alpha-Synuclein affects the MAPK pathway and accelerates cell death.

Insoluble alpha-synuclein accumulates in Parkinson's disease, diffuse Lewy body disease, and multiple system atrophy. However, the relationship between its accumulation and pathogenesis is still unclear. Recently, we reported that overexpression of alpha-synuclein affects Elk-1 phosphorylation in cultured cells, which is mainly performed by mitogen-activated protein kinases (MAPKs). We further examined the relationship between MAPK signaling and the effects of alpha-synuclein expression on ecdysone-inducible neuro2a cell lines and found that cells expressing alpha-synuclein had less phosphorylated MAPKs. Moreover, they showed significant cell death when the concentration of serum in the culture medium was reduced. Under normal serum conditions, the addition of the MAPK inhibitor U0126 also caused cell death in alpha-synuclein-expressing cells. Transfection of constitutively active MEK-1 resulted in MAPK phosphorylation in alpha-synuclein-expressing cells and improved cell viability even under reduced serum conditions. Thus, we conclude that alpha-synuclein regulates the MAPK pathway by reducing the amount of available active MAPK. Our findings suggest a mechanism for pathogenesis and thus offer therapeutic insight into synucleinopathies.

Insoluble ␣-synuclein accumulates in Parkinson's disease, diffuse Lewy body disease, and multiple system atrophy. However, the relationship between its accumulation and pathogenesis is still unclear. Recently, we reported that overexpression of ␣-synuclein affects Elk-1 phosphorylation in cultured cells, which is mainly performed by mitogen-activated protein kinases (MAPKs). We further examined the relationship between MAPK signaling and the effects of ␣-synuclein expression on ecdysone-inducible neuro2a cell lines and found that cells expressing ␣-synuclein had less phosphorylated MAPKs. Moreover, they showed significant cell death when the concentration of serum in the culture medium was reduced. Under normal serum conditions, the addition of the MAPK inhibitor U0126 also caused cell death in ␣-synuclein-expressing cells. Transfection of constitutively active MEK-1 resulted in MAPK phosphorylation in ␣-synuclein-expressing cells and improved cell viability even under reduced serum conditions. Thus, we conclude that ␣-synuclein regulates the MAPK pathway by reducing the amount of available active MAPK. Our findings suggest a mechanism for pathogenesis and thus offer therapeutic insight into synucleinopathies.
␣-Synuclein was first identified in the purified synaptic vesicles of a torpedo (1). It is a small acidic protein composed of 140 amino acid residues. It has seven incomplete repeats of 11 amino acids with a core of KTKEGV at the amino terminus, whereas the carboxyl terminus has no known structural elements. Although its function is still unclear, there is accumulating evidence that ␣-synuclein is the main structural component of the insoluble filaments that form the Lewy bodies of Parkinson's disease (PD) 1 as well as those of dementia with Lewy bodies (2)(3)(4) in addition to the glial cytoplasmic inclusions of multiple system atrophy (MSA) (5,6).
Protein aggregation and filament formation are hallmarks of many neurodegenerative diseases (e.g. the neurofibrillary tangles and amyloid plaques of Alzheimer's disease and the neuronal intranuclear inclusions observed in polyglutamine diseases). The formation of these disease-specific amyloid structures is thought to play certain roles in the pathogenesis of neurodegenerative diseases (7). Thus, ␣-synuclein might play an important role in PD, dementia with Lewy bodies, and MSA, which are all classified as synucleinopathies (8,9). Although most cases of these diseases are sporadic and do not involve mutations in the ␣-synuclein gene, it has been shown that A30P and A53T mutations in the ␣-synuclein gene can cause a rare familial form of Parkinson's disease (10,11). Thus, the features of these mutant gene products might differ from wild type (WT) ␣-synuclein.
In vitro generated ␣-synuclein, especially that of the A53T mutant, forms aggregates under certain conditions in a nucleation-dependent manner (12,13). This aggregation is accelerated by copper (14), ferric iron (15), and cytochrome c (16). Aggregate formation has been attributed to a unique sequence within ␣-synuclein that is not shared by its homologues ␤-synuclein and ␥-synuclein (17). In vivo studies have shown that mice and flies made to overexpress ␣-synuclein within their neurons develop a neurological disorder that mimics Parkinson's disease (18,19). Oxidative stress has been shown to accelerate cell death in cultured cells (20 -23), and recently, it has been reported that accumulated ␣-synuclein in diseased brain tissue is nitrated (24,25). These findings give rise to the idea that, in synucleinopathies, abnormal mitochondrial stress might lead to cell death.
Although knowledge about the aggregate formation of ␣-synuclein is accumulating, its exact function is still unknown. It has been reported that ␣-synuclein shares a functional homology with 14-3-3 protein and binds to MAPK (26). We have also reported that ␣-synuclein directly binds ERK2 and forms a complex with Elk-1, which is also an ERK2 substrate (27). To further analyze the functional relationship between ␣-synuclein and MAPKs, we generated inducible stable cell lines of ␣-synuclein and studied their properties. In the present study, we report that ␣-synuclein inhibits MAPK signaling and accelerates cell death following serum reduction. Moreover, this phenomenon can be partially reversed by the introduction of constitutively active MEK-1.

EXPERIMENTAL PROCEDURES
All reagents and chemicals were purchased from Nacalai Tesque Inc. (Kyoto, Japan), except as otherwise noted. All experiments were done at least three times unless otherwise noted.
Statistical Analysis-All statistical analysis was performed using Stat View J5.0, and significances were determined by the Bonferroni/ Dunn test.
cDNAs and Plasmid Constructs-␣-synuclein, ␤-synuclein, and ␥-synuclein cDNA were amplified from the human cDNA library. Known human mutations of ␣-synuclein were generated using polym-* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Cell Culture and Establishment of Inducible Cell Lines-neuro2a and HEK293 cells were cultured in Dulbecco's modified Eagle's medium (Sigma) plus 10% adult bovine serum (BIO Whittaker, Walkersville, MD) at 37°C in an atmosphere of 95% air and 5% CO 2 . For generation of stable cell lines, neuro2a cells were co-transfected with pVgRXR (Invitrogen) and a pIND (Sp1)-hygro vector containing wild type, A30P mutant, or A53T mutant ␣-synuclein and selected in 100 g/ml of zeocin (InvivoGen, San Diego, CA) and 25 g/ml hygromycin B (Invitrogen).
Clones showing the highest level of ␣-synuclein expression and the tightest regulation by immunoblot analysis were chosen for further study. Clones were named Nmock, Nwt, NA30P, and NA53T, following transfection of their cDNAs. Each experiment was done with several clones to confirm the results. For cDNA transfection, cells were transfected with 1 g of plasmid DNA and 3 l of Transfast reagent (Promega)/1 ϫ 10 4 cells suspended in Opti-MEM (Life Technologies). After 1 h of incubation, medium was changed to Dulbecco's modified Eagle's medium containing the appropriate concentration of serum. To evaluate cell viability, cells were grown under 0.5% serum-reduced conditions for an indicated length of time.
Immunocytochemistry-MC36 antibodies were generated by injecting the ␣-synuclein-specific sequence, MPVDPSSEAYEMPSE peptide, into rabbits and through affinity purification using a peptide-bound column. Cultured cells, plated onto a Lab-TEK chamber slide 2 well (Nalge Nunc, Naperville, IL), were fixed with 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4) and followed by permeabilization with 0.25% Triton X-100 in phosphate-buffered saline (PBS). Chamber slides were then incubated with 5% bovine serum albumin in PBS for 1 h and incubated overnight with primary antibody MC36 diluted in PBS (1: 1000) at 4°C. The slides were then washed with PBS and incubated with Alexa flour 488 anti-rabbit antibody (Molecular Probes, Inc., Eugene, OR) diluted in PBS (1:500). The slides were mounted in 50% glycerol in PBS with 10 mM n-propyl gallate and observed using an Olympus SV-300 confocal microscope (Olympus Optical, Tokyo, Japan).
Immunoblotting-Cells were lysed with PBS containing Complete (Roche Molecular Biochemicals), 20 mM Na 3 VO 4 , and 10 mM NaF. For SDS-PAGE, protein concentration was determined using a Coomassie Protein Assay Reagent (Pierce). 40 g/lane was run on a 5/20% gradient or 12.5% polyacrylamide gels and transferred to Immobilon (Millipore, Bedford, MA). The membrane was then incubated for 1 h in 5% skim milk in TBST (50 mM Tris-HCl, pH 7. Following this, the membrane was washed, incubated for 1 h in 1:3000 peroxidase-coupled secondary antibodies (Amersham Pharmacia Biotech) in TBST, washed, and visualized using the ECL reagent (Amersham Pharmacia Biotech).
Immunoprecipitation-Subconfluent Nmock, Nwt, NA30P, and NA53T cells in 9-cm culture plates, induced with 1 M ponasterone A (Invitrogen) over 48 h, were harvested, washed once with PBS, and lysed on ice for 30 min with immunoprecipitation buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1 mM EDTA, 20 mM Na 3 VO 4 , 10 mM NaF, Complete (Roche Molecular Biochemicals), and 1% Triton X-100). The lysate was then centrifuged (20,000 ϫ g, 15 min), and the supernatant was incubated with 10 l of protein G-Sepharose (Life Technologies, Inc.) bound with 3 l of preimmune serum or 1 g of S1 antibody (27) and 10 l of protein G. Then the supernatant was incubated for 2 h at room temperature. Each sample was then washed four times with immunoprecipitation buffer, washed once with PBS, subjected to SDS-PAGE, and immunoblotted.
MAPK Pathway Stimulation-Cells incubated with 1 M of ponasterone A for 48 h were grown in 1% serum for 8 h and stimulated with 100 ng/ml of epidermal growth factor (EGF) (Peprotech EC, London, United Kingdom), 10 g/ml anisomycin, or 20 J/m 2 of UV exposure using a Stratalinker UV cross-linker (Stratagene, La Jolla, CA). For immunoblot of the phosphorylated proteins and RNA preparation, cells were harvested after 15 min of stimulation. For immunoblot of the c-Fos protein, cells were harvested after 1 h of stimulation.
GST Pull-down-BL21 (DE3) pLysS containing pGEX-6P (Amersham Pharmacia Biotech), and pGEX-6P ␣-synuclein (wild type) were cultured and induced by the addition of 0.1 mM isopropyl-␤-D-thiogalactopyranoside. Cells were lysed and sonicated and then centrifuged at 20,000 ϫ g for 30 min at 4°C. Glutathione-Sepharose 4B (Amersham Pharmacia Biotech) was added to the supernatant, which was then washed three times with HNTG buffer (25 mM HEPES-KOH, pH 7.5, 100 mM NaCl, 0.1% Triton X-100, 10% glycerol, 5 mM MgCl 2 , 1 mM dithiothreitol, and Complete). Then 10 l of either GST or GST ␣-synuclein-bound beads was added to ϳ10% of the cell lysate from subconfluent neuro2a cells within the 9-cm cultured plate for 2 h at 4°C. The samples were then washed three times with HNTG buffer and once with PBS and then subjected to SDS-PAGE and followed and immunoblotted.
Apoptosis Detection-Terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick-end labeling (TUNEL) staining was performed using a TACS apoptosis detection kit (Trevigen, Gaithersburg, MD), as described by the manufacturer. DNA ladder was detected by following a previously outlined protocol (28).
Northern Blotting-50 g of the total RNA obtained from Nmock, Nwt, NA30P and A53T cells using Trizol (Life Technologies), was run on 1% agarose gel containing 2.2 M formaldehyde and transferred to a Hybond N ϩ membrane (Amersham Pharmacia). The membrane was then probed with a digoxigenin-labeled (digoxigenin high prime DNA labeling kit; Roche Molecular Biochemicals) c-fos (exon 4 probe digested with AccI and AvaI) or 18 S RNA probe and visualized by chemiluminescence as indicated by the manufacturer.

␣-Synuclein Overexpression Affects Cell Viability in neuro2a
Cells-First we investigated whether the overexpression of ␣-synuclein affects cell viability in neuro2a cells. Cells were transfected with WT ␣-synuclein and both of the ␣-synuclein mutants (A30P and A53T). MTT assays were performed after 48 h of transfection. As shown in Fig. 1, A and B, the transient overexpression of WT, A30P mutant, and A53T mutant ␣-synuclein in transfected cells in 10% serum significantly reduced cell viability. To examine whether this phenomenon was specific to neuro2a cells, we performed a similar experiment with human embryonic kidney (HEK293) cells. However, HEK293 cells did not show a change in cell viability with ␣-synuclein transfection (data not shown). To confirm that the reduction of cell viability was not simply due to an overexpression of the proteins, we transfected the ␣-synuclein N-terminal ␣-Synuclein Affects MAPK Pathway fragment, the C-terminal fragment, the ␤-synuclein, and the ␥-synuclein to observe the effects. N-terminal ␣-synuclein and ␥-synuclein transfection significantly reduced cell viability, whereas transfection of C-terminal ␣-synuclein and ␤-synuclein had no effect on cell viability (Fig. 1, C and D).
Generation of Stable ␣-Synuclein-inducible neuro2a Cell Lines-To further investigate the mechanism by which decreased cell viability results from ␣-synuclein expression, we constructed stable neuro2a cell lines with an ecdysone-inducible system. Cells were stably co-transfected with a pVgRXR and pIND (Sp1)-hygro vector containing WT, A30P mutant, and A53T mutant ␣-synuclein. The cells became capable of expressing ␣-synuclein upon the addition of ponasterone A to their culture medium. Fig. 2A shows the expression patterns of ␣-synuclein in ponasterone A-treated cells. Each ␣-synucleincarrying clone showed a similar expression pattern after being induced. Expression of ␣-synuclein was detected after 24 h of induction with 1 M of ponasterone A and increased with time. However, cells that were not induced also expressed a small amount of ␣-synuclein due to leakage (Fig. 2B). ␣-Synuclein expression was observed by immunocytochemistry (Fig. 2C).
We then measured the cell viability of these cell lines by MTT assay. Cells grown in 10% serum did not show a significant difference in cell viability upon the addition of 1 M pon-asterone A (Fig. 3A). However, when grown in 0.5% serum, the viability of cells expressing ␣-synuclein was significantly reduced, as judged by the MTT assay, with 1 M ponasterone A treatment (Fig. 3B). Under serum-reduced conditions, the ratio of surviving cells was determined by trypan blue staining, and significant cell death was observed (Fig. 3C). TUNEL staining also showed a significant increase in TUNEL-positive ␣-synuclein-expressing cells (Fig. 3, D and E). DNA extracted from serum-reduced, ponasterone A-treated NA53T cells showed a typical DNA ladder (Fig. 3F).
␣-Synuclein Overexpression Suppressed Phosphorylation of MAPKs but Not MAPK Kinase Phosphorylation-Since serum reduction can cause a down-regulation of MAPK signaling in cells, we focused on MAPK pathway abnormalities and screened the phosphorylation status of the molecules involved in MAPK pathways. We found that, in ␣-synuclein-expressing cells, phosphorylated forms of MAPKs, such as ERK1/2, p38 MAPK, and SAPK/JNK, were all reduced even under normal serum conditions (Fig. 4A), whereas phosphorylated MAPKK levels, including those of phosphorylated MEK-1, MKK-3/6, and SEK-1, were similar among the cells. Phosphatases, such as MKP-1, MKP-2, and calcineurin, also showed no difference in expression levels among the cells. Cells that were not induced showed no variation in the phosphorylation status of ␣-Synuclein Affects MAPK Pathway their MAPKs (data not shown). To clarify the relationship between ␣-synuclein expression and MAPK suppression, NA53T cells were induced with various concentrations of ponasterone A, and the amount of phosphorylated ERK1/2 was evaluated. As shown in Fig. 4B, an increase in ␣-synuclein expression caused a reduction in phosphorylated ERK1/2. Next, we estimated the activity of MAPK by incubating recombinant ERK2 with cell lysates from ␣-synuclein-expressing cells. Fig. 4C shows that the cell lysate of ponasterone Atreated NA53T cells cultured in normal serum showed a significant decline in ERK2 phosphorylation activity compared with the cell lysate of Nmock cells.
␣-Synuclein-expressing Cells Lack MAPK Response to Extracellular Stimuli-We next examined the response of MAPKs to extracellular stimuli. Similar responses were observed among those cells that were not treated with ponasterone A upon stimulation after 8 h of exposure to serum that had been reduced to 1% (Fig. 6A). Cells incubated with 1 M ponasterone A for 48 h were grown for 8 h in 1% serum and stimulated with 100 ng/ml EGF, 10 g/ml anisomycin, or 20 J/m 2 UV exposure. As shown in Fig. 6B, cells expressing ␣-synuclein showed an increase in MEK-1 phosphorylation after EGF stimulation but did not show an increase in ERK1/2 phosphorylation, as determined by immunoblotting. Northern blotting revealed that they also lacked c-fos gene expression. A decreased expression of c-Fos protein was also confirmed by immunoblotting (Fig.  6C). Stimulation by anisomycin activated MKK-3/6 and SEK-1 (data not shown), but phosphorylation of p38 MAPK (Fig. 6B) and SAPK/JNK (data not shown) was not observed. UV stimulation caused phosphorylation of p38 MAPK in Nmock and Nwt cells but not in NA30P and NA53T cells.
Constitutively Active MAPK Restores Decreased Cell Viability in ␣-Synuclein-overexpressed Cells-To confirm that suppression of the MAPK pathway results in a decrease in cell viability, Nmock, Nwt, NA30P, and NA53T cells were induced with ponasterone A and incubated with the MEK inhibitor U0126 in 10% serum. As shown in Fig. 7A, the MTT assay of Nwt, NA30P, and NA53T cells treated with 10 M U0126 and 1 M ponasterone A showed a significant decrease in their viability. Cells that were not treated with ponasterone A showed no significant change in cell viability even with U0126 treatment (data not shown).
To recover MAPK phosphorylation in ␣-synuclein-expressing cells, we used constitutively active MEK-1 to phosphorylate inactive ERK1/2. neuro2a cells transiently transfected with Myc-tagged constitutively active MEK-1 cultured in 0.5% serum overnight showed increased levels of phosphorylated ERK1/2 compared with an empty vector (Fig. 7B). As can be seen in Fig. 7C, even when ␣-synuclein is induced and serum is ␣-Synuclein Affects MAPK Pathway reduced to 0.5%, constitutively active MEK-1 can increase the phosphorylation of ERK1/2 in Nmock, Nwt, NA30P, and NA53T cells. We investigated the effect of constitutively active MEK-1 on cell viability and found that constitutively active MEK-1 can partially restore cell viability in ␣-synucleinexpressing cells under serum-reduced conditions (Fig. 7D).
To examine the effects of MEK-1 in a transient system, we co-transfected neuro2a cells cultured in 10% serum with ␣-synuclein and constitutively active MEK-1. As shown in Fig.  7E, constitutively active MEK-1 can restore cell viability, even in a transient system. DISCUSSION ␣-Synuclein Suppresses the MAPK Pathway-In this study, we first found that transient overexpression of ␣-synuclein affects the cell viability of neuro2a cells, as measured by an MTT assay. This effect was only observed with ␣-synuclein and not its homologue ␤-synuclein, and its N-terminal portion seemed to play a more important role than that of the Cterminal portion. To evaluate this in detail, we generated stable inducible ␣-synuclein and ␤-synuclein cell lines. When cultured under normal serum conditions, the cells showed no change, even with induction by ponasterone A. This result seemed rather contradictory to the results of the transient transfection study. However, we thought that a low level of ␣-synuclein might be expressed without induction due to un-avoidable leakage, thereby conferring some resistance to cells under normal serum conditions. When ␣-synuclein was induced and the serum concentration was reduced, significant cell death was observed in Nwt, NA30P, and NA53T cells, compared with Nmock cells.
In cultured cells, a reduction in the concentration of serum within the medium results in a down-regulation of the MAPK pathway. This transmits a survival signal from a cell's surface receptor to its nucleus via the phosphorylation of various proteins (30 -32). Thus, we investigated the status of MAPKs within cells and found that ␣-synuclein-expressing cells have less phosphorylated MAPK than control cells under normal serum conditions. As shown in Fig. 4, not only the phosphorylation of classical MAPK ERK1/2, but also the phosphorylation of p38 MAPK and SAPK/JNK, is attenuated by overexpression of ␣-synuclein. In our inducible cell lines, suppression of MAPK phosphorylation increased as the level of ␣-synuclein expression increased. This supports the idea that overexpression of ␣-synuclein affects the phosphorylation status of MAPK. Induction of ␣-synuclein decreased MAPK phosphorylation as well as c-fos gene expression when attempts were made to stimulate these through exposure to extracellular agents, such as EGF, anisomycin, and UV light (31,(33)(34)(35)(36). Moreover, an in vitro kinase study using cell lysate confirmed a decrease in MAPK phosphorylation in ␣-synuclein-expressing cells.

␣-Synuclein Affects MAPK Pathway
␣-Synuclein is shown to inhibit PKC activity (26) in cultured cell lines. Possibly, our down-regulation of MAPK might be the result of PKC inhibition. However, PKC is an upstream regulator of the MAPK pathway (37,38). Thus, our observation that MEK-1 phosphorylation was not affected by ␣-synuclein induction supports our idea that ␣-synuclein has a direct, rather than indirect, effect on MAPK phosphorylation. From these results and considerations, we concluded that ␣-synuclein suppresses the MAPK pathway.
␣-Synuclein Affects the Phosphorylation of MAPKs by MAPKKs-To clarify which component of the MAPK pathway is affected, we looked upstream of MAPK. We found that phos- Immunoblotting was done using anti-phospho-ERK1/2 and anti-ERK1/2 antibodies. Cell lysate from NA53T showed less ERK2 phosphorylation activity than lysate from Nmock.
We then studied the mechanism governing this inhibition and found that ␣-synuclein binds to ERK1/2, p38 MAPK, and SAPK/JNK. We previously reported that the aa 68 -133 sequence of mouse ERK2 is required for ERK2 to bind with ␣-synuclein (27). SAPK/JNK and p38 MAPK each have a homologous sequence to the ␣-synuclein binding region of ERK2 (Fig. 8), so it can be presumed that ␣-synuclein binds to MAPKs through this homologous sequence. The N-terminal fragment (aa 1-74) of ␣-synuclein is responsible for the observed decrease in cell viability (Fig. 1) and is also important for binding with ERK2 (27). To exclude the possibility of increased dephos-phorylation of MAPKs, we investigated the status of several MAPK phosphatases such as MKP-1 and MKP-2 as well as the phosphatases of MAPK substrates such as calcineurin (46,47). However, we could not find any difference in their levels of expression among the cells.
It is known that ␣-synuclein can be a substrate for casein kinase and G protein-coupled receptor kinases at serine 129 residue (48,49) and by c-Src and Fyn kinase at the tyrosine 125 residue (50,51). However, in the latter report, MAPK, Lyn, PYK2, FAK, SAPK/JNK, and Cdk5 are proved not to phosphorylate ␣-synuclein, but there are no reports of whether MAPKK phosphorylates ␣-synuclein. Thus, we could not exclude the possibility that ␣-synuclein acts as a competitive substrate for MAPKK. Based on the direct binding of ERK2 to ␣-synuclein, it is more plausible that ␣-synuclein affects the phosphorylation of MAPKs by MAPKKs through binding to MAPKs. If so, then why can constitutively active MEK-1 phosphorylate ERK1/2? From our result, we believe that constitutively active MEK-1 is capable of overcoming ␣-synuclein inhibition.
MAPK Down-regulation by ␣-Synuclein Affects Cell Viability-Our next point of interest was the relationship between MAPK attenuation and decreased cell viability. To clarify this, we initially demonstrated that ␣-synuclein-expressing cells are FIG. 6. The MAPKs of Nwt, NA30P, and NA53T cells do not respond to extracellular stimulation. A, cells responded to stimulation before they were induced. Cells that were not treated with 1 M ponasterone A responded to EGF and anisomycin after 8 h in serum reduced to 1%. B, stimulation with EGF, anisomycin, and UV after induction with ponasterone A. Nmock, Nwt, NA30P, and NA53T cells were treated with 100 ng/ml EGF or 10 g/ml anisomycin and harvested after 15 min. EGF stimulation resulted in an increase in the phosphorylated form of MEK-1 in each clone. However, ERK1/2 was not phosphorylated in Nwt, NA30P, or NA53T cells. To study c-fos expression, total RNA was obtained from EGF-treated cells and hybridized with c-fos or 18 S RNA probes for Northern blot. Nwt, NA30P, and NA53T cells did not show c-fos expression following EGF stimulation. Stimulation with anisomycin showed similar results. Nwt, NA30P, and NA53T cells did not express phosphorylated p38 MAPK. In UV stimulation, cells were exposed to 20 J/m 2 of UV light and harvested after 30 min. NA30P and NA53T cells did not express phosphorylated p38 MAPK, whereas Nmock and Nwt cells had similar amounts of phosphorylated p38 MAPK. C, immunoblot of c-Fos in NA53T cells. c-Fos protein was not induced with EGF treatment. Cells were harvested after 1 h of 100 ng/ml EGF treatment. sensitive to MAPK attenuation by exposing Nmock, Nwt, NA30P, and NA53T cells to the MEK inhibitor U0126 (52) under normal serum conditions. Treatment of the cells with this inhibitor significantly decreased the cell viability of ␣-synuclein-expressing cells, in comparison with control cells. This result supports the idea that cells depend on a survival signal under normal serum conditions, although basal MAPK activity is low in these cells (Fig. 4A). Thus, under normal serum conditions, cells are sensitive to additional MAPK down-regulation by serum reduction. If ␣-synuclein expression by ponasterone A induction is not enough to cause cell death due to a complete suppression of the MAPK pathway, the cells do not die. The significance of the effect of ␣-synuclein expression on the p38 MAPK and SAPK/JNK pathways is unclear. These kinases are also regulated by growth factors, but these are mainly activated under conditions of stress (43,53,54). We have confirmed the effect of stress on our cells by using hydrogen peroxide as a stress inducer. We found that Nwt, NA30P, was immunoblotted with anti-Myc antibody, anti-phosphorylated ERK1/2 antibody, and anti-ERK1/2 antibody. C, effect of constitutively active MEK-1 on uninduced cells. Nmock, Nwt, NA30P, and NA53T cells that were not treated with ponasterone A were transfected with pcDNA4/MycHisMEK-1 (S218D, S221D) in 0.5% serum. Cell lysate was collected after 48 h of transfection and immunoblotted with antiphosphorylated ERK1/2 antibody or MC36. D, loss of cell viability in ␣-synuclein-expressing cells due to serum reduction is reversed by constitutively active MEK-1. Nmock, Nwt, NA30P, and NA53T cells treated with 1 M ponasterone A for 48 h were transfected with pcDNA4/Myc-His mock or MEK-1 (S218D, S221D), and serum was reduced to 0.5%. A MTT assay was performed after 48 h of transfection. *, p Ͻ 0.001 versus pcDNA mock vector-transfected cells. E, constitutively active MEK-1 reversed the decline in cell viability in a transient system. neuro2a cells were co-transfected with pCI ␣-synuclein, pcDNA4/Myc-His mock, or MEK-1 (S218D, S221D). Serum was kept at 10% for 48 h, and a MTT assay was performed. *, p Ͻ 0.01; **, p Ͻ 0.05; ***, p Ͻ 0.0001 versus mock pcDNA vector-transfected cells. and NA53T cells were all more vulnerable to oxidative stress than control cells (55). Since oxidative stress caused by hydrogen peroxide activates p38 MAPK and SAPK/JNK (56 -58), ␣-synuclein might block this reaction and disrupt equilibration of the MAPK pathway, thereby causing cell death. ␣and ␤-synuclein are known to inhibit phospholipase D activity in vitro (59), which is intimately related to ERK1/2 function (60,61). Considering this along with our own results, ␣-synuclein might bind to a number of molecules in order to regulate signal transduction.
Constitutively Active MEK-1 Restores Cell Viability by Upregulating the MAPK Pathway-Constitutively active MEK-1 is known to act like phosphorylated MEK-1 (62). Since constitutively active MEK-1 can phosphorylate MAPK in ␣-synuclein-expressing cells (Fig. 6), we used this to restore cell viability. As expected, constitutively active MEK-1 restored MAPK phosphorylation and rescued Nwt, NA30P, and NA53T cells. When co-transfected with ␣-synuclein in a transient system, a similar result was obtained. These results strongly suggest that the MAPK pathway is involved in the decreased cell viability induced by ␣-synuclein.
Implications for Normal ␣-Synuclein Function and Disease Pathogenesis-In the present study, we could not find aggregations of ␣-synuclein in cultured cells, even when it was overexpressed. To aggregate, it might be necessary for ␣-synuclein to be modified, for example, by nitration (24,25) or by phosphorylation (48,50,51). However, once aggregates are formed, they might behave in the same manner as overexpressed ␣-synuclein to recruit MAPKs and other molecules, leading to a down-regulation of signal transduction pathways. The fact that glial cytoplasmic inclusions in MSA contain MAPK and Cdk5 (63) supports this idea.
Our study revealed that mutant ␣-synuclein showed a stronger inhibitory effect on MAPK phosphorylation than wild type at UV stimulation, suggesting the possibility that this inhibitory effect might be more pathogenic in familial Parkinson's disease with those mutations. However, the inhibitory effect by ␥-synuclein is perplexing. The accumulation of ␣-synuclein but not ␥-synuclein is observed in PD and MSA. Thus, if the accumulation of ␣-synuclein affects MAPK phosphorylation, the inhibitory effect of ␣-synuclein is pathogenic in PD and MSA. Further study will be necessary for the effect of ␣-synuclein aggregates on MAPK phosphorylation.
As a result of studies involving knock-out mice, ␣-synuclein is thought to be a negative regulator of dopamine release at synaptic terminals (64). The mechanism of this inhibition is still unclear, but there are several clues. Activation of ERKs is required for catecholamine secretion in bovine adrenomedullary chromaffin cells (65). Furthermore, in PC12 cells, activation of MEK and ERK signaling is required for calcium-dependent dopamine release (66). Combining these findings with our own results, ␣-synuclein might regulate synaptic transmission through MAPK regulation.
Restoration of cell viability using constitutively active MEK-1, following damage from ␣-synuclein overexpression, provides a therapeutic strategy by which to regulate the MAPK pathway in order to rescue cells from death in synucleinopathies such as exist in PD and MSA.