Corticotropin-releasing Hormone Induces Fas Ligand Production and Apoptosis in PC12 Cells via Activation of p38 Mitogen-activated Protein Kinase*

Recent experimental findings involve corticotropin-releasing hormone (CRH) in the cellular response to noxious stimuli and possibly apoptosis. The aim of the present work was to examine the effect of CRH on apoptosis and the Fas/Fas ligand system in an in vitro model, the PC12 rat pheochromocytoma cell line, which is widely used in the study of apoptosis and at the same time expresses the CRH/CRH receptor system. We have found the following. CRH induced Fas ligand production and apoptosis. These effects were mediated by the CRH type 1 receptor because its antagonist antalarmin blocked CRH-induced apoptosis and Fas ligand expression. CRH activated p38 mitogen-activated protein kinase, which was found to be essential for CRH-induced apoptosis and Fas ligand production. CRH also promoted a rapid and transient activation of ERK1/2, which, however, was not necessary for either CRH-induced apoptosis or Fas ligand production. Thus, CRH promotes PC12 apoptosis via the CRH type 1 receptor, which induces Fas ligand production via activation of p38.

Several lines of evidence suggest that corticotropin-releasing hormone (CRH) 1 may play a role in the cellular response to noxious stimuli that promote neuron death (1). Indeed, CRH contributes to hippocampal ischemic injury, an effect prevented by the CRH antagonist ␣-helical CRH (ahCRH) (2). Similarly, astressin, a potent CRH antagonist, exerts a considerable neuroprotective effect on hippocampal cell damage following kainic acid-induced excitotoxic seizures (3). In animal models of induced status epilepticus, CRH causes neuronal loss in limbic structures, including the CA3 region of the hippocampus characterized by pyramidal cell apoptosis (4). Furthermore, administration of CRH to the brain of immature rats is associated with progressive hippocampal CA3 neuron apoptosis independent of glucocorticoids (5). These phenomena involve cell apoptosis. However, there is no information regarding the effects of CRH on the apoptotic machinery.
The Fas/Fas ligand system controls apoptosis of several types of immune cells and possibly of epithelial and neural cells including cells in hippocampus and cortex (6,7). The aim of the present work was to examine the effect of CRH on apoptosis and Fas ligand expression in a well established in vitro model, the PC12 rat pheochromocytoma cell line, which has characteristics of epithelial and neuronal cells and is widely used as a model in the study of apoptosis. PC12 cells express the CRH/ CRH receptor and the Fas/Fas ligand systems, providing a physiological model for the study of the effects of CRH on apoptosis and the intracellular signaling cascade involved (8,9). When PC12 cells that are differentiated by nerve growth factor (NGF) are deprived of growth factors, they undergo apoptosis via expression of Fas ligand (10). Expression of Fas ligand and apoptosis in differentiated PC12 cells and primary cultures of rat sympathetic neurons depends on activation of stress-activated protein (SAP) kinase, p38 mitogen-activated protein kinase (MAPK), and c-Jun N-terminal kinase (JNK) (10 -12), whereas apoptosis of non-neuronally differentiated PC12 cells does not depend on JNK activation (12). Activation of p38 MAPK plays a central role in both undifferentiated and differentiated PC12 as well as in neuronal cell apoptosis because inhibition of this kinase promotes cell survival (13). Furthermore, withdrawal of survival factors from NGF-differentiated PC12 cells causes activation of p38 MAPK and downregulation of extracellular signal regulated kinases (ERK1/2) MAPK resulting in apoptosis. Activation of ERK1/2 induced by NGF promotes PC12 cell proliferation and survival (14,15). Indeed, sustained activation of ERK1/2 MAPK initiates proliferating and antiapoptotic signals, whereas rapid and transient activation correlates with apoptosis (16).
In the first part of this work we examined the effect of CRH on apoptosis measured as the DNA fragmentation rate or fluorescent staining of apoptotic bodies. CRH was applied in the presence or absence of CRH antagonists ahCRH (9 -41) and antalarmin, a pyrrolopyrimidine compound that antagonizes CRH type 1 receptor (CRHR1)-mediated effects of CRH, including pituitary ACTH release, stress behaviors, and acute inflammation (17)(18)(19). In the second part of this work we examined the mechanism through which CRH achieves its proapoptotic effect. We examined the effect of CRH in the presence or absence of its antagonists on Fas ligand production at the protein level using fluorescence-activated cell sorter analysis, Western blotting, or immunofluorescence and at the mRNA level using reverse transcription-PCR. In the third part of this work we analyzed the signaling pathways involved in CRH-induced Fas ligand production. Thus, we examined the effect of CRH on p38 MAPK and ERK1/2 activation and the effect of a MAPK-ERK kinase (MEK1) inhibitor and a p38 MAPK inhibitor on CRHinduced apoptosis and Fas ligand production.

PC12 Cell Culture
PC12 cells were obtained from three sources: Dr. M. Greenberg (Children's Hospital, Boston, MA), early passages from the late Dr. G. Guroff (Section on Growth factors, NICHD, NIH), and the American Type Culture Collection. All cells responded similarly to CRH. Cells were grown in RPMI 1640 containing 2 mM L-glutamine, 15 mM HEPES, 100 units/ml penicillin, 0.1 mg/ml streptomycin, 10% horse serum, and 5% fetal calf serum (all purchased from Invitrogen) at 5% CO 2 and 37°C. On day 0 of each experiment the initial culture media were changed either with complete media or with serum-free media supplemented with 0.1% bovine serum albumin fraction V. Apoptosis was measured at different time points ranging from 3 h to 5 days. For the evaluation of the early effect of CRH on protein phosphorylation, cells were plated for 12 h in serum-free media and then exposed to CRH for 0, 5, 10, and 20 min, and 1 and 2.5 h.

Apoptosis Detection Assays
Quantitative Measurement of Apoptosis-PC12 cells were plated in 96-well plates at an initial concentration of 30,000 cells/well. After 3 days the number of PC12 cells doubled. At that point the culture media were changed to serum-free media. As expected, PC12 cells undergo apoptosis when cultured under serum-free conditions. Apoptosis was measured by direct determination of nucleosomal DNA fragmentation with the "Cell Death Detection enzyme-linked immunosorbent assay plus" kit (Roche Molecular Biochemicals). Cells were harvested and lysed according to the manufacturer's protocol. The mono-and oligonucleosomes contained in the cell lysates were determined using an antihistone-biotin antibody. The data are expressed in photometric units. Each unit corresponds to ϳ12,500 apoptotic cells. The concentration of the nucleosomes was determined photometrically using 2,2Ј-azino-di[3ethylbenzthiazolin-sulfonate] as substrate for the biotin present on the antibody. The optical density was read on a Dynatech MicroElisa reader (Chantilly, VA) at a wavelength of 405 nm.
DAPI Staining-Apoptosis was also quantitated by assessing nuclear changes using the nuclear binding dye DAPI. After treatment, cells were fixed and permeabilized with 2% (w/v) formaldehyde and 0.4% (w/v) Triton X-100, respectively. Staining was performed with 1 g/ml DAPI in PBS for 3 min. The percentage of apoptotic cells was determined by analyzing 300 cells under a fluorescence microscope (Zeiss, Germany).

Flow Cytometric Analysis of Fas Ligand Expression
In brief, cells were washed twice with PBS, prefixed, and permeabilized as described above. Blocking was performed with 1% fetal calf serum in PBS for 15 min. Then, the primary antibody (anti-mouse Fas ligand) was added and incubated overnight at 4°C. Cells were washed twice with PBS and incubated with a fluorescein isothiocyanate-labeled rabbit anti-mouse IgG and analyzed on a flow cytometer (Epics Elite Coulter, U. K.).

Western Blot Analysis
After stimulation cells were harvested and lysed in a lysis buffer containing 50 mM Tris-HCl, pH 8, 150 mM NaCl, 0.1% SDS, 0.5% sodium deoxycholate, 1% Nonidet P-40, and freshly added proteinase inhibitors phenylmethylsulfonyl fluoride (10 g/ml) and 1 g/ml aprotinin for the detection of Fas and Fas ligand. Lysis was performed for 30 min on ice with occasional vortexing. To detect the phosphorylated and nonphosphorylated forms of p38 MAPK and ERK1/2, cells were lysed in 62.5 mM Tris-HCl, pH 6.8, 10% glycerol, 2% SDS, and freshly added inhibitors phenylmethylsulfonyl fluoride (10 g/ml), 0.5 mM dithiothreitol, and 50 mM Na 2 F. Subsequently, cells were sonicated for 5 s on ice. Solid cellular debris was removed by centrifugation at 12,000 ϫ g for 15 min. Lysates were collected and stored at Ϫ80°C. The protein concentration of each lysate was measured by a modification of the Bradford Coomassie Brilliant Blue G-250 method using bovine serum albumin fraction V as standard (20). SDS-PAGE sample loading buffer was added in 10 g of protein from each lysate electrophoresed through a 12% SDS-polyacrylamide gel as described previously (21). Protein was transferred to nitrocellulose membranes, using an LKB electroblot transfer system (LKB, Bromma, Sweden). Membranes were processed according to standard Western blotting procedures. To detect protein levels, membranes were incubated with the appropriate antibodies and then exposed to Kodak X-Omat AR films. A PC-based Image Analysis program was used to quantify the intensity of each band (Image Analysis, Inc., Ontario, Canada).
To normalize for protein content the blots were stripped in stripping buffer containing 62.5 mM Tris-HCl, pH 6.7, 2% SDS, 100 mM ␤-mercaptoethanol and stained with anti-actin antibody; the concentration of each target protein was normalized versus actin. Where p38 MAPK and ERK1/2 was measured, membranes were first probed for the phosphorylated form of the protein, then stripped and probed for the total protein. The intensity of the bands was quantified using the Bio-Rad imaging system, and the quantity of the phosphorylated proteins was expressed as the ratio of the phosphorylated divided by the total protein in each case.
Immunoprecipitation-Protein lysates were incubated with immobilized anti-phospho-p38 MAPK (Thr-180/Tyr-182) monoclonal antibody in p38 MAPK assay, or c-Jun fusion protein beads in SAP kinase/JNK assay. Immune complexes were precipitated, and kinase assays were performed according to the manufacturer (Cell Signaling, MA).

Immunofluorescence
PC12 cells were grown on glass slides and treated with CRH and its antagonists for 2.5 h. Control cells were left untreated. At the end of the incubation periods, cells were fixed in 2% (w/v) formaldehyde for 10 min, permeabilized in 0.2% (w/v) Triton X-100 for 10 min, blocked in 1% fetal calf serum in PBS for 15 min, and then incubated overnight at 4°C with a mouse anti-rat Fas ligand antibody diluted 1/100 in PBS containing 1% fetal calf serum. Samples were washed with PBS and incubated for 1 h with fluorescein isothiocyanate-conjugated secondary antimouse antibody diluted 1/100. Cells were analyzed in a confocal laser scanning microscope (Leica TCS-NT).

Reverse Transcription-PCR for Fas Ligand
Primers for Fas ligand were 5Ј-CAG CCC CTG AAT TAC CCA TGT C-3Ј (sense) and 5Ј-CAC TCC AGA GAT CAA AGC AGT CC-3Ј (antisense). Primers for ␤-actin were 5Ј-CAT CCT GTC GGC AAT GCC AGG-3Ј (sense) and 5Ј-CTT CCT GGG CAT GGA GTC CTG-3Ј (antisense) (10). Total cellular RNA was isolated using Trizol reagent (Invitrogen). After reverse transcription, the cDNA product was amplified by PCR, at 35 cycles, annealing to temperature of 60°C for Fas ligand (55°C for ␤-actin). It should be noted that Fas ligand mRNA amplification at 35 cycles is still at the exponential phase. 10 l of the ampli-fied products (507 bp for Fas ligand and 174 bp for ␤-actin) were separated on a 2% agarose gel and visualized by ethidium bromide staining.

Statistical Analysis
Results are presented either as absorbance on a Microelisa reader or as optical density in Western blots normalized/total cellular protein or as percentage of parallel control cells that were treated with the vehicle only. For the statistical evaluation of our data we have used analysis of variance, post hoc comparison of means followed by two multiple comparison tests: the Fisher's least significance difference and the Newman-Keuls test. For data expressed as percent changes we have used the nonparametric Kruskal-Wallis test for several independent samples.

RESULTS
CRH Induces PC12 Cell Apoptosis-The effect of CRH was measured on accelerated apoptosis from serum deprivation. PC12 cells were treated with CRH, and apoptosis was measured at several time points. Apoptosis became significant at 24 h and peaked at 72 h (Fig. 1A). More specifically, exposure of 60,000 cells to CRH for 24 h increased the number of apoptotic cells to 7,621 Ϯ 223 (mean Ϯ S.E., n ϭ 15) compared with 4,932 Ϯ 411 of parallel controls (Table I). At 72 h the number of apoptotic cells was 22,215 Ϯ 888 (n ϭ 21) (51.27 Ϯ 1.7% of the remaining cell population) in the CRH-exposed cells compared with 14,273 Ϯ 1,170 (30.74 Ϯ 2.12% of the remaining cell population) of parallel controls. Apoptotic cells were counted at the same time points using DAPI staining giving similar results (data not shown). A representative example of apoptotic bodies is shown in Fig. 1D.
The effect of CRH was more profound on the 3rd day. We therefore tested whether the effect of CRH was dose-dependent on the 3rd day of exposure. Indeed, the effect of CRH was dose-dependent peaking between 10 Ϫ9 and 10 Ϫ8 M (Fig. 1B). More specifically, 10 Ϫ9 M CRH increased apoptosis to 1.90 Ϯ 0.13 photometric units (n ϭ 21 of seven independent experiments, p Ͻ 0.001) compared with 1.14 Ϯ 0.09 photometric units of control cells, i.e. treated with the CRH diluent only. Similarly, 10 Ϫ8 M CRH increased apoptosis to 1.79 Ϯ 0.23 photometric units (n ϭ 21 of seven independent experiments, p Ͻ 0.001) compared with parallel controls.
Treatment of cells with 10 Ϫ8 M antalarmin decreased apo-ptosis induced by 10 Ϫ9 M CRH to 76 Ϯ 17% photometric units (n ϭ 5 of two independent experiments) compared with 167 Ϯ 11% of parallel control cells, i.e. cells treated with CRH alone (p Ͻ 0.05) (Fig. 1C), suggesting that the proapoptotic effect of CRH was mediated by the CRHR1 receptor. In the presence of 10 Ϫ6 M antalarmin, 10 Ϫ9 M CRH-induced apoptosis was suppressed further, reaching 53 Ϯ 7% of parallel controls (n ϭ 5 of two independent experiments, p Ͻ 0.01) (Fig. 1C). It should be noted that a higher concentration of CRH (10 Ϫ6 M) required a higher concentration of antalarmin to have its effect suppressed, suggesting that there is a stoichiometric balance between the two molecules (data not shown). In contrast, ahCRH required higher concentrations than antalarmin to reverse CRH-induced apoptosis in PC12 cells. Treatment of PC12 cells with 10 Ϫ9 M CRH requires 10 Ϫ6 M ahCRH to reduce apoptosis to the basal levels, e.g. 83 Ϯ 20% of the control cells (n ϭ 5 of two independent experiments, statistically significant compared with cells exposed to CRH alone, p Ͻ 0.05) (Fig. 1C).
Because antalarmin is mainly a CRHR1 antagonist and the effect exerted by CRH was completely inhibited by antalarmin, we assume that CRH-induced apoptotic signals are mediated via CRHR1. Antalarmin alone at 10 Ϫ6 M had a low but significant suppressive effect on serum deprivation-induced apo-FIG. 1. CRH induces apoptosis. Panel A, time-response curve of PC12 cells exposed to serum-deprived RPMI 1640 supplemented with vehicle (control) or 10 Ϫ9 M CRH for a time period ranging from 3 to 120 h. Apoptosis was measured as described under "Experimental Procedures." Panel B, dose-response curve on the 3rd day of CRH exposure. Apoptosis is depicted as a percentage of parallel controls. Panel C, effect of CRH receptor antagonists on CRH-induced apoptosis. Cells were treated for 3 days with 10 Ϫ9 M CRH with and without antagonists in the concentrations noted on the graph. Panel D, representative photograph of apoptotic bodies of PC12 cells after treatment with 10 Ϫ9 M CRH. Nuclei were stained with DAPI and examined under a fluorescence microscope. Note several fragmented nuclei (apoptotic bodies). Data are expressed as photometric units (panel A) or percentages of parallel controls exposed only to vehicles (panels B and C). *p Ͻ 0.05, **p Ͻ 0.01, or ***p Ͻ 0.001 denote significant statistical differences compared with parallel controls (panels A and B) or with cells exposed to CRH alone (panel C). ptosis to 86 Ϯ 6% (n ϭ 14 of five independent experiments, p Ͻ 0.05) of parallel controls (Fig. 1C). CRH Induces Fas Ligand Production-Serum deprivation induced the expression of Fas ligand in a time-dependent manner ( Fig. 2A, Western blot data) in agreement with previously published data (10). To test whether CRH promotes apoptosis via Fas ligand production, the Fas ligand protein levels were measured on PC12 cells exposed to CRH. CRH induced a rapid increase in the concentration of Fas ligand, peaking at 2.5 h ( Fig. 2A). Thus, 10 Ϫ9 M CRH induced-Fas ligand production at 2.5 h was 164 Ϯ 12% (Fas ligand:actin ratio, n ϭ 13 of four independent experiments, p Ͻ 0.001) of parallel controls, i.e. cells exposed to vehicles only. The stimulatory effect of CRH was detected for 24 h.
Staining of PC12 cells with antibody against Fas ligand showed dense cytoplasmic staining of cells exposed to CRH 10 Ϫ9 M for 2.5 h (Fig. 3B). Furthermore, flow cytometry analysis of PC12 cells exposed to CRH 10 Ϫ9 M confirmed the Western blot data (Fig. 4). The analysis of the mRNA of Fas ligand by reverse transcription-PCR showed that CRH stimulated Fas ligand expression (Fig. 5). Indeed, Fas ligand mRNA started to increase at 6 h of exposure to CRH and remained elevated at 24 h.
The effect of CRH on Fas ligand production was blocked completely by antalarmin, suggesting that CRH acts via CRHR1 to activate intracellular pathways that lead to Fas ligand induction. Exposure of our cells to 10 Ϫ8 M antalarmin suppressed CRH-induced Fas ligand production from 164 Ϯ 12% (Fas ligand:actin ratio, n ϭ 13 of four independent experiments) to 64 Ϯ 27% compared with control cells (Fas ligand: actin ratio, n ϭ 6 of three independent experiments, statistically significant compared with cells treated with CRH, p Ͻ 0.05) (Figs. 2B, 3D, and 4E). Treatment with antalarmin alone had no significant effect on the basal levels of Fas ligand (Fig.  2B). The expression levels of CD95/Fas did not change during serum deprivation with or without CRH stimulation (data not shown). The preceding observations suggest that CRH initiates signals through CRHR1 which lead to Fas ligand induction and apoptosis.
p38 MAPK Mediates the Proapoptotic Effect of CRH-In this group of experiments we examined the effect of CRH on p38 MAPK and the role of the latter in CRH-induced apoptosis because p38 MAPK plays a central role in regulating apoptosis in PC12 cells (22), also affecting Fas ligand production (10). For this purpose we used SB203580, a specific inhibitor of p38 MAPK. Treatment of PC12 cells with different doses of SB203580 inhibited CRH-induced apoptosis (Fig. 6A). The presence of the 20 M SB203580 resulted in inhibition of the effect initiated by 10 Ϫ9 M CRH of apoptosis from 167 Ϯ 11% to 112 Ϯ 9% (n ϭ 6 of two independent experiments, statistically significant compared with cells exposed to CRH alone, p Ͻ 0.05).
In a second set of experiments, we examined the role of p38 MAPK in CRH-induced Fas ligand production. Thus, whereas treatment with 10 Ϫ9 M CRH increased the concentration of Fas ligand to 164 Ϯ 12% of parallel controls (Fig. 6B), the presence of SB203580 diminished the effect of CRH to 118 Ϯ 3% of parallel controls (n ϭ 4 of two independent experiments, statistically significant compared with cells exposed to CRH alone, p Ͻ 0.05).
To confirm that CRH activates p38 MAPK, PC12 cells were treated with CRH, and p38 MAPK activity (Fig. 7A) and phosphorylation (Fig. 7B) were measured at different time points. CRH exerted a rapid but transient activation of p38 MAPK which peaked 10 min after treatment and came back to base-line levels after 1 h (Fig. 7D). To test whether the phosphorylation of p38 MAPK is provoked by CRH, cells were incubated with CRH and FIG. 2. Effect of CRH on Fas ligand (FasL) production, Western blot data. Panel A, PC12 cells were treated with vehicle (control) or 10 Ϫ9 M CRH in serum-free medium for 2.5, 24, and 72 h. Cells were lysed, and supernatants were electrophoresed and probed by Western analysis with anti-Fas ligand monoclonal antibody followed by antiactin antibodies. CRH increased Fas ligand peaking at 2.5 h of exposure. Panel B, the stimulatory effect of CRH was blocked by the specific CRHR1 antagonist antalarmin. Cells were exposed for 2.5 h to 10 Ϫ9 M CRH with and without 10 Ϫ8 M antalarmin, and the Fas ligand concentration was measured as described above. Data are expressed as a percentage of parallel controls exposed only to vehicles and represent the mean of four independent experiments (panel A) or three independent experiments (panel B). *p Ͻ 0.05 and ***p Ͻ 0.001 denote a significant statistical difference compared with controls (panel A) or with cells exposed to CRH alone (panel B). supplemented with SB203580 for 10 min, 1 h, and 2.5 h. Treatment with SB203580 blocks the CRH-mediated enhancement, bringing the levels of p38 MAPK back to serum deprivation levels (Fig. 7C). Cumulative data are shown on Fig. 7D.
CRH Induces ERK1/2 Phosphorylation, Which Is Not Associated with Fas Ligand Expression or Apoptosis-Activation of ERK1/2 is known to participate in apoptosis. Indeed, transient activation leads to proapoptotic signals, whereas prolonged activation of ERK1/2 is involved in mediating mitogenic and antiapoptotic signals that could be blocked by PD98059. To determine a possible involvement of ERK1/2 in mediating CRH-initiated signals to induce apoptosis, we treated PC12 cells with PD98059 (MEK1 inhibitor) and measured apoptosis as described previously. Treatment of PC12 cells with PD98059 did not affect CRH-induced apoptosis. PC12 cells were treated with PD98059 and stimulated with CRH for a period of 1-3 days, and apoptosis was evaluated. Our data revealed that inhibition of the ERK1/2 MAPK kinases with PD98059, a MEK1 inhibitor, had no significant effect on CRH-induced apoptosis (Fig. 8A). Similarly, treatment with the same inhibitor did not alter Fas ligand production (Fig. 8B). CRH receptors were shown previously to activate ERK1/2 in different cell types through both CRHR1 and CRHR2 after stimulation by the CRH homolog urocortin (23,24). Thus, we tested whether CRH activates ERK1/2 phosphorylation in PC12 cells. Indeed, CRH directly activated the ERK1/2 MAPK pathway in PC12 cells. Cells were treated with CRH, and the levels of ERK1/2 MAPK phosphorylation at different time points were measured by Western blotting using a specific anti-phospho-ERK1/2 MAPK antibody. Phosphorylation of ERK1/2 MAPK was induced quickly after treatment with 10 Ϫ9 M CRH, showing a 2.5-fold induction within 10 min which is reduced to base-line levels within 20 min (Fig. 8C), indicating that CRH exerts a rapid and transient activation of ERK1/2. The transient activation induced by CRH may initiate proapoptotic signals that are not inhibited by PD98059, which blocks the endogenous sustained ERK1/2 activation by blocking MEK1. Alternatively, CRH-induced ERK1/2 phosphorylation may lead to events that are not essential for Fas ligand expression and apoptosis. As a result there is no change in Fas ligand protein expression levels and, therefore, no difference in the apoptosis levels with or without inhibition of ERK1/2 (Fig. 8, A and B). DISCUSSION We have found that CRH promotes PC12 cell apoptosis via its CRHR1 receptor. Its apoptosis involves activation of p38 MAPK and the Fas/Fas ligand system. Activation of p38 MAPK appears to be crucial for the proapoptotic effect of CRH because inhibition of the action of p38 MAPK blocked the stimulatory effect of CRH on both Fas ligand production and apoptosis. The effect of CRH appears to be mediated by the CRHR1 because it was inhibited by the CRHR1 antagonist antalarmin. Based on our data we propose the following sequence of events: CRH activates the CRHR1 transmitting signals, which results in phosphorylation/activation of p38 MAPK leading to Fas ligand production, which accelerates apoptosis (Table II). Our data suggest that p38 MAPK has a central role in CRH-induced Fas ligand production and apoptosis. The association is supported by previously published reports showing that p38 MAPK affects Fas ligand expression and apoptosis in PC12 cells (10,22). In NGF-differentiated PC12 cells, withdrawal of serum and NGF results in up-regulation of p38 MAPK and JNK, leading to Fas ligand expression and apoptosis. Furthermore, inhibition of p38 MAPK promotes survival in primary neuronal cell cultures and in NGF-differentiated PC12 cells, emphasizing the significance of p38 MAPK in apoptosis (13).
CRH receptors are not death receptors per se, i.e. they cannot directly induce activation of apoptotic mechanisms. CRH receptors are G protein-coupled receptors (25) that initiate signals through several intracellular pathways such as that of cAMP, protein kinase C, and MAPKs (23,26), leading to activation of transcription factors such as the cAMP response element-binding protein (27). The presence of a particular subtype of CRH receptor and the cell type can determine activation of a specific intracellular signaling pathway by CRH. For example, forced expression of CRH receptors in HEK293 and Chinese hamster ovary cells, which do not normally express these receptors, leads to activation of ERK1/2 MAPK and protein kinase C (23). In this model, urocortin activates ERK phosphorylation through both CRH receptors, but CRH fails to do so. Similarly, in primary cardiac myocytes in culture which do not express CRHR1, urocortin promotes cell survival and induces a rapid phosphorylation of ERK1/2 (26). Finally, in human cytotrophoblast cells, a model in which CRH and CRH receptors play a physiological role, CRHR1 stimulation leads to the formation of inositol trisphosphate and protein kinase C activation but fails to induce cAMP formation (28).
In the present work we employed the model of undifferentiated, chromaffin-like PC12 cells that undergo apoptosis under serum starvation conditions. In this system p38 MAPK is an essential component for Fas ligand induction after CRH exposure. Even though JNK plays an important role in Fas ligand production in NGF-differentiated PC12 cells, it does not appear to be involved in chromaffin-like PC12 cells because treatment with the inhibitor CEP-1347, which indirectly blocks JNK, does not prevent serum starvation-induced apoptosis (12). Indeed CRH did not provoke changes in JNK phosphorylation or JNK kinase activity in PC12 cells (data not shown). After treatment of Chinese hamster ovary cells with CRH, its receptors failed to FIG. 6. p38 MAPK is required for CRH-induced Fas ligand (FasL) production and apoptosis. SB203580 (a p38 MAPK inhibitor) completely inhibited CRH-induced apoptosis (panel A) and CRHinduced Fas ligand production (panel B), both in a statistically significant manner. Western analysis of cells exposed to 10 Ϫ9 M CRH with and without 20 M SB203580 for 2.5 h is shown. The levels of Fas ligand and actin were evaluated after incubation with specific antibodies. Data are expressed as percentages of parallel controls exposed only to vehicles (mean of two independent experiments). *p Ͻ 0.05 and **p Ͻ 0.01 denote a significant statistical difference compared with cells treated with 10 Ϫ9 M CRH.
FIG. 7. CRH activates p38 MAPK. PC12 cells were serum starved for 12 h and then exposed to 10 Ϫ9 M CRH for 5 min up to 24 h. Panel A, p38 MAPK activity; panel B, p38 MAPK phosphorylation. The levels of kinase activity were measured using anti-phospho-ATF-2 monoclonal antibody. The levels of MAPK phosphorylation were measured by specific anti-phospho-p38 MAPK and anti-total p38 MAPK antibodies. p38 MAPK was activated rapidly within 10 min of exposure and returned to base line after 1 h (panels A and B). SB203580 abolishes the stimulatory effect of CRH on p38 MAPK phosphorylation (panel C). Cumulative data are presented in panel D. Data are expressed as percentages of parallel controls exposed only to vehicles and represent the mean of three independent experiments. *p Ͻ 0.05 denotes a significant statistical difference compared with parallel controls. activate JNK, indicating that JNK is not a target of the signals initiated by the CRHR1 receptor (23). These observations suggest a signaling difference between neuronal-like differentiated and chromaffin-like undifferentiated PC12 cells according to which different pathways may be employed in each case for the induction of apoptosis.
The ERK1/2 system does not appear to be involved in CRHinduced apoptosis. It should be noted that both CRHR1 and CRHR2 activate ERK1/2 when force expressed in Chinese hamster ovary cells. ERK1/2 is important in the protection of PC12 cells from apoptosis, thus acting as an opposing pathway to the proapoptosis signal of p38 MAPK (22,29). Characteristic of this system is that sustained activation of ERK1/2 correlates with cell survival in several cell types, whereas transient activation is associated with induction of apoptosis. Inhibition of sustained ERK1/2 activation abolishes the antiapoptotic effect and promotes apoptosis. Here we show that CRH exerts a transient activation of ERK1/2, which, however does not appear to be associated with apoptosis because treatment with the MEK1 inhibitor PD98059 had no effect on CRH-induced apoptosis or CRH-induced Fas ligand production. The effect of CRH on ERK1/2 may be important for other effects of CRH on chromaffin cells including regulation of catecholamine production and secretion.
Our data showing that CRH induced apoptosis of PC12 cells which derive from neural crest are in agreement with recently published reports suggesting that CRH may play an important role in neuronal cell survival. Indeed, it has been shown that CRH provokes hippocampal CA3 neuron loss independently of glucocorticoids (5). This deleterious effect of CRH in hippocampus appears to be direct and paracrine, i.e. from locally synthesized CRH by CRHergic neurons localized in several areas of the developing rat hippocampus including the CA3 pyramidale and oriens strata, the lacunosum-moleculare of Ammon's horn, and the granule cell layer and hilus of dentate gyrus (30).
However, the effect of CRH on cell survival appears to be more complicated than a mere inducer of apoptosis because in primary neuronal cell cultures, CRH exerts a protective effect against cell death caused by amyloid-␤ peptides, lipid peroxidation, or glutamate (31). The protective effect of CRH is blocked by CRH receptor antagonists.
We consider the proapoptotic effect of CRH to be physiologically relevant for at least two reasons. First, the range of CRH concentrations found to be effective on apoptosis was similar to the range believed to be present within the adrenal medulla (32)(33)(34)(35)(36). Second, normal chromaffin cells, like the PC12 cells, have specific CRH binding sites (37,38). The major sources of adrenal CRH are the medullary chromaffin cells (9), adrenal resident immune cells (39), and preganglionic nerve terminals. This characteristic of normal adrenal chromaffin cells appears to be retained by at least some pheochromocytomas because several human pheochromocytomas and the human (KAT45) and rat (PC12) pheochromocytoma cell lines produce CRH (9, 36, 39 -44). The physiological role of adrenal CRH appears to be confined within the gland being paracrine or autocrine. It is now believed that a CRH-containing system exists within the adrenal gland regulating adrenal pre-and postnatal growth, daily steroidogenic activity, catecholaminergic tone, and response to stress (9,(45)(46)(47). Furthermore, adrenal CRH appear to be crucial in the adrenal cortical cell response to pituitaryderived systemic ACTH because antalarmin attenuates adrenal responsiveness to ACTH (19). Thus, induction of apoptosis by CRH may be an additional regulatory mechanism within the adrenal gland.
In conclusion, our data indicate that CRH is an inducer of cell apoptosis in the PC12 rat pheochromocytoma cells. The proapoptotic effect of CRH is mediated by CRHR1 and involves activation-phosphorylation of the p38 MAPK, which induces Fas ligand production causing acceleration of apoptosis.
FIG. 8. CRH activates ERK1/2, which is not required for the induction of Fas ligand or apoptosis. Cells were exposed to 10 Ϫ9 M CRH with and without 50 M PD98059 in serum-deprived RPMI 1640 medium for 3 days and lysed as described under "Experimental Procedures." Panel A, measurement of apoptosis. Panel B, Fas ligand production expressed as the Fas ligand:actin ratio. Panel C, cells were exposed to 10 Ϫ9 M CRH for 10 min, 20 min, 1 h, and 2.5 h. Western blot analysis was performed using antibodies specific for phosphorylated and total ERK1/2. CRH provoked a transient activation of ERK1/2. Data are expressed as percentage of parallel controls exposed only to vehicles and represent the mean of two independent experiments. **p Ͻ 0.01 denotes a statistical significance compared with parallel controls, i.e. cells exposed only to vehicles.