HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 20, 17649-17656, May 17, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/20/17649    most recent M111704200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Wei, W. Articles by Kusiak, J. W. Search for Related Content PubMed PubMed Citation Articles by Wei, W. Articles by Kusiak, J. W. Social Bookmarking                      What's this?

Signaling Events in Amyloid -Peptide-induced Neuronal Death and Insulin-like Growth Factor I Protection^*

Wanli Wei, Xiantao Wang, and John W. Kusiak

From the Molecular Neurobiology Unit, Laboratory of Cellular and Molecular Biology, NIA, Intramural Research Program, National Institutes of Health, Baltimore, Maryland 21224

Received for publication, December 7, 2001, and in revised form, February 28, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Amyloid -peptide (A) is implicated as the toxic agent in Alzheimer's disease and is the major component of brain amyloid plaques. In vitro, A causes cell death, but the molecular mechanisms are unclear. We analyzed the early signaling mechanisms involved in A toxicity using the SH-SY5Y neuroblastoma cell line. A caused cell death and induced a 2- to 3-fold activation of JNK. JNK activation and cell death were inhibited by overexpression of a dominant-negative SEK1 (SEK1-AL) construct. Butyrolactone I, a cdk5 inhibitor, had an additional protective effect against A toxicity in these SEK1-AL-expressing cells suggesting that cdk5 and JNK activation independently contributed to this toxicity. A also weakly activated ERK and Akt but had no effect on p38 kinase. Inhibitors of ERK and phosphoinositide 3-kinase (PI3K) pathways did not affect A-induced cell death, suggesting that these pathways were not important in A toxicity. Insulin-like growth factor I protected against A toxicity by strongly activating ERK and Akt and blocking JNK activation in a PI3K-dependent manner. Pertussis toxin also blocked A-induced cell death and JNK activation suggesting that G_i/o proteins were upstream activators of JNK. The results suggest that activation of the JNK pathway and cdk5 may be initial signaling cascades in A-induced cell death.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Alzheimer's disease (AD)¹ brain is characterized by the selective loss of synapses and neurons and the presence of amyloid plaques composed primarily of aggregated amyloid -peptide (A) 40 to 42 amino acids in length (1, 2). A is derived from the proteolytic processing of the amyloid precursor protein (APP) and is hypothesized to be the toxic agent in AD. Previous reports have suggested that A-induced cell death involves oxidative stress and disturbance of intracellular calcium homeostasis (1, 3). A can cause cell death by both apoptotic and necrotic mechanisms in vitro and evidence for both types of cell death has been reported in AD brain (4-6). However, the underlying mechanisms of toxicity and the neuronal cellular signaling cascades activated by A are not fully understood.

Mitogen-activated protein kinase (MAPK) family members, including extracellular signal-regulated kinase (ERK), c-Jun N-terminal kinase (JNK), and p38 MAPK have been proposed to be important signaling components linking extracellular stimuli to cellular responses. The ERK pathway plays a major role in regulating cell growth and differentiation (7). JNK and p38 are highly activated in response to a variety of stress signals, including tumor necrosis factor, ultraviolet irradiation, and hyperosmotic stress (8, 9). Their activation is most frequently associated with induction of apoptosis. The serine/threonine kinase Akt/protein kinase B is activated via a phosphoinositide 3-kinase (PI3K)-dependent signaling pathway when cells or tissues are exposed to growth factors, insulin, and certain cytokines (10). Akt has received widespread attention as an important anti-apoptotic protein through which various survival signals suppress cell death induced by growth factor withdrawal, cell cycle disruption, or detachment of cells from their extracellular matrix (7, 12). The role of MAPKs and PI3K/Akt pathways in A toxicity is unclear.

Cyclin-dependent kinase 5 (cdk5) is a small serine/threonine kinase, which is required for normal development of the mammalian central nervous system. p35 and its cleavage product p25 regulate cdk5 activity. Whereas p35/cdk5 interactions are necessary for proper development and other functions of the mature nervous system, p25 binds with high affinity to cdk5 and constitutively activates cdk5, causing hyperphosphorylation of tau, collapse of the cytoskeleton, and cell death. p25 has been found to accumulate in neurons in the brains of patients with AD (13). Persistent activation of cdk5 has been associated with AD (14).

It is unclear whether A interacts with cell surface receptors or cytoplasmic effectors to cause cell death. A interaction with the receptor for advanced glycation end products on neurons elicits an NF-B-dependent increase in macrophage colony-stimulating factor release and cellular oxidative stress (15). This response may involve multiple signaling pathways, including the small GTPases Rho and Rac (16). A binding to the 75-kDa neurotrophin receptor (p75^NTR) also activates NF-B in human neuroblastoma cells and leads to apoptotic cell death (17). In rat cortical neurons and NIH-3T3 cells expressing p75^NTR, A specifically bound with high affinity to p75^NTR and caused apoptotic death (18, 19). Two recent reports have described the binding of A to cell surface APP (20) and to a novel protein, -amyloid peptide binding protein (BBP) (21). In both of these cases A treatment of cells expressing these proteins causes apoptotic cell death. A recent report has shown that aggregated A and calcitonin interact directly with G proteins to stimulate a high affinity GTPase activity, which correlates with the toxicity of the peptides (22). The G proteins activated by both peptides are G_o and G_i. G protein-coupled receptors, upon activation, can regulate JNK activity (23, 24). Recent evidence suggests that both G and free subunits activate JNK in a manner dependent upon Rac1 and CdC42 (25). Thus, the possibility exists that A may interact with G protein-coupled receptors or G proteins directly to activate them and lead to JNK activation and cell death.

Insulin-like growth factor (IGF-I) has been shown to protect cells against numerous stressors (26). It has been reported that IGF-I protects hippocampal neurons from both A and amylin toxicity (27). IGF-I also protects cells transfected with familial AD mutant forms of APP from apoptotic cell death (28). Furthermore, IGF-I suppresses JNK activation induced by several stressors, including anisomycin and tumor necrosis factor-, and inhibits cell death. The initial signaling involved in this protection has been shown to involve both ERK and PI3K-dependent pathways (29).

A interaction with neurons may disrupt the normal homeostasis of cell survival and cell death signaling pathways to favor the latter. In this study we sought to determine the cell signaling cascades that are activated upon A stimulation of neuronal cells and to determine the mechanism by which IGF-I may protect against A-induced death. Our results showed that A treatment of SH-SY5Y cells caused cell death by apoptosis in response to activation of JNK. JNK activation was important for this death, because inhibition of JNK activity blocked A-induced death. Pertussis toxin (PTX) treatment protected against A-induced death suggesting that G protein activation upstream of JNK may be important in A-mediated cell death.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Reagents-- Media and N2 supplements for cell culture were from Invitrogen (Carlsbad, CA). A peptides, including A 25-35, A 1-40, A 1-42, and A 42-1 were purchased from Biosource International (Camarillo, CA). A 1-40, A 1-42, and A 42-1 were dissolved in hexafluoroisopropanol and dried under a stream of argon gas. The dried peptides were then dissolved in water at 1 mM and incubated at 37 °C for 48-72 h before use. A 25-35 was dissolved in water and incubated at 37 °C for 24 h before use. Anti-p38, anti-phospho-p38, anti-phospho-Akt, and anti-Akt1 antibodies were purchased from Cell Signaling Technology (Beverly, MA). Anti-JNK1 antibody was obtained from Santa Cruz Biotechnology (Santa Cruz, CA). U0126 and anti-phospho SAPK/JNK, anti-phospho-ERK1/2, and anti-ERK1/2 antibodies were purchased from Promega (Madison, WI). Wortmannin and butyrolactone I (BT-I) were purchased from Calbiochem (La Jolla, CA).

Cell Cultures and Transfection-- Undifferentiated human neuroblastoma SH-SY5Y cells were grown in 50% minimal essential medium, 50% F-12 nutrient mixture, with 10% fetal bovine serum, 1× minimal essential medium non-essential amino acids, and 1× antibiotic-antimycotic (100 units/ml penicillin, 100 µg/ml streptomycin, and 2.5 µg/ml Fungizone) at 37 °C under 5% CO₂/95% air. In cell viability assays, cells were treated with various peptides in serum-free media containing N2 supplements. In signaling experiments, cell were incubated in serum-free media for 16 h and then treated with various peptides for the indicated time periods.

Transfections were performed with LipofectAMINE (Invitrogen) according to the manufacturer's protocol. SH-SY5Y cells were transfected with either pcDNA3.zeo or pcDNA.zeo-SEK1-AL constructs, kindly provided by Dr. James R. Woodgett. Pooled cells stably expressing pcDNA3.zeo or pcDNA.zeo-SEK1-AL were selected by adding 200 µg/ml Zeocin (Invitrogen) for 2 months.

Measurement of Cell Death-- Trypan Blue exclusion was used to measure cell death by counting the number of dead (blue) and live cells in the cultures after A peptide treatment. A cell death detection ELISA kit (Roche Molecular Biochemicals, Indianapolis, IN) was used to detect apoptosis after A peptide treatment. The assay is based on a quantitative sandwich ELISA using antibodies directed against DNA and histones to detect mono- and oligonucleosomes in the cytoplasm of cells undergoing apoptosis. The ELISA was carried out according to the manufacturer's protocol. Hoechst labeling of cells to detect nuclei was performed as described previously (30). In brief, prior to staining, the cells were fixed with 4% paraformaldehyde for 30 min at room temperature. Hoechst was added to the fixed cells for 30 min, after which they were examined by fluorescence microscopy. Apoptotic cells were identified by condensation and fragmentation of nuclei. Percentage of apoptotic cells was calculated as the ratio of apoptotic cells to total cells counted × 100. A minimum of 400 cells was counted for each treatment.

Immunoblot Analysis-- Cells were harvested in 300 µl of lysis buffer A (20 mM Hepes, pH 7.4, 2 mM EGTA, 50 mM -glycerol phosphate, 1% Triton X-100, 10% glycerol, 1 mM dithiothreitol (DTT), 1 mM phenylmethylsulfonyl fluoride, 10 µg/ml leupeptin, 10 µg/ml aprotinin, 1 mM Na₃VO₄, and 5 mM NaF). The resulting lysates were resolved on 4-12% NuPAGE Bis-Tris gels (30 µg/lane) and transferred onto polyvinylidene difluoride membranes (Millipore, Bedford, MA). The membranes were blocked in TBST (10 mM Tris-HCl, pH 7.4, 150 mM NaCl, 0.1% Tween 20) containing 5% non-fat milk and then probed with different antibodies. Proteins were detected by using enhanced chemiluminescence (ECL) reagents (PerkinElmer Life Sciences, Boston, MA).

Immunoprecipitations and Kinase Assays-- Cells were harvested in 500 µl of lysis buffer A. JNK/SAPK was immunoprecipitated by the addition of 1 µg of anti-JNK antibody to the cell lysates for 3 h, with the addition of 40 µl of a 50% slurry of Protein A-Sepharose during the final hour of incubation. The beads were pelleted by centrifugation and washed three times each in lysis buffer A and wash buffer (100 mM Tris/HCl, pH 7.6, 500 mM LiCl, 0.1% Triton X-100, 1 mM DTT). The beads were left as a 1:1 suspension in assay buffer and 20 µl (0.3 mg/ml) of GST-c-Jun was added. Kinase reactions were initiated by the addition of 15 µl of solution containing 50 mM MgCl₂, 500 µM ATP, 10 µCi of [-³²P]ATP (3000 Ci/mmol) and incubated at 30 °C for 20 min. Reactions were stopped by the addition of Laemmli sample buffer and boiling for 5 min. Samples were separated on 12% SDS-PAGE, and the dried gels were subjected to autoradiography. Quantification was performed with a PhosphorImager (Molecular Dynamics, Sunnyvale, CA).

Data Analysis-- Quantitative data are expressed as arithmetic means ± S.E. based on at least three separate experiments performed in duplicate. The difference between two groups was statistically analyzed by Student's t test or an analysis of variance (one-way analysis of variance). A p value of <0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

A Toxicity-- Previous results suggest that aggregated A peptide is highly toxic to a variety of cultured primary neurons and neuronal cell lines. SH-SY5Y cells were treated with aggregated A peptides for 48 h in serum-free media containing N2 supplements. A 1-42 and A 25-35 were dose-dependently toxic (Fig. 1A), each causing up to 50-60% cell death at 50 µM. In contrast, reverse peptide A 42-1 was not toxic nor was A 1-40 at low concentrations. Only at 50 µM, A 1-40 caused ~30% cell death. We also used an apoptotic cell death detection ELISA (Fig. 1B) and Hoechst labeling (data not shown) to measure apoptosis and found that A 1-42 and A 25-35, at low concentrations (~5 µM), caused the majority of the cells to die by apoptosis. A 1-40 at all concentrations tested did not cause apoptotic cell death. Treatment with high doses of A peptides caused both apoptosis and necrosis.

View larger version (23K): [in this window] [in a new window]   Fig. 1.   A toxicity. SH-SY5Y cells were treated with A peptides for 48 h at 37 °C, and Trypan blue exclusion (A) and ELISA (B) were used to determine cell death. Data are means ± S.E. for three separate experiments performed in duplicate.

A Activates ERK and Akt Signaling Pathways in SH-SY5Y Cells-- To investigate whether A treatment led to ERK and Akt activation, lysates from cells treated with A peptides for various times were subjected to Western blot analysis using anti-phospho-ERK and -Akt antibodies to detect activated ERK and Akt, respectively (Fig. 2, A and B). Duplicate blots were probed with antibodies recognizing total ERK1/2 and Akt1 to verify equal protein loading in the samples. As shown in Fig. 2A (upper panel), treatment with 4 µM A 1-42, which resulted in mainly apoptotic death, led to increased phosphorylation of ERK1/2 within 15 min and this phosphorylation remained elevated for 2 h. Treatment with 4 µM A 1-42 also weakly activated Akt, and this activation was sustained over the following 24-h period (Fig. 2B). Activation of ERK and Akt pathways have been shown to promote cell survival/proliferation after growth factor stimulation and to play a protective role after oxidant treatment (7, 31). To evaluate whether these weak ERK and Akt activations affected A-induced apoptosis, we used specific inhibitors of MEK1 (U0126) and PI3K (wortmannin), which are highly selective in their inhibition of the ERK and Akt pathways. Cells were pretreated with 5 µM U0126 or 200 nM wortmannin for 1 h prior to addition of 4 µM A 1-42 for 24 h. Treatment of cells with these two inhibitors during exposure to A 1-42 completely abolished the A-evoked ERK and Akt activation (see Fig. 6C below) and slightly increased the amount of apoptotic cell death compared with A treatment alone (Fig. 2D). The similar amounts of cell death measured by both Trypan blue exclusion (Fig. 2D, top panel) and Hoechst labeling (Fig. 2D, lower panel) suggested that most of the death induced by 4 µM A 1-42 is apoptotic. These results also suggested that this level of ERK and Akt activation was ineffective in protecting against A-induced apoptosis.

View larger version (44K): [in this window] [in a new window]   Fig. 2.   Effect of A 1-42 on ERK and Akt phosphorylation in SH-SY5Y cells. A, Western blot analysis shows ERK1/2 phosphorylation in cells treated with 4 µM A 1-42 for the indicated times after a 16-h serum starvation period. B, Western blot analysis shows Akt phosphorylation in cells treated as in panel A. C, Western blot analysis shows no p38 kinase phosphorylation in cells treated as in panel A. Lane P, positive control, lysates of SH-SY5Y cells treated with 300 µM arsenite for 1 h. The blots are representative of three separate experiments. Upper panels in A-C are Western blots using phosphorylation-specific antibodies; the lower panels are Western blots with antibodies recognizing total ERK1/2, Akt1, and total p38 kinase. D, cells were either left untreated or pretreated with 200 nM wortmannin or 5 µM U0126 for 1 h followed by treatment for 24 h with or without 4 µM A 1-42. Cell death was measured by Trypan blue exclusion (top panel) and Hoechst nuclear labeling (lower panel). *, p < 0.05 versus untreated cells.

p38 kinase has been implicated in apoptotic cell death in some cell systems and is activated by several types of cell stress (32). To evaluate the role of p38 kinase on A-induced neuronal death, p38 phosphorylation was measured at various times after A treatment. Total p38 protein levels were monitored using antibodies that recognized both the phosphorylated and the unphosphorylated forms of the protein. As shown in Fig. 2C, p38 kinase was not activated during a 24-h exposure to 4 µM A 1-42. This result suggested that p38 kinase activation was not involved in A-induced neuronal death.

Activation of JNK Is Critical for A-induced Neuronal Death-- To study the role of JNK in A-induced neuronal death, we examined JNK activity in SH-SY5Y cells following exposure to A for various lengths of time. JNK activity was assessed by using an immunocomplex kinase assay with GST-c-Jun-(1-135) fusion protein as a substrate. Western blotting of immunocomplexes to detect JNK1 was performed to verify the presence of equal amounts of the protein. JNK was rapidly activated at early time points by both A 1-42 (Fig. 3A) and A 25-35 (Fig. 3B). A 25-35 at 40 µM also activated JNK at late time points (15-24 h, Fig. 3B) when significant cell death occurred. To investigate the functional consequences of JNK activation by A, we transfected SH-SY5Y cells with plasmids containing either empty vector or a dominant-negative SAPK/ERK kinase 1 (SEK1-AL) mutant, the upstream kinase responsible for the activation of JNK (33). A pool of stably transfected cells constitutively expressing HA-tagged mutant SEK1-AL was identified by Western blotting using an anti-HA antibody (Fig. 4A) and analyzed for their response to A treatment. Control cells transfected with the empty vector showed a similar pattern of JNK activation as that of naive cells after 40 µM A 25-35 stimulation with a maximal 3-fold activation seen at 15 min. In contrast, cells that overexpressed mutant SEK1-AL exhibited no JNK activation up to 30 min after 40 µM A 25-35 (Fig. 4B) and 4 µM A 1-42 treatment (data not shown). To determine whether this decreased JNK activity correlated with suppression of A-induced neuronal death, cell death was assessed by Trypan blue exclusion (Fig. 4, C and D) and by Hoechst labeling (Table I) after A treatment of mutant SEK1-AL and control cells. Cells overexpressing mutant SEK1-AL were significantly protected by between 48 and 70% against A 25-35- and A 1-42-induced cell death especially at low concentrations compared with cells transfected with vector alone. These results suggested that JNK activation is an important and early event in A-induced neuronal death.

View larger version (69K): [in this window] [in a new window]   Fig. 3.   A induces JNK activation in SH-SY5Y cells. A, kinase assay shows JNK activation in cells treated with 4 µM A 1-42 and B, with 40 µM A 25-35 at various times after a 16-h serum starvation period. Upper panels are results of JNK kinase assays; lower panels are Western blots of immunoprecipitates with an antibody recognizing total JNK1. The results are representative of three separate experiments.

View larger version (30K): [in this window] [in a new window]   Fig. 4.   Expression of dominant-negative SEK1-AL inhibits JNK activation by A. A, Western blot analysis of SH-SY5Y cells stably expressing HA-tagged SEK1-AL using an anti-HA antibody. B, kinetics of JNK activation and its attenuation in SEK1-AL-expressing cells treated with 40 µM A 25-35 by using kinase assay. C and D, SEK1-AL-expressing SH-SY5Y cells showed decreased cell death in response to A treatment. Control cells (empty vector) and cells expressing SEK1-AL were treated with A 25-35 (C) and A 1-42 (D) for 48 h. Cell death was measured by Trypan blue exclusion. Data are means ± S.E. for six separate experiments performed in duplicate.

View this table: [in this window] [in a new window]   Table I Effect of BT-I on SEK1-AL cells against A-induced apoptosis SH-SY5Y cells stably expressing SEK1-AL or empty vector (5YV) were either left untreated or treated with BT-I and/or A for 24 h at 37°C, and Hoechst nuclear labeling was used to determine apoptosis. Data are means ± S.E. for three separate experiments performed in duplicate. Protection percentage was calculated according to the formula: [(A treatment of 5YV=Control) =(treatment of BT-I or SEKI-AL cell = Control)/ (A treatment of 5YV= Control) × 100%.

Effect of BT-I on A-induced Toxicity in SEK-1 AL-transfected Cells-- Because SEK1-AL-expressing cells were only partially protected against A-induced death, yet they had a complete inhibition of JNK activity, we wanted to know if other kinase pathways were important in this process. We sought to investigate if cdk5 activity was required for A toxicity, because prolonged activation of cdk5 has been associated with AD (13). Cdk5 activity is found only in neurons and is controlled by a neuron-specific regulatory protein, p35, and its cleavage product p25. Treatment of control SH-SY5Y cells with a cdk5 inhibitor, BT-1, partially protected against apoptotic cell death induced by low concentrations of A peptides. When added to cells overexpressing SEK1-AL, BT-1 completely protected these cells against low dose A toxicity, suggesting that activation of both the JNK and cdk5 pathways is required for A-induced apoptosis (Table I).

PTX Protects Against A-induced Neuronal Death by Inhibiting A-induced JNK Activation-- To determine the role of G proteins in A-induced neuronal death, we used PTX, a specific inhibitor of G_i/o proteins. Cells were pretreated with PTX for 1 h prior to addition of A peptides and cell death was measured 24 h later by Trypan blue exclusion. PTX treatment of cells alone did not cause cell death, but the total cell numbers were about 20% less than those in untreated cultures, suggesting reduced proliferation. PTX protected against both A 1-42- and A 25-35-induced cell death (Fig. 5A). To determine the effects of G proteins on signaling events in A-induced neuronal death, cells were treated with 50 µM PTX and 4 µM A 1-42 for 5 min. Lysates of treated cells were subjected to Western blot analysis using anti-phospho-ERK1/2, -Akt, and -JNK antibodies to detect phosphorylated ERK1/2, Akt, and JNK, respectively. PTX did not alter the A-induced ERK and Akt activation (Fig. 5B, upper and middle sets of panels). However, PTX inhibited the A-induced JNK activation by 70% (Fig. 5B, lower set of panels). These results suggested that G_i/o protein activation was involved in A-induced neuronal death. Further, G protein activation appeared to be an upstream event in the JNK activation by A.

View larger version (39K): [in this window] [in a new window]   Fig. 5.   Effect of PTX on A toxicity in SH-SY5Y cells. A, cells were either left untreated or pretreated with 50 µM PTX for 1 h, then treated with 4 µM A 1-42, 5 µM A 25-35, or 30 µM A 25-35 for 24 h. Cell death was measured by Trypan blue exclusion. *, p < 0.05 versus untreated cells; #, p < 0.05 versus A-treated cells. B, effect of PTX on ERK, Akt, and JNK activation after A treatment. Cells were serum-starved for 16 h and then were either left untreated or pretreated with 50 µM PTX for 1 h followed by 4 µM A 1-42 treatment for 5 min. Cells were harvested, and the cell lysates were subjected to Western blot analysis of phospho-ERK, -Akt, and -JNK and total ERK, Akt1, and JNK1.

IGF-I Protects Against A-induced Neuronal Death by Activating Akt and ERK-- To address if IGF-I protection against A toxicity occurred by activation of either PI3K/Akt or ERK pathways or if it occurred by inhibition of JNK activation, we assessed the effect of IGF-I on A-induced neuronal death and signaling events. Cells were treated with IGF-I and A 1-42 (4 µM) or A 25-35 (5 µM) for 48 h. Cell death was determined by Trypan Blue exclusion and Hoechst labeling. IGF-I protected cells from A 1-42- and A 25-35-induced death in a dose-dependent manner (Fig. 6A, left panel). 20 nM IGF-I almost completely prevented A-induced apoptosis at low doses of peptides (Fig. 6A, right panel).

View larger version (22K): [in this window] [in a new window]   Fig. 6.   Effect of IGF-I on A toxicity in SH-SY5Y cells. A, untreated cells and cells that were treated with IGF-I together with or without A 1-42 at 4 µM and A 25-35 at 5 µM for 48 h were assessed for cell death as measured by Trypan blue exclusion (left) and Hoechst labeling (right). *, p < 0.05 versus untreated cells; #, p < 0.05 versus A-treated cells. B, cells were left untreated or pretreated with 200 nM wortmannin or 5 µM U0126 for 1 h followed by 4 µM A 1-42 and 10 nM IGF-I treatment for 48 h. The cell death was measured by Hoechst labeling. *, p < 0.05 versus untreated cells; #, p < 0.05 versus A treated cells; **, p < 0.05 versus A+IGF-I treated cells. C, effect of IGF-I on ERK and Akt activation after A treatment. Cells were either left untreated or pretreated with 200 nM wortmannin or 5 µM U0126 for 1 h and followed by 4 µM A 1-42 and 10 nM IGF-I for 5 min after 16-h serum starvation period. Cells were harvested, and cell lysates subjected to Western blot analysis of phosphorylated and total ERK and Akt. D, effect of IGF-I on JNK activation after A treatment. Following a 16-h serum starvation period, cells were either left untreated or pretreated with 200 nM wortmannin or 5 µM U0126 for 1 h, followed by treatment with 4 µM A 1-42 and/or 10 nM IGF-I for 5 min. Cells were harvested and cell lysates subjected to JNK kinase assay.

To determine whether PI3K/Akt or ERK pathways mediated the protective effect of IGF-I on A-induced cell death, cells were preincubated with either wortmannin or U0126 for 1 h prior to addition of 4 µM A1-42. Pre-treatment with either wortmannin or U0126 totally abolished the protective effect of IGF-I (Fig. 6B) and blocked the IGF-I-evoked Akt and ERK activation, respectively (Fig. 6C). These results strongly suggest that IGF-I-induced phosphorylation of Akt and ERK was necessary for protection from A-induced death.

Because we found that JNK activation plays a critical role in A-induced death, we examined the effects of IGF-I on JNK activation following exposure to A. Treatment with 10 nM IGF-I inhibited A 1-42-induced JNK activation (Fig. 6D). To determine whether the effect of IGF-I on suppression of A-induced JNK activity involved the MEK1/ERK or PI3K/Akt pathways, cells were pretreated with either U0126 or wortmannin for 1 h before addition of IGF-I and A peptide for 5 min. U0126 did not alter the IGF-I effect on A-induced JNK activity (Fig. 6D). However, wortmannin blocked IGF-I suppression of A-induced JNK activity by 90% (Fig. 6D). Thus, IGF-I inhibition of A-induced JNK activation occurred via a PI3K/Akt-dependent signaling pathway. Our results suggest that IGF-I protected against A-induced death by regulating multiple signaling pathways, including a potent, direct activation of cell survival pathways (PI3K/Akt and ERK), and the inhibition of the JNK cell death pathway in a PI3K-dependent manner.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The major findings of this study are: First, JNK was rapidly activated by A treatment and that this activation appeared to be critical for A-induced neuronal death. Second, IGF-I protected cells against A-induced cell death, in part by a mechanism that involved inhibition of JNK activation via a PI3K-dependent pathway. Third, PTX protected against A-induced cell death. Because PTX specifically inhibits G_i/o proteins, this result suggests that A-induced cell death may be caused, at least in part, by activation of a G protein-coupled receptor or by direct activation of these G proteins.

JNK is a stress-activated protein kinase, and its activation has been associated with the induction of apoptosis by several types of environmental stress in neuronal cells (9, 34, 35). Our results in the present study suggest that JNK activation was critical for A-induced neuronal death. A induced a 2- to 3-fold activation of JNK. JNK activation occurred rapidly, as early as 5 min after addition of A to the cells suggesting the possibility of a receptor-mediated response. Overexpression of a dominant interfering form of SEK1 (MKK4), the upstream kinase of JNK, treatment with IGF-I, or pretreatment with PTX, each prevented the cell death induced by A and inhibited JNK activation. Our results are consistent with recent reports showing that JNK activation is involved in A-induced apoptosis of PC12 cells and cortical neurons (36, 37). Several groups have shown that A induction of neuronal apoptosis is inhibited when c-Jun function is blocked (37, 38). Other evidence has shown that JNK and c-Jun are activated in degenerating or apoptotic neurons in AD brains (39-43). Taken together, these findings suggest that activation of the JNK/c-Jun pathway may be an important early event in A-induced cell loss in AD.

Overexpression of a dominant negative SEK1 construct completely blocked A-evoked JNK activation and caused 48-70% protection against A-induced cell death. Treatment of cells overexpressing SEK1-AL with BT-I, a cdk5 inhibitor, totally protected against low dose A-induced apoptosis, suggesting that cdk5 and JNK activation initiated distinct processes in A toxicity. However, our results do not exclude the possibility that other mechanisms may also mediate A-induced neuronal loss in AD.

We also examined the effects of A on other kinase pathways. In SH-SY5Y cells, A weakly activated both ERK1/2 and Akt. However, this activation did not appear to be required for A-induced cell death, because U0126 and wortmannin completely inhibited ERK1/2 and Akt activation, respectively, and slightly increased the amount of A-induced apoptosis. Activation of both ERKs and Akt are important in cellular responses to a variety of extracellular stimuli. ERK activation in response to growth factors and neurotrophic factors contributes to cell proliferation, growth, and differentiation, but the role of ERK activation in response to A peptides in neurons is unclear. Two groups have reported that A induces a sustained increase in ERK phosphorylation in mature hippocampal neurons and that an MEK1 inhibitor prevents tau phosphorylation and neurite degeneration caused by A (44, 45). On the other hand, Ekinci and Abe showed that A did not affect ERK phosphorylation in either cortical or hippocampal neurons (46, 47). The effect of A on microglial cells seems to be more consistent and caused activation of these cells and the release of nitric oxide and chemokines (48, 49). Activation of Akt protects cells from apoptotic signals such as growth factor withdrawal, cell cycle disruption, and cell detachment (50, 51). The role of Akt activation after A treatment is unknown; the weak activation noted in the present study may reflect an ineffective cellular protective response of the cells to a toxic stimulus. Interestingly, because U0126 and wortmannin treatment led to a slight increase in A-induced apoptosis, this result also suggested that the basal activity of these pathways is important in cell viability. Finally, p38 kinase has also been implicated in apoptotic cell death in some cell systems (32). Similar to our present results, Troy et al. (36) recently reported no activation of p38 kinase by A in their studies suggesting that this kinase was not critical for A-induced neuronal death.

IGF-I is a growth factor important in early developmental processes in the central nervous system and in peripheral tissues. In brain, IGF-I has potent neuroprotective and neurotrophic effects (52). IGF-I and its receptors are highly concentrated in the hippocampus, an area of the brain severely affected in AD (53, 54). Previous reports have shown that IGF-I is protective against A- and amylin-induced cell death (27), but the signaling mechanisms involved have not been explored. We showed that IGF-I protected against A-induced cell death and strongly activated Akt and ERK. Specific inhibitors of PI3K and MEK1 blocked this activation and abolished the IGF-I protection of A-induced cell death, indicating that activation of Akt and ERK was important for IGF-I protection. IGF-I blocked JNK activation, and wortmannin abolished this effect suggesting that IGF-I inhibited A-induced JNK activation by a PI3K-dependent pathway. U0126 did not block IGF-I suppression of A-evoked JNK activation but still abolished the IGF-I protective effect (Fig. 6B) and ERK activation (Fig. 6C). This suggested that ERK activity was required for the cell survival effect of IGF-I even in the absence of JNK activation. Moreover, this suggested that A also induced cell death by a mechanism other than JNK activation, possibly by activation of cdk5 (13, 55, 56). Our results are consistent with recent reports of JNK regulation in HEK 293 cells. In this case IGF-I suppresses anisomycin- and tumor necrosis factor--induced JNK activation in a PI3K/Akt-dependent manner (29). Furthermore, IGF-I protects cultured cerebellar granule cells from low potassium-induced apoptosis in a PI3K/Akt-dependent manner (58). In contrast, Cheng and Feldman (59) recently showed that IGF-I protects SH-SY5Y cells from high glucose-induced programmed cell death; in this case the JNK activation induced by high glucose is inhibited by IGF-I in an ERK-dependent manner, because PD98059 prevents the IGF-I-induced JNK inhibition. Thus, the neuroprotective effects of IGF-I may be due in part to its regulation and integration of early signaling events in several kinase cascades. The mechanism of the PI3K-dependent JNK inhibition by IGF-I is unclear but may involve negative regulation of MAP kinase kinase 3/4 or JNK perhaps by activation of a MAPK phosphatase (60). In this regard, a recent report (61) showed that activation of Akt negatively regulates the upstream kinase of JNK, SEK1, by phosphorylation and inhibits apoptosis. Furthermore IGF-I also has been shown to activate ERK in a PI3K-dependent manner at the level of Raf-1 (an mitogen-activated protein kinase kinase kinase) in neuroectodermal cells (62). This PI3K-dependent ERK activity may then act downstream of JNK to suppress JNK activity. Thus ERK-dependent protection by IGF-I, although not inhibiting JNK activation, may act at a point downstream of JNK where survival and death/stress signaling converge to determine the ultimate fate of the cell (59).

Activation of G protein-coupled receptors by peptides, hormones, neurotransmitters, and growth factors stimulates families of G proteins, which in turn, control divergent cellular activities, including protein phosphorylation, gene transcription, cytoskeletal reorganization, secretion, and membrane depolarization (63, 64). It has become clear that both the G and G subunits of G proteins are involved in regulating multiple intracellular signaling cascades (65). We showed that PTX, a specific G_i/o inhibitor, protected against A-induced cell death, suggesting that G_i/o proteins were activated after A treatment. We also showed that PTX treatment inhibited A-induced JNK activation, suggesting that G protein activation is an upstream event in JNK activation. Our results are also consistent with reports showing that G proteins regulate JNK activity in other systems (23, 24, 66, 67). The activation of JNK by G proteins may proceed via the interaction of G subunits with two guanine nucleotide exchange factors including Rac1 and Cdc42 (25). It remains to be determined whether A activates G proteins directly or via receptor-mediated processes. Previous results have shown that fibrillar or protofibrillar forms of A increase high affinity GTPase activity of purified G_o and G_i, which correlates with A toxicity, suggesting that A may directly interact with G proteins (22). A also may interact with cell surface receptors such as p75^NTR or APP to initiate a G protein-dependent toxic mechanism (17, 20). A recent report described a novel protein, BBP, that bound A with high affinity in vitro (21). Upon expression in cell culture, BBP sensitized cells to apoptotic death induced by A. Interestingly, BBP contains a G protein-coupling domain similar to the seven-transmembrane domain G protein-coupled receptors, and the toxic effects of A are abrogated by treatment with PTX. Other evidence has also shown that APP itself interacts with G_o protein to activate GTPase activity (68). Interestingly, several recent reports have shown that overexpression of familial AD mutant forms of APP and presenilin 2 cause apoptotic cell death, which is prevented by PTX (11, 57, 69). Both familial AD APP and presenilin 2 overexpression are known to increase production of A 1-42, which may in turn interact with and activate G proteins (57). Thus, A may interact with G protein-coupled receptors or G proteins directly to activate them and lead to JNK activation and cell death.

In conclusion, we demonstrate here that activation of the JNK pathway was critical in A toxicity. Activation of cdk5 may also be important in this cytotoxic response. Activation of G_i/o proteins may be one of several important upstream events in A toxicity. The neurotrophic factor IGF-I protected against A-induced neuronal death via activation of ERK and Akt and the inhibition of JNK activity. These results elucidate signaling mechanisms of A toxicity and identify possible therapeutic target pathways for preventing neuronal death and supporting neuronal survival in neurodegenerative diseases.

	ACKNOWLEDGEMENTS

We thank Darrell Norton for support during the course of these studies and Drs. Yusen Liu, Ron Wange, and Myriam Gorospe for helpful discussions.

	FOOTNOTES

^* This research was funded by the Intramural Research Program of NIA, National Institutes of Health.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence may be addressed: Molecular Neurobiology Unit, Laboratory of Cellular and Molecular Biology, Intramural Research Program, Gerontology Research Center, National Institute of Aging, NIH, 5600 Nathan Shock Dr., Baltimore, MD 21224-6825. Tel.: 410-558-8467; Fax: 410-558-8386; E-mail: jk133r@nih.gov (J. W. K.) or weiwa@grc.nia.nih.gov (W. W.).

Published, JBC Papers in Press, March 6, 2002, DOI 10.1074/jbc.M111704200

	ABBREVIATIONS

The abbreviations used are: AD, Alzheimer's disease; A, amyloid -peptide; APP, amyloid precursor protein; MAPK, mitogen-activated protein kinase; ERK, extracellular signal-regulated kinase; MEK1, MAPK/ERK kinase 1; JNK, c-Jun N-terminal kinase; PI3K, phosphoinositide 3-kinase; cdk5, cyclin-dependent kinase 5; p75^NTR, 75-kDa neurotrophin receptor; BBP, -amyloid peptide binding protein; IGF-I, insulin-like growth factor; PTX, pertussis toxin; BT-I, butyrolactone I; DTT, dithiothreitol; SYV, SH-SY5Y cells stably transfected with pcDNA3.zeo empty vector; SAPK, stress-activated protein kinase; SEK1, SAPK/ERK kinase 1; ELISA, enzyme-linked immunosorbent assay; GST, glutathione S-transferase.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES