Abstract
Hyperphosphorylation of tau is pivotally involved in the pathogenesis of Alzheimer’s disease (AD) and related tauopathies. Glycogen synthase kinase-3β (GSK-3β) and protein phosphate 2A (PP2A) are crucial enzymes to regulate tau phosphorylation. GSK-3β activity is regulated by its inhibitory phosphorylation at Ser9. We previously reported the cross-talk between GSK-3β and PP2A signaling and showed that PP2A could dephosphorylate GSK-3β at Ser9. Here, we investigated the dephosphorylation of GSK-3β in brain extracts in the presence of phosphatase inhibitors and found that a PP2A-like phosphatase activity was required for dephosphorylation of GSK-3β at Ser9. PP2A interacted with GSK-3β and suppressed its Ser9 phosphorylation in vitro and in HEK-293FT cells. Activity of PP2A negatively correlated to the level of phosphorylated GSK-3β in kainic acid-induced excitotoxic mouse brain. Alteration of methylation of the catalytic subunit of PP2A (PP2Ac) at Leu309 did not affect GSK-3β phosphorylation. These findings suggest that Leu309 methylation is not required for PP2A to dephosphorylate GSK-3β at Ser9.
INTRODUCTION
Glycogen synthase kinase-3 (GSK-3) is a ubiqui-tously-expressed proline-directed serine/threonineprotein kinase that phosphorylates more than 40 substrates involved in regulation of a large number of cellular processes that include body pattern formation, energy metabolism, cell cycle, apoptosis, and neuronal cell development [1]. The two isoforms of GSK-3, GSK-3α and GSK-3β, are highly homologous within their kinase domains. In the human brain, GSK-3β is more abundant than GSK-3α [2]. GSK-3β is one of the major kinases of the microtubule-associated protein tau. It phosphorylates tau protein at multiple sites both in vitro and in vivo [3–5]. Increased active form of GSK-3β in Alzheimer’s disease (AD) [6, 7] has been shown to be associated with tau hyperphosphorylation and amyloidogenic processing of amyloid-β protein precursor in AD [8].
GSK-3 is constitutively active in resting cell and most commonly inhibited by phosphorylation of a serine residue, Ser21 in GSK-3α and Ser9 in GSK-3β, at the N-terminal domain. This phosphorylation of the N-terminal serine can be catalyzed by kinases from different signaling pathways [9–11]. For example, protein kinase B/Akt, a key mediator of PI3K pathway, phosphorylates Ser9 and suppresses GSK-3β activity [9]. Dysfunction of PI3K signaling pathway increases the kinase activity of GSK-3β and consequently leads to abnormal tau hyperphosphorylation [12, 13]. Thus, multiple inhibitors of GSK-3β are considered as potential therapeutic compounds for neurodegenerative diseases [8, 14].
The inhibitory phosphorylation of GSK-3 can be reversed by protein phosphatases (PPs), like PP1 [15] and PP2A [16, 17], which together account for more than 90% of the total serine/threonine phosphatase activity in mammalian brain [18, 19]. PP1 and PP2A activities are significantly downregulated in AD brain [20]. Decreased activity of PP2A might be responsible for hyperphosphorylation of tau in AD brain, since PP2A accounts for over 70% of tau phosphatase activity in the human brain [19].
PP2A consists of scaffold (A), catalytic (C), and regulatory (B) subunits. Activity of PP2A is regulated by posttranslational modifications at the C-terminus of the catalytic subunit (PP2Ac). For instance, Tyr307 or Thr304 phosphorylation inhibits PP2A holoenzyme assembly [21–23]. PP2Ac Leu309 methylation is required for the assembly of core enzyme with B subunit PR55, through which tau is dephosphorylated [23, 24]. The reversible methylation and demethylation of Leu309 is catalyzed by leucine carboxyl methyltransferase-1 (LCMT-1) [25] and protein phosphatase methylesterase-1 (PME-1) [26], respectively. Increased level of demethylated PP2Ac may contribute to compromised dephosphorylation of hyperphosphorylated tau in AD brain [27, 28].
Our previous study showed that GSK-3β regulated PP2Ac methylation by suppressing the expression of PME-1 and phosphorylating LCMT-1 and that PP2A dephosphorylated GSK-3β at Ser9 [29]. In this study, we determined the roles of PP1 and PP2A in dephosphorylating GSK-3β in cultured cells and in mouse brain extracts. We found that PP2A, rather than PP1, was the main phosphatase of GSK-3β. Phosphorylation of GSK-3β at Ser9 was not affected by Leu309 methylation of PP2Ac, suggesting Leu309 methylation of PP2Ac is not required for its activity to dephosphorylate Ser9 of GSK-3β.
MATERIALS AND METHODS
Plasmids, antibodies, and other reagents
pCI/HA-PP2Ac, pDsRed-N1/PP2Ac, pCI/HA-PP1, and pDsRed-N1/PP1 were generated by subcloning human PP2Ac (PP2Acα) and PP1 (PP1α) cDNAs into corresponding vectors, respectively. pCI/HA-PME-1 and pCI/HA-LCMT-1 constructs were generated as previously described [29]. Short interfering RNAs (siRNAs) of LCMT-1 and PME-1 were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA).
Rabbit polyclonal anti-pSer21/pSer9-GSK-3α/β, rabbit polyclonal anti-pSer9-GSK-3β, rabbit polyclonal anti-GSK-3α/β, and rabbit monoclonal anti-GSK3β were from Cell Signaling Technology (Beverly, MA, USA). Mouse monoclonal anti-PP1 and mouse monoclonal anti-PP2Ac were purchased from BD Bioscience (San Jose, CA, USA). Mouse monoclonal anti-PP2Ac, anti-demethylated (dm)-PP2Ac and PP2A holoenzyme were from Millipore (Billerica, MA, USA). Mouse monoclonal and rabbit polyclonal anti-HA were from Sigma (St. Louis, MO, USA). Rabbit polyclonal anti-GAPDH was from Santa Cruz Biotechnology. Horseradish peroxidase-conjugated anti-mouse and anti-rabbit IgG were obtained from Jackson ImmunoResearch Laboratories (West Grove, PA, USA). Alexa 488-conjugated goat anti-mouse IgG and Alexa 555-conjugated goat anti-rabbit IgG were from Molecular Probes (Carlsbad, CA). Enhanced chemiluminescence (ECL) kit was from Thermo Fisher Scientific (Rockford, IL, USA).
Animals
Adult FVB mice were purchased from Charles River Laboratory and Nanjing Animal Model Center. The animals were housed in a 12-h light/dark schedule with free access to food and water. Animal use was in full compliance with the NIH guidelines and was approved by our institutional Animal Care and Use Committees.
Cell culture and transfection
HEK-239FT, HeLa, and N2a cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum at 37°C in 5% CO2 atmosphere. All transfections were performed with Lipofectamine 3000 (Invitrogen, Carlsbad, CA, USA) or FugenHD (Promega, Madison, WI, USA) according to the manufacturer’s instructions.
Inhibition of GSK-3β dephosphorylation with phosphatase inhibitors in mouse brain homogenates
Adult male FVB mouse brains were homogenized with ice-cold buffer containing 50 mM Tris-HCl, pH 7.4, 8.5% sucrose, 0.25% sodium deoxycholate, 0.1% TritonX-100, 0.1% NP-40, 2 mM EDTA, and protease inhibitor cocktail (Roche, Indianapolis, IN, USA). Concentration of extracts was determined by Bradford protein assay (Bio-Rad, Hercules, CA, USA) and adjusted to 4 mg/ml. The extracts were incubated with DMSO, okadaic acid (OA) (Sigma, MO, USA) or calyculin A (Cal A) (Cell signaling Technology) for 15 min at 30°C. Reactions were stopped by adding 4 × Laemmli sample buffer and then subjected to western blots. Band intensities in blots were quantified using ImageJ software (NIH, Bethesda, MD, USA). The IC50 values were calculated using the Graphpad Prism ® (GraphPad Software, La Jolla, CA).
In vitro dephosphorylation assay
HEK-293FT cells were transfected with pCI/HA-GSK-3β for 48 h and lysed in lysis buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 0.25% sodium deoxycholate, 0.1% TritonX-100, 0.1% NP-40, 50 mM NaF, 1 mM Na3VO4, 2 mM EDTA, 1 mM PMSF, and protease inhibitor cocktail). The 16,000 × g extracts were prepared from the lysates. HA-GSK-3β protein was immunoprecipitated from the cell extracts with anti-HA coupled protein G-agarose beads and incubated with PP2A holoenzyme at room temperature for 10 min in reaction buffer (50 mM Tris-HCl, pH 7.4, 10 mM β-mercaptoethanol, 10 mM MgCl2, 1.0 mM EGTA). The phosphorylation level of GSK-3β was analyzed by western blots developed with anti-pSer9-GSK-3β.
Co-immunoprecipitation
HEK-293FT cells were transfected with pCI/HA-GSK-3β for 48 h as described above. The cells were washed twice with PBS and lysed by sonication in lysis buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 50 mM NaF, 1 mM Na3VO4, 2 mM EDTA, 1 mM PMSF, and protease inhibitor cocktail). Insoluble materials were removed by centrifugation; the supernatants were pre-absorbed with protein-G-agarose beads (Thermo) and incubated with anti-HA coupled protein G-agarose beads overnight at 4°C. The beads were washed twice with lysis buffer and twice with TBS. The bound proteins were eluted from the beads by boiling in Laemmli sample buffer. The samples were subjected to western blot developed with the indicated primary antibodies.
Immunofluorescence
HeLa cells were plated on poly-L-lysine-coated glass coverslips in 12-well plates. Two days later, the cells were washed with PBS and fixed with 4% paraformaldehyde in PBS for 30 min at room temperature. After washing with PBS, the cells were blocked with 10% goat serum in 0.2% Triton X-100-PBS for 2 h at 37°C, and incubated with rabbit anti-GSK-3β (27C10, 1:200), mouse anti-PP2A antibody (1D6, 1:200) or mouse anti-PP1 antibody (1E9, 1:200) overnight at 4°C. After washing and incubating with secondary antibodies (Alexa 488-conjugated goat anti-mouse IgG and Alexa 555-conjugated goat anti-rabbit IgG, 1:1000) plus TO-PRO-3 iodide at room temperature, the cells were washed with PBS, mounted with Fluoromount-G, and visualized with a confocal microscope (Nikon, Brighton, MI).
Excitotoxicity induced by kainic acid injection
Male FVB mice were injected intraperitoneally with a single dose of kainic acid (KA; 20 mg/kg body weight) as described previously [30]. The mice were sacrificed immediately before and 2.5, 6, 10, 24, and 36 h after injection. Brain homogenates and extracts were prepared from the forebrain as described [30], and then subjected to western blots and PP2A activity assays.
PP2A activity assays
PP2A was immunoprecipitated from the brain extracts of KA-injected mice with monoclonal anti-PP2A. The immunoprecipitated PP2A was incubated with 0.2 mg/mL 32P-tau in a reaction buffer containing 50 mM Tris-HCl, pH 7.4, 1.0 mm MnCl2, and 10 mM β-mercaptoethanol at 30°C for 20 min as previously described [19, 30]. The dephosphorylation reaction was terminated, and the released 32Pi was determined by Cerenkov counting after separation from 32P-tau by ascending paper chromatography, as describedpreviously [31].
Statistical analysis
Quantitative data are presented as means ± S.D. and were analyzed by unpaired two-tailed Student’s t test. The calculated p-values are indicated in figure legends. For analysis of the correlation between PP2A activity and phosphorylated GSK-3β in mouse brains, the Pearson product-moment correlation coefficient r was calculated.
RESULTS
PP2A dephosphorylates GSK-3β at Ser9
PP1 and PP2A show different sensitivity to phosphatase inhibitors, such as OA and Cal A. OA inhibits PP2A at subnanomolar concentrations [IC50 (half maximal inhibitory concentration) = 0.1 nM] and PP1 at over 100-fold higher concentrations (IC50 = 15–20 nM). Cal A, a potent inhibitor of both phosphatases, inhibits PP1 and PP2A with IC50 values of 2 nM and 0.5–1 nM, respectively. Thus, these differences can be used to distinguish PP1 and PP2A activities in cell extracts [32, 33]. To identify the protein phosphatase that dephosphorylates GSK-3β in the brain, mouse brain extracts were treated in the absence and presence of various concentrations of OA (Fig. 1A) or Cal A (Fig. 1B). We found that 50% inhibition of pS9-GSK-3β dephosphorylation was achieved at a relatively low concentration of 5 nM OA or 10 nM Cal A (Fig. 1D), whereas 50% inhibition of phospho-Ser21-GSK3α (pS21-GSK-3α) dephosphorylation required a higher concentration (10 nM OA or 20 nM Cal A) (Fig. 1C). These results suggest that a PP2A-like phosphatase activity is responsible for pS9-GSK-3β dephosphorylation in brain extracts. In contrast, pS21-GSK-3α might be dephosphorylated mainly by a PP1-like phosphatase activity in the brain.
To confirm that PP2A dephosphorylates GSK-3β, we immunopurified HA-tagged GSK-3β from HEK-293FT cells with anti-HA and incubated it with purified PP2A for 10 min at 30°C. Then, phosphorylation of Ser9-GSK-3β was analyzed by western blots. The level of pSer9-GSK-3β was significantly decreased after incubation with PP2A (Fig. 1E), suggesting that PP2A could efficiently dephosphorylate it in vitro.
To further understand the roles of PP2A and PP1 in dephosphorylation of GSK-3β at Ser9 in cultured cells, we overexpressed PP1 or PP2Ac (catalytic subunit of PP2A) in HEK-293FT cells and measured phosphorylation of S9/S21-GSK-3β/α. We found that 2.5 fold overexpression of PP1 did not affect the phosphorylation of either GSK-3α or GSK-3β (Fig. 2A–C). Overexpression of PP2Ac suppressed endogenous PP2Ac expression (Fig. 2D, E) and reduced the level of pS9-GSK-3β but not pS21-GSK-3α (Fig. 2D, F). These results supported that PP2A, but not PP1, is the major phosphatase of pS9-GSK-3β. Moreover, four-fold overexpression of PP2Ac only caused 20% reduction of pS9-GSK-3β (Fig. 2D), which suggests that PP2A activity is tightly regulated in cells.
PP2A and PP1 interact with GSK-3β
To investigate the interaction between GSK-3β and the protein phosphatases, we overexpressed HA tagged GSK-3β in HEK-293FT cells and then immunoprecipitated GSK-3β with anti-HA. We found that either PP2Ac or PP1 was co-immunoprecipitated by GSK-3β (Fig. 3A and B).
To study whether the PP2A interacts with GSK-3β in intact cells, we immunostained HeLa cells using specific antibodies against PP2Ac, PP1 and GSK-3β. GSK-3β was predominantly localized in the cytoplasm and was also present in the nucleus (Fig. 3C). PP2Ac and PP1 were almost equally distributed in the cytoplasm and nucleus, partially co-localized with GSK-3β (Fig. 3C). These results supported that both PP2A and PP1 could interact with GSK-3β in cells.
Dephosphorylation of GSK-3β at Ser9 by PP2A is independent of its methylation at Leu309
PP2Ac Leu309 methylation has been reported to play a crucial role in the assembly of PP2A/Bα holoenzyme [23, 24], the primary tau phosphatase in brain. Reduction of PP2Ac methylation leads to down-regulation of PP2A/Bα, thereby resulting in increased tau phosphorylation in vivo and in AD [34]. To investigate the role of PP2Ac methylation in dephosphorylation of GSK-3β at Ser9, we overexpressed PME-1 or LCMT-1 in N2a cells and then measured the demethylation of PP2A and phosphorylation of GSK-3β. We found that over-expression of PME-1 or LCMT-1 induced a significant increase or decrease in the level of demethylated PP2Ac at Leu309 (dm-PP2Ac), respectively. Surprisingly, the level of pS9-GSK-3β was not altered in either case (Fig. 4A). To confirm that changes in dm-PP2Ac level did not affect pS9-GSK-3β dephosphorylation, we used siRNA to knockdown PME-1 or LCMT-1 specifically in HEK-293FT cells and measured the level of pS9-GSK-3β. Similarly, the level of pS9-GSK-3β showed no apparent difference between siPME-1, siLCMT-1, and control siRNA transfected cells (Fig. 4B). These results suggest that PP2Ac methylation may not be necessary for the dephosphorylation of GSK-3β at Ser9.
Phosphorylation of GSK-3β at Ser9 is negatively correlated with PP2A activity
KA is a neuroexcitotoxic amino acid acting on glutamic kainate receptors. It is known that PP2A is first activated and then inhibited during KA-induced excitotoxicity [30]. To investigate effect of PP2A methylation on pS9-GSK-3β phosphorylation in vivo, we measured the levels of pS9-GSK-3β and DM-PP2Ac in the excitotoxic mouse brains induced by KA at various time points after the injection. We found that PP2Ac methylation was not altered during the first phase (within 6 h post-injection) after KA injection (Fig. 5A, B), whereas Ser9 phosphorylation of GSK-3β was significantly decreased at 2.5–6 h (Fig. 5C, D). After 10 h post-injection (the second phase), a gradual increase in dm-PP2Ac level was detected. During the same period of time, the reduced level of pS9-GSK-3β recovered to normal. Dephosphorylation of pS9-GSK-3β occurred much earlier than the demethylation of PP2Ac in response to KA administration, further supporting that GSK-3β dephosphorylation is independent of PP2Ac methylation in vivo. However, the phosphatase activity of PP2A was first increased within 6 hours and then decreased after 10 hours post-injection (Fig. 5E), which coincided with the changes of phosphorylation of GSK-3β at Ser9 and negatively correlated to the changes of pS9-GSK-3β (Fig. 5F). These results suggest that (i) PP2A activity is responsible for pS9-GSK-3β dephosphorylation in vivo and (ii) PP2Ac methylation does not affect the dephosphorylation of pS9-GSK-3β during KA-induced excitotoxicity.
DISCUSSION
PP2A and PP1 are the most abundant serine/threonine phosphatases in brain [18, 19]. Although several studies have reported that PP2A and PP1 could dephosphorylate the inhibitory phospho-Ser9 and activate GSK-3β [15–17, 29], it remains unclear which of these phosphatases regulates GSK-3β phosphorylation in brain and what a mechanism is involved. In the present study, we found that PP2A, rather than PP1, was the main phosphatase of GSK-3β. First, pS9-GSK-3β dephosphorylation was inhibited by relatively low concentrations of the phosphatase inhibitor OA or Cal A, indicating a PP2A-like phosphatase activity was required for pS9-GSK-3β dephosphorylation in mouse brain extracts. Second, PP2A could dephosphorylate pS9-GSK-3β in vitro. Third, overexpression of PP2A, but not PP1, resulted in a reduction of GSK-3β phosphorylation at Ser9 in cultured cells. Finally, PP2A activity was negatively correlated to pS9-GSK-3β phosphorylation during KA-induced excitotoxic damage.
A previous study reported that a preferential dephosphorylation of phospho-GSK-3β by PP1 phosphatase and dephosphorylation of phospho-GSK-3α by PP2A based on the different sensitivities of PP1 and PP2A to OA treatment [33]. In their model of COS-7 cells (African green monkey kidney fibroblast-like cells), the relative level of pS9-GSK-3β was approximately 5-fold higher than that of pS21-GSK-3α, indicating a stronger regulation of GSK-3β phosphorylation might be involved in the cell line. However, in mouse brain extracts, relative phosphorylation of GSK-3α at Ser21 and GSK-3β at Ser9 are at a similar level (Fig. 1A), which ensured consistent substrate concentrations for the protein phosphatase inhibition assay. Moreover, Cal A also inhibited pS9-GSK-3β dephosphorylation at relatively low concentrations and suppressed pS21-GSK-3α dephosphorylation at high concentrations, consistent with the results of OA treatment. These findings further supported that pS9-GSK-3β was dephosphorylated by a PP2A-like phosphatase in mouse brain extracts and suggested that pS9-GSK-3β dephosphorylation might be differentially regulated in COS-7 cells and brain extracts.
Both PP2A and PP1 were reported to dephosphorylate GSK-3β at Ser9 in vitro [15–17]. In the present study, we found that overexpression of PP2A, but not PP1, altered GSK-3β phosphorylation in HEK-293FT cells, suggesting that PP1 may not dephosphorylate pS9-GSK-3β in HEK-293FT cells effectively. It has been shown that PP1 target proteins contain an Arg/Lys-Val/Ile-Xaa-Phe motif. PP1 interact with target proteins via this conservative motif. GSK-3β does not contain this PP1 acting motif, also suggesting GSK-3β may not interact with PP1 directly. However, we cannot exclude the possibility that PP1 could also play a minor role in regulating GSK-3β phosphorylation as PP1 also co-localized and interacted with GSK-3β in cultured cells through the regulatory protein.
Holoenzyme of PP2A consists of A, C, and B subunit. Regulatory B subunits determines subcellular localization and substrate specificity. There are four types of B subunits through which PP2A acts on different substrates. Thus, overexpression of PP2Ac without increase of B subunit results in a 20% reduction of GSK-3β phosphorylation, supporting critical role of B subunits. Leu309 methylation of PP2Ac is essential for the assembly of PP2A holoenzyme containing a B subunit PR55, which targets tau substrate [23, 24]. The levels of LCMT-1 and methylated PP2Ac are decreased, concomitant with hyperphosphorylation of tau in AD brains [34–36]. Deregulation of PP2A methylation also disrupts the interaction between PP2A and tau [37], and alters tau distribution in neuroblastoma cells [38]. These previous studies have shown that methylation of PP2Ac is essential for it to dephosphorylate tau. In addition, PP2A activity is suppressed by the endogenous inhibitors, I1PP2A and I2PP2A, which are both upregulated and can lead to tau hyperphosphorylation in AD brain [39–42].
To our surprise, results of the present study indicated that unlike tau, phosphorylation of GSK-3β at Ser9 is regulated by PP2A in a mechanism independent of its Leu309 methylation. Changes of PP2Ac methylation induced by overexpression or knock-down of PME-1 or LCMT-1 did not affect the levels of phosphorylated GSK-3β at Ser9. After KA treatment, pS9- GSK-3β level was reduced (at 2.5 h post-injection) prior to the changes in the level of demethylated-PP2Ac (at 10 h post-injection), suggesting that the rapid dephosphorylation of GSK-3β was independent of PP2A methylation during the first phase of excitotoxicity. Although the levels of demethylated-PP2Ac were not altered, PP2A activity was significantly enhanced during the first phase [30], implying other post-translational modification(s) of PP2A might be involved in modulating PP2A phosphatase activity in the rapid response to KA treatment.
Although PP2Ac methylation is important for the association of PR55/B subunits with the core enzyme [23, 24], it is not absolutely required for the recruitment of the PR61/B’ or PR72/B” subunit families [24]. All the three families of PP2A B-type subunits, PR55/B, PR61/B’, and PR72/B”, are expressed in the brain [43–45]. Besides PR55/B subunits [34, 43], PR61/B’ subunits may also possess a brain-specific function for its high expression levels in mouse brain [44]. In the present study, we found that the dephosphorylation of GSK-3β at Ser9 by PP2A is the methylation insensitive. It has been shown that knockout of PR61/B’δ in mouse brain increases GSK-3β phosphorylation [46], suggesting PR61/B’δ might be the regulatory subunit of PP2A to target GSK-3β. Thus, altered methylation of PP2A by GSK-3β may not affect PP2A activity toward GSK-3β.
Besides methylation, PP2A activity is regulated by phosphorylation of its catalytic subunit at Thr304 and Tyr307. The phosphorylation affects the interaction between PP2Ac and regulatory B subunits [21–24]. GSK-3β promotes methylation of PP2Ac via PME-1 and LCMT-1 and phosphorylation of PP2Ac at Tyr307 via PP1B [47, 48]. Thus, GSK-3β may enhance PP2A activity via the methylation and suppresses PP2A activity via Tyr307 phosphorylation in the C-tail sensitive B subunits mediated PP2A. It has been shown that PR61/B’δ does not require methylated PP2A nor an intact C-terminal PP2Ac tail to form holoenzymes [46]. Therefore, if dephosphorylation of GSK-3β by PP2A is mediated PR61/B’δ, GSK-3β phosphorylation may not be affected by the C-tail post-translational modification. However, the effect of phosphorylation of PP2Ac at Thr304 or Try307 on GSK-3β remains to be determined.
In conclusions, we found that PP2A was the major phosphatase of GSK-3β in cultured cells and in brain extracts. PP2A activity was negatively correlated with Ser9 phosphorylation of GSK-3β during KA-induced excitotoxicity. PP2A methylation was not required for GSK-3β dephosphorylation. Our results suggested that dephosphorylation of GSK-3β by PP2A was regulated through a mechanism different from that of tau.
Footnotes
ACKNOWLEDGMENTS
This work was supported in part by Nantong University and the New York State Office of People with Developmental Disabilities (OPWDD), and by grants from the U.S. Alzheimer’s Association (Grants IIRG-10-173154), the National Natural Science Foundation of China (Grant 81030059 and 81300978), Graduate research and innovation projects (YKC13050), and the Priority Academic Program Development of Jiangsu Higher Education institutions (PAPD).
