Nitric Oxide Induces Tau Hyperphosphorylation Via Glycogen Synthase Kinase-3B Activation

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

FEBS Letters 579 (2005) 6230–6236 FEBS 30081

Nitric oxide induces tau hyperphosphorylation via glycogen synthase kinase-3b activation

Yong-Jie Zhang, Ya-Fei Xu, Ying-Hua Liu, Jun Yin, Jian-Zhi Wang* Pathophysiology Department, Key Laboratory of Neurological Disease of Hubei Province, Tongji Medical College, Hua-Zhong University of Science and Technology, Wuhan 430030, PR China

Received 11 August 2005; revised 14 September 2005; accepted 14 September 2005

Available online 17 October 2005

Edited by Sandro Sonnino

implicated in the formation of NFT, and nitric oxide is an acti- Abstract Nitric oxide is associated with neurofibrillary tangle, which is composed mainly of hyperphosphorylated tau in the vator of p21ras [7–9]; the co-expression of NOS and p21ras in brain of AlzheimerÕs disease (AD). However, the role of nitric tangle-containing pyramidal neurons in early stage of AD has oxide in tau hyperphosphorylation is unclear. Here we show that also been reported [2]. More recently, it was reported that tau nitric oxide produced by sodium nitroprusside (SNP), a recog- hyperphosphorylation with a concurrent elevation of NOS nized donor of nitric oxide, induces tau hyperphosphorylation mRNA and protein level was observed by co-culture of Ab- at Ser396/404 and Ser262 in HEK293/tau441 cells with a simul- stimulated astrocytes with primary hippocampal neurons, taneous activation of glycogen synthase kinase-3b (GSK-3b). and this hyperphosphorylation of tau was eliminated by addi- Pretreatment of the cells with 10 mM lithium chloride (LiCl), tion of NOS inhibitor [10]. All these studies strongly suggested an inhibitor of GSK-3, 1 h before SNP administration inhibits that nitric oxide may be involved in tau hyperphosphorylation GSK-3b activation and prevents tau from hyperphosphorylation. and neurofibrillary tangle formation in AD brain. However, This is the first direct evidence demonstrating that nitric oxide induces AD-like tau hyperphosphorylation in vitro, and GSK- the direct evidence for the role of nitric oxide in AD-like tau 3b activation is partially responsible for the nitric oxide-induced hyperphosphorylation and the possible underlying mechanism tau hyperphosphorylation. It is suggested that nitric oxide may is still missing. be an upstream element of tau abnormal hyperphosphorylation In the present study, we treated HEK293/tau441 cells in AD. directly with sodium nitroprusside (SNP), a widely used nitric 2005 Published by Elsevier B.V. on behalf of the Federation of oxide donor with short turnover [11–13]. We found that nitric European Biochemical Societies. oxide induced tau hyperphosphorylation at Ser396/404 and Ser262 with a concurrent GSK-3b activation, and inhibition Keywords: Nitric oxide; Sodium nitroprusside; Tau; of GSK-3 eliminated the effect of nitric oxide on tau hyper- Hyperphosphorylation; Glycogen synthase kinase-3b phosphorylation at Ser396/404.

2. Materials and methods 1. Introduction 2.1. Chemicals and antibodies Rabbit polyclonal antibody (pAb) against GSK-3b and pAb against Neurofibrillary tangle (NFT) is one of the most prominent phospho-GSK-3b (Serine 9) were purchased from Cell Signaling Tech- histopathological features in AlzheimerÕs disease (AD). Accu- nology, Inc. (Beverly, MA, USA). pAb Ser262 against tau phosphory- mulating evidence has demonstrated that abnormal hyper- lated at Ser262 was purchased from Biosource Inc, (Carmarillo, CA, USA). pAb R134d against total tau was a gift from Dr. I Grundke- phosphorylation of microtubule-associated protein tau, Iqbal (New York State Institute for Basic Research, Staten Island, which aggregates into paired helical filaments (PHF), is the NY, USA). Monoclonal antibody (mAb) PHF-1 against tau phos- main element of NFT [1]. However, it is still not fully under- phorylated at Ser396/404 was a gift from Dr. P Davies (Albert Einstein stood for the upstream effectors and the mechanisms leading College of Medicine, Bronx, NY, USA). mAb Tau-1 against tau to tau hyperphosphorylation and formation of neurofibrillary unphosphorylated at Ser199/202 was from Chemicon International Inc., (Temecula CA, USA). Goat anti-rabbit or goat anti-mouse per- degeneration. oxidase-conjugated secondary antibody, bicinchoninic acid kit and The following studies show that nitric oxide (NO) is relevant chemiluminescent substrate kit were from Pierce Chemical Company with AD pathology: different isoforms of nitric oxide synthase (Rockford, IL, USA). Sodium nitroprusside (SNP) was from Axxora (NOS), the enzyme catalyzing the production of nitric oxide, Life Science Inc. (San Diego, CA, USA). Nitric oxide detection kit was purchased from Nanjing Jiancheng Biologic Engineering Co. are coexisted with NFT in tangle-bearing neurons of AD brain (Nanjing, P. R. China). mAb DM1A against a-tubulin, Lithium chlo- [2–5]; the dimethylargininase, a zinc protein involved in the ride (LiCl) and other chemicals were purchased from Sigma Chemical regulation of nitric oxide synthase, is specifically elevated in Co. (St. Louis, MO, USA). the cytoplasm of tangle- and pretangle-bearing neurons in AD brain [6]; elevated p21ras expression in AD brain has been 3. Cell culture, treatment and lysate preparation

*Corresponding author. Fax: +86 27 83693883. Human Embryonic Kidney (HEK) 293 cells (a kind gift E-mail address: [email protected] (J.-Z. Wang). from Dr. H. Xu, The Burnham Institute, San Diego, USA)

0014-5793/$30.00 2005 Published by Elsevier B.V. on behalf of the Federation of European Biochemical Societies. doi:10.1016/j.febslet.2005.09.095 Y.-J. Zhang et al. / FEBS Letters 579 (2005) 6230–6236 6231 were stably transfected with the longest human tau cDNA 6. Statistical analysis (HEK293/Tau441) were seeded at 3 · 105 cells/well in 6-well plates in DMEM in the presence of 200 lg/ml G418 with Data were analyzed using SPSS 10.0 statistical software. The 10% fetal bovine serum (FBS, vol/vol) in a humidified atmo- one-way ANOVA procedure followed by LSDÕs post hoc tests sphere of 5% CO2 at 37 C. Cells were used for experiment was used to determine the statistical significance of differences 36 h after seeding. of the means. SNP was freshly resolved in medium before use and the cells were cultured with different concentrations of SNP for 15 or 30 min as indicated in the figures. To ensure the changes are 7. Results caused by nitric oxide but not by other chemicals released from the donor, we designed the control experiments by using light- 7.1. Nonlinear dose-dependent release of nitric oxide from depleted SNP, i.e., by extensive exposure of SNP to light for 14 SNP in culture medium days to eliminate nitric oxide from the product [14]. To con- The level of nitric oxide in the culture medium was mea- firm the effect of GSK-3b, we used LiCl (10 mM) 1 h before sured by Griess Reaction [15] after treatment of the cells SNP administration to inhibit GSK-3. with different doses (0, 0.25, 0.5, 1, 4, and 8 mM) of SNP Cells were rinsed twice in ice-cold phosphate-buffered saline for 15 and 30 min, respectively. It was shown that the level (PBS, pH 7.5) and lysed with buffer containing 50 mM Tris–Cl, of nitrite, a stable metabolite representing the level of nitric pH 8.0, 150 mM sodium chloride, 1% NP-40, 0.5% sodium oxide released in the system, was increased in a nonlinear deoxycholate, 0.1% sodium dodecyl sulfate, 0.02% sodium dose-dependent manner with SNP concentration (Fig. 1), azide, 100 lg/ml phenylmethysulfonyl fluoride, and 10 lg/ml which is consistent with previous report [16,17]. A signifi- protease inhibitors (leupeptin, aprotinin and pepstatin) fol- cantly increased level of nitrite (P < 0.05) was shown when lowed by sonication for 5 s on ice. After centrifugation at the concentration of SNP reached 0.25 mM or higher and 12000 · g for 5 min at 4 C, supernatants were fetched out about 80–85% of the nitric oxide was released in the first and added with equal volume of 2· Laemmli sample buffer 15 min after SNP administration. Pretreatment of the cells (125 mM Tris–HCl, pH 6.8, 8% SDS, 17% glycerol, 10% by LiCl showed no remarkable effect on the release of nitric b-mercaptoethanol, and 0.05% bromophenol blue). Samples oxide from SNP (Fig. 1). were boiled for 10 min before electrophoresis. Protein concentration was estimated by BCA kit. 7.2. Nitric oxide induces tau hyperphosphorylation at Ser396/ 404 and Ser262 in HEK293/tau441 cells 4. Measurement of nitric oxide production The level of tau phosphorylation was measured after the cells were treated with SNP at different concentrations (0, Nitric oxide generated from SNP was determined indirectly 0.25, 0.5, 1, 4, and 8 mM) for 15 and 30 min, respectively. through the measurement of nitrite, a stable metabolite of As shown in Fig. 2, enhanced staining of tau at PHF-1 nitric oxide, by the Griess Reaction [15]. This assay relies on and pS262 epitopes, representing tau hyperphosphorylation a simple colorimetric reaction between nitrite and Griess re- at Ser396/404 (PHF-1) and Ser262 (pS262) (P < 0.05), was agent (0.1% N-(naphthyl)ethylenediamine dihydrochloride, shown after SNP treatment. No obvious change was seen 1% sulfanilamide in 5% phosphoric acid) to produce a pink at Tau-1 epitope (Ser199/202) as well as the level of total azo product. Briefly, a 50 ll aliquot of medium was mixed with tau (probed by R134d) (Fig. 2A). Quantitative analysis an equal volume of Griess reagent and incubated for 10 min at showed that the elevation of tau phosphorylation was not room temperature. The absorbance was read at 550 nm using a linear with increase of SNP concentration (Fig. 2B). To en- microtiter plate reader and compared with that of standard sure that the altered phosphorylation of tau is due to nitric sodium nitrite solution for quantitation. oxide but not other chemicals released from SNP, we repeated the same experiment using nitric oxide-depleted SNP (dSNP) (see Section 2) and found that the dSNP 5. Western blotting showed no obvious effect on tau phosphorylation (Fig. 2C). This data confirmed the role of nitric oxide in The cell lysates with equal amount of protein (i.e., 8 lg for tau hyperphosphorylation. R134d, Tau-1 and DM1A, 10 lg for GSK-3b,30lg for pS262 and pGSK-3b, and 40 lg for PHF-1) were firstly sepa- 7.3. Nitric oxide activates GSK-3b in HEK293/tau441 cells rated by 10% SDS–polyacrylamide gel electrophoresis (PAGE) As glycogen synthase kinase-3b (GSK-3b) is the major ki- and then electrophoretically transferred to nitrocellulose mem- nase responsible for tau phosphorylation at Ser396/404 (rec- brane. Tau was probed with antibodies R134d (1:5000), Tau-1 ognized by PHF-1) [18], we then estimated the activity of (1:30000), PHF-1 (1:500), pS262 (1:1000), GSK-3b was probed GSK-3b by measuring the ratio of Ser9-phosphorylated with anti-total-GSK-3b antibody (1:1000) and anti-GSK-3b (i.e., inactivated form of the kinase [19]) to total level of antibody specific to Ser9-phosphorylated form (1:1000), GSK-3b. As shown in Fig. 3A, the immunoreaction of a-tubulin was probed with DM1A (1:500). Immunoreactive Ser9-phosphorylated GSK-3b was significantly decreased in bands were then visualized by enhanced chemiluminescent a dose-dependent manner and such decrease is much more substrate kit and exposed to CL-XPosure film. Protein bands obvious in the first 15 min after SNP administration. These were quantitatively analyzed by Kodak Digital Science 1D results indicate that nitric oxide activates GSK-3b by reduc- software (Eastman Kodak Company, New Haven, CT, tion in phosphorylation at Ser9 under the experimental con- USA), and were expressed as sum optical density. dition used. No obvious change was seen in total level of 6232 Y.-J. Zhang et al. / FEBS Letters 579 (2005) 6230–6236

Fig. 1. Nonlinear dose-dependent release of nitric oxide from SNP in culture medium. Nitrite, the stable metabolite of nitric oxide, in the culture media was measured by Griess Reaction. The results are expressed as the means ± S.D. (N = 6); \P < 0.05, \\P < 0.01 vs. corresponding control (i.e., no SNP treatment at 15 and 30 min, respectively).

Fig. 2. Nitric oxide induces tau hyperphosphorylation at Ser396/404 and Ser262 in HEK293/tau441 cells. Equal amount of protein (i.e., 8 lg for R134d, Tau-1 and DM1A, 40 lg for PHF-1 and 30 lg for pS262) were loaded each lane for Western blot (A,C) and quantitative analysis (B). The level of Ser396/404 and Ser262 epitopes-phosphorylated tau was signiﬁcantly elevated, but no obvious change was seen in total tau (R134d) as well as Tau-1 epitope-dephosphorylated tau and tubulin (DM1A). dSNP showed no obvious eﬀect on tau phosphorylation (C). The results are expressed as the means ± S.D. (N = 3); \P < 0.05, \\P < 0.01 vs. controls (see Fig. 1). Y.-J. Zhang et al. / FEBS Letters 579 (2005) 6230–6236 6233

Fig. 3. Nitric oxide activates GSK-3b in HEK293/tau441 cells. Equal amount of protein (10 lg for GSK-3b, and 30 lg for Ser9-phosphorylated GSK-3b) were applied for Western blot (A,C) and for quantitative analysis (B). The significantly decreased ratio of Ser9-phosphorylated GSK-3b (i.e., the inactivated form) to total GSK-3b represents significantly increased activity of the enzyme after SNP treatment. dSNP showed no obvious effect on the phosphorylation level of Ser9-GSK-3b. The results were expressed as the means ± S.D. (N = 3); \P < 0.05, \\P < 0.01 vs. controls (see Fig. 1).

GSK-3b (probed by pAb against GSK-3b). The same exper- 8. Discussion iment was repeated by using nitric oxide-depleted SNP (dSNP) and found that the dSNP has no obvious effect on Abnormal hyperphosphorylation of tau is a pivotal lesion in GSK-3b Ser9 phosphorylation (Fig. 3C). These data sug- the pathogenesis of AlzheimerÕs diseases (AD). The identifica- gested that the activation of GSK-3b may be involved in ni- tion of upstream factors that induce tau abnormal hyper- tric oxide-induced tau hyperphosphorylation. phosphorylation and the underlying molecular mechanisms are among the major goals of research on AD. Increasing evi- 7.4. Inhibition of GSK-3b by LiCl partially attenuates nitric dence has shown that nitric oxide may be involved in the for- oxide-induced tau phosphorylation at Ser396/404 in mation of neurofibrillary degeneration in AD brain [2–6,10]; HEK293/tau441 cells however, direct evidence and the underlying mechanisms sup- To further confirm the role of GSK-3b in nitric oxide- porting this notion are still lacking. Here we show that nitric induced tau hyperphosphorylation, we treated the cells with oxide can induce tau hyperphosphorylation at Ser396/404 10 mM LiCl, a widely used GSK-3 inhibitor [20,21], for 1 h and Ser262 sites in HEK293/Tau441 cells and GSK-3b activa- before SNP administration. We observed that the immuno- tion is at least partially responsible for nitric oxide-induced reaction of Ser9-phosphorylated GSK-3b was dramatically Alzheimer-like tau hyperphosphorylation at Ser396/404 sites. increased in the SNP group pretreated with LiCl (P < 0.05) Nitric oxide is produced by NOS. It is reported that during with no obvious change in the level of total GSK-3b.At the chronic progress of AD, overproduced b-amyloid (Ab) the same time, hyperphosphorylation of tau at PHF-1 sites can activate microglia and astrocytes and thus leads to NOS- (P < 0.01) was also partially restored in LiCl treated group, mediated production of nitric oxide [22–25]. Due to its free but no obvious change in total tau was seen (Fig. 4). It is radical properties, nitric oxide may in turn mediate Ab toxicity also observed that LiCl can also inhibits the basal GSK- and cause oxidative damage as seen in AD [26]. Taken together 3b activity showing by increased level of Ser9-phosphory- the above information and the observation from this study, we lated GSK-3b (Fig. 4). This data confirmed that GSK-3b propose that tau hyperphosphorylation seen in AD brain activation is largely responsible for the nitric oxide-induced might be a downstream consequence of the reciprocal nitric tau hyperphosphorylation. As GSK-3b does not phosphory- oxide-Ab mediated toxicity. late Ser262 in vitro, we have not detected whether LiCl also We choose SNP, a recognized donor of nitric oxide, as a di- attenuated tau hyperphosphorylation at this site. rect source of nitric oxide in this study. By incubating the cells 6234 Y.-J. Zhang et al. / FEBS Letters 579 (2005) 6230–6236

Fig. 4. Inhibition of GSK-3b by LiCl partially attenuates nitric oxide-induced tau phosphorylation at Ser396/404 in HEK293/tau441 cells. The cells were pretreated with 10 mM LiCl for 1 h before SNP administration. The Western blots (A) and quantitative analysis (B) show that LiCl restores Ser9-phosphorylated GSK-3b to normal level and concurrently prevents tau from SNP-induced hyperphosphorylation at PHF-1 epitope. The results are expressed as the means ± S.D. (N = 3); \P < 0.01 vs. only SNP treatment of the corresponding doses. #P < 0.05, ##P < 0.01 vs. only LiCl treatment of the corresponding doses. with different concentrations of SNP, we observed that the The level of nitric oxide and its derivatives, including perox- amount of nitrite, a stable metabolite used to evaluate the level ynitrite, is elevated during the chronic process of oxidative of nitric oxide, released from SNP to the culture medium was stress in AD [33,34]. The relationship between nitric oxide or increased dose dependently. LiCl, a GSK-3 inhibitor, showed oxidative stress and GSK-3b is not currently understood. It no significant effect on the production of nitric oxide from is reported that peroxynitrite induces nitration of regulatory SNP. With this data, we can use SNP as a liable nitric oxide subunit p85 of PI 3-kinase [35] and abolishes its interaction donor to study the effect of nitric oxide on tau phosphorylation with the catalytic subunit p110, and thus inhibits the activity and the underlying mechanisms. of PI 3-kinase [36]. PI 3-kinase is known to stimulate Akt/ After exposure of the cells to SNP, tau hyperphosphoryla- PKB activity, which in turn inhibits GSK-3b by phosphorylat- tion at PHF-1 and Ser262 sites was observed. As GSK-3b is ing it at Ser9 [19]. Therefore, during the long process of oxida- the major kinase in phosphorylating PHF-1 sites (Ser396/ tive stress in AD, nitric oxide-derived species peroxynitrite 404) both in vitro and in vivo [18,27] and it is also a major may overactivate GSK-3b by downregulating the PI 3-kinase kinase implicated in abnormal tau hyperphosphorylation in pathway [36] and thus lead to tau hyperphosphorylation at AD brain [28–31], we then detected the activity of GSK-3b GSK-3b sites. This may partially explain the mechanism that by activity-dependent antibody. We observed that the inacti- nitric oxide activates GSK-3b. vated form (i.e., Ser9-phosphorylated form) of GSK-3b was We also noticed that treatment of the cells with SNP induced decreased in a dose-dependent manner after SNP treatment, enhancement in tau phosphorylation at Ser396/404 and and such decline was especially dramatic in the first 15 min Ser262, but neither of these sites phosphorylation correlated of SNP treatment, suggesting activation and involvement of well with the concentration and duration of SNP exposure, GSK-3b in nitric oxide-induced tau hyperphosphorylation. i.e., the hyperphosphorylation of tau was not positively corre- This involvement of GSK-3b was further confirmed by lated with the increased level of nitric oxide. Recent study has pretreatment of the cells with GSK-3 inhibitor, LiCl. As demonstrated that nitric oxide may regulate kinase activity 80–85% of the nitric oxide was released in the first 15 min, with distinct concentration threshold [37], suggesting that the we speculated that it might be the quick production of nitric kinases involved in this nitric oxide-induced tau hyperphosph- oxide rather than the total amount of the molecule responsible orylation may possess different sensitivity to the amount of ni- for the activation of GSK-3b. This is consistent with the view- tric oxide. As tau phosphorylation is regulated by a balanced point that it is the rate of nitric oxide flux but not the net regulation of various kinases and phosphatases, we believe amount of the molecule determining the final effect of nitric that in addition to GSK-3b, other kinase(s) must have also oxide [32]. Further investigation is needed to verify whether participated in nitric oxide-induced tau hyperphosphorylation, nitric oxide regulates GSK-3b activation in a flux-dependent and this is especially true for Ser262, which is not a direct manner in vivo. GSK-3b site [38]. Y.-J. Zhang et al. / FEBS Letters 579 (2005) 6230–6236 6235

In conclusion, we have demonstrated in the present study [15] Green, L.C., Wagner, D.A., Glogowski, J., Skipper, P.L., that nitric oxide induces tau hyperphosphorylation at Ser396/ Wishnok, J.S. and Tannenbaum, S.R. (1982) Analysis of nitrate, 404 through GSK-3b activation in vitro. Nitric oxide may be nitrite, and [15N] nitrate in biological fluids. Anal. Biochem. 126, 131–138. an upstream element of tau hyperphosphorylation in AD [16] Motterlini, R., Foresti, R., Intaglietta, M. and Winslow, R.M. pathology. (1996) NO-mediated activation of heme oxygenase: endogenous cytoprotection against oxidative stress to endothelium. Am. J. Acknowledgements: This work was supported in part by grants from Physiol. 270, H107–H114. Committee of Science and Technology of China (No. [17] Sanchez, J.J., Abreu, P. and Gonzalez, M.C. (2002) Sodium 2006CB500703) and from the Natural Science Foundation of China nitroprusside stimulates L-DOPA release from striatal tissue (No. 30430270). We thank Dr. K. Iqbal, I. Grundke-Iqbal and C.X. through nitric oxide and cGMP. Eur. J. Pharmacol. 438, 79– Gong from NYS Institute for Basic Research for scientific discussion 83. and reagents support. We also thank Dr. P. Davies from Albert Ein- [18] Cho, J.H. and Johnson, G.V. (2003) Glycogen synthase kinase stein College of Medicine for PHF-1 antibody. 3beta phosphorylates tau at both primed and unprimed sites: differential impact on microtubule binding. J. Biol. Chem. 278, 187–193. [19] Grimes, C.A. and Jope, R.S. (2001) The multifaceted roles of References glycogen synthase kinase 3beta in cellular signaling. Prog. Neurobiol. 65, 391–426. [1] Grundke-Iqbal, I., Iqbal, K., Quinlan, M., Tung, Y.C., Zaidi, [20] Ryves, W.J. and Harwood, A.J. (2001) Lithium inhibits glycogen M.S. and Wisniewski, H.M. (1986) Microtubule-associated pro- synthase kinase-3 by competition for magnesium. Biochem. tein tau. A component of Alzheimer paired helical filaments. J. Biophys. Res. Commun. 280, 720–725. Biol. Chem. 261, 6084–6089. [21] Eickholt, B.J., Walsh, F.S. and Doherty, P. (2002) An inactive [2] Luth, H.J., Holzer, M., Gertz, H.J. and Arendt, T. (2000) pool of GSK-3 at the leading edge of growth cones is implicated in Aberrant expression of nNOS in pyramidal neurons in Alzhei- Semaphorin 3A signaling. J. Cell Biol. 157, 211–217. merÕs disease is highly co-localized with p21ras and p16INK4a. [22] DiPatre, P.L. and Gelman, B.B. (1997) Microglial cell activation Brain Res. 852, 45–55. in aging and Alzheimer disease: partial linkage with neurofibril- [3] Vodovotz, Y., Lucia, M.S., Flanders, K.C., Chesler, L., Xie, lary tangle burden in the hippocampus. J. Neuropathol. Exp. Q.X., Smith, T.W., Weidner, R., Mumford, J., Webber, R., Neurol. 56, 143–149. Nathan, C., Roberts, A.B., Lippa, C.F. and Sporn, M.B. [23] Fukumoto, H., Asami-Odaka, A., Suzuki, N. and Iwatsubo, T. (1996) Inducible nitric oxide synthase in tangle-bearing neurons (1996) Association of A beta 40-positive senile plaques with of patients with AlzheimerÕs disease. J. Exp. Med. 184, 1425– microglial cells in the brains of patients with AlzheimerÕs disease 1433. and in non-de mented aged individuals. Neurodegeneration 5, 13– [4] Law, A., Gauthier, S. and Quirion, R. (2001) Say NO to 17. AlzheimerÕs disease: the putative links between nitric oxide and [24] Laskowitz, D.T., Matthew, W.D., Bennett, E.R., Schmechel, D., dementia of the AlzheimerÕs type. Brain Res. Brain Res. Rev. 35, Herbstreith, M.H., Goel, S. and McMillian, M.K. (1998) Endog- 73–96. enous apolipoprotein E suppresses LPS-stimulated microglial [5] Thorns, V., Hansen, L. and Masliah, E. (1998) nNOS expressing nitric oxide production. NeuroReport 9, 615–618. neurons in the entorhinal cortex and hippocampus are affected in [25] Vitek, M.P., Snell, J., Dawson, H. and Colton, C.A. (1997) patients with AlzheimerÕs disease. Exp. Neurol. 150, 14–20. Modulation of nitricoxide production in human macrophages by [6] Smith, M.A., Vasak, M., Knipp, M., Castellani, R.J. and Perry, apolipoprotein-E and amyloid-beta peptide. Biochem. Biophys. G. (1998) Dimethylargininase, a nitric oxide regulatory protein, in Res. Commun. 240, 391–394. Alzheimer disease. Free Radic. Biol. Med. 25, 898–902. [26] Koppal, T., Drake, J., Yatin, S., Jordan, B., Varadarajan, S., [7] Dawson, T.M., Sasaki, M., Gonzalez-Zulueta, M. and Dawson, Bettenhausen, L. and Butterfield, D.A. (1999) Peroxynitrite- V.L. (1998) Regulation of neuronal nitric oxide synthase and induced alterations in synaptosomal membrane proteins: insight identification of novel nitric oxide signaling pathways. Prog. into oxidative stress in AlzheimerÕs disease. J. Neurochem. 72, Brain Res. 118, 3–11. 310–317. [8] Lander, H.M., Ogiste, J.S., Teng, K.K. and Novogrodsky, A. [27] Liu, S.J., Zhang, A.H., Li, H.L., Wang, Q., Deng, H.M., (1995) p21ras as a common signaling target of reactive free Netzer, W.J., Xu, H. and Wang, J.Z. (2004) Overactivation of radicals and cellular redox stress. J. Biol. Chem. 270, 21195– glycogen synthase kinase-3 by inhibition of phosphoinositol-3 21198. kinase and protein kinase C leads to hyperphosphorylation of [9] Yun, H.Y., Gonzalez-Zulueta, M., Dawson, V.L. and Dawson, tau and impairment of spatial memory. J. Neurochem. 87, T.M. (1998) Nitric oxide mediates N-methyl-D-aspartate receptor- 1333–1344. induced activation of p21ras. Proc. Natl. Acad. Sci. USA 95, [28] Pei, J.J, Braak, E., Braak, H., Grundke-Iqbal, I., Iqbal, K., 5773–5778. Winblad, B. and Cowburn, R.F. (1999) Distribution of active [10] Saez, T.E., Pehar, M., Vargas, M., Barbeito, L. and Maccioni, glycogen synthase kinase 3beta (GSK-3beta) in brains staged for R.B. (2004) Astrocytic nitric oxide triggers tau hyperphosphory- Alzheimer disease neurofibrillary changes. J. Neuropathol. Exp. lation in hippocampal neurons. In Vivo 18, 275–280. Neurol. 58, 1010–1019. [11] Fitzgerald, R.S., Shirahata, M. and Chang, I. (2005) The effect of [29] Shiurba, R.A., Ishiguro, K., Takahashi, M., Sato, K., Spooner, a nitric oxide donor, sodium nitroprusside, on the release of E.T., Mercken, M., Yoshida, R., Wheelock, T.R., Yanagawa, H., acetylcholine from the in vitro cat carotid body. Neurosci. Lett. Imahori, K. and Nixon, R.A. (1996) Immunocytochemistry of tau 385, 148–152. phosphoserine 413 and tau protein kinase I in Alzheimer [12] William, M., Vien, J., Hamilton, E., Garcia, A., Bundgaard, H., pathology. Brain Res. 737, 119–132. Clarke, R.J. and Rasmussen, H.H. (2005) The nitric oxide donor [30] Mandelkow, E.M., Drewes, G., Biernat, J., Gustke, N., Van Lint, sodium nitroprusside stimulates the Na+–K+ pump in isolated J., Vandenheede, J.R. and Mandelkow, E. (1992) Glycogen rabbit cardiac myocytes. J. Physiol. 565, 815–825. synthase kinase-3 and the Alzheimer-like state of microtubule- [13] Paul, V. and Ekambaram, P. (2005) Effects of sodium nitroprus- associated protein tau. FEBS. Lett. 314, 315–321. side, a nitric oxide donor, on gamma-aminobutyric acid concen- [31] Avila, J., Lucas, J.J., Perez, M. and Hernandez, F. (2004) Role of tration in the brain and on picrotoxin-induced convulsions in tau protein in both physiological and pathological conditions. combination with phenobarbitone in rats. Pharmacol. Biochem. Physiol. Rev. 84, 361–384. Behav. 80, 363–370. [32] Bishop, A. and Anderson, J.E. (2005) NO signaling in the CNS: [14] Clarke, A., Desikan, R., Hurst, R.D., Hancock, J.T. and Neill, from the physiological to the pathological. Toxicology 208, 193– S.J. (2000) NO way back: nitric oxide and programmed cell 205. death in Arabidopsis thaliana suspension cultures. Plant J. 24, [33] Hensley, K., Maidt, M.L., Yu, Z., Sang, H., Markesbery, W.R. 667–677. and Floyd, R.A. (1998) Electrochemical analysis of protein 6236 Y.-J. Zhang et al. / FEBS Letters 579 (2005) 6230–6236

nitrotyrosine and dityrosine in the Alzheimer brain indicates signaling in retinal endothelial cells via PI 3-kinase tyrosine region-speciﬁc accumulation. J. Neurosci. 18, 8126–8132. nitration. J. Cell Sci. 118, 243–252. [34] Smith, M.A., Richey Harris, P.L., Sayre, L.M., Beckman, J.S. [37] Thomas, D.D., Espey, M.G., Ridnour, L.A., Hofseth, L.J., and Perry, G. (1997) Widespread peroxynitrite-mediated damage Mancardi, D., Harris, C.C. and Wink, D.A. (2004) Hypoxic in AlzheimerÕs disease. J. Neurosci. 17, 2653–2657. inducible factor 1alpha, extracellular signal-regulated kinase, and [35] Hellberg, C.B., Boggs, S.E. and Lapetina, E.G. (1998) Phospha- p53 are regulated by distinct threshold concentrations of nitric tidylinositol 3-kinase is a target for protein tyrosine nitration. oxide. Proc. Natl. Acad. Sci. USA 101, 8894–8899. Biochem. Biophys. Res. Commun. 252, 313–317. [38] Gong, C.X., Liu, F., Grundke-Iqbal, I. and Iqbal, K. (2005) Post- [36] el-Remessy, A.B., Bartoli, M., Platt, D.H., Fulton, D. and translational modiﬁcations of tau protein in AlzheimerÕs disease. Caldwell, R.B. (2005) Oxidative stress inactivates VEGF survival J. Neural. Transm. 112, 813–838.