Statin's Excitoprotection Is Mediated by sAPP and the Subsequent Attenuation of Calpain-Induced Truncation Events, Likely via Rho-ROCK Signaling

Statin's Excitoprotection Is Mediated by sAPP and the Subsequent Attenuation of Calpain-Induced Truncation Events, Likely via Rho-ROCK Signaling
of 11
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
  Neurobiology of Disease Statin’s Excitoprotection Is Mediated by sAPP and theSubsequent Attenuation of Calpain-Induced TruncationEvents, Likely via Rho-ROCK Signaling TaoMa, 1,2 YongBoZhao, 2 Young-DonKwak, 3 ZhangminYang, 4 RobertThompson, 3 ZhijunLuo, 4 HuaxiXu, 3 andFrancesca-FangLiao 1 1 Department of Pharmacology, University of Tennessee Health Science Center, College of Medicine, Memphis, Tennessee 38163,  2 Department of Neurology,First People’s Hospital of Shanghai Jiaotong University, Shanghai 200080, China,  3 Department of Neuroscience and Aging Center, Burnham Institute forMedical Research, La Jolla, California 92037, and  4 Department of Biochemistry, Boston University School of Medicine, Boston, Massachusetts 02118 The widely used cholesterol-lowering drugs, statins, were reported to reduce the incidence of stroke and the progression of Alzheimer’sdisease. However, little is known on how statins exert these beneficial effects. In this study, we investigated the molecular mechanismsunderlyingtheneuroprotectiveactionsofstatinsinprimaryculturedcorticalneurons.Wefoundthatchronictreatmentofneuronswitha low dosage of two CNS-permeable statins (lovastatin and simvastatin) selectively reduced NMDA-induced cell death but not thecaspase-mediated apoptosis. The protective effects of stains were inhibited by mevalonate, a PI3K inhibitor, and tyrphostin AG538,suggesting roles for cholesterol and insulin/IGF-1 signaling in the neurotoxic response. We further demonstrate that statins blockcalcium-dependentcalpainactivation,resultingincompletesuppressionofproteintruncationeventsonmultiplecalpainsubstratesthatare involved in neuronal death including CDK5 coactivator p35 cleavage to p25, GSK3 and  -catenin. This is followed by reduced andincreasednucleartranslocationofp25and  -catenin,respectively.Underexcitotoxicconditions,theactivitiesofCDK5and  -cateninareexclusively regulated by calpain-mediated cleavage while apoptosis modulates  -catenin mainly through phosphorylation. Strikingly,our data demonstrate that the calpain-blocking effect of statins is largely mediated by stimulation of    -secretase cleavage of APP,resulting in increased secretion of its soluble form, sAPP. Finally, our data suggest that statin-regulated sAPP secretion occurs viaactivation of the PI3K pathway and inhibition of ROCK signaling. Altogether, our study provides novel insights into statin-mediatedneuronal excitoprotection through both cholesterol-dependent and -independent mechanisms and links them to calpain-mediatedneuronal death. Introduction Excitotoxicitymediatedbyglutamate-gatedionchannelsisawelldocumented form of neuronal death caused by brain ischemia(Sattler and Tymianski, 2001), which has also been associatedwith several neurodegenerative diseases such as Alzheimer’s dis-ease, Parkinson’s disease, and Huntington’s disease (Choi, 1988,1995). Elevated extracellular glutamate has long been recognizedasahallmarkphenomenonduringneuronalexcitotoxicity(Choi,1988, 1995). Although excitotoxicity is triggered by an exagger-ated and prolonged rise in intracellular Ca 2  , little is knownabout the subsequent events that ultimately lead to cell death.Excessive glutamate triggers massive Ca 2  influx through NMDAreceptors (NMDARs), which in turn can activate the Ca 2  -dependent protease calpains. Calpains are likely to be involved inprocessing of numerous enzymes and cytoskeletal components,thereby linking their activity to a variety of intracellular events im-plicated in excitotoxicity-related conditions such as hypoxia, isch-emia,epilepsy,andAlzheimer’sdisease(RayandBanik,2003).Hydroxymethylglutaryl-coenzyme A (HMG-CoA) reductaseinhibitors, known as “statins,” have been highly effective in low-ering serum cholesterol and reducing the incidence of coronary events. Epidemiological studies have also supported a beneficialeffect in the human subjects taking statins through a reducedprevalence of AD and stroke (Switzer and Hess, 2006; Whitfield,2006; Miida et al., 2007). A limited number of clinical andexperimental studies have been published to elucidate themultiple mechanisms,dependentandindependentofstatins’anti-cholesterol effect (Delanty et al., 2001). In addition to its widely studied effect on APP processing and reducing A   production,which is likely mediated via its action on protein isoprenylation(Cordle et al., 2005; Cole and Vassar, 2006), statins have also beenreportedtobeexcitoprotective(Zaccoetal.,2003;Bo¨seletal.,2005).However,thedetailedmechanismsarelargelyundefined.Cortical neuronal culture model of NMDA toxicity has beenusedextensivelytoinvestigatethemechanismsofneuronalinjury  Received Dec. 24, 2008; revised May 12, 2009; accepted July 31, 2009.This work was supported by National Institutes of Health Grants R01 NS054880 (F.-F.L.), R01 AG021173 (H.X.),R01NS046673(H.X.),R01AG030197(H.X.),andR01CA118918(Z.L.),andbyAlzheimer’sAssociationInvestigator-Initiated Research Grant IIRG-06-26070 (F.-F.L.). We thank Traci Fang for assisting in preparation of primaryneurons.Correspondence should be addressed to Dr. Francesca-Fang Liao, Department of Pharmacology, University of Tennessee Health Science Center, College of Medicine, 874 Union Avenue, Memphis, TN 38163. © 2009 Society for Neuroscience 0270-6474/09/2911226-11$15.00/0 11226  •  The Journal of Neuroscience, September 9, 2009  •  29(36):11226–11236  and to test/screen for neuroprotective agents. Excitotoxic neuro-nal death induced by NMDA has been shown to occur throughboth necrosis and apoptosis, with apoptosis being the predomi-nant form when the insults are relatively mild (30–300   M NMDA) (Bonfoco et al., 1995). Using this model, we examinedthe excitoprotective effect of statins. Chronic treatment with st-atins protected cortical neurons against NMDA toxicity. Neuro-protectionwassubstantiallyabolishedbycotreatmentwitheither1 m M  mevalonate or cholesterol, suggesting that the neuropro-tective effect of statins is mediated by inhibition of   de novo cholesterol synthesis. In addition, our data revealed a cholesterol-independent mechanism by which statin excitoprotection involvesstimulation of soluble APP secretion, which is likely modulated by Rho-ROCK signaling and subsequent attenuation of calcium-dependentcalpainactivation. MaterialsandMethods  Antibodies and chemicals.  Spectrin    II C-3 (sc-48382), CDK5 (cyclin-dependent kinase 5) C-8 (sc-173), p35 C-19 (sc-820), IGF-IR     C-20(sc-713), and   -actin (sc-1615) antibodies were from Santa Cruz Bio-technology. Glycogen synthase kinase 3    (GSK3  , 9315), pGSK3  (Ser-9, 9323), GSK3  (9338), pGSK3  (Ser-21, 9316), pGSK3  /  (Ser-21/9, 9327), AKT (2966), pAKT/Ser-473 (4058), phospho-  -catenin(Ser33/37/Thr41, 9561), insulin receptor  (3025) antibodies were fromCell Signaling Technology.   -Catenin (C-terminal, 610153) was fromBD Transduction Laboratories. The monoclonal antibody 22C11,MAP-2 (MAB 3418), NeuN (MAB377), and pY (4G10) were from Mil-lipore. Alexa 488-conjugated anti-mouse IgG and Alexa-594-conjugatedanti-rabbit IgG were from Invitrogen.Lovastatin (LOV), simvastatin, mevalonic acid (MVA), cholesterol,NMDA, glycine, I-OMe-tyrphostin AG 538, farnesyl pyrophosphate(FPP), geranylgeranyl pyrophosphate (GGPP), 4  ,6  -diamidino-2-phenylindole (DAPI), trypan blue, poly- D -lysine, and SP600125 wereobtained from Sigma.  In situ  cell death detection kit (terminal deoxynu-cleotidyl transferase-mediated biotinylated UTP nick end labeling,TUNEL) was obtained from Roche. The chemical compounds includingLY294002, wortmannin, APV, staurosporine (STS), BH3I-1, camptoth-ecin (CPT), Y27632, FTI-277, FTase inhibitor II, FPT inhibitor II,TAPI-2, PD98059, U0126, SU203580, Z-VAD-FMK, calpain inhibitorI/  N  -acetyl-Leu-Leu-norleucinal, calpastatin/CS peptide, and PD150606were obtained from Calbiochem. A  25-35  peptides were from Bachem. Primary neuronal cell culture.  Primary cortical neurons were isolatedand purified from embryos of Sprague Dawley rats at embryonic day 17(E17) as described previously (Han et al., 2005). Isolated primary neu-rons were plated onto coverslips precoated with poly- D -lysine (100   g/ml) at a density of 75,000 per well in 24-well plates for staining or platedinto 6-well plates at a density of 600,000 per well or 100 mm dishes at adensity of 3,000,000 per dish precoated with poly- D -lysine for Westernblot analysis. The cultures were maintained in serum-free Neurobasalmedium and were treated with 5   M  AraC to inhibit proliferation of non-neuronal cells. All experiments presented in this work were per-formedonpureneuronalcells(  95%neuronalpurityassessedbystain-ing with neuronal marker proteins: neuronal-specific nuclear protein/NeuN and microtubule-associated protein-2/MAP-2) after 14 DIV.  NMDA-induced neuronal cell death.  The cultures were maintained inserum-free Neurobasal medium for 2 weeks to allow development of NMDAreceptorsbeforebeingchallengedwithNMDA(100  M ,15min)in Mg 2  -free Eagle’s balanced salt solution (EBSS) containing 1.8 m M CaCl 2  and 100   M  glycine. After NMDA exposure, cells were gently washed with EBSS/1.8 m M  CaCl 2  and 1.2 m M  MgCl 2  and returned to thesrcinal culture medium for an additional 16–24 h at 37°C with 9.6%CO 2  before being assessed for cell death. Neurons were also treated with200n M STS(6h),50  M BH3I-1(6h),10  M CPT(6h),or25  M A  25-35 peptide (24 h). To test the effect of LOV, 500 n M  LOV was added 3 dbefore NMDA exposure except as otherwise indicated in some experi-ments. LY294002, APV, Y27632, PD98059, U0126, SU203580, Z-VAD-FMK, calpain inhibitors, and SP600125 were added 1 h before NMDAexposure.Aftertreatmentwithcelldeath-inducingorblockingagents,celldeathwas assessed for necrosis and apoptosis. In most experiments, cell sur-vival was assessed by trypan blue exclusion. After staining with trypanblue at 0.4% for 5 min, cells were then washed with PBS and fixed in 4%paraformaldehydefor10min.Trypanblue-positivecellswerecountedin10 randomized fields. Apoptotic neurons were identified by TUNEL,followed by counterstaining with membrane-permeable DAPI (0.1   g/ml, 5 min). In addition to TUNEL, apoptosis was also assessed by con-densed morphological changes of the nuclei. Immunofluorescence staining.  Two week-cultured neurons seeded oncoverslips were fixed with 4% paraformaldehyde/PBS at room tempera-ture for 15 min and permeabilized in 0.2% Triton X-100/PBS for 5 min.After washing in PBS, The neurons were blocked with PBS containing10% goat serum at room temperatures for 60 min in a humid chamber.Primary antibodies, including NeuN (1:500), MAP-2(1:500), p35(1:200), and   -catenin (1:250), diluted in the blocking buffer were ap-plied to the specimens and incubated overnight at 4°C overnight. Afterextensive washes, Alexa 488-conjugated anti-mouse or Alexa-594-conjugated anti-rabbit secondary antibody (1:500) was applied and in-cubated for 1 h at room temperature, followed by counterstaining withDAPI. Slides were mounted in Fluoromount medium, and immunoflu-orescent signals were observed under confocal microscopy (LSM510;Zeiss).ForMAP-2staining,thelengthofneuriteswasquantifiedfrom50oftheMAP-2-positivecells,andthenumberofprocessesprojectingfromeachpositivecellwasalsoquantified.Forp35/25staining,p35/25immu-nofluorescence signals in the nucleus were compared with those in thecytoplasm, and the neurons showing higher fluorescence signals in thenucleus than in the surrounding cytoplasm were counted as “nuclearp35/25-positive cells.” For  -catenin staining, immunofluorescence sig-nals in the nucleus were quantified. Western blot analysis.  The assays were performed as described (Han etal., 2005). Primary antibodies used were as follows: spectrin    II C-3(1:1000), cleaved caspase-3 (1:1000), CDK5 C-8 (1:1000), p35 C-19 (1:1000), GSK3  /   (1:1000), GSK3   (1:1000), pGSK3   (1:1000, Ser-9),AKT (1:1000), pAKT (1:1000, Ser-473),  -catenin (1:1000, C-terminal),phosphor-  -catenin (1:1000, Ser33/37/Thr41), IR-  (1:1000), IGF-IR   (1:1000),pY(1:1000),and  -actin(1:5000).Insomeexperiments,West-ern blots were scanned and protein bands were quantified using ScionImage software. Mouse monoclonal antibody 22C11 (1:200) was used todetect sAPP secretion from the cultured conditioned media. Detection of cytosolic    -catenin.  Cells were collected in a lysis buffercontaining 20 m M  HEPES, pH 7.4, 2 m M  MgCl 2 , 1 m M  EDTA, 100 m M KCl, 0.5 m M  DTT, and a cocktail of protease inhibitors. Cell mixtureswere incubated on ice for 20 min and homogenized using a Douncehomogenizer for 25 strokes followed by centrifugation at 15,000   g   for30 min. Equal amounts of lysate samples were then subjected to SDS-PAGE and Western analysis. Preparation of dominant-negative mutant of ROCK1-expressing adenovi-rus.  The cDNA for human ROCK1 was purchased from Open Biosystems.The dominant-negative mutant (DN ROCK) was generated according toItoh et al. (1999). Accordingly, the cDNA encoding the amino acids 1-477was amplified by PCR using the following primers: (1) 5  forward primer,ATGTCGACTGGGGACAGTTTTGAG;(2)3  reverseprimer,TCATCTA-GATTTCTTCTTTGATTTCCCTCTTC (ending at amino acid 477); (3)mutatedforwardprimer,AAGGTATATGCTATGATG(forK-to-Mmuta-tion)CTTCTCAGC;and(4)mutatedreverseprimer,GCTGAGAAGCAT-CATAGCATATACCTT. The PCR product was then cloned into thepCR-BluntII-Topo vector (Invitrogen) and verified by sequencing. The in-sert was subsequently subcloned into pAdTrack-CMV at the NotI andEcoRVsites,whichwasfurtherusedfortransienttransfectionandadenovi-rus preparation. Adenoviral genome was made by carrying out recombina-tion of pAdTrack-CMV-ROCK1 with pAdEasy1 and transfection intoHEK293 cells, as described by He et al. (1998). The virus was amplified tworoundsandpurifiedbyanionicexchangecartridge(Sartorius). Statistical analysis.  Data are presented as mean  SD. For statisticalcomparison,theStudent’s t  testwasused.  p valuessmallerthan0.05wereconsidered to be statistically significant. Ma et al. • Novel Excitoprotective Mechanisms of Statins J. Neurosci., September 9, 2009  •  29(36):11226–11236  • 11227  Results Chronicstatintreatmentatlowdosageprotectsprimary corticalneuronsagainstNMDA-induced excitotoxicity  It has been reported that statins are excitoprotective in neuronsinsulted by NMDA. Various statins have been compared and ithas been shown that the protective potency is proportional totheirabilitytoinhibittheHMG-CoAreductaseactivity(Zaccoetal., 2003). To study the mechanisms underlying statin’s protec-tiveeffect,wetreatedprimaryculturedcorticalneurons(14DIV)with two different statins that are known to be the most perme-able to the CNS, lovastatin and simvastatin. We sought to inves-tigate these statins using a chronic treatment regimen with low dosages(  1  M )tomimictheclinicalsetting.Allworkpresentedherein was done in 2-week-cultured primary cortical neurons toallowexpressionofafullspectrumoftheNMDAreceptors.Sim-ilar results were obtained with lovastatin and simvastatin, wetherefore only present data on lovastatin.Inapilotstudy,weusedaclassictrypanblueassaytoassesscellviability and tested two statins at various concentrations rangingfrom 0.1 to 1   M . Concentrations of    2   M  resulted in acutecytotoxicity whereas the low concentrations,  0.5–1.0  M , pro-videdtheoptimalmaximumprotectionwhenneuronswerechal-lenged with NMDA (100   M ), as assessed by trypan bluepermeability. At an optimal concentration (0.5  M ), statins con-ferred protection in NMDA-treated neurons of   60%, compa-rable to that of APV, a specific NMDAR antagonist (Fig. 1  A ).Interestingly, the protection was only achieved with either statinafter chronic treatment (3–5 d). These optimized conditions of lowdosageandchronictreatment(0.5  M and3d)arerelevanttothe clinical regimen and were used consistently throughout ourstudies. As previously reported (Bonfoco et al., 1995; Han et al.,2005), exposure to 100  M  NMDA for 15 min resulted in  70%neuronal death with the majority being apoptosis, as evidencedby the TUNEL-positive staining (Fig. 1 B ). The apoptosis wassignificantly reduced by statin pretreatment (Fig. 1 B ), an ef-fect comparable to that of three calpain inhibitors,  N  -acetyl-Leu-Leu-norleucinal, calpastatin/CS peptide, and PD150606(data not shown). In contrast, NMDA-induced neuronaldeath was not prevented by Z-VAD-fmk, a pan caspase 3 in-hibitor (data not shown).We also examined morphological alterations on apoptoticcellsusingafluorescentmicroscope,whichwerecharacterizedby  Figure1.  LovastatinprotectsNMDA-inducedexcitotoxicityinprimaryculturedcorticalneurons.  A ,Representativemicrographsoftrypanblue-positivecells.Neuronswerepretreatedwith500n M LOVfor3dor100  M APV,aspecificNMDARantagonist,for30minorleftuntreatedbeforea15minexposureto100  M NMDAand100  M glycine.Viabilitywasmeasuredbycellcountingaftertrypanbluestainingat24hafterNMDAexposure. B ,RepresentativephotographsandquantitativeassessmentofapoptoticcelldeathbyTUNELassay.Dataarepresentedasthemean  SDfromthree independent experiments.  *p  0.001 versus untreated;  **p  0.001 versus NMDA.  C  , Photomicrographs showing fluorescence staining of MAP-2 with DAPI. Primary cortical ratneurons were pretreated with 500 n M  lovastatin or vehicle for 3 d before exposure to NMDA for 15 min. After fixation, cells were stained with the neuronal markers (MAP-2 and NeuN,green) to delineate the cell morphology, followed by counterstaining with membrane-permeable DAPI (blue). The length of neurites was quantified from 50 of the MAP-2-positive cells,and the number of processes projecting from each positive cell was also quantified. Data are presented in the bar graph as the mean  SD from three independent experiments. *p  0.001 versus untreated;  **p  0.001 versus NMDA. 11228  •  J. Neurosci., September 9, 2009  •  29(36):11226–11236 Ma et al. • Novel Excitoprotective Mechanisms of Statins  condensed nuclei and cell shrinkage. Using microtubule associ-ated protein 2 (MAP2) as a neuron-specific marker, we assessedNMDA-receptor-mediated neurotoxicity in neuron-enrichedcultures.Wefoundthat,comparedwithLDHassay,MAP2stain-ing plus DAPI for nuclei morphology was more sensitive andreliable in evaluating neuronal damage and death. As shown inFigure 1 C  , 30   M  NMDA exposure for 15 min resulted in analmost complete loss of dendritic structures of all the neurons.Chronic pretreatment of neurons with either statin for 3 d pre-vented dendritic loss to a great extent (  60% protection), asquantified by both the number of dendrites per neuron and thelength of the remaining dendrites (Fig. 1 C  ). LovastatinselectivelyprotectsneuronsagainstNMDA-inducedexcitotoxicitybutnotagainstcaspase-mediatedapoptosis—aspecificeffectofcalpaininhibition To test whether statins are universal inhibitors of apoptosis orspecific to NMDA-induced neurotoxicity, we assessed their pro-tective effects on cell death induced by different proapoptoticagents, including STS (200 n M ), a potent protein kinase C inhib-itor, the cell-permeable Bcl-2 homology (GH)-3 domain inhibi-tor BH3I-1, the DNA topomerase inhibitor CPT, or the amyloidpeptide (A  25-35 ). All these agents are known to induce classicapoptotic cascades involving cytochrome  c   release from themitochondria and the subsequent caspase activation. To oursurprise, statins only exhibited inhibitory effect on NMDM-induced cell deaths without effect on caspase-mediated apo-ptosis (Fig. 2  A ).Overactivation of NMDA receptors is known to mediate ex-citotoxicity due to excessive entry of calcium, leading to the acti-vationofseveralcalcium-dependentenzymes.Oneofthemisthefamily of calcium-activated proteases, calpains that appear toplayadominantroleinexcitotoxicneuronaldeath.Wethereforeexamined the effect of statins on NMDA-induced cleavage of spectrin  ,acytoskeletalproteinthatisamajorsubstrateforbothcalpain and caspases, albeit at differential cleavage sites. Indeed,unlike several classic proapoptotic agents tested that inducedcaspase-mediated pathways, NMDA induced primarily calpainactivation, as evidenced by calpain-specific cleavage of spectrinandp35top25conversion(Fig.2 B , C  ).Notably,lovastatintreat-ment remarkably suppressed calpain activation in response toNMDA exposure (Fig. 2 C  , lanes 1–4) but with almost no effecton caspase-3 activation (lanes 5–12).Calpains are also known to alter properties of many othersubstrate proteins by cleavage regulation. To validate statin’s ef-fectoncalpainactivity,weinvestigatedtheproteolytictruncationof three additional substrates that are crucial mediators in neu-ronal cell death. They include p35 (CDK5 coactivator), GSK3,and  -catenin. LovastatinsuppressesCDK5activationinducedby NMDA We examined statin’s effect on preventing the calpain-mediatedcleavage from p35 to p25. Upon NMDA treatment, the cleavagetook place within the first few minutes (Fig. 3  A ) as shown by atime course study. Increases in levels of the truncated products(mostly N terminus) appeared within 5 min of NMDAR activa-tion and were stable for periods   30 min (Fig. 3  A ). Statinpretreatment completely inhibited the conversion from p35 top25 as detected by immunoblot analysis (Fig. 3 B ) and alsoprevented the p25 translocation into the nuclei upon NMDAinsult (Fig. 3 C  ). LovastatinsuppressesNMDA-inducedGSK3  activationandpreserves  -catenin We next examined the consequences of the other two effectormolecules downstream of both PI3K/Akt and Wnt signalingpathways upon statin-mediated neuroprotective action. The bi-ological activities of both GSK3   and   -catenin are classically recognized to be regulated by their specific phosphorylationevents. However, they have recently been found to also be regu-lated by calpain-mediated truncation events, resulting in modu-lated protein activities (Abe and Takeichi, 2007; Gon˜i-Olover etal., 2007). We thus investigated the relationship between thesetwo regulatory mechanisms.Upon treatment of cells with NMDA, a detectable increase incalpain-mediated truncation of GSK3 was found (Fig. 4  A ). Con-currently, we observed a decrease in Ser-9 phosphorylation, anevent which is associated with elevated kinase activity (Fig. 4 B ). Figure2.  Lovastatin selectively protects neurons against NMDA-induced excitotoxicity butnot against caspase-mediated apoptosis.  A , Quantitative assessment of trypan blue-positivecells.Neuronswerepretreatedwith500n M LOVfor3dorleftuntreatedbeforeexposureto100  M  NMDA, 200  M  STS, 50  M  BH3I-1, 10  M  CPT, or 25  M  A  25-35  peptide. Viability wasmeasured by cell counting after trypan blue staining. Trypan blue-positive cells were quanti-tatedasdescribedinMaterialsandMethods.Dataarepresentedasthemean  SDfromthreeindependentexperiments. *p  0.001versusuntreated; **p  0.001versusNMDA. B ,NMDAinduces significant calpain activation compared with other stresses. Neuronal lysates wereprepared1or6hafterexposuretoSTS,CPT,A  25-35 peptide,orNMDA.RepresentativeWesternblot autographs of spectrin and cleaved caspase-3 are shown.  C  , Lovastatin selectively sup-pressescalpainactivationbutnotcaspase-3activation.Neuronswerepretreatedwithorwith-out500n M lovastatinfor3dbeforeanexposuretoNMDA,STS,BH3I-1,CPT,orA  25-35 peptide.Representative Western blots of spectrin, p35/25, and cleaved caspase-3 are shown. Ma et al. • Novel Excitoprotective Mechanisms of Statins J. Neurosci., September 9, 2009  •  29(36):11226–11236  • 11229  Allthesechangeswerereversedbystatintreatment.Noeffectwasfound from statin treatment on the pSer-21 of GSK3    whichinversely correlates with its kinase activity (data not shown). It isnot clear whether this protein truncation event is related to itsSer-9phosphorylation,thoughitwasreportedthatthetruncatedGSK3 is associated with elevated kinase activity (Gon˜i-Olover etal., 2007).WenextexaminedwhetherchangesinGSK3  concurredwithalteration of the levels of    -catenin, a crucial transcriptional ac-tivatorwhoseturnoverandtranslocationtothenucleusisknownto be mostly regulated by GSK3   phosphorylation. We firsttested whether the turnover of    -catenin differed betweenNMDA-treated and statin-rescued neurons. Indeed, statinscouldrescuedegradationof   -catenininducedbyNMDA,result-ing in elevated  -catenin levels (Fig. 5  A ). Moreover, as reported(Hagenetal.,2004),wefoundthatcytosolic  -catenincorrelatedwith its nuclear level, reflective of its transcriptional activity (Fig.5 B , C  ). Most strikingly, we found that   -catenin was regulatedsolely by the calpain-mediated truncation events (Fig. 5 C  ) ratherthan by phosphorylation, which is typically induced by conven-tional apoptotic stimuli (Fig. 5 D ). Statin exerted no effect on thephosphorylated   -catenin, which was selectively induced by staurosporine but not by NMDA. The question remains of whether   -catenin is regulated exclusively by calpain-mediatedtruncation and whether this is dependent on the GSK3 phos-phorylation and/or its truncation mechanism. Lovastatin’sexcitoprotectiveeffectismediatedviainsulinsignaling,regulatedbyPI3K/Aktactivationpathways We then investigated which signaling pathway(s) underlies sta-tin’s excitoprotective mechanism. Since statins are reported toactivate the phosphatidylinositol-3-kinase (PI3K) pathway, re-sulting in the phosphorylation of Akt (Kureishi et al., 2000) inendothelial cells, we tested whether a similar regulatory pathway exists in statin-elicited neuronal protection against excitotoxicinsult.We first tested a panel of pharmacological inhibitors againstPI3K, MARK, Jun/JUNK, and tyrphostin AG538 and found thatonly the PI3K-specific inhibitor LY294002 and AG538 com-pletely abolished statin-mediated excitoprotection (Fig. 6  A ),whereas the other inhibitors (PD98059/MEK1/2, U0126/MEK1/2,SU203580/p38MARK,andSP600125/Jun/JNK)exhibitednosig-nificant effect (data not shown). Consistently, treatment of cellswithstatinsresultedinsustainedandprolongedactivationofAkt, Figure 3.  Lovastatin suppresses CDK5 activation induced by NMDA in primary neurons.  A ,The time course of p35 to p25 conversion after NMDA exposure.  B , Lovastatin pretreatmentcompletely inhibits the NMDA-induced conversion from p35 to p25. The protein level of CDK5wasnotchanged. C  ,Cellswerefixedandimmunostainedforp35/p25(red),followedbycoun-terstaining with DAPI (blue). The p35/p25 immunofluorescent signals in the nucleus werecomparedwiththoseinthecytoplasm,andtheneuronsshowinghigherfluorescencesignalsinthenucleusthaninthesurroundingcytoplasmwerecountedas“nuclearp35/25-positivecells.”Representativepositivecellsareshownbythearrowheads.Dataarepresentedasthemean  SD.  *p  0.001 versus untreated;  **p  0.001 versus NMDA. Figure 4.  Lovastatin suppresses NMDA-induced GSK3 truncation and GSK3  activation inprimaryneurons.  A ,LovastatinpretreatmentattenuatestheGSK3truncationinducedbyNMDAexposure. The representative Western blots and densitometry analysis of GSK3 truncation areshown. Data are presented as the mean  SD.  *p  0.001 versus untreated;  **p  0.001versus NMDA.  B , Lovastatin suppresses GSK3   activation upon NMDA. GSK3   activity wasmeasured by the Ser 9 phosphorylation which inversely correlates with its enzymatic activity.DensitometryanalysisofpGSK3  /GSK3  ratioisshowninthebargraph.Dataarepresentedasthe mean  SD.  *p  0.001 versus untreated;  **p  0.001 versus NMDA. 11230  •  J. Neurosci., September 9, 2009  •  29(36):11226–11236 Ma et al. • Novel Excitoprotective Mechanisms of Statins
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!