassette (26), leading to the activation of mitogen-activated protein kinase (MAPK). Inhibition of MEK, the upstream MAPK kinase, blocks NGF-induced differentiation (43), thus suggesting that MAPK plays a critical role in cell differentiation. However, MAPK activation is not absolutely required for differentiation of PC12 cells, since bone morphogenic protein 2 can induce differentiation in the absence of MAPK activation (18). NGF also leads to the activation of both the p38 kinase (39) and c-Jun N-terminal kinase (JNK) (17, 38). In addition, phosphatidylinositol 3-kinase (PI3K) is activated and required for NGF-mediated differentiation (19, 22, 25). PI3K is also required for NGF survival signaling (59).

The protein kinase C (PKC) superfamily, composed of 11 isoforms (51), has been implicated in mediating NGF responses as well. PC12 cells express all 11 isoforms of PKC, and each is activated in response to NGF (56, 57), implying that each plays a role in mediating NGF responses. Inhibition of PKC by sphingosine blocks NGF-induced neurite outgrowth (16), and microinjection of PKC antibodies inhibits NGF-induced neurite outgrowth and c-Fos expression (3). However, downregulation of PKC with chronic phorbol ester treatment, resulting in removal of classical and nonclassical PKC (cPKC and nPKC) pools, has no effect on NGF-induced neurite outgrowth (46) or NGF-induced MAPK activation (33). We demonstrated that the phorbol ester-sensitive PKC isoforms (, , , , and ) were not required for NGF differentiation and further demonstrated that NGF activated the phorbol ester-insensitive atypical PKC (aPKC) isoforms, / and  (9, 56). Moreover, removal of aPKCs was observed to block NGF-induced differentiation of PC12 cells only in the absence of other PKCs, demonstrating a hierarchal relationship between aPKCs and other PKC isoforms activated by NGF (9). Recently, FEZ1 (fasciculation and elongation protein zeta 1), a brain-specific transcript which is the mammalian homologue of UNC-76, a protein involved in axonal outgrowth and fasciculation in Caenorhabditis elegans, has been identified as a PKC- binding protein whose expression in PC12 cells stimulates differentiation (28). These findings further underscore the importance of aPKC in control of neuronal responses.

Overexpression of aPKCs enhances NGF responsiveness as well as survival of differentiated cells through an NF-B pathway (58). These findings are in keeping with the ability of aPKCs to bind IKK and directly regulate the B pathway (29). Members of the aPKC subfamily comprised of iota/lambda and zeta isoforms are highly homologous. Furthermore, aPKCs are activated by a wortmannin-sensitive, PI3K-dependent pathway (2) involving phosphorylation of aPKC by PDK1 (31). Additionally, we have recently shown that aPKCs can bind Src, resulting in tyrosine phosphorylation of aPKC (49). Little is understood regarding the precise placement of the aPKCs in relation to known components of the NGF survival/differentiation signaling pathway(s). To broaden our understanding of aPKC's position in these pathways, we undertook a detailed study to map the relationship of PKC- to components of the Src-Ras/Raf-1/MAPK, p38, and JNK pathways utilizing NF-B as a downstream target. In addition, the relationship of PI3K to this signal network, as well as its role in activation of aPKC, Ras-MAPKs, and NF-B, was studied. We find that both Ras and Src lie upstream of aPKC, regulated by PI3K. In PC12 cells, neither aPKC nor PI3K is directly required for activation of MAPK. Furthermore, NGF-induced activation of NF-B required Src. Our findings reveal that two discrete NGF signaling pathways are operative; differentiation employs the Ras-Src/aPKC cassette, whereas survival is restricted to the Src-PI3K/aPKC pathway. Atypical PKC- emerges as a component of both differentiation and survival signaling pathways regulated by a novel Src kinase pathway.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Materials. NGF was purchased from Harlan BioProducts for Science (Indianapolis, Ind.). The p38 inhibitor SB203580, the MEK inhibitor PD98059, Ras antibody, pseudosubstrate peptide to cPKC, and chelerythrine chloride were from Calbiochem (La Jolla, Calif.). The phosphospecific antibodies to MAPK, JNK, and p38 were obtained from New England Biolabs, Beverly, Mass. Raf-1, MEK, PKC-, IKK, and c-Jun (S-63, KM-1) antibodies and recombinant MEK were purchased from Santa Cruz Biotechnology, Santa Cruz, Calif. Antibody to c-Src and p34 Src peptide substrate was obtained from Upstate Biotechnology, Lake Placid, N.Y. Both myristoylated and nonmyristoylated forms of PKC- pseudosubstrate peptide were synthesized by the Center for Macromolecular Structure, University of Kentucky, Lexington. LY294002 and wortmannin were purchased from BioMol, Indianapolis, Inc. NF-B oligonucleotide was from Promega (Madison, Wis.). PC12 cells overexpressing c-Src under the control of the cytomegalovirus promoter (Src cells) and Src cells (Src DN2 expressing the K295R mutant [kinase dead] form of chicken Src) were kindly provided by Simon Halegoua (Stony Brook, N.Y.). PC12 cells (M-M17-26) expressing the dominant inhibitory mutant RasN17, which interferes with normal Ras function, were provided by Geoffery Cooper (Harvard Medical School, Boston, Mass.). PC12 cells expressing a temperature-sensitive v-Src were provided by Gordon Guroff (National Institutes of Health). Jorge Moscat (Centro de Biología Molecular, Universidad Autonoma, Madrid, Spain) kindly provided aPKC constructs. Polyclonal antibody to JNK and purified glutathione S-transferase fused to amino acids 1 to 79 of c-Jun [GST-cJun(1-79)] was a gift from Guisheng Zhou, Baylor College of Medicine, Houston, Tex.

Cell culture. PC12 cells were grown on collagen-coated plastic culture dishes in Dulbecco's modified Eagle's medium with 10% horse serum and 5% fetal calf serum, as described elsewhere (56-58). Prior to treatment with NGF, the cells were starved overnight by incubation in a ratio of 1 ml of conditioned medium to 5 ml of serum-free medium.

aPKC activity. PC12 cells were incubated for different times with NGF, extracted in lysis buffer (50 mM Tris [pH 7.5], 150 mM NaCl, 1% Triton X-100, 2 mM EDTA, 1 mM EGTA, protease inhibitors), and immunoprecipitated with affinity-purified rabbit polyclonal antibody (20). Immunoprecipitates were washed five times with lysis buffer containing 0.5 M NaCl. For immune complex kinase assay, the samples were resuspended in kinase buffer (35 mM Tris-HCl [pH 7.5], 10 mM MgCl₂, 0.5 mM EGTA, 0.1 mM CaCl₂, 1 mM p-nitrophenyl phosphate) containing 1 µg of myelin basic protein (MBP) and 5 µCi (100 µM) of [-³²P]ATP for 30 min at 30°C in a final volume of 40 µl. To assess the specificity for PKC-, assays were done in the presence or absence of 40 µM pseudosubstrate peptide (SIYRRGARRWRKL) or irrelevant control peptide (IETVDNKASTRAY), both preincubated for 5 min. Reactions were terminated by addition of concentrated sample buffer and separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) followed by exposure to X-ray film and quantitation with a computer-interfaced densitometer. MBP phosphorylation was normalized to the amount of PKC- in each sample and thus expressed as specific activity in arbitrary units.

Coimmunoprecipitation of PKC- with Ras or Src. The procedure used was essentially as previously described by Diaz-Meco et al. (11). PC12 cells were lysed with a mixture of 30 mM HEPES (pH 7.5), 1% Triton X-100, 10% glycerol, 10 mM NaCl, 5 mM MgCl₂, 1 mM Na₃VO₄, 2 mM phenylmethylsulfonyl fluoride (PMSF), 25 mM NaF, 1 mM EGTA, and protease inhibitors. The lysates were immunoprecipitated by addition of pan-Ras antibody. Immune complexes were recovered by addition of protein A-agarose followed by incubation overnight at 4°C. The beads were washed twice with 20 mM Tris-HCl (pH 7.5)-500 mM NaCl-1% Triton X-100-0.1% -mercaptoethanol, followed by a final wash in buffer containing 10 mM Tris-HCl (pH 7.5), 5 mM MgCl₂, 1 mM EDTA, 25 mM NaF, 100 µM Na₃VO₄, leupeptin (20 µg/ml), pepstatin A (1 µg/ml), and aprotinin (4 µg/ml). The immune complexes were separated by SDS-PAGE on a 12% gel, transferred to nitrocellulose, and Western blotted with PKC-, Ras, or Src antibody; proteins were identified by enhanced chemiluminescence.

Purification of aPKCs. In brief, Sf9 cells were infected with recombinant virus. The cells were lysed in a mixture of 20 mM Tris-HCl (pH 7.5), 5 mM EDTA, 43 mM -mercaptoethanol, 1% Triton X-100, PMSF, and protease inhibitors. The lysate was clarified by centrifugation at 10,000 × g for 30 min, and aPKC was purified as previously described (60).

In vitro phosphorylation of MEK-1. Phosphorylation of MEK was conducted using immunoprecipitated Raf-1 to which purified PKC- was added at various concentrations. The reactions were conducted in the presence or absence of PKC- inhibitor pseudosubstrate peptide (SIYRRGARRWRKL). Kinase reactions were performed for 30 min at 30°C in 50 µl of this buffer with 10 µCi of [-³²P]ATP, with or without 1 µg of the natural substrate MEK1. Reactions were terminated by the addition of SDS sample buffer and analyzed on SDS-10% polyacrylamide gels.

Raf-1 protein kinase activity. PC12 cells were stimulated with NGF and lysed in buffer containing 10 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 0.2 mM Na₃VO₄, 0.2 mM PMSF, 1% Triton X-100, 0.5% NP-40, and protease inhibitors. Lysates were clarified by centrifugation and precleared twice for 1 h with 25 µl of protein A-Sepharose. Raf-1 was immunoprecipitated by addition of Raf-1 antibody and inverted end-over-end for 1 h at 4°C, followed by addition of 30 µl of anti-goat antibody for an additional 1.5 h. The immune complex was washed three times in lysis buffer and twice in piperazine-N,N'-bis(2-ethanesulfonic acid) (PIPES; pH 7)-10 mM MnCl₂-1 µg of aprotinin/ml. Kinase reactions were performed for 30 min at 30°C in 50 µl of this buffer with 10 µCi of [-³²P]ATP, with or without 1 µg of the natural substrate MEK1 (48, 53). Reactions were terminated by the addition of SDS sample buffer and analyzed on SDS-10% polyacrylamide gels.

Activation of Raf-1, p38, MAP, and JNK. Measurement of Raf-1 activation was also determined by electrophoretic mobility shift assay (EMSA) and visualized by Western blotting, indicative of Raf-1 phosphorylation and activation (53). p38, MAP, and JNK were determined by immunoblotting with phosphospecific antibodies. The blots were stripped by addition of 0.1 M glycine (pH 2) followed by neutralizing and reprobed with antibody to p38, MAP, and JNK. Alternatively, the activity of JNK was also measured by immune complex kinase assay as previously described with GST-cJun(1-79) as the substrate (17, 37). The radioactive protein bands were scanned with a computer-interfaced densitometer to determine the relative changes in substrate phosphorylation.

NF-B EMSA. Cell extracts were prepared in high-salt detergent buffer (Totex; 20 mM HEPES [pH 7.9], 350 mM NaCl, 20% [vol/vol] glycerol, 1% [wt/vol] NP-40, 1 mM MgCl₂, 0.5 mM EDTA, 0.1 mM EGTA, 0.5 mM dithiothreitol [DTT], 0.1% PMSF, 1% aprotinin) (58). The cells were harvested by centrifugation, washed in ice-cold phosphate-buffered saline, resuspended in 4 volumes of Totex buffer, incubated on ice for 30 min, and then centrifuged for 5 min at 13,000 × g at 4°C. The protein content of the supernatant was determined, and equal amounts of protein (20 µg) were added to a reaction mixture containing 20 µg of bovine serum albumin, 2 µg of poly(dI-dC), 2 µl of buffer D+ (20 mM HEPES [pH 7.9], 20% glycerol, 100 mM KCl, 0.5 mM EDTA, 0.25% NP-40, 2 mM DTT, 0.1% PMSF), 4 µl of buffer F (20% Ficoll 400, 100 mM HEPES, 300 mM KCl, 10 mM DTT, 0.1% PMSF), and 100,000 cpm of a ³²P-labeled oligonucleotide (5' AGTTGAGGGGACTTTCCCAGGC 3') in a final volume of 20 µl. Samples were incubated at room temperature for 25 min. For supershift assays, 2 to 5 µg of antibody was added to the protein and allowed to incubate overnight, followed by inclusion in an assay. Excess AP-1 (5' CGCTTGATGAGTCAGCCGGAA 3') or NF-B oligonucleotide was included as a negative control. The samples were resolved on a 6% Tris-glycine polyacrylamide gel. The gel was dried and exposed to X-ray film for 24 to 72 h.

Coprecipitation of IB, Src, and PKC-. A previously defined (1) coprecipitation protocol was used. In brief, PC12 cells were washed in phosphate-buffered saline and lysed in ice-cold lysis buffer (20 mM Tris [pH 7.8], 150 mM NaCl, 1 mM CaCl₂, 0.2% deoxycholic acid, 0.2% NP-40, 1 mM Na₃VO₄, protease inhibitors). The lysates were incubated with anti-IKK followed by protein A, and the beads containing the complexes were washed extensively with lysis buffer.

Measure of Src activity. As previously described (1), Src was immunoprecipitated and washed in kinase lysis buffer (1.5% NP-40, 150 mM NaCl, 25 mM Tris [pH 8], 25 mM NaF, 100 mM NaVO₃) and then preincubated with 15 ml of kinase reaction buffer (25 mM HEPES [pH 7.4], 150 mM NaCl, 5 mM MnCl₂, 100 µM NaVO₃), for 10 min at 22°C. The kinase reaction was initiated by addition of 5 µCi of [-³²P]ATP (6,000 Ci/mmol), 5 µM ATP, and 1 µg of p34 c-Src and was terminated after 2 min by addition of an equal volume of 2× SDS sample buffer; the mixture was boiled, dried, and exposed to X-ray film.

Cell survival and differentiation. To assess cell death or differentiation, PC12 cells were plated and after 3 days washed five times with serum-free medium (59). Cultures were treated in the presence or absence of 100 ng of NGF per ml. After 48 h, the cells and their medium were collected. Trypan blue was added to examine cell viability. An aliquot of the cells was counted, and the percentage cells that stained trypan blue positive was determined. Alternatively, PC12 differentiation was determined by scoring for neurite outgrowth 6 days after addition of NGF. The percentage cells that possessed neurites was determined by counting treatments in triplicate. A minimum of 500 cells were counted within a treatment group. The percentage of cells with neurites was calculated based on an individual cell bearing one process with an extension which was greater in length than two cell diameters. Clumped cells were not included in the scoring process.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Activation of aPKC requires Ras, Src and PI3K. PKC- antibody obtained from Santa Cruz has been previously shown not to cross-react with other PKC isoforms or with the highly homologous PKC- (40). To validate that the immune complex kinase assay accurately measures PKC- activity, several approaches were undertaken. Previously, we have shown that NGF-induced PKC- activation results in an approximately twofold increase in aPKC activity using an in situ assay (56). Similarly, upon stimulation of PC12 cells with NGF, the activity of PKC-, measured in the immune complex kinase assay, was increased approximately twofold (Table 1). Thus, measure of NGF-activated PKC- using this assay was similar to previous findings employing a separate method. The aPKCs lack the C2 domain and consequently do not bind, nor are they downregulated by phorbol esters (51). Prolonged treatment with phorbol myristate acetate (PMA) results in PC12 cells that are devoid of cPKC and nPKC isoforms (56). Treatment of PC12 cells with 1 µM PMA for 48 h prior to NGF stimulation failed to alter the level of PKC- activity. Peptides complementary to the pseudosubstrate region have previously been shown to serve as highly competitive selective inhibitors of specific PKC isoforms (30). To further validate the assay of PKC-, the cells were treated with pseudosubstrate peptides to which an N-terminal myristic acid had been added to facilitate their diffusion through the plasma membrane. Treatment of the cells with pseudosubstrate peptide to the cPKC isoforms failed to inhibit recovery of NGF-stimulated PKC- activity, whereas treatment with pseudosubstrate peptide to atypical PKC- diminished the activity of NGF-stimulated immune-complexed kinase enzyme. Last, PKC- antibody was preincubated with peptide antigen prior to immunoprecipitation. In this case, recovery of NGF-stimulated PKC- was blocked by inclusion of the peptide. Collectively, these findings validate the measure of PKC- activity by the immune complex kinase assay.

View this table: [in this window] [in a new window]   TABLE 1.   Validation of immune complex kinase assay for measure of PKC- activity

View larger version (17K): [in this window] [in a new window]   FIG. 1.   Effects of pharmacological agents on aPKC activity. PC12 cells were pretreated with either wortmannin (Wort; 50, 100, or 200 nM), LY294002 (Ly; 25, 50, or 100 µM), or chelerythrine chloride (CH; 3, 6, or 9 µM) for 1 h prior to addition of NGF (100 ng/ml) for 15 min. PC12 cell lysates (400 µg) were analyzed for aPKC activity in triplicate by immune complex kinase assay using MBP as substrate. Data shown are the means ± standard error of the means of three independent experiments. N, NGF.

View larger version (29K): [in this window] [in a new window]   FIG. 2.   PKC- coassociates with Ras and Src. (A) PC12 cells were stimulated with NGF (100 ng/ml) as indicated for 0 to 15 min. Cell lysates (500 µg) were prepared and immunoprecipitated (IP) with antibody to either Ras or Src followed by Western blotting (WB) with PKC- antibody. Included in the analysis as a standard was a PC12 cell whole cell lysate (WCL; 70 µg). As positive control, the cell lysates (50 µg) were analyze by immunoblotting with antibody to Ras, Src, or PKC-. (B) Control, Ras, or Src PC12 cells were stimulated with NGF (100 ng/ml) for 0 to 60 min followed by analysis of aPKC activity in triplicate by immune complex kinase assay using MBP as substrate. This experiment was repeated two other times with similar results.

Raf-1 is regulated by Ras independently of Src and PKC-. To further dissect the aPKC signaling pathway, the relationship of Src/Ras and PKC- to NGF-induced activation of Raf-1 was studied. Overexpression of Raf-1 has been shown to stimulate neurite outgrowth of PC12 cells (54), demonstrating a sufficiency of overexpression requirement of Raf-1 for the differentiation response. In addition, NGF activates the Raf-MAPK cascade (55). Using a cotransfection approach, aPKC has also been reported to activate MEK independently of Raf-1 (48), suggesting that a Raf-1-independent pathway also exists. Therefore, we used several approaches to explore the regulation of NGF-induced activation of Raf-1 by PKC-. As a control, Raf-1 activation by NGF was validated in cells that were lacking Ras or Src activity (Fig. 3A). Raf activation occurs via phosphorylation; therefore, a hypershift of p74 Raf-1 mobility on SDS-PAGE can be used as an indicator of Raf-1 activation. Cell lysates prepared from PC12, Src, or Ras cells stimulated with NGF were analyzed for Raf activation by EMSA (Fig. 3A). NGF-induced Raf-1 activity was dependent on Ras and independent of Src, thus validating assay of Raf-1 activity by gel shift and accounting for any possible differences in clonal variants of these PC12 cells. Agents that were shown to inhibit activation of aPKC (Fig. 1) had little effect on NGF-induced activation of Raf-1 (Fig. 3B). The inability of aPKC to modulate Raf-1 was further confirmed by transfecting mutants for either PKC- or PKC- that confer a dominant negative phenotype (5, 6). Cell lysates obtained from transfected cells stimulated with NGF were confirmed to possess diminished levels of aPKC activity (data not shown). An absence of aPKC activity had no effect on NGF-induced Raf-1 activation (Fig. 3C). Moreover, overexpression of aPKCs failed to enhance NGF-induced activation of Raf-1 (58). These findings (Fig. 3B and C) were likewise validated by measure of Raf-1 activity in an immune complex kinase assay using MEK as the substrate: inhibitors of aPKC or PI3 kinase had little effect on Raf-1 kinase (data not shown). Collectively, these findings demonstrate that aPKC does not directly influence NGF-induced activation of the Raf-1 node of the MAPK signaling cascade in PC12 cells and are in line with previous work defining MEK as the critical point of regulation (48).

View larger version (42K): [in this window] [in a new window]   FIG. 3.   Positioning of Ras, Src, and aPKC relative to Raf-1. (A) Control, Ras, or Src PC12 cells were stimulated with NGF (100 ng/ml) for 0 to 30 min. After lysis, equal protein aliquots were resolved by SDS-PAGE (7.5% polyacrylamide) and then immunoblotted with Raf-1 antibody. (B) PC12 cells were pretreated with either wortmannin (Wort; 50 or 100 nM), LY294002 (Ly; 25 or 50 µM), or chelerythrine chloride (CH; 3 or 6 µM) for 1 h prior to addition of NGF (100 ng/ml) for 15 min. After lysis, equal protein aliquots were resolved by SDS-PAGE (7.5% polyacrylamide) and then immunoblotted with Raf-1 antibody. C, control; N, NGF. (C) PC12 cells were transfected with control construct, mutant PKC- (Imut), or mutant PKC- (Zmut). Thereafter, the cells were stimulated with NGF (100 ng/ml) for 0 or 15 min, as indicated. After lysis, equal protein aliquots were resolved by SDS-PAGE (7.5% polyacrylamide) and then immunoblotted with Raf-1 antibody. Samples were scored + based on the ability of NGF to stimulate Raf-1 hyperphosphorylation/activation as indicated by an increase in relative molecular weight or gel shift (---). These experiments were repeated two other times with similar results. Bands at 66 kDa are indicated on the right.

View larger version (25K): [in this window] [in a new window]   FIG. 4.   PKC- activates MEK. (A) PC12 cells were stimulated with NGF (100 ng/ml) for 15 min followed by immunoprecipitation (IP) of Raf-1 and subjected to an in vitro kinase reaction as indicated with [-32P]ATP in the presence of 1 µg of recombinant MEK1. The samples were separated by SDS-PAGE (10% polyacrylamide) followed by autoradiography. Autophosphorylated PKC- and phosphorylated MEK1 are shown. Sizes are indicated in kilodaltons. (B) MEK1 activity was measured using immune complex kinase assay as previously described (48). The relative changes in activity was normalized to that obtained with NGF treatment. Similar results were obtained in two other experiments. N, NGF.

aPKCs and PI3K fail to regulate p38, MAPK, or JNK. In other systems, aPKCs have been reported to lie upstream of MAPK (5, 6) as well as JNK (58). Overexpression of aPKCs in PC12 cells was observed to slightly enhance MAPK and markedly enhanced NGF-activated JNK (58). In addition, aPKC was directly capable of activating JNK, thus positioning aPKCs upstream of JNK. Using a complementary pharmacological approach, we mapped the relationship of Ras, Src, and aPKC to downstream activation of p38, MAPK, and JNK. NGF has been shown to stimulate the activation of all three kinases, JNK, MAPK, and p38 (19, 22, 39). Activation of MAPK by NGF was Ras dependent and slightly independent of Src, whereas activation of p38 was independent of Ras and dependent on Src (Fig. 5A). Since NGF-induced aPKC activity was shown to be chelerythrine chloride and PI3K sensitive (Fig. 1), these inhibitors were used to explore the relative position of aPKC to p38, MAPK, and JNK. Pretreatment of PC12 cells with these inhibitors failed to diminish NGF-stimulated MAPK or p38 activities (Fig. 5B), consistent with a previous study reporting a lack of effect of PI3K on NGF-stimulated MAPK activity (25). Due to the poor quality of the blots obtained with the phospho-JNK antibodies, the effects of aPKC inhibitors on NGF-stimulated JNK kinase activity were examined using an immune complex kinase assay with GST-cJun(1-79) as the substrate. NGF stimulated JNK activation ~2-fold, similar to what has been previously reported (17). Activation of JNK by NGF was dependent on Ras, Src, aPKC, and PI3K. In addition, constitutive activation of v-Src had a profound effect on the activity of JNK (Table 2). These observations were confirmed by measure of activated c-Jun phosphorylated at S63 (data not shown). Placement of aPKC upstream of JNK parallels our previous findings using a genetic approach (58).

View larger version (35K): [in this window] [in a new window]   FIG. 5.   Positioning of Ras, Src, and aPKC relative to MAPK and p38. (A) Control, Ras, or Src PC12 cells were stimulated with NGF (100 ng/ml) for 0 to 30 min. After lysis, equal protein aliquots (70 µg) were resolved by SDS-PAGE (12% polyacrylamide) and then immunoblotted with either anti-phospho-MAPK or -p38 antibody. The blots were stripped and reprobed with a nonphospho-specific antibody to MAPK or p38. (B) PC12 cells were pretreated with either wortmannin (Wort; 50 or 100 nM), LY294002 (Ly; 25 or 50 µM), or chelerythrine chloride (CH; 3 or 6 µM) for 1 h prior to addition of NGF (100 ng/ml) for 15 min. After lysis, equal protein aliquots (70 µg) were resolved by SDS-PAGE (12% polyacrylamide) and then immunoblotted with anti-phospho-MAPK or -p38 antibody. This experiment was repeated three times with similar results. C, control; N, NGF.

NGF-induced activation of NF-B is dependent on PKC-, Src, JNK, and PI3K and independent of Ras. aPKC lies upstream of NF-B in the NGF signaling pathway (58), consistent with its placement in other signaling paradigms (5, 10, 12). One essential step in the activation of NF-B is the phosphorylation of IB by either IKK or IKK. Moreover, aPKC has been shown to lie upstream of IKK, regulating the kinase via phosphorylation (29), thus documenting direct interaction between aPKC and the NF-B pathway. To further dissect the relationship of aPKC to this downstream transcription target, the requirement of Ras or Src for NGF-induced activation of NF-B was examined. NGF-induced activation of NF-B has previously been characterized as prolonged (29). Activation of NF-B by NGF was observed to require Src but not Ras (Fig. 6A). Consistent with a Src requirement, overexpression of Src enhanced basal NF-B levels. To determine the requirement of either JNK, MAPK, or p38 kinase, PC12 cells were pretreated with specific pharmacological agents that target these signal transduction pathways. JNK is specifically inhibited by curcumin (8), MEK and MAPK are inhibited by PD98059 (43), and p38 is inhibited by SB202190 (39). The ability of these agents to impair NGF-induced activation of NF-B was evaluated. Inhibition of JNK markedly inhibited NGF-induced NF-B (Fig. 6B). Activation of NF-B was sensitive to MEK and relatively insensitive to p38. The role of PI3K in mediating NGF-induced NF-B was likewise examined (Fig. 6C). Pretreatment with LY294002 or wortmannin, potent inhibitors of PI3K, resulted in a dose-dependent inhibition of NGF-induced NF-B activity.

View larger version (43K): [in this window] [in a new window]   FIG. 6.   Positioning of Ras, Src, and aPKC relative to NF-B. (A) PC12, Ras, Src, or cells overexpressing c-Src were stimulated with NGF (100 ng/ml) for 0 to 6 h. After lysis, equal protein aliquots were subjected to EMSA analysis. (B) PC12 cells were pretreated with curcumin (Cu; 5 or 20 µM), PD98059 (PD; 15 or 60 µM), or SB202190 (SB; 5 or 20 µM) followed by addition of NGF (100 ng/ml) for 3 h. After lysis, equal protein aliquots were subjected to EMSA analysis. (C) PC12 cells were pretreated with either wortmannin (Wort; 25 to 200 nM) or LY294002 (Ly; 12.5 to 100 µM) for 1 h prior to addition of NGF (100 ng/ml) for 3 h. After lysis, equal protein aliquots were subjected to EMSA analysis. This experiment was repeated twice with similar results.

Src regulates formation of an IKK-PKC- complex and mediates the upstream portion of an NF-B survival signaling pathway. aPKC can form a complex with IKK, phosphorylating the kinase to lead to activation of the NF-B pathway (29). Moreover, Src regulates aPKC by increasing its tyrosine phosphorylation state and modulating enzyme activity (49; Vandenplas et al., submitted). Since Src cells fail to activate NF-B, we hypothesized that Src may play a role in the coupling of aPKC with the NF-B pathway via modulating the interaction of tyrosine-phosphorylated PKC- with IKK. To test this idea, IKK was immunoprecipitated after NGF stimulation from equivalent protein of control and Src PC12 cells (Fig. 7). IKK was constitutively associated with PKC-, Src, and IB, whereas in Src cells a drastic reduction in the recovery of PKC- complexed with IKK was observed, even though Src/IB levels were similar between both sets of immunoprecipitates (compare Fig. 7). Pretreatment of PC12 cells with genistein, a potent and selective tyrosine kinase inhibitor, likewise reduced the amount of PKC- complexed to IKK (data not shown). Thus, interaction of IKK with PKC- was dependent on Src tyrosine kinase activity as determined by two independent means. Activation of NF-B is preceded by signal-induced IB degradation. After NGF treatment, PKC- dissociates from IB (Fig. 8A) in a time frame coincident with activation of PKC- and Src-induced tyrosine phosphorylation of PKC-, along with parallel activation of NF-B (58; Vandenplas, submitted). To test the requirement for Src activity in this process, the association of PKC-/IB was examined in Src cells and TrkA cells (Fig. 8B). Upon treatment of the cells with NGF, PKC- dissociates from the IB complex (Fig. 8A and B). However, in cells that were deficient in Src, no such dissociation was observed; rather, a low level of basal coassociation between PKC- and IB was noted, consistent with the amount of PKC-/IB observed in the IKK complex in the absence of Src activity (Fig. 7). Treatment with genistein likewise blocked dissociation of PKC-/IB. These results demonstrate that Src tyrosine directly participates in the coupling of PKC- to the NF-B pathway. Moreover, Src activity was required to stimulate translocation of p65/NF-B to the nucleus (Fig. 8C). Last, the degradation of IB was examined in Src cells. NGF failed to stimulate IB degradation in Src (37%) compared to TrkA (74%) cells (Fig. 8D). Taken together, the uncoupling of IKK/PKC- (Fig. 7) parallels a lack of NGF-stimulated NF-B activation (Fig. 6 and 8C, and 8D) and diminished PKC- activity observed in Src cells (Fig. 1). These data demonstrate that Src tyrosine kinase plays a role in modulating the interaction between IKK and PKC-, whereas an absence of Src activity abrogates the signal-induced dissociation of PKC- and IB. Thus, regulation of PKC-/IB interaction would thus account for the absence of NF-B DNA binding previously observed in PC12 cells deficient in Src activity (Fig. 6A).

View larger version (49K): [in this window] [in a new window]   FIG. 7.   Src regulates the association of IKK with aPKC. PC12 cells, parental or Src, were treated with NGF (100 ng/ml) for the times indicated, and lysates (500 µg) were immunoprecipitated with anti-IKK. The immunoprecipitates (IP:IKK) or lysates (50 µg), as a positive control, were Western blotted (WB) with PKC-, Src, and IB antibodies. This experiment was repeated twice with similar results.

View larger version (30K): [in this window] [in a new window]   FIG. 8.   Src is required for PKC-/IB coassociation. (A) PC12 cells were treated with NGF (100 ng/ml) for the times indicated, and lysates were immunoprecipitated (IP) with anti-IB. The precipitates were Western blotted (WB) with PKC- antibody. (B) PC12 cells, cells pretreated with genistein (Gen; 20 µM), or Src cells were treated with NGF (100 ng/ml) for the times indicated, and lysates were immunoprecipitated with anti-PKC-. The precipitates were blotted with IB. (C) PC12 or Src cells were exposed to NGF (100 ng/ml) for 30 min. Nuclei were isolated and immunoblotted with anti-p65/NF-B antibody. (D) PC12 or Src cells were exposed to NGF (100 ng/ml) for 30 min. Whole cell lysates were prepared (50 µg) and immunoblotted with antibody to IB. The blots were scanned, and the relative change in intensity of IB is shown in parentheses.

View larger version (11K): [in this window] [in a new window]   FIG. 9.   Model illustrating the relative positioning of aPKC within the NGF signaling cascades for survival and differentiation. aPKC can activate MEK and lies upstream of JNK. Both Ras and Src lie upstream of aPKC, as does PI3K, whereas aPKC lies upstream of NF-B, directly interacting with IKK. Differentiation requires components of both the Src and Ras pathways. The differentiation pathway overlaps with components of the survival pathway through Src, aPKC, and PI3K. Survival signaling is Ras independent but dependent on PI3K, Src, aPKC, and NF-B. Our findings support a model whereby aPKC occupies a critical node, capable of interacting with Ras-MAPK through modulation of MEK and JNK and upstream of NF-B. NGFR, NGF receptor.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

In this study, we show that a PKC does not reside as a component of the Ras-MAPK pathway but instead is a component of a parallel, distinct signaling pathway. Our data demonstrate that this pathway is regulated in a novel fashion by c-Src. Both Src and Ras have been shown to induce neurite extension similar to NGF (4, 42, 50). A Src-Ras signaling cassette for NGF signaling was initially proposed by microinjection studies, with Src positioned upstream of Ras (26). Persistent activation of the Ras-Raf-MAPK pathway was observed to be insufficient to sustain PC12 differentiation (52). In addition, activation of pp60 Src was also required for PC12 differentiation (26). In lieu of a linear Src-Ras cascade, these data support existence of a parallel Src cascade as being important for PC12 differentiation. Support for such a parallel cascade has likewise been obtained in H19-7 cells, which are rat E17 hippocampal cells that have been conditionally immortalized with a retroviral vector expressing a temperature-sensitive simian virus (27). At the nonpermissive temperature, when T antigen is inactivated, the cells differentiate. Under these conditions, no activation of Ras-Raf-MAPK by v-Src was detected. Similar observations were likewise made in this study, whereby v-Src failed to induce the MAPK pathway but significantly enhanced JNK activity. Thus, Src and Ras appear to activate two discrete pathways, JNK and MAPK. Since both Ras and Src pathways are required for differentiation to ensue, the observation of inhibition of MEK and abrogation of neurite outgrowth (43) implies that MEK is a likely convergent point at which the two pathways overlap. Synergistic induction of neurite outgrowth by v-Src and activated MEK would support the existence of a separate pathway that can cross talk at the level of MEK to support differentiation of PC12 cells. Our findings are consistent with previous work in other systems (6, 48, 53), demonstrating that aPKC can modulate MEK. Thus, aPKC acting as a MEK kinase may directly serve as a convergence point for the Ras/MAPK and Src/JNK pathways.

Overexpression of aPKC has previously been shown to modulate the JNK pathway, whereas removal of aPKC using a full-length antisense construct blocked NGF-induced JNK activity (58). Consistent with these findings, inhibition of aPKC by pharmacological manipulation likewise blocked NGF-induced activation of JNK. Thus, in PC12 cells, aPKC positions itself upstream of JNK. In addition, JNK activation by NGF is influenced by PI3K, as both wortmannin and LY294002 impaired NGF effects, which is in keeping with the ability of Src but not Ras to regulate NGF-induced JNK activity. This finding is consistent with the ability of v-Src expression to enhance activation of JNK, concomitant with formation of neurites. Although JNK has been more commonly studied in stress signaling, it is also activated by epidermal growth factor (34). Previous studies (25) revealed that overexpression of PI3K resulted in activation of JNK but not MAPK. Our findings position PI3K upstream of JNK. We conclude that Src, PI3K, aPKC, and JNK are components of a newly defined signaling cascade, consistent with the demonstrated ability of JNK to modulate neurite outgrowth (23) as well as survival and activation of NF-B (32). The adapter protein Crk is also involved in TrkA signaling (36). In keeping with the ability of v-Src to modulate JNK, JNK activation by v-Src is blocked by dominant negative mutants of Crk (13). On the basis of these findings, we propose that a Src-PI3K-aPKC-JNK-NF-B pathway exists, comprising the survival signaling cascade in PC12 cells.

While survival is Ras independent, confirming observations made by others (59), Src is required for survival responses. In keeping with Src's role in cell survival, overexpression of Src was observed to enhance cell survival in a serum-free environment to the extent that overexpression of PKC- did (58). In addition, Src overexpression resulted in constitutive activation of NF-B. Moreover, only removal of Src or aPKCs drastically promoted cell death. These results demonstrate that neuroprotection can be mimicked by NF-B activation and that cell viability is reduced by NF-B inhibition. In this regard, inhibition of PI3K blocked cell survival and NGF-induced NF-B activation. These results strongly support a pathway for NF-B activation exerted by NGF that involves PI3K, Src, and aPKC, which collaborate to provide a neuroprotective function. This finding is in keeping with constitutive activation of NF-B being capable of promoting resistance to apoptosis (15), and also the ability of NF-B activation to promote survival in other systems (21). NGF induction of the NF-B pathway is profoundly influenced by JNK compared to MAPK or p38, thus suggesting that modulation of JNK in PC12 cells may have a more direct influence on survival than would alterations in MAPK or p38. This is in fact the case, as inhibition of JNK abrogated NGF-induced B activity and likewise blocked cell survival as well as differentiation. Therefore, we hypothesize that overexpression of any element of the Src-PI3K-aPKC-JNK pathway would promote increased NF-B activity and cell survival.

Previous studies have shown that aPKC plays a critical role in B-dependent transcription. Recent findings document that aPKCs directly regulate IKK in vitro and in vivo (29). Moreover, removal of aPKC was likewise observed to block NGF-induced NF-B activation, whereas overexpression of aPKC enhanced NF-B activity and survival of PC12 cells in a serum-free environment (58). The results presented herein further define a critical role of Src in activation of NF-B and place Src, as well as PKC-, within a common signaling cascade regulating NF-B. This is in keeping with recent observations made by this lab, demonstrating that Src binds aPKC and is likewise tyrosine phosphorylated (49; Vandenplas et al., submitted). Moreover, tyrosine phosphorylation of aPKC is mediated by Src independently of PI3K (Vandenplas, submitted for publication); likewise, PI3K modulates aPKC activity independently of an effect on Src. Thus, PI3K lies in a pathway distinct from Src, although tied to regulation of NF-B through aPKC. In addition, evidence presented herein supports Src-induced modulation of aPKC as contributing to regulation of coupling between IKK/aPKC, NF-B, and cell survival. In the absence of Src activity, aPKC fails to couple with IKK and likewise blocks NF-B activation via inhibiting dissociation from IB the resting complex, resulting in a direct effect on both survival and differentiation. Thus, Src emerges as a critical regulator necessary for coupling of PKC- to the B pathway. In this regard, overexpression of Src, analogous to PKC- overexpression (58), enhances NF-B activity as well as survival responsiveness.

Association of aPKC with the B pathway in other cells occurs through p62, the aPKC binding protein (44), which localizes the PKC- to the receptor complex (47). Since Src modulates aPKC interaction with IKK, it stands to reason that Src may also play a role in directing the association of p62 with the receptor complex. Studies are under way to address this possibility. FEZ1, another recently identified aPKC binding protein (28), may serve to scaffold critical elements of the NGF receptor with the differentiation pathway, since this protein is homologous to UNC-76, a protein involved in axonal outgrowth and fasciculation in C. elegans. The differences between the two binding proteins, p62 and FEZ1, and their relationship to NGF responses warrant consideration since the scaffolding of particular signal cascades and the factors that regulate these interactions are likely to be critical to directing signal specificity. It is possible that the role of each binding protein is to scaffold aPKC to a particular signal pathway.

Src family kinases have previously been implicated in regulation of ostoeclastogenesis (1), T-cell function (14), and memory impairment (35). Similar to observations made herein, tumor necrosis factor induction of NF-B in murine bone marrow macrophages is likewise mediated by Src, through formation of a long-lived complex between IB and Src (1). In T cells, v-Src is capable of activating NF-B (14). Also, an absence of Src family member Fyn has been associated with abnormal hippocampal development, defective long-term potentiation, and impaired memory (35), processes which have been coupled to the activation of aPKCs. Thus, regulation of aPKCs by Src-mediated tyrosine phosphorylation may be common to a number of other systems.

In summary, our study reveals that aPKC, regulated by a novel Src-kinase pathway, plays a critical role in activation of NF-B. Due to the position that aPKC occupies, it plays a role in modulating both differentiation as well as survival signaling pathways. Thus, our observations are consistent with aPKC playing a critical role in regulating NGF responses in PC12 cells.