Insulin-like Growth Factor I-mediated Activation of the Transcription Factor cAMP Response Element-binding Protein in PC12 Cells

IGF-I is known to support growth and to prevent apoptosis in neuronal cells. Activation of the nuclear transcription factor cAMP response element-binding protein (CREB) has emerged as a central determinant in neuronal functions. In the present investigation, we examined the IGF-I-mediated phosphorylation and transcriptional activation of CREB in rat pheochromocytoma (PC12) cells, a cellular model for neuronal differentiation, and defined three distinct postreceptor signaling pathways important for this effect including the p38 mitogen-activated protein kinase (MAPK) pathway. CREB phosphorylation at serine 133 and its transcriptional activation as measured by a CREB-specific Gal4-CREB reporter and the neuroendocrine-specific gene chromogranin A was induced 2–3.3-fold by insulin-like growth factor (IGF)-I. This activation was significantly blocked (p < 0.001) by the dominant negative K-CREB or by mutation of the CRE site. IGF-I stimulated chromogranin A gene expression by Northern blot analysis 3.7-fold. Inhibition of MAPK kinase with PD98059, PI 3-kinase with wortmannin, and p38 MAPK with SB203580 blocked IGF-I-mediated phosphorylation and transcriptional activation of CREB by 30–50% (p < 0.001). Constitutively active and dominant negative regulators of the Ras and PI 3-kinase pathways confirmed the contribution of these pathways for CREB regulation by IGF-I. Cotransfection of PC12 cells with p38β and constitutively active MAPK kinase 6 resulted in enhanced basal as well as IGF-I-stimulated chromogranin A promoter. IGF-I activated p38 MAPK, which was blocked by the inhibitor SB203580. This is the first description of a p38 MAPK-mediated nuclear signaling pathway for IGF-I leading to CREB-dependent neuronal specific gene expression.

Insulin-like growth factor-I (IGF-I) 1 is a growth-promoting polypeptide with diverse cellular functions. IGF-I is involved in the growth and differentiation of various cell types such as muscle and adipocytes (1)(2)(3). Although liver is the primary source of circulating IGF-I, significant expression of this growth factor is seen in various tissues including brain, where it is known to exert autocrine and paracrine functions (4,5). IGF-I has been shown to stimulate neurite outgrowth and promote survival of neurons in culture (4,5).
IGF-I exerts its cellular effects through its type I IGF receptor which resembles the insulin receptor in structural as well as functional aspects (reviewed in Refs. 6 and 7). This heterotetrameric transmembrane glycoprotein consists of two ␣and two ␤-subunits. The ␤-subunit has intrinsic tyrosine kinase activity that is stimulated when IGF-I binds to the ␣-subunits. The receptor tyrosine kinase in turn phosphorylates intracellular substrates such as insulin receptor substrates 1 and 2 and Shc (7,8). The tyrosine phosphorylation sites on these docking proteins recruit Src homology 2-containing proteins such as Grb2, Nck, Crk, SHP2, and the p85 subunit of PI 3-kinase. From this intermediary complex of signaling proteins, two significant pathways emerge. One pathway activates extracellular signal-regulated kinase 1/2 (ERK1/2) through Ras/Raf/MEK, and another pathway proceeds through PI 3-kinase. IGF-I has been studied extensively in the PC12 cell line, a model of neuronal tissue. In these cells, IGF-I promotes growth and proliferation, primarily via activation of the ERK pathway (9). For the prevention of apoptosis, IGF-I requires the PI 3-kinase pathway (10).
One of the common nuclear targets of tyrosine kinase signaling cascades is CREB, the Ca 2ϩ /cyclic AMP response elementbinding protein. CREB is a 43-kDa nuclear transcription factor belonging to the CREB/ATF family (11). Activation of CREB by forskolin, a potent stimulator of cAMP, stimulates PC12 cell differentiation to a sympathetic neuron-like phenotype with neurite extension (12,13). NGF and IGF-I regulation of CREB is essential for neuronal plasticity, full axonal development, memory consolidation, and neuroprotection (14 -20). IGF-I is known to regulate a number of CREB response element (CRE)containing genes including bcl-2 and c-fos (21,22). CREB is constitutively expressed, and it binds to the specific sequence, 5Ј-TGACGTCA-3Ј known as CRE. Phosphorylation on the serine 133 residue of CREB increases its transcriptional activity. This phosphorylation does not alter the binding of CREB to CRE, but it increases its association with adapter proteins such as CREB-binding protein, leading to the activation of transcriptional machinery. CREB was initially identified as a substrate for PKA and a mediator of cAMP-regulated gene expression (23). Later studies showed that CREB can be phosphorylated and activated by multiple signaling pathways including ERK, protein kinase C, calcium/calmodulin-dependent protein kinases, and p38 MAPK (12, 24 -26). Thus, diverse signaling pathways, many of which are activated by IGF-I, are capable of regulating this transcription factor, which plays a role in neuronal growth and survival.
Chromogranin A is an acidic glycoprotein present in secretory granules of the neuroendocrine system (27). The promoter region of this gene has a conserved CRE site, which is essential for transactivation in PC12 cells (28). Chromogranin A, being a physiologically relevant CRE-dependent gene, can serve as a read-out for IGF-I-mediated gene regulation in PC12 cells. We recently demonstrated an ERK-dependent CREB activation by insulin in Hep-G2 and 3T3-L1 cell lines (29). Since CREB is capable of regulating many important functions in neuronal tissues, we investigated whether CREB was important for IGF-I-mediated gene regulation in PC12 cells. The objectives of this investigation were to (a) examine whether IGF-I stimulates the phosphorylation and transcriptional activation of the nuclear transcription factor CREB in PC12 cells, (b) gain insight into the mechanism by which IGF-I-mediated signal transduction pathways lead to the activation of CREB, and (c) assess the impact of IGF-I on the neuronal specific CRE-dependent gene chromogranin A.
We demonstrate that IGF-I increases CREB serine 133 phosphorylation and transcriptional activation of CREB reporter systems and the neuronal specific gene chromogranin A through at least three pathways: PI 3-kinase, MEK/ERK, and p38 MAPK. The novel finding of this study is the contribution of the p38 MAPK-mediated signaling pathway to the regulation of CREB-dependent gene expression by IGF-I.

EXPERIMENTAL PROCEDURES
Materials-PD98059, wortmannin, and rapamycin were purchased from Biomol (Plymouth Meeting, PA). SB203580 and SB202190 were obtained from Calbiochem. Cell culture media and supplies were from Life Technologies, Inc. and Gemini Bio Products, Inc. (Calabasa, CA). The plasmid for the expression of the chimeric protein (Gal4-CREB) consisting of the DNA binding domain of Gal4 and the transactivation domain of CREB and the expression vector for Gal4-CREB protein with serine to alanine substitution at position 133 were a generous gift from Dr. William J. Roesler (University of Saskatchewan, Saskatoon, Canada). An expression vector for the luciferase reporter gene driven by the enhancerless thymidine kinase (TK) promoter linked to four copies of Gal4 regulatory sequence (pGal 4 -TK-Luc) was provided by Dr. James Hoeffler (Invitrogen, San Diego, CA). Three constructs of mouse chromogranin A promoter linked to luciferase in the promoterless luciferase reporter vector pXP1 were provided by Dr. Daniel O'Connor (San Diego, CA). The full-length promoter pXP1133 contained 1133 bp in the 5Јflanking region. The CRE-containing truncated promoter, which maintains the minimal neuroendocrine specificity, is in pXP77; the CREmutated version of pXP77 is pXPM41. Constitutively active and dominant negative Ras and Raf-1 were obtained from Arthur Gutierrez-Hartmann (University of Colorado Health Sciences Center, Denver, CO) and Ulf Rapp (Strathlenkunde, Germany). For the PI 3-kinase, SR␣-wild type p85, and SR␣-⌬p85 were provided by Dr. Masato Kasuga (Kobe, Japan). The constitutively active form of MAPK kinase 6 was obtained from Joel Raingeaud (Institut Curie, Orsay, France), and the pcDNA3-p38␤ was provided by Jiahuai Han (San Diego, CA). The luciferase assay kit was purchased from Analytical Luminescence Laboratory (San Diego, CA). Antibodies specific for CREB, P-CREB (Ser 133 ), and phospho-ATF-2 and the ATF-2 fusion protein were obtained from New England Biolabs (Beverly, MA). Antibody to p38 MAPK (C-20) was obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA), and the dual phosphorylation site-specific antibody to p38 MAPK was a gift from Dr. Eric Schaefer (Promega). Plasmids for transfection experiments were purified using Qiagen's (Valencia, CA) Maxi kit. All other fine chemicals were purchased from Sigma.
Cell Culture-Rat pheochromocytoma (PC12) cells (provided by Dr. Gary Johnson (Denver, CO) and Drs. Derek LeRoith and Marcelina Parrizas (NIDDK, National Institutes of Health, Bethesda, MD) were maintained in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum, 5% heat-inactivated horse serum, 100 g/ml streptomycin, and 100 microunits/ml penicillin at 37°C. The cells were cultured in 60-mm dishes for immunoblotting experiments and in 6 ϫ 35-mm wells for transfection studies. Medium was changed every second day. Confluent cell cultures were split 1:4 and used for the experiments 4 days later. The cells were fasted for 5 h by maintaining them in the medium containing 0.1% fetal bovine serum and 0.05% heatinactivated horse serum before treatment with growth factors and other agents in the experiments for measuring CREB phosphorylation. The stock solutions of pharmacological inhibitors such as PD98059, wortmannin, and SB203580 were prepared in Me 2 SO at a concentration of 1000-fold, so that when they were added to the culture medium, the concentration of Me 2 SO was below 0.1%.
Immunoblotting-Immunoblotting for CREB, phospho-CREB, p38 MAPK, and dual phospho-p38 MAPK were carried out as described previously (29,30). PC12 cells cultured in 60-mm dishes were fasted for 5 h before each experiment. After preincubation with inhibitors for 30 min and incubation with growth factors for appropriate duration, the cells were washed twice with ice-cold PBS, and total cell lysates were prepared by scraping the cells with 200 l of 1ϫ Laemmli sample buffer containing 100 mM dithiothreitol. The proteins were resolved on 12% SDS-polyacrylamide gels and transferred to polyvinylidene difluoride membranes. The blots were blocked with TBST (20 mM Tris-HCl, pH 7.9, 8.5% NaCl, and 0.1% Tween 20) containing 5% nonfat dry milk (blotting grade) at room temperature for 1 h. The blots were then treated with the primary antibody for P-CREB in TBST containing 5% bovine serum albumin at 4°C overnight. After three washes with blocking buffer, the blots were incubated with anti-rabbit IgG conjugated to alkaline phosphatase for 1 h at room temperature. This was followed by three washes with blocking buffer, two washes with 10 mM Tris-HCl (pH 9.5), 10 mM NaCl, 1 mM MgCl 2 , and a 5-min incubation with diluted CDP-Star reagent (New England Biolabs, Beverly, MA) and then exposed to x-ray film. The membranes were then stripped in the buffer containing 62.5 mM Tris-HCl, pH 6.7, 2% SDS, and 100 mM ␤-mercaptoethanol and reprobed with antibody for CREB by a similar procedure. The intensity of the bands was quantitated by scanning. The extent of CREB phosphorylation was measured by calculating the ratio of P-CREB and CREB bands.
For phospho-p38 and p38, the blots were incubated with the primary antibody against dually phosphorylated p38 (Promega anti-active p38; 140 ng/ml) in TBST for 1 h at room temperature. After three washes with TBST, the blots were incubated with anti-rabbit IgG conjugated to horseradish peroxidase for 1 h at room temperature. This was followed by three washes in TBST and incubation with diluted ECL chemiluminescent reagent for 1 min. The membranes were then stripped and reprobed with antibody against p38 (C-20, Santa Cruz Biotechnology; 50 ng/ml).
p38 MAPK Assay-The cells were treated with IGF-I and inhibitor as described in the figure legends. After washing the cells with PBS, 200 l of ice-cold cell lysis buffer (20 mM Tris (pH 7.5), 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM ␤-glycerophosphate, 1 mM sodium orthovanadate, 10 g/ml leupeptin, 500 nM okadaic acid, and 1 mM phenylmethylsulfonyl fluoride) was added. The cells were scraped, lysed by sonication, and centrifuged for 20 min to collect the supernatant. The lysate (300 g) was mixed with 4 l of p38 MAPK antibody overnight at 4°C. Protein A-Sepharose (20 l) was added and gently rocked for 3 h at 4°C. After centrifugation, the pellet was washed twice with cell lysis buffer and twice with kinase assay buffer (25 mM Tris (pH 7.5), 5 mM ␤-glycerophosphate, 2 mM dithiothreitol, 0.1 mM sodium orthovanadate, 10 mM MgCl 2 ). The pellet was suspended in 30 l of kinase buffer with 200 M ATP and 2 g of ATF-2 fusion protein and incubated for 30 min at 30°C. The reaction was terminated by the addition of 10 l of 4ϫ Laemmli sample buffer. These samples were electrophoresed and immunoblotted with antibody to phospho-ATF-2. The intensities of the bands were measured by scanning.
Isolation of Total RNA and Northern Blot Analysis-PC12 cells (90% confluence) were cultured in fasting medium in the absence and presence of 100 ng/ml IGF-I for 24 h. Total RNA was isolated from these cells using the Qiagen RNeasy kit. RNA samples were fractionated on denaturing 1.2% agarose-formaldehyde gels and transferred to Hybond Nϩ membrane. The 1.6-kilobase pair XhoI/EcoRI insert of rat chro-mogranin A cDNA probe was labeled with thermostable alkaline phosphatase using the AlkPhos-Direct kit from Amersham Pharmacia Biotech. Hybridization, washing, and detection by CDP-Star were performed according to manufacturer's protocol. The blots were stripped and reprobed with labeled ␤-actin by a similar protocol. The expression of chromogranin A was normalized to ␤-actin expression.
Transfection Procedure-The PC12 cells were cultured to 60 -80% confluence for transfection experiments in 6 ϫ 35-mm plates. For each well, 2 g of plasmids and 20 g of LipofectAMINE reagents (Life Technologies, Inc.) were used as per the manufacturer's instructions. The plasmid containing the ␤-galactosidase gene driven by the SV 40 promoter was included to normalize the transfection efficiency. DNA and the LipofectAMINE reagent were diluted separately in 100 l of serum-free medium without antibiotics, mixed together, and incubated at room temperature for 30 min. The culture plates were washed with PBS and 800 l of serum, and antibiotic-free medium was added. The 200 l of the plasmid LipofectAMINE mixture was then added to each well, and the plates were incubated at 37°C for 4 h. Then 1.0 ml of high serum medium (20% fetal bovine serum and 10% heat-inactivated horse serum) was added and incubated for approximately 40 h before induction with growth factors for luciferase. After 4 h of induction, the cells were washed in PBS and lysed with 100 l of reporter lysis buffer. In the case of chromogranin A promoter constructs, the induction was 24 h after transfection for a period of 30 h. The cells were lysed by freezing and thawing, and lysate was centrifuged at 14,000 rpm for 30 min. The supernatant was used for the assay of luciferase and ␤-galactosidase. Luciferase assays were carried out using the enhanced luciferase assay kit (Analytical Luminescence Laboratory, San Diego, CA) on a Monolight 2010 luminometer. The ␤-galactosidase assay was performed according to the method of Wadzinski et al. (31).
Statistical analysis was carried out by Student's t test.

Dose-and Time-dependent Phosphorylation of CREB at Ser 133 by IGF-I-
The nuclear transcription factor CREB was phosphorylated in a time-dependent manner in response to IGF-I. There was a 2.3-fold (p Ͻ 0.001) increase in CREB phosphorylation at serine 133 ( Fig. 1A) at 10 min when PC12 cells were stimulated with IGF-I (100 ng/ml). The phosphorylation returned to near basal level by 120 min. The protein level of CREB did not change during this 2-h period in the presence of IGF-I. The time course of CREB phosphorylation mediated by IGF-I was comparable with that of insulin in 3T3-L1 fibroblasts as reported earlier (29). The dose-response curve shows a significant increase in CREB phosphorylation at 10 ng/ml (p Ͻ 0.05) with dose-dependent increases at higher concentrations (Fig. 1B). We observed an additional band with the antibody specific for the phosphorylation sequence around serine 133 of CREB. Serine 63 phosphorylation of ATF-1 is known to be detected by the same antibody as serine 133-phosphorylated CREB, since they are 100% homologous for this consensus phosphorylation sequence (32). The phosphorylation pattern of ATF-1 was parallel to that of CREB in terms of both intensity and time course.
Multiple Signaling Pathways Are Involved in IGF-I-mediated Phosphorylation of CREB-To explore the role of different signaling pathways in CREB phosphorylation, we examined the effect of PD98059 (an inhibitor of MEK) and wortmannin FIG. 1. Dose-and time-dependent CREB phosphorylation mediated by IGF-I. PC12 cells were cultured in 60-mm dishes to near confluence and then maintained in serum-free medium for 5 h. They were treated with 100 ng/ml IGF-I for varying periods of time, from 10 to 120 min (A). In another set of experiments, the fasted cells were exposed to increasing concentrations of IGF-I for 10 min (B). The cells in both experiments were washed in ice-cold PBS at the end of incubation period, and the cell lysates were prepared by the addition of 200 l of warm 1ϫ Laemmli sample buffer followed by sonication. The samples containing equal amounts of proteins were electrophoresed and immunoblotted with the antibody specific for CREB phosphorylated at Ser 133 . The membranes were stripped and reprobed with CREB antibody. Quantitation of specific bands was done by scanning densitometry. The ratio of phosphorylated CREB over the nonphosphorylated form was calculated, and this value for the untreated cells was taken as 1. The results are the mean Ϯ S.E. of four independent experiments.
(an inhibitor of PI 3-kinase) on IGF-I-induced CREB phosphorylation. In addition, we examined the effects of SB203580 and SB202190, two specific inhibitors of p38 MAPK (33,34). This protein kinase has been shown to mediate the effects of fibroblast growth factor and NGF on CREB phosphorylation. As shown in Fig. 2A, preincubation of PC12 cells with PD98059 (30 M) or wortmannin (100 nM) resulted in a 25-30% decrease in IGF-I-mediated CREB phosphorylation at serine 133. The addition of 30 M PD98059 decreased the induction of ERK1/2 MAPK activity as detected by dual phospho-ERK antibody (Promega, Madison, WI) (data not shown). The findings with wortmannin clearly indicate that the PI 3-kinase mediates IGF-I-induced nuclear signaling in addition to the regulation of cytosolic proteins such as glycogen synthase. The inhibitors of p38 MAPK, SB203580 (10 M), and SB202190 (10 M) were able to decrease the CREB phosphorylation stimulated by IGF-I significantly (p Ͻ 0.001), suggesting a novel pathway for nuclear signaling of this growth factor. Rapamycin, an inhibitor of p70 S6 kinase, which is one of the downstream components of the PI 3-kinase signaling system, had no significant impact on IGF-I-induced CREB phosphorylation.
The neurotrophic actions of IGF-I are similar to those of NGF in PC12 cells. Therefore, we compared the effects of IGF-I and NGF on CREB phosphorylation. The partial reduction in NGFmediated CREB phosphorylation in the presence of individual inhibitors was similar to that of IGF-I (data not shown) with the minor difference that PD98059 was more effective than wortmannin in decreasing NGF-induced CREB phosphoryla-tion. In some experiments, combinations of inhibitors were shown to block the effects of growth factors completely (Fig.  2B). For example, in the presence of wortmannin (100 nM) and SB203580 (10 M), IGF-I did not increase CREB phosphorylation above basal level. The inhibitors PD98059 (30 M) and SB203580 (10 M) used together blocked NGF action. For the remainder of the studies, parallel experiments were conducted with NGF as a control. The NGF data will only be presented for selected experiments.
IGF-I Activates p38 MAPK in PC12 Cells-The experiments with pharmacological inhibitors demonstrated that multiple signaling pathways mediate IGF-I-induced phosphorylation of CREB, including ERK, PI 3-kinase, and p38 MAPK. Growth factors have been shown to activate p38 in neuronal cell lines (26,35,36). Because the pharmacological inhibition of p38 with the SB compounds suggested a role for p38, we examined p38 activity in response to IGF-I in these cells. To examine whether IGF-I activates p38 MAPK in PC12 cells, experiments were carried out to measure the formation of phospho-p38 MAPK, the active form of this enzyme. As shown in Fig. 3A, IGF-I increased the phosphorylation of p38 significantly (p Ͻ 0.001) over the untreated cells in 5 min and slowly decreased over the remaining 2-h incubation period. Sodium arsenite is a known stimulator of p38 MAPK activity, and it serves as a control for the immunoprecipitation kinase assay. When p38 MAPK was assayed in PC12 cells treated with IGF-I (100 ng/ml) and sodium arsenite (300 M) by immunoprecipitation followed by phosphorylation of ATF-2 phosphoprotein, 78 and 145% in- creases in the enzyme activity were observed, respectively. Pretreatment of cells with the inhibitor SB203580 (10 M) decreased the stimulated enzyme activity significantly (p Ͻ 0.01).
IGF-I-mediated CREB Phosphorylation Does Not Involve cAMP-dependent Protein Kinase (Protein Kinase A)-CREB was initially described as a substrate for protein kinase A (23). Therefore, we assessed the role of protein kinase A in IGF-I induced CREB phosphorylation using H89, a pharmacologic inhibitor that specifically inhibits this kinase (data not shown). This inhibitor decreased the formation of P-CREB mediated by dibutyryl cAMP (500 M) and forskolin (10 M) significantly (p Ͻ 0.001). H89 did not block IGF-I and NGF-mediated increases in P-CREB formation.
IGF-I Mediated CREB Phosphorylation Leads to Its Transcriptional Activation-CREB phosphorylation at serine 133 is essential for transcriptional activation, but under some conditions this phosphorylation is inadequate to drive transcription (37). Thus, it was essential to determine whether IGF-I-mediated CREB phosphorylation enhanced its transcriptional activation. For initial experiments, we used a Gal4-TK-Luc reporter system specific for the transactivation of CREB. Since endogenous transcription factors do not bind to the promoter pGal4-TK-Luc, the increase in luciferase activity in the presence of IGF-I is a measure of the stimulation of the transactivational potency of the Gal-CREB chimeric protein through phosphorylation. PC12 cells were transiently transfected with an expression vector for a chimeric protein consisting of the Gal4 DNA binding domain linked to the transactivation domain of CREB and a plasmid containing the luciferase reporter gene linked to an enhancerless thymidine kinase promoter and four copies of Gal4-responsive sequences. By this approach, one could eliminate other CRE-binding endogenous transcription factors binding to the reporter gene. This permits evaluation of the pathways leading specifically to CREB activation. IGF-I increased transcription in a dose-dependent manner to a maximum of 3-fold in this system (Fig. 4A). No transcriptional activation was noted in the control experiments without the Gal4-CREB chimeric protein (results not shown). To optimize cell viability after transient transfection, the cells were maintained in serum during the induction with growth factors, because PC12 cells undergo programmed cell death with serum withdrawal. This contributes to the high basal CREB transcriptional activity and also represents the normal physiological context. Consistent increases over the physiological background were noted in response to IGF-I.
To determine the functional significance of phosphorylation of serine 133 by IGF-I for activation of this transcriptional reporter, cotransfection experiments were carried out using the expression vector for Gal4-CREB in which serine was replaced with alanine at position 133. When cotransfected with pGal4-TK-Luc, this mutated fusion protein did not induce luciferase expression significantly when compared with the wild type Gal4-CREB (Fig. 4B). Treatment with growth factors did not further enhance the luciferase expression.  (Fig. 5A). Rapamycin did not have any effect on the transcriptional activation by IGF-I. In the case of NGF, significant decreases (25-35%) in transcriptional activation were exerted by the inhibitors PD98059 (30 M), wortmannin (100 nM), SB203580 (10 M), and SB202190 (10 M) (data not shown). As with IGF-I, rapamycin had no effect on NGF action. The inhibition of transcriptional activation by these inhibitors was partial when used alone. Parallel to our observation of the impact of combined inhibitors in Fig.  2, IGF-I-induced luciferase production was decreased to the basal level when the cells were preincubated with wortmannin and SB203580 together (Fig. 5B). These findings clearly demonstrate that IGF-I uses novel signaling pathways to increase the transcriptional activation of CREB in PC12 cells.

IGF-I Activates Transcription of CRE-containing Chromogranin A Promoter
Constructs-Once the transcriptional activating potential of IGF-I on CREB had been determined using the Gal4-TK-Luc reporter system, we did a series of experiments to assess the physiological relevance of CREB activation by IGF-I. For these experiments, we employed the neuronal specific CRE-containing chromogranin A promoters. The promoter with 1133 bp of 5Ј-flanking region was stimulated 3.3-fold by IGF-I, whereas the truncated promoter with the CRE (pXP77) was activated 2.5-fold. To determine whether this activity was dependent upon CREB, we did a series of experiments cotransfecting a dominant negative CREB, K-CREB, as well as the truncated chromogranin A promoter with the CRE site mutated (Fig. 6A). The dominant negative K-CREB decreased basal and total IGF-I stimulation compared FIG. 3. Activation of p38 MAPK by IGF-I. PC12 cells (90% confluent) were fasted for 5 h and exposed to IGF (100 ng/ml) for varying time periods (A). The cells were washed and harvested for immunoblotting with the antibody to phospho-p38 MAPK. The membranes were then stripped and reprobed with the antibody to p38 MAPK. In some experiments, the fasted cells were preincubated in the absence (o) and presence (f) of 10 M SB203580 for 30 min followed by incubation with IGF (100 ng/ml) for 10 min or sodium arsenite for 300 M for 1 h (B). The activity of p38 MAPK was measured by immunoprecipitating the cell lysates with p38 MAPK antibody and phosphorylating ATF-2 fusion protein followed by immunoblotting phospho-ATF-2. Quantitation of the bands was done by scanning densitometry. The values are the means of three observations. with controls, as did the CRE mutant. Some increase in activity was seen in response to IGF-I, indicating both CREB-dependent and CREB-independent regulation of chromogranin A by IGF-I. We also noted a 3.7-fold increase in chromogranin A mRNA when PC12 cells were exposed to 100 ng/ml of IGF-I for 24 h (Fig. 6B), demonstrating that the endogenous gene and the reporter constructs respond similarly.

Impact of Constituitively Active or Dominant Negative Ras and Raf-1 on IGF-I-mediated Stimulation of Chromogranin
A-Pharmacological inhibition of MEK using PD98059 (Figs. 2 and 5) indicated a role for the ERK1/2 MAPK cascade in IGF-I regulation of CREB activity. To confirm these data using another strategy, we cotransfected PC12 cells with a truncated CREBresponsive chromogranin A reporter construct and plasmids for constitutively active Ras and Raf-1 (pSVRas and pRSV BxBraf, respectively) or dominant negative Ras (pZCRN17Ras). Activation of the Ras 3 Raf 3 MEK 3 ERK pathway using the constitutively active iosforms of either Ras or Raf-1 (Fig. 7A) led to a significant increase in basal chromogranin A activity, demonstrating responsiveness to this pathway. IGF-I treatment gave an additional stimulation of 1.5-fold over the high basal level, suggesting that there are other pathways in addition to ERK1/2 that contribute to the IGF-I response. Dominant negative Ras decreased basal activity with a restoration toward basal upon exposure to IGF-I. Taken together, these data support a role for ERK1/2 activation of chromogranin A by IGF-I. They also suggest that additional pathways are involved.

Role of PI 3-Kinase in the IGF-I-mediated Activation of Chromogranin A-
The PI 3-kinase inhibitor data suggested that one of the additional pathways involved PI 3-kinase. As demonstrated in Figs. 2A and 5A, wortmannin, an inhibitor of PI 3-kinase, interferes with both CREB phosphorylation and its transcriptional activation by IGF-I. Hence, we examined the effect of transient transfection of wild-type and dominant negative p85 subunits (courtesy of Dr. Masato Kasuga, Kobe, Japan) in PC12 cells. The wild type p85 subunit exerted a small increase in basal and IGF-I-mediated transcriptional activation, whereas ⌬p85, the kinase-dead PI 3-kinase isoform, inhibited IGF-I activation of chromogranin A (p Ͻ 0.001). These results demonstrate a PI 3-kinase-dependent activation of chromogranin A in PC12 cells.

MAPK Kinase 6 and p38␤ Enhance the Activation of Chromogranin A Promoter by IGF-I-Previous
studies have indicated that, among p38 MAPK isozymes, the ␤ isoform is involved in the hypertrophic action. Hence, we cotransfected the PC12 cells with p38␤ and the constitutively active form of its upstream kinase, MAPK kinase 6, and examined the promoter activity of chromogranin A (pXp77). The stimulation of p38␤ resulted in the increase of basal and IGF-I-induced chromogranin A promoter activity by 80 -90%. This increase was significantly (p Ͻ 0.001) blocked when the cells were preincubated with the p38 MAPK inhibitor SB203580 (10 M). The results of this experiment further support the role of the p38 MAPK pathway in IGF-I-mediated activation of the nuclear transcription factor CREB. DISCUSSION In this investigation, we demonstrate that IGF-I stimulates phosphorylation of the nuclear transcription factor, CREB, the Ca 2ϩ /cAMP response element-binding protein, at serine 133 in PC12 cells. This post-translational modification leads to an increase in CREB's transcriptional activity as demonstrated by the Gal4-TK-Luc reporter system. IGF-I is also capable of regulating chromogranin A, a neuroendocrine-specific gene, by a CREB-dependent mechanism. Using specific inhibitors such as PD98059, SB203580, or wortmannin or by cotransfecting constitutively active and dominant negative components of the Ras and PI 3-kinase pathways, we demonstrate that IGF-I-mediated CREB activation proceeds through three distinct pathways involving ERK, PI 3-kinase, and p38 MAPK. IGF-I is known to exert its actions on cellular proliferation, survival, and differentiation through the ERK and PI 3-kinase pathways. In this study, we show for the first time that some of the CREB-dependent gene regulatory actions of IGF-I proceed through the p38 MAPK pathway.
In neuronal cells, the nuclear transcription factor CREB plays a central role in several critical functions. It is important for protein synthesis-dependent long term memory formation, since targeted mutation of CREB leads to a decrease in long term memory in mice (14,15). Hormonal regulation of dentritic spine formation in cultured hippocampal neurons requires the phosphorylation of CREB (19). In PC12 cells, a cell culture FIG. 4. Transcriptional activation of CREB by IGF-I in PC12 cells. PC12 cells were cultured in 6 ϫ 35-mm wells to around 70% confluence. The cells were cotransfected with pGal4-TK-Luc, pRSV-Gal4-CREB-341, and pRSV ␤-galactosidase in the medium containing no serum and antibiotics by the LipofectAMINE transfection method for 4 h. For each well, 2 g of plasmids and 20 g of LipofectAMINE reagent were used. Induction for luciferase with different agents for 4 h was carried out 48 h after the initiation of transfection. A, the transfected cells were incubated with varying doses of IGF-I as indicated. B, in the transfection protocol for this experiment either pRSV-Gal4-CREB or pRSV-Gal4-CREB S133A was used, and later the transfected cells were induced with 100 ng/ml IGF or 50 ng/ml NGF. Cell lysates were prepared, and the transcription was measured by assaying the luciferase activity by the procedure described under "Experimental Procedures." In these lysates, ␤-galactosidase activity was also assayed to correct for the efficiency of transfection. The transcription mediated by IGF-I and other agents was expressed as -fold induction over the basal transcription in transfected but untreated cells. Results are means Ϯ S.E. of three independent experiments. model of neurons, the CREB/ATF-1 family of transcription factors are needed for the neurite outgrowth (38). Dominant negative ATF-1 blocks cAMP-induced neurite formation by inhibiting cAMP-mediated CREB activation in these cells (38). Interference of CREB and other ATF family members with E1A viral antigen also blocks PC12 cell differentiation (13). Additionally, CREB is critical for the induction of immediate early gene c-fos by NGF (12). The promoter regions of several neuronal specific genes such as chromogranin A (CgA) and vgf contain CREB response elements. Because of the diverse neuronal responses that require CREB, it is important to understand the specific mechanisms whereby IGF-I, an important neurotrophin, regulates CREB dependent transcription.
CREB is regulated by multiple factors in PC12 cells including forskolin, NGF, epidermal growth factor, and 12-O-tetradecanoylphorbol-13-acetate (12,28). These diverse stimuli can result in divergent cell fates including proliferation and differentiation. With respect to the diversity of factors that can impact CREB-and CRE-regulated transcription, we present important new information on CREB regulation by IGF-I in PC12 cells. We present convincing data that IGF-I treatment at a physiologically relevant concentration leads to transcriptionally important phosphorylation of CREB at serine 133. Additionally, CREB activation plays a major role in the IGF-Imediated regulation of the neuronal specific chromogranin A gene. We see an impact of IGF-I on both the chromogranin A promoter and induction of chromogranin A mRNA. The experiments described define many parallels between NGF and IGF-I for CREB regulation in this cell line. Both agents employ multiple signaling pathways for CREB regulation. From inhibitor studies, it appears that the dominant pathways for IGF-1 are p38 MAPK and PI 3-kinase, whereas MEK is the dominant pathway for NGF with a contribution from p38 MAPK. This role of p38 MAPK activation by growth factors is a new and rapidly evolving area of research. The current data do not permit a detailed comparison between the IGF-I and NGF, but the differences noted between the two neurotrophins could provide insight into their divergent impacts on cell fate, survival, and proliferation. IGF-I has been shown to have significant neurotrophic actions such as survival and regeneration of neurons (4,5). In diabetes, IGF-I activity is decreased in neuronal tissues, and this could contribute to the development of diabetic neuropathy (39). IGF-I is being considered as a potential therapeutic agent Total RNA was isolated from these cells using Qiagen's RNeasy kit. RNA 10-g samples were resolved on formaldehyde-agarose gels and transferred to Hybond Nϩ membranes and probed with the chromogranin A cDNA probe labeled with alkaline phosphatase and detected with the CDP-Star system. The blots were then stripped and reprobed with labeled ␤-actin by a similar procedure. Two representative blots from the set of five are shown here.
in the treatment of neurodegenerative diseases (40). For these reasons, it is essential to understand the mechanism by which this growth factor regulates gene expression in neuronal cells. Our present findings show that IGF-I mediated the activation of the nuclear transcription factor CREB through multiple signaling pathways and that this leads to enhanced expression of a neuroendocrine-specific gene, CgA. We chose to examine CgA because it is known to be regulated in a CREB-dependent manner in PC12 cells (27,28). This tool establishes a neuronal context for the experiments designed to define the important signaling pathways. CREB is known to bind the promoters of mouse and human CgA (27,41). In the present investigation, we observed an enhanced expression of CgA in IGF-I-treated PC12 cells, as measured by the Northern blot analysis of mRNA. Further, IGF-I-mediated activation of CREB through multiple signaling pathways leads to the stimulation of fulllength as well as the CRE-containing truncated promoter of CgA. Studies with dominant negative K-CREB and the promoter containing mutated CRE clearly demonstrate that CREB plays a significant role in mediating IGF-I action. These findings indicate that IGF-I-stimulated CREB activation is involved in the physiologically relevant gene expression in neuronal cells.
Our observation that IGF-1 stimulates the phosphorylation of nuclear transcription factor CREB through a p38 MAPK is new, and the physiological relevance remains to be determined.
To briefly review the current understanding of the p38 MAPK family, they are a family of serine/threonine kinases, activated by dual phosphorylation on threonine and tyrosine residues. In mammalian cells, three distinct MAPKs have been identified: ERK1/2, stress-activated protein kinase/c-Jun N-terminal kinase, and RK cytokine suppressive anti-inflammatory drug binding protein p38 MAPK. The pathways mediated by c-Jun N-terminal kinase and p38 MAPK have been shown to play a significant role in stress-mediated signal transduction. The p38 MAPK is associated with apoptosis, and it opposes the actions of ERK in PC12 cells (30,42). However, p38 and ERK MAPKs cooperate in the transcriptional activation of c-fos in response to UV irradiation (43). Further, p38 MAPK participates in the protein phosphorylation cascade resulting from activation of growth factor/hormone receptors. For example, fibroblast growth factor activates p38 MAPK in SK-N-MC cells, and insulin stimulates this kinase in L6 muscle cells (26,44). NGF has been shown to activate CREB through the ERK as well as p38 MAPK pathway (35). IGF-I stimulates p38 MAPK activity in SH-SY5Y neurobastoma cells (36). In a recent study, Scrimgeour et al. (45) used a mutant of the IGF-I receptor in which tyrosines at positions 1250 and 1251 in the carboxyl-terminal region had been replaced to demonstrate that some of the actions of IGF-I could involve a third pathway other than ERK and PI 3-kinase pathways. We support this possibility by showing in this study that IGF-I does activate CREB by a third pathway involving p38 MAPK. IGF-I-mediated phosphorylation and activation of CREB decreased significantly in the presence of SB203580. This pyridinyl imidazole derivative has been shown to be specific for p38 MAPK, and it did not have any inhibitory action toward 12 other protein kinases tested in vitro (34). Several isoenzymes of p38 MAPK have been identified that are likely to have differential actions (46,47). Wang et al. demonstrated in cardiomyocytes that p38␤ mediates hypertrophic response, whereas p38␣ induces apoptosis (48). In a recent study, this ␤ isoform has been also shown to provide creases in CREB activation by IGF-I and NGF were calculated after correcting for transfection efficiency. The values represent means Ϯ S.E. of three observations. FIG. 7. Activation of chromogranin A promoter by IGF through multiple signaling pathways. PC12 cells (70% confluence) cultured in 6 ϫ 35-mm wells were transfected with pxP77 and pRSV ␤-galactosidase along with indicated plasmids for 4 h by the Lipo-fectAMINE transfection method using 2 g of total plasmids and 20 g of LipofectAMINE reagent. In these transfection experiments, plasmids for the modulation of signaling pathways involving Ras (A), PI 3-kinase (B), and p38␤ (C) were also included. After 1 day of transfection, the cells were cultured in the absence (o) and presence (f) of 100 ng/ml IGF-I (A and B) or as indicated (C) for 30 h. The activities of luciferase and ␤-galactosidase were measured in the cell lysates. The -fold in-protective effect against apoptotic signals (49). We have also observed in the present study an increase of IGF-I-induced chromogranin A promoter activity when the PC12 cells were cotransfected with p38␤ and the constitutively active form of MAPK kinase 6. IGF-I has been shown to activate the ERK and PI 3-kinase pathways in several cell types including PC12 cells (1,6,50,51). In the present study, the MEK inhibitor (PD98059) partially decreases the phosphorylation and activation of CREB mediated by IGF-I. Figs. 2 and 5 demonstrate only a partial, albeit significant, inhibition of IGF-I-mediated CREB phosphorylation and activation using PD98059. Cotransfection experiments with constitutively active Ras and Raf-1 demonstrate a role for this pathway in the activation of CgA. However, IGF-I was able to drive transcription of CgA even in the face of constitutively active Ras and Raf-1. These data suggest that the Ras 3 Raf-1 3 MEK pathway is important but is not the dominant pathway for IGF-1-mediated CREB activation. In a recent study, however, we observed that the ERK pathway plays a major role in the insulin-mediated CREB activation in HepG2 and 3T3-L1 cell line, suggesting cell-specific variations in signaling pathways. The PI 3-kinase pathway, which is critical for IGF-I-mediated neuronal survival under stress conditions, strongly contributes to IGF-I-mediated CREB activation. The importance of PI 3-kinase for IGF-I activation of CREB is clearly demonstrated by the wortmannin inhibitor studies. It is further supported by cotransfection of ⌬p85 PI 3-kinase, the kinase-dead mutant, which ablates IGF-I-mediated stimulation of CgA. Akt, the downstream component of the PI 3-kinase pathway, is known to regulate the covalent modification of cytosolic proteins such as glycogen synthase and the proapoptotic protein BAD in PC12 cells (51,52). In the present study, we demonstrate the involvement of this pathway in the nuclear actions of IGF-I. It has been previously shown that the expression of Bcl-xL and the bcl-2 are increased by IGF-I (21,53). These proteins belong to the Bcl-2 family, which is known to protect the cells from programmed cell death (21). It is possible that IGF-I-mediated CREB activation is involved in the regulation of the expression of bcl-2, since it is a CREB-dependent gene (54).
To summarize, the data presented in this paper demonstrate IGF-1 activation of CREB, an important transcription factor for neurotrophin activity. This activation employs signaling pathways mediated by PI 3-kinase and p38 MAPK. Future studies will explore the implications of each of these signaling pathways for CREB-responsive genes important for cell cycle regulation, survival, and differentiation.