Stimulation of Protein Kinase C Modulates Insulin-like Growth Factor-1-induced Akt Activation in PC12 Cells*

Activation of protein kinase C (PKC) plays an important role in the negative regulation of receptor signaling, but its effect on insulin-like growth factor-1 (IGF-1) receptor signaling remains unclear. In this study, we characterized the intracellular pathways involved in IGF-1-induced activation of Akt and evaluated the effects of the PKC activator phorbol 12-myristate 13-ace-tate (PMA) on the Akt activation by IGF-1. IGF-1 induced a time- and concentration-dependent activation of Akt. The effect of IGF-1 was blocked by the phosphatidylinositide 3-kinase (PI3K) inhibitors LY294002 (50 m M ) and wortmannin (0.5 m M ), but not by the MEK inhibitor PD98059 (50 m M ) or the p70 S6 kinase pathway inhibitor rapamycin (50 n M ), suggesting that the stimulation of Akt by IGF-1 is mediated by the PI3K pathway. Interest-ingly, cotreatment with PMA (400 n M ) attenuated IGF-1- induced activation of Akt. The attenuation was blocked completely by the PKC inhibitor GO6983 (0.5 m M ), but only partially by the MEK inhibitor PD98059 (50 m M ), indicating that MAPK-dependent and -independent pathways are involved. PMA induced the activation of PKC in PC12 cells, and this induction was blocked by GO6983. These data further support the role of PKC in the effect of PMA. Moreover, PKC d is likely involved in the action of PMA on the basis of data obtained using isoform-specific inhibitors such as rottlerin. PMA also decreased IGF-1-induced tyrosine phosphorylation of insulin receptor substrate-1 and its association with PI3K. Taken together, these results suggest, for the first time, that stimulation of PKC modulates IGF-1-induced activation of Akt. dye-binding method with bovine serum albumin as the standard. PKC Kinase Assay— PKC activity in post-nuclear cell lysates and subcellular fractions was determined using a commercial PKC assay kit (Signa TECT TM PKC assay system) as suggested by Promega. The activity of PKC was obtained by subtracting the radioactivity of the reaction in the absence of phospholipids (control buffer) with 100 m M PKC peptide inhibitor from that of the reaction in the presence of phospholipids (activation buffer) without the PKC inhibitor. Results are presented as percent of activity from untreated cells (control). Western Blotting— Western blotting was performed as described earlier with some modifications (28). Briefly, treated cells from different experimental conditions were rinsed twice with ice-cold Hanks’ balanced buffer and lysed in either sample buffer (62.5 m M Tris-HCl, pH 6.8, 2% (w/v) SDS, 1% glycerol, 50 m M dithiothreitol, and 0.1% (w/v) bromphenol blue) or radioimmune precipitation assay buffer (50 m M Tris-HCl, pH 8.0, 150 m M NaCl, 1 m M EDTA, 1% Igepal CA-630, 0.1% SDS, 50 m M NaF, 1 m M NaVO 3, 5 m M phenylmethylsulfonyl fluoride, 10 m g/ml leupeptin, and 50 m g/ml aprotinin). Samples with equal amounts of protein were then separated by 4–20% polyacrylamide gel electro- phoresis, and the resolved proteins were electrotransferred to Hy-bond-C nitrocellulose. Membranes were incubated with 5% nonfat milk and 2% bovine serum albumin in TBST (10 m M Tris-HCl, pH 8.0, 150 m M NaCl, and 0.2% Tween 20) fo r 1 h atroom temperature and incu- bated with the appropriate primary antibody at 4 °C

Although IRS-1 is crucial for IGF-1 receptor signaling, its role in the activation of Akt remains unclear. It has recently been reported that the protein kinase C (PKC) activator phorbol 12-myristate 13-acetate (PMA) decreases insulin-induced tyrosine phosphorylation of IRS-1 and its ability to bind and activate PI3K (25)(26)(27), thus raising the possibility that activation of PKC may be involved in the regulation of Akt activation by IGF-1. In the study, using rat pheochromocytoma (PC12) cells, a model of neuronal differentiation, we have shown that IGF-1 induces the activation of Akt by a PI3K pathway and that the stimulation of PKC modulates the intracellular signaling of IGF-1 at different levels of the signaling cascade such as IRS-1 and Akt.
Cell Culture-PC12 and NIH 3T3 cells were kindly provided by Dr. Gordon Guroff (NICHD, National Institutes of Health) and cultured as described previously (28). In brief, PC12 cells, human teratocarcinoma cells (NT2), and NIH 3T3 cells were maintained in 75-cm 2 flasks in high glucose Dulbecco's modified Eagle's medium supplemented with 5% (v/v) fetal bovine serum, 5% horse serum, 100 g/ml streptomycin, and 100 units/ml penicillin (for NIH 3T3 cells, with 5% fetal bovine serum only). The cells were incubated at 37°C in a 5% CO 2 humidified atmosphere. The stock culture was routinely subcultured at a 1:5 ratio at 1-week intervals.
Treatments-Before each experiment, the cells were detached using 5 mM EDTA in Hanks' balanced buffer and seeded in 12-or 6-well plates (coated with 10 g/ml poly-D-lysine) at a density of 4 -8 ϫ 10 5 cells/well in 2% serum medium for 48 h. The culture medium was replaced with Dulbecco's modified Eagle's medium 2 h before the desired reagents were added. To study the effect of IGF-1 on the activation of Akt, cells were treated with 10 nM IGF-1 for 10 min, except for the time course study, for which 10 and 100 nM IGF-1 were used. Alternatively, cells were pretreated with wortmannin (0.5 M, 20 min), LY294002 (50 M, 20 min), rapamycin (50 nM, 20 min), GO6976 (0.5 M, 20 min), GO6983 (0.5 M, 20 min), rottlerin (5-40 M, 20 min), GF10923X (2 M, 20 min), cell-permeable myristoylated PKC peptide inhibitor (12.5-100 M, 60 min), U0126 (20 M, 30 min), and PD98059 (50 M, 40 min), followed by stimulation with 10 nM IGF-1. In the experiments involving PKC, 400 nM PMA or phorbol 12,13-didecanoate was added to the cells 2.5 min prior to IGF-1 stimulation. All experiments were carried out independently at least three to four times. Subcellular Fractions-Subcellular fractions of PC12 cells were prepared as described (29) with minor modifications. Briefly, cells grown in 30-mm culture dishes at a density of ϳ3 ϫ 10 6 cells/dish were exposed to PMA with or without different PKC inhibitors. Treated cells were immediately placed on ice and rinsed with 1 ml of pre-cooled Hank's balanced buffer. Then 0.8 ml of buffer A (20 mM Tris-HCl, pH 7.5, 10 mM KCl, 1 mM EDTA, 1 mM dithiothreitol, 50 mM NaF, 1 mM NaVO 3, 0.5 mM phenylmethylsulfonyl fluoride, 10 g/ml leupeptin, and 1 g/ml aprotinin) was added to each dish, and cells were incubated on ice for 20 min. After incubation, cells were scraped and homogenized (with 15 strokes) on ice in a pre-cooled Dounce homogenizer. Cell lysates were centrifuged at 2000 ϫ g for 10 min at 4°C; and supernatants, collected as post-nuclear cell lysates, were used to determine total PKC activity. Post-nuclear cell lysates were centrifuged at 100,000 ϫ g for 1 h, and the resulting supernatant was used as the cytosolic fraction. The pellet was rinsed with buffer A, extracted with buffer A containing 0.5% Triton X-100, and then centrifuged at 14,000 ϫ g for 10 min. The resulting supernatant was saved as the membrane fraction. Protein concentration was determined using the Bio-Red dye-binding method with bovine serum albumin as the standard.
PKC Kinase Assay-PKC activity in post-nuclear cell lysates and subcellular fractions was determined using a commercial PKC assay kit (Signa TECT TM PKC assay system) as suggested by Promega. The activity of PKC was obtained by subtracting the radioactivity of the reaction in the absence of phospholipids (control buffer) with 100 M PKC peptide inhibitor from that of the reaction in the presence of phospholipids (activation buffer) without the PKC inhibitor. Results are presented as percent of activity from untreated cells (control).
Western Blotting-Western blotting was performed as described earlier with some modifications (28). Briefly, treated cells from different experimental conditions were rinsed twice with ice-cold Hanks' balanced buffer and lysed in either sample buffer (62.5 mM Tris-HCl, pH 6.8, 2% (w/v) SDS, 1% glycerol, 50 mM dithiothreitol, and 0.1% (w/v) bromphenol blue) or radioimmune precipitation assay buffer (50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 1 mM EDTA, 1% Igepal CA-630, 0.1% SDS, 50 mM NaF, 1 mM NaVO 3, 5 mM phenylmethylsulfonyl fluoride, 10 g/ml leupeptin, and 50 g/ml aprotinin). Samples with equal amounts of protein were then separated by 4 -20% polyacrylamide gel electrophoresis, and the resolved proteins were electrotransferred to Hybond-C nitrocellulose. Membranes were incubated with 5% nonfat milk and 2% bovine serum albumin in TBST (10 mM Tris-HCl, pH 8.0, 150 mM NaCl, and 0.2% Tween 20) for 1 h at room temperature and incubated with the appropriate primary antibody at 4°C overnight. The membranes were then washed twice with TBST and probed with horseradish peroxidase-conjugated anti-rat/mouse/rabbit/goat secondary antibody at room temperature for 1 h. Membranes were finally washed several times with TBST to remove unbound secondary antibodies and visualized by the ECL detection kit (Amersham Pharmacia Biotech, Ontario, Canada). A part of the SDS gel was stained with Coomassie Blue to ensure the use of equal amounts of proteins. The phosphorylation of Akt/MAPK/PKC␦ was determined by Western blotting with anti-phospho-Akt, anti-phospho-ERK, or anti-phospho-PKC␦ antibody. Blots were subsequently stripped and reprobed with anti-Akt/anti-ERK/anti-PKC␦ antibody to make sure that equal amounts of Akt/ERK/ PKC␦ were present. Quantification of the blots was performed using the MCID image analyzer system (Imaging Research Inc., Ontario, Canada). The phosphorylation of Akt was normalized to the amounts of Akt present in each band, and IGF-1-stimulated Akt phosphorylation was calculated as the level of Akt phosphorylation in IGF-1-stimulated cells versus untreated cells. For experiments on the subcellular redistribution of PKC stimulated by PMA, subcellular fractions were prepared as described above, and the presence of each PKC isoenzyme was detected by Western blotting with the corresponding anti-PKC isoenzyme antibodies.
Determination of Tyrosine Phosphorylation of the IGF-1 Receptor and IRS-1 and Their Interactions with PI3K by Immunoprecipitation-PC12 cells were treated with 10 nM IGF-1 for 8 min with or without 400 nM PMA and rinsed with cold phosphate-buffered saline. After centrifugation at 1000 ϫ g for 5 min at 4°C, cell pellets were lysed on ice in cold pre-radioimmune precipitation assay buffer for 20 min. Cell lysates were then pelleted at 13,000 ϫ g for 10 min, and the concentration of protein in each sample was determined using the Bio-Red dye-binding method with bovine serum albumin as the standard. The supernatant with an equal amount of protein was incubated overnight at 4°C with anti-IGF-1 receptor, anti-IRS-1, or anti-PI3K antibody. Formed immunocomplexes were isolated by protein A/G PLUS-agarose (Santa Cruz Biotechnology) and separated by 4 -20% SDS-polyacrylamide gel electrophoresis, and then tyrosine phosphorylation was determined by Western blotting with a mixture of anti-phosphotyrosine antibodies 4G10 and PY99. Blots were stripped and reprobed with anti-PI3K or anti-IRS-1 antibody to evaluate the interaction of the IGF-1 receptor and IRS-1 with PI3K. Finally, the blots were reprobed with anti-IGF-1 receptor, anti-IRS-1, or anti-PI3K antibody to ensure the presence of equal amounts of proteins. Quantification of the blots was performed as described above, and the amount of p85 PI3K associated with IRS-1 or tyrosine phosphorylation of IRS-1 was normalized to the total protein of p85 PI3K or IRS-1 present in each lane.
Akt in Vitro Kinase Assay-The Akt kinase assay was performed as recommended by New England Biolabs Inc. with some modifications. Briefly, cells were treated with IGF-1 with or without PMA, and Akt was separated by immunoprecipitation with anti-Akt antibody as described above. The immunoprecipitates were then washed four times with radioimmune precipitation assay buffer and once with kinase buffer (25 mM Tris-HCl, pH 7.5, 5 mM ␤-glycerophosphate, 2 mM dithiothreitol, 0.1 mM Na3VO 4 , and 10 mM MgCl 2 ). Subsequently, an in vitro kinase reaction was carried out in 40 l of kinase buffer containing precipitated Akt, 200 M ATP, and 0.2 g of GSK3␣ fusion protein as substrate. Following a 30-min incubation at 30°C, the reaction was stopped by addition of 10 l of 5ϫ reduced SDS sample buffer. Akt activity was then determined by Western blotting by measuring the phosphorylation of GSK3␣ fusion protein with anti-phospho-GSK3␣/␤ antibody. The blots were quantified by scanning the bands of phosphorylated GSK3␣ fusion protein as described above, and IGF-1 stimulation was calculated as the level of GSK3␣ fusion protein phosphorylated in IGF-1-stimulated cells versus untreated cells.
Statistical Analysis-Data are expressed as the means Ϯ S.E. A one-way analysis of variance with the Newman-Keul test was used to calculate differences among means, and p Ͻ 0.05 was considered significant.

IGF-1 Induces a Time-and Concentration-dependent Activation of Akt in PC12
Cells-IGF-1 exhibits anti-apoptotic function by activating Akt kinase in both neural and non-neural cells (14,15). Activation of Akt requires the phosphorylation of Thr-308 and Ser-473 in the Akt␣ molecule (16,17). In this study, phosphorylation of Ser-473 was used to evaluate the activation of Akt. PC12 cells were treated with 100 nM IGF-1 for different times (0 -80 min) or with various concentrations of IGF-1 (0.33-100 nM) for 10 min. IGF-1 induced a time-and concentration-dependent activation of Akt kinase in PC12 cells. At 100 nM IGF-1, Akt activation was evident at 2.5 min, peaked at ϳ5 min, and remained stable for at least 80 min (Fig. 1, A and B). Similar results was obtained by treatment with 10 nM IGF-1 (Fig. 1E). Treatment of PC12 cells with 1.0 -100 nM IGF-1 caused a 3-6-fold increase in Akt phosphorylation at Ser-473. Activation of Akt was observed at a minimal concentration of 1.0 nM IGF-1 and reached maximal levels at ϳ10 nM IGF-1 (Fig. 1, C and D).
Activation of Akt by IGF-1 Is Mediated by a PI3K Pathway-After establishing that IGF-1 can activate Akt, we studied the signaling pathways mediating the action of IGF-1. PC12 cells pretreated with different kinase inhibitors were stimulated with 10 nM IGF-1 for 10 min, and then phosphorylation of Akt and MAPKs was evaluated by Western blotting using antiphospho-Akt or anti-phospho-ERK antibody. 10 nM IGF-1 caused a 3-6-fold increase in the phosphorylation of Akt and MAPKs (Fig. 2, A and B, sixth lane versus first lane). Pretreatment of the cells with the PI3K inhibitors LY294002 (50 M) and wortmannin (0.5 M) blocked IGF-1-induced activation of Akt ( Fig. 2A, seventh and eighth lanes versus sixth lane). Inhibitory effects were concentration-dependent (Fig. 3) and specific, as neither LY294002 nor wortmannin altered IGF-1-induced activation of MAPK (Fig. 2B, seventh and eighth lanes versus sixth lane). In contrast to the PI3K inhibitors, the MEK inhibitor PD98059 (50 M), an upstream inhibitor of MAPK, did not influence the action of IGF-1 on Akt phosphorylation ( Fig. 2A, ninth lane versus sixth lane), but blocked IGF-1-

FIG. 1. IGF-1 induces a timedependent (A, B, and E) and concentrationdependent (C and D) activation of Akt in PC12 cells.
Cell were treated with 10 nM (E) and 100 nM (A and B) IGF-1 for various times or with different concentrations of IGF-1 for 10 min (C and D), and then the activation of Akt was determined by Western blotting using anti-phospho-Akt antibody. The blots represent prototypical examples of experiments that were replicated at least three times. Quantification data for A and C are shown in the histograms in B and D, respectively. CTL, control.

FIG. 2. IGF-1-induced activation of Akt in PC12 cells is dependent on PI3K.
Following treatment with different kinase inhibitors, PC12 cells were exposed to 10 nM IGF-1, and then phosphorylation of Akt/ERK was determined by Western blotting with anti-phospho-Akt/ ERK antibody. As evident from the blots, the PI3K inhibitors LY294002 and wortmannin inhibited IGF-1-induced activation of Akt (A), but not the activation of MAPK (B). The blots represent prototypical examples of experiments that were replicated three to four times.

FIG. 3. Concentration-dependent effects of the PI3K inhibitors LY294002 (A) and wortmannin (B) on IGF-1-induced Akt activation.
Following pretreatment with different concentrations of LY294002 or wortmannin, PC12 cells were stimulated with 10 nM IGF-1, and the activation of Akt was determined. Each inhibitor, as evidenced from the blots, blocked IGF-1-induced activation of Akt in a concentration-dependent manner. The blots represent prototypical examples of experiments that were replicated at least three times.
induced activation of MAPK (Fig. 2B, ninth lane versus sixth lane). Moreover, the p70 S6 kinase inhibitor rapamycin (50 nM) failed to alter IGF-1-induced Akt or MAPK activation (Fig. 2, A and B, tenth lane versus sixth lane). These data suggest that PI3K is located upstream of Akt in the IGF-1 receptor-induced intracellular signaling pathway.
Protein Kinase C Activator PMA Attenuates IGF-1-induced Activation of Akt-Previous reports have shown that the PKC activator PMA attenuates insulin-induced IRS-1 tyrosine phosphorylation and its association with PI3K by activating PKC/ MAPK (25)(26)(27). To evaluate the role of PKC in IGF-1-induced activation of Akt, PC12 cells were treated with 10 nM IGF-1 with or without 400 nM PMA. IGF-1, as expected, induced a 5-10-fold increase in Akt phosphorylation (Fig. 4A, third lane versus first lane). Cotreatment of the cells with IGF-1 and PMA caused a 55 Ϯ 6% decrease in IGF-1-induced Akt phosphorylation (Fig. 4A, fourth lane versus third lane), whereas PMA itself had no effect on the activation of Akt (second lane versus first lane). Similar results were obtained using another PKC activator, phorbol 12,13-didecanoate (data not shown). PMAinduced attenuation of Akt activation was concentration-dependent (Fig. 4E), with a minimal effect seen at 100 nM and reaching the maximal level at 400 nM PMA. Consistent with this result, an in vitro kinase assay with GSK3␣ fusion protein as substrate also showed that PMA decreased IGF-1-induced activation of Akt by 43 Ϯ 5% in PC12 cells (Fig. 4B, fourth lane versus third lane). In contrast to the Akt activation, PMA treatment did not alter IGF-1-induced CREB phosphorylation, thus indicating that PMA selectively affects Akt (but not CREB) phosphorylation induced by IGF-1 (30).
PMA Decreases IGF-1-induced Tyrosine Phosphorylation of IRS-1 and Its Association with PI3K-Having established that PMA is capable of attenuating IGF-1-induced activation of Akt, we investigated next whether this is the consequence of decreased IGF-1-induced tyrosine phosphorylation of IRS-1 and its association with PI3K. PC12 cells were treated with IGF-1 and PMA, and IRS-1 was separated by immunoprecipitation with antibodies to IRS-1. Tyrosine phosphorylation of IRS-1 and its association with PI3K were evaluated by Western blotting using anti-phosphotyrosine or anti-PI3K antibody. IGF-1 caused tyrosine phosphorylation of IRS-1 and its association with PI3K (Fig. 5A, third lane versus first lane). Cotreatment with PMA decreased IGF-1-induced IRS-1 tyrosine phosphorylation (25%) and PI3K (43%) associated with IRS-1 induced by IGF-1 (Fig. 5A, fourth lane versus third lane).
To confirm this effect of PMA, PC12 cells were treated with IGF-1 and PMA, and p85 PI3K was separated by immunoprecipitation with anti-PI3K antibody. Tyrosine phosphorylation of IRS-1 and its association with PI3K were determined by Western blotting using anti-phosphotyrosine or anti-IRS-1 antibody. Consistent with earlier results, PMA decreased tyrosine phosphorylation of IRS-1 and its association with PI3K (Fig.  5B, fourth lane versus third lane). IGF-1 also induced some association of the IGF-1 receptor with PI3K (5-10% of that of IRS-1), which seemed to be unaffected by PMA treatment (data not shown).
PMA Activates MAPK and Attenuates IGF-1-induced Akt Activation-It has been reported that PMA regulates insulin signaling by activating a PKC/MAPK pathway (25)(26)(27). Accordingly, the effects of PMA and IGF-1 on the activation of MAPK and Akt were investigated. Both IGF-1 and PMA caused an increase in the phosphorylation of MAPK, whereas the activation of Akt was induced only by IGF-1 (Fig. 6, second and third  lanes versus first lane). Interestingly, coexposure to PMA and IGF-1 increased MAPK activation, but decreased the phosphorylation of Akt (Fig. 6, fourth lane versus third lane). To establish further a role of PKC/MAPK, PC12 cells were pretreated with a PKC inhibitor, GO6983 (31), and a MEK inhibitor, PD98059 (32). GO6983 blocked the effect of PMA on the activation of Akt and MAPK induced by IGF-1 (Fig. 7A, fourth lane  8A) or for 10 min with or without pretreatment with GO6983 (Fig. 8, B and C), and post-nuclear cell lysates (Fig. 8, A and B) and the membrane fraction (Fig. 8C) were prepared. PKC activity was measured as described under "Experimental Procedures." 400 nM PMA time-dependently activated PKC in PC12 cells, with an increase in PKC activity observed at 2.5 min, peaking at 5 min, and remaining stable for at least 20 min (Fig.  8A). Fig. 8B shows that the activation of PKC by PMA was blocked by 0.5 M GO6983. As shown in Fig. 8C, PMA also induced PKC activity, an effect reversed by GO6983 in the membrane fraction.
PMA Induces the Subcellular Redistribution of PKC Isoenzymes-PKC is composed of a family of at least nine related isoforms that can be broadly divided into three groups. Members of the first group of isoenzymes (␣, ␤I, ␤II, and ␥) are activated by Ca 2ϩ , phosphatidylserine, and diacylglycerol and are referred to as conventional PKCs. Members of the second group (␦, ⑀, , and ) are stimulated by diacylglycerol and phosphatidylserine, but do not require Ca 2ϩ for activation, and are referred to as unconventional PKCs. Members of the last group ( and ) are activated by phosphatidylserine alone and thus are considered to be atypical PKCs (35). In the PC12 cells that we used, PKC␣, -␦, -⑀, and -are expressed (28). The effects of PMA treatment on the subcellular redistribution of PKC␣, -␦, -⑀, and -were studied next to establish further the activation of PKC and to obtain data on the various isoforms that may involved in the effect of PMA. Therefore, PC12 cells were treated with 400 nM PMA for 10 min, and the distribution of various PKC isoforms was detected as described above. Fig. 9 shows that in response to our PMA treatment, all studied isoforms translocated to the membrane fraction, albeit with different extents (⑀ Ͼ ␦ Ͼ ␣ and ), suggesting that these various isoforms were all activated by PMA. These data are consistent with an early study showing that PMA time-dependently induced membrane translocation of all PKC isoforms in PC12 cells (34).
PMA Down-regulation Inhibits the Effect of PMA on the Activation of Akt Induced by IGF-1-Knowing that PMA can activate various isoforms of PKC in PC12 cells, we next studied their respective role in the effect of PMA. As PKC is known to be resistant to PMA down-regulation (29,34), this approach was used first. PC12 cells were treated with 1 M PMA for 24 h, and the residual effect of an acute PMA treatment (400 nM) on the activation of Akt by IGF-1 was determined. Fig. 10A shows that a 24-h exposure to a high concentration of PMA attenuated the subsequent effect of PMA on the activation of Akt induced by IGF-1. This result suggests that PMA-sensitive PKC isoenzymes are involved. Consistent with this, GF10923X, a PKC isoenzyme inhibitor of PKC␣, -␤, -␥, -␦, and -⑀ (36,37), inhibited the effect of PMA (Fig. 10B).
Rottlerin, a PKC␦-specific Inhibitor, Blocks the Effect of PMA on Akt-To identify the individual isoenzyme that may be involved in the effect of PMA, different PKC isoenzyme inhibitors were used. GO6983 (inhibitor of PKC␣, -␤, -␥, -␦, and -) blocked the effect of PMA on the activation of Akt induced by IGF-1, whereas the Ca 2ϩ -dependent PKC isoform inhibitor GO6976 (inhibitor of PKC␣, -␤, and -␥) (38) had no effect (Fig.  10B). These data, along with the resistance of PKC to downregulation, suggest that PKC␦ is likely to be the isoform that mediates the effect of PMA. This hypothesis is supported further by the finding that rottlerin, an inhibitor of PKC␦ (39), with an IC 50 of 3-6 M for PKC␦ and much lower affinities for other isoforms, inhibited the effects of PMA on the activation of Akt induced by IGF-1 in a concentration-dependent manner (5-20 M) (Fig. 10C). Table I provides a summary of the above  results. PMA Time-dependently Induces the Phosphorylation of PKC␦ in PC12 Cells-Having established that PKC␦ may be involved in the effect of PMA on the activation of Akt, we studied the activation of PKC␦ stimulated by PMA using Western blotting with an anti-phospho-PKC␦ antibody that specifically recognizes Thr-505 in the active loop of PKC␦. PMA time-dependently induced the phosphorylation of PKC␦ in PC12 cells (Fig.  11A), whereas the activation of PKC␦ induced by PMA was blocked by PMA down-regulation (Fig. 11B).
PMA Inhibits IGF-1-induced Akt Activation in NIH 3T3 and NT2 Cells-To establish whether PMA, apart from PC12 cells, can attenuate IGF-1-induced activation of Akt in other cell lines, NIH 3T3 and NT2 cells were treated with 10 nM IGF-1 in the presence or absence of 400 nM PMA, and the activation of Akt was measured as described above. PMA also inhibited IGF-1-induced Akt activation in both cell lines (Fig. 12). DISCUSSION In this study, we have shown that IGF-1 stimulates the activation of Akt in PC12 cells via a PI3K pathway. Additionally, we have provided evidence that PMA, an activator of PKC, negatively modulates IGF-1 receptor signaling, including the activation of Akt.
It was reported earlier that IGF-1 protects PC12 cells from apoptosis by activating the PI3K pathway (40,41). However, the effect of IGF-1 on the activation of Akt was not known. We have shown here that IGF-1 activated Akt in a time-and concentration-dependent manner. The effect of IGF-1 was found to be inhibited by the PI3K inhibitors wortmannin (42) and LY294002 (43), thus indicating that the PI3K is located upstream of Akt. This is supported by earlier reports that have shown that overexpression of a dominant-negative PI3K mutant or mutation of the tyrosine residues in the platelet-derived growth factor receptor that bind PI3K inhibited the activation

TABLE I Effects of PKC isoform-specific inhibitors and PMA down-regulation on PKC isoforms and the inhibitory effect of PMA on the activation of Akt (IEP) in PC12 cells
Ϫ ϩ a A plus sign indicates that the treatment inhibited isoform activity and the inhibitory effect of PMA on the activation of Akt (IEP) whereas a minus sign indicates a lack of effect. of Akt (44,45). Interestingly, in contrast to PI3K inhibitors, neither the MEK inhibitor PD98059 (32) nor the p70 S6 kinase inhibitor rapamycin (46) affected the activation of Akt induced by IGF-1. These results suggest that IGF-1-induced activation of Akt, as reported for other growth factors, is mediated by the PI3K (but not the MAPK) pathway (42,44,45,47).
PMA, an activator of protein kinase C, was found to attenuate IGF-1-induced stimulation of Akt in several cell lines, including PC12 and NT2 cells. It has been reported earlier that PKC modulates insulin receptor signaling by interacting with the receptor substrate IRS-1 via Ser/Thr phosphorylation (25-27, 48, 49). In this study, IGF-1 was shown to induce tyrosine phosphorylation of the IGF-1 receptor and IRS-1 as well as their association with PI3K. However, the amount of PI3K associated with the IGF-1 receptor was found to be significantly less compared with that associated with IRS-1, suggesting that the PI3K/Akt kinase pathway is mediated primarily by IRS-1 in PC12 cells. This is supported by the evidence that PMA, which reduced IGF-1-induced Akt activation, attenuated PI3K association with IRS-1, but not with the IGF-1 receptor. These data are consistent with recent reports that indicate that PMA does not affect insulin receptor tyrosine phosphorylation or its kinase activity, but reduces tyrosine phosphorylation of IRS-1 and its association with PI3K (26,50).
The IRS-1 molecule contains at least 30 Ser/Thr residues that could be phosphorylated by different Ser/Thr kinases (25,26,(51)(52)(53). Evidence suggests that four serine residues are located in the motif Pro-X-Ser-Pro, a consensus sequence recognized by members of the MAPK family (54). All four of these serines are located exactly four amino acids downstream of potential tyrosine phosphorylation sites, and one (Ser-612) is close to the major PI3K-docking site, Tyr-608. It was reported earlier that PKC can phosphorylate Ser-612 possibly by stimulating MAPK activity, leading to the inhibition of receptor signaling mediated via the IRS-1/PI3K pathway (25,27). If this is indeed the case, it is likely that the negative regulation of IGF-1-induced Akt activation by PMA, as observed in this study, is mediated via MAPK-induced phosphorylation of Ser-612. This event leads to the subsequent inhibition of phosphorylation of tyrosine 608 in IRS-1 by the receptor kinase and then decreases the ability to activate PI3K signaling. This is supported by two lines of evidence. First, the MEK kinase inhibitors PD98059 and U0126 attenuated the effects of PMA on IGF-1-induced activation of Akt; and second, PMA reduced IGF-1-induced tyrosine phosphorylation of IRS-1 and its association with PI3K. However, based on the evidence that MEK kinase inhibitors, unlike PKC inhibitors, only partially block the effect of PMA, it is likely that MAPK-dependent and -independent pathways are involved in the negative regulation by PMA of IGF-1-induced Akt activation.
A plethora of experimental approaches have shown that Akt can be activated via at least two pathways, namely PI3K-dependent and -independent pathways (15,(55)(56)(57). Growth factor-induced signals leading to Akt activation are transduced via the PI3K pathway, whereas stress-, okadaic acid-, and cAMP-induced activation of Akt is PI3K-independent. Recently, it has been shown that the ␣, ␦, and isoforms of PKC are able to interact directly with Akt most likely by an association with the pleckstrin homology domain of Akt (56). Furthermore, PKC has been shown to negatively regulate Akt activity under in vivo conditions (58). These findings raise the possibility that PMA-stimulated PKC, apart from its effect on MAPK/IRS-1, is able to negatively regulate Akt activity in PC12 cells. As all these isoforms of PKC are expressed in the PC12 cells used (28), it was deemed important to establish if one isoform was more specifically involved. PMA down-regulation blocked the effect of PMA on IGF-1-induced activation of Akt. Since PKC is resistant to PMA down-regulation (28,34), it is thus unlikely that this isoform is involved. This hypothesis is supported further by the facts (a) that GF10923X (blocker of PKC␣, -␤, -␥, -␦, and -⑀) inhibited the effect of PMA and (b) that the cell-permeable myristoylated PKC peptide inhibitor at concentrations of 25-100 M had no effect on PMA regulation of IGF-1-induced activation of Akt. In fact, our data using various isoform inhibitors suggest a preferential role for PKC␦ in our system. Indeed, this hypothesis is supported by the following results. (a) PMA activated this isoform, as it induced its phosphorylation and subcellular redistribution; (b) PMA-induced down-regulation deleted this isoform and blocked PMA-induced phosphorylation of PKC␦ and the effect of PMA on Akt; (c) PKC isoform-specific inhibitors (GO6983 and GF10923X) with the reported ability to inhibit PKC␦ blocked the effect of PMA, whereas GO6976, a PKC inhibitor with no effect on PKC␦, was much less effective; and (d) rottlerin, a rather selective PKC␦ inhibitor, significantly attenuated the effect of PMA. Taken together, these results suggest that PKC␦ is involved in the effect of PMA on Akt activation by IGF-1 and are consistent with a very recent study showing opposing effects on cell apoptosis between overexpression of Akt and PKC␦ in COS cells (59). However, a minor participation of other isoforms cannot be totally excluded, as rottlerin did not fully block the effect of PMA.
Although the underlying mechanism remains to be established, it is suggested that PKC may phosphorylate Akt, leading to the inhibition of further phosphorylation of Akt by PDK1/2 and its kinase activity (58). A similar mechanism has FIG. 12. Effects of PMA on IGF-1-induced activation of Akt in NT2 and NIH 3T3 cells. NT2 and 3T3 cells were treated with 10 nM IGF-1 with or without 400 nM PMA, and then activation of Akt was determined as described under "Experimental Procedures." As in PC12 cells, PMA attenuated the activation of Akt induced by IGF-1 in both NT2 and 3T3 cells. been reported for the regulation of Btk activity by PKC (60). In this scenario, it is possible that inhibition of IGF-1-induced Akt stimulation by PMA, as observed in the present study, is mediated, at least in part, by a direct interaction between PKC and Akt. As PKC inhibited constitutively activated Akt in COS cells, it is unlikely that the effect of PKC is being mediated by a structural hindrance of the phosphorylation sites of PDK1/2 (58). Alternately, it is possible that the phosphorylation of Akt by PKC may change its conformation, which subsequently inactivates the kinase and blocks further phosphorylation of Akt by PDK1/2. Since little is known regarding the negative regulation of the Ser/Thr kinase involved in Akt phosphorylation, it remains to be established if it is indeed the mechanism by which PKC can inhibit the activation of Akt.
Our results indicate that PMA negatively regulates IGF-1induced Akt activation by stimulating MAPK-induced phosphorylation of Ser-612 in IRS-1 and also possibly by a direct action of PKC on the Akt kinase. Prior studies have shown that the negative regulation of insulin receptor signaling by PKC involves the activation of MAPK and the phosphorylation of Ser-612 in IRS-1 (25)(26)(27). Thus, it appears that an activator of PKC is able to inhibit IGF-1 and insulin receptor signaling by overlapping as well as non-overlapping signaling pathways. Apart from PKC, a number of other factors including plateletderived growth factor and endothelin-1 have been shown to inhibit the ability of the insulin receptor to tyrosine-phosphorylate IRS-1 by serine phosphorylation (25)(26)(27)51). Interestingly, endothelin-1, an activator of PKC, has been reported to regulate insulin signaling via the MAPK pathway, whereas platelet-derived growth factor acts via a distinct mechanism, possibly by the activation of a serine kinase modulated by the Akt pathway that requires serine phosphorylation of distinct residues (Ser-632, Ser-662, and Ser-731) in the IRS-1 molecule (27). More recently, tumor necrosis factor-␣ has been shown to inhibit IGF-1 receptor signaling by suppressing the phosphorylation of IRS-2 and the subsequent inhibition of PI3K activity (61,62). Given the established role of Akt in survival-linked signaling pathways, the negative regulation of IGF-1 receptor signaling by PKC may have important functional consequences not only during normal development, but also in certain neurodegenerative disorders such as amyotrophic lateral sclerosis, Alzheimer's disease, and cerebral ischemia. For example, it was recently reported that IGF-1 can protect/rescue neurons against ␤-amyloid toxicity (63, 64) as well as ischemic events induced by the occlusion of cerebral arteries. Under such circumstances, it will be important to establish whether a shift in the balance between IGF-1 and PKC stimulators regulating Akt activation plays a significant role in determining the survival of neurons, thus providing an opportunity to assess the potential therapeutic relevance of IGF-1 or its mimetics in the treatment of neurodegenerative disorders.
In summary, we have shown that the activation of Akt by IGF-1 is mediated by a PI3K (but not a MAPK) pathway. Additionally, PMA, an activator of PKC, negatively regulates IGF-1-induced Akt activation, most likely by stimulating MAPK-induced phosphorylation of Ser-612 in IRS-1 as well as by a direct action of PKC on Akt kinase. Taken together, these data provide evidence that PKC activators can have a direct effect on IGF-1 receptor signaling.