@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix skos: . vivo:departmentOrSchool "Non UBC"@en, "Medicine, Faculty of"@en ; edm:dataProvider "DSpace"@en ; ns0:identifierCitation "Molecular Brain. 2012 Apr 26;5(1):13"@en ; dcterms:contributor "University of British Columbia. Djavad Mowafaghian Centre for Brain Health"@en ; ns0:rightsCopyright "Wang et al.; licensee BioMed Central Ltd."@en ; dcterms:creator "Wang, Min"@en, "Li, Shupeng"@en, "Zhang, Hongyu"@en, "Pei, Lin"@en, "Zou, Shengwei"@en, "Lee, Frank J S"@en, "Wang, Yu T"@en, "Liu, Fang"@en ; dcterms:issued "2016-01-11T23:16:30Z"@*, "2012-04-26"@en ; dcterms:description "Over-activation of AMPARs (α−amino-3-hydroxy-5-methylisoxazole-4-propionic acid subtype glutamate receptors) is implicated in excitotoxic neuronal death associated with acute brain insults, such as ischemic stroke. However, the specific molecular mechanism by which AMPARs, especially the calcium-impermeable AMPARs, induce neuronal death remains poorly understood. Here we report the identification of a previously unrecognized molecular pathway involving a direct protein-protein interaction that underlies GluR2-containing AMPAR-mediated excitotoxicity. Agonist stimulation of AMPARs promotes GluR2/GAPDH (glyceraldehyde-3-phosphate dehydrogenase) complex formation and subsequent internalization. Disruption of GluR2/GAPDH interaction by administration of an interfering peptide prevents AMPAR-mediated excitotoxicity and protects against damage induced by oxygen-glucose deprivation (OGD), an in vitro model of brain ischemia."@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/56402?expand=metadata"@en ; skos:note "RESEARCH Open AccessDirect interaction between GluR2 and GAPDHregulates AMPAR-mediated excitotoxicityMin Wang1, Shupeng Li1, Hongyu Zhang1, Lin Pei1, Shengwei Zou1, Frank J S Lee1, Yu Tian Wang2 andFang Liu1,3,4*AbstractOver-activation of AMPARs (α−amino-3-hydroxy-5-methylisoxazole-4-propionic acid subtype glutamate receptors) isimplicated in excitotoxic neuronal death associated with acute brain insults, such as ischemic stroke. However, thespecific molecular mechanism by which AMPARs, especially the calcium-impermeable AMPARs, induce neuronaldeath remains poorly understood. Here we report the identification of a previously unrecognized molecularpathway involving a direct protein-protein interaction that underlies GluR2-containing AMPAR-mediatedexcitotoxicity. Agonist stimulation of AMPARs promotes GluR2/GAPDH (glyceraldehyde-3-phosphatedehydrogenase) complex formation and subsequent internalization. Disruption of GluR2/GAPDH interaction byadministration of an interfering peptide prevents AMPAR-mediated excitotoxicity and protects against damageinduced by oxygen-glucose deprivation (OGD), an in vitro model of brain ischemia.IntroductionGlutamate is the principal excitatory neurotransmitter inthe brain and is involved in numerous physiologicalfunctions including neuronal circuit development, learn-ing and memory [1]. Glutamate-induced neurotoxicity isimplicated in neuropathological disorders such as strokeand epilepsy [2]. The effects of glutamate are mediatedvia two major subfamilies of ligand-gated ion channels:NMDAR (N-methyl-D-aspartate receptor) and AMPAR[3]. AMPAR mediates fast synaptic transmission atexcitatory synapses, while NMDAR is critical in producinga number of different forms of synaptic plasticity [1]. Inneurons, mature AMPA receptors are found as tetramersconsisting of various combinations of GluR1 to GluR4subunits [4], each of which has the same topology: threetransmembrane domains and one membrain re-entrantloop. All subunits are permeable to both Na+ and Ca2+ ionswith the exception of GluR2, which is uniquelyimpermeable to Ca2+. The majority of AMPA receptorsin vivo contain GluR2 subunits whose ion selectivity isdominant over other subunits [5].The accumulation of glutamate, which occurs immedi-ately after ischemia, results in excessive stimulation ofglutamate receptors and leads to neurotoxicity [6,7].NMDAR-mediated neurotoxicity is dependent uponextracellular Ca2+ and is likely mediated by Ca2+ influxdirectly through receptor-gated ion channels [6,7].AMPAR is also tightly associated with a selective patternof neuronal loss in certain brain areas following bothglobal and focal ischemia [8-20]. Similar to what isreported for NMDAR, excitotoxicity mediated byAMPAR lacking the GluR2 subunit is thought to bedependent on ion influx (Ca2+, Zn2+) through AMPARchannels following agonist stimulation [19-21]. However,as most native AMPARs in the hippocampus contain theGluR2 subunit and therefore are likely impermeable toCa2+ [22-26], it is still unclear how activation of theGluR2-containing AMPAR leads to neuronal cell death.Protein-protein interactions with the AMPAR havebeen reported to affect function of AMPAR, amongwhich the best characterized ones, such as GRIP (glutamatereceptor interacting protein), ABP (AMPAR-bindingprotein), SAP97 (synapse-associated protein-97), PICK1(protein interacting with C kinase-1), stargazin, NSF(N-ethylmaleimide-sensitive factor) and AP2 (adaptorprotein-2) [27-34], bind to the intracellular carboxylterminus of AMPAR. They regulate AMPAR functionin a variety of ways, including modulation of AMPAR* Correspondence: f.liu.a@utoronto.ca1Department of Neuroscience, Centre for Addiction and Mental Health,Toronto, Canada3Brain Research Center, University of British Columbia, Vancouver, CanadaFull list of author information is available at the end of the article© 2012 Wang et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative CommonsAttribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in anymedium, provided the original work is properly cited.Wang et al. Molecular Brain 2012, 5:13http://www.molecularbrain.com/content/5/1/13subcellular localization, clustering and/or trafficking.Recent studies have demonstrated that NARP (neuronalactivity-regulated pentraxin) and N-cadherin interact withthe amino terminus (NT) of AMPAR subunits and play animportant role in AMPAR clustering [35] as well asdendritic spine formation [36]. In the present study, wehave identified a new AMPAR-interacting partner, GAPDH.We show that secreted GAPDH binds specifically to theextracellular NT domain of the GluR2 subunit, a processwhich is promoted by AMPAR activation. Disruption ofGluR2/GAPDH interaction prevents AMPAR-mediatedexcitotoxicity and protects against damage in OGD model.ResultsGluR2 subunit directly interacts with GAPDH via itsY142-K172 region of N-terminusTo identify potential proteins that may interact with theNT domain of AMPAR subunits, we used GST-fusionproteins GST-GluR1NT (A19-E538) and GST-GluR2NT(V22-E545) to affinity “pull-down” proteins from solubilizedrat hippocampal tissues along with GST alone as a control.The precipitated proteins were then identified byCoomassie brilliant blue staining following SDS-PAGE.A prominent protein band of ~37 kD was specificallyprecipitated by GST-GluR2NT, but not by GST aloneor GST-GluR1NT (Figure 1A). Mass spectrometryanalysis (LC-MS/MS, Protana [now Transition Thera-peutics]) of this protein band identified three fragmentsthat were homologous to and covered 17% of thesequences within rat GAPDH (VIISAPSADAPMFVMGVNHEK; VIHDNFGIVEGLMTTVHAITATQK; VPTPNVSVVDLTCR). These results suggested that theGluR2 subunit might form a protein complex withGAPDH through its NT domain. We then confirmedthe GluR2/GAPDH interaction with affinity purificationexperiments using GST-GluR2NT, GST-GluR2CT (I833-I883) and GST alone. Subsequent Western blot analysisusing a GAPDH antibody confirmed the associationbetween GAPDH and GluR2NT, but not GluR2CT(Figure 1B).Before conducting further experiments, we examinedwhether GluR2/GAPDH complex exists in vivo. Asshown in Figure 1C, the GluR2 antibody was able to co-immunoprecipitate (Co-IP) GAPDH from solubilizedproteins extracted from rat hippocampal tissues confirm-ing the in vivo association between GluR2 and GAPDH. Inorder to smooth the way for the following functionalstudies, three GluR2NT GST-fusion proteins (GluR2NT1:V22-S271, GluR2NT2: K272-I421, GluR2NT3: L422-E545) wereconstructed (Figure 1D) and utilized in affinity purificationexperiments to delineate the region (s) of GluR2NTinvolved in the interaction with GAPDH. As shown inFigure 1E, GST-GluR2NT1, but not GST-GluR2NT2,GST-GluR2NT3 or GST alone, precipitated GAPDHindicating that the GluR2 subunit interacts with GAPDHthrough its NT region V22-S271. A series of truncations ofthe GluR2NT1 region were then created to map the sitethat interacts with GAPDH (Figure 1D). As shown inFigure 1F and 1G, GST-GluR2NT1-3 (H122-K172) andGST-GluR2NT1-3–2 (Y142-K172) were able to precipitateGAPDH from rat hippocampal tissues.While these results suggested the existence of theGluR2/GAPDH complex, it did not clarify whether thisGluR2/GAPDH complex was formed through either adirect interaction or was mediated indirectly by otheraccessory binding proteins. Therefore we performedin vitro binding assays to examine whether GAPDH andthe GluR2 subunit directly interact with each other. Asshown in Figure 1H, in vitro translated [35 S]-GAPDHprobe bound with GST-GluR2NT1 but not with GST-GluR2NT2, GST-GluR2NT3 or GST alone, indicating thespecificity of the direct protein-protein interactionbetween GAPDH and GluR2NT1. Consistent with theresults from affinity purification experiments, the in vitrotranslated [35 S]-GAPDH probe only hybridized withGST-GluR2NT1-3 and GST-GluR2NT1-3–2, (Figure 1I, J).Together, these data provided in vitro evidence thatGAPDH forms a direct protein-protein interaction withthe GluR2 subunit through the Y142-K172 region of theGluR2NT.Agonist-facilitated GluR2/GAPDH complex formationoccurs extracellularlyAs the NT region of GluR2 locates extracellularly, wethen investigated whether the GluR2/GAPDHinteraction occurs extracellularly by performing cellsurface biotinylation experiments in primary culture ofrat hippocampus, in which cell surface proteins of neuronswere labeled with sulfo-NHS-LC-biotin. As shown inFigure 2A, the GluR2 antibody precipitated GAPDH fromthe biotinylated (B, cell surface) fraction, but failed to pulldown GAPDH from the non-biotinylated (NB, intracellular)fraction, suggesting that the GluR2/GAPDH complexformation occurs extracellularly. Consistent with ourfindings, a previous study demonstrated that GAPDH wasconstitutively secreted into the extracellular space inseveral mammalian cell lines including HEK-293 T cellsand neuro-2a cells [37]. We therefore speculated thatGAPDH might be secreted into the extracellular spaceand form a protein complex with GluR2NT. To test ourhypothesis, we first confirmed GAPDH secretion in ourcell lines by immunoprecipitating GAPDH from theconditioned medium (incubation with neurons/cells for24 hours) of hippocampal primary cultures with aprimary antibody against GAPDH. As shown inFigure 2B, GAPDH was immunoprecipitated fromconditioned medium, but not from fresh medium. Tofurther exclude the possibility that the observedWang et al. Molecular Brain 2012, 5:13 Page 2 of 12http://www.molecularbrain.com/content/5/1/13GAPDH in the conditioned medium resulted from celllysis, conditioned media from non-transfected HEK-293 T cells and from cells expressing GluR1/2 subunitswere collected, concentrated and examined by Westernblot analyses using anti-GAPDH and anti-α-tubulinantibodies. As shown in Figure 2C, regardless ofGluR1/2 subunit expression, GAPDH was detectedfrom both conditioned media and cell lysates, whereasα-tubulin (a cytoplasmic protein marker) was onlydetected from cell lysates, indicating that the GAPDHfound in the conditioned medium is secreted from cellsand is not a contaminant due to cell lysis.Furthermore, we examined the effect of the AMPARactivation on the formation of GluR2/GAPDH complex.By conducting Co-IP experiments, we found thatAMPAR activation with either 100 μM glutamate inHEK-293 T cells expressing GluR1/2 subunits or 100 μMkainic acid (KA) in hippocampal neurons facilitated theBanti-GAPDH35E GFAnti-GAPDH 35IH J[35S]-GAPDH 35GluR2NTV22 E545GluR2NT1V22 S271GluR2NT2K272 I421GluR2NT3L422 E545V2 F7A7 T12H12 K17W17 D221Q22 S27H12 E141Y14 K172GluR2NT1-3-1GluR2NT1-3-2GluR2NT1-1GluR2NT1-2GluR2NT1-3GluR2NT1-4GluR2NT1-5DA35C IPAnti-GAPDHAnti-GluR2351302 122122232 1Figure 1 Identification and characterization of GluR2/GAPDH interaction. A, Coomassie blue stained SDS-PAGE gel of the protein(s)selectively affinity pulled down by GST-GluR2NT, GluR1NT and GST alone from solubilized rat hippocampal lysates. Protein of interest: ~37 kDa.B, Western blot analysis of rat hippocampal proteins affinity purified by GST-GluR2NT, GST-GluR2CT and GST from solubilized rat hippocampallysates and immunoblotted with primary antibody against GAPDH. C, Co-immunoprecipitation of GAPDH by the GluR2 primary antibody fromsolubilized rat hippocampus. D, Schematic representation of GST-fusion proteins encoding truncated GluR2NT segments. E-G, Western blot analysisof rat hippocampal proteins affinity purified by (E) GST-GluR2NT1, GST-GluR2NT2 GST-GluR2NT3 and GST; (F) GST-GluR2NT1-1, GST-GluR2NT1-2, GST-GluR2NT1-3, GST-GluR2NT1-4, GST-GluR2NT1-5 and GST; (G) GST-GluR2NT1-3–1, GST-GluR2NT1-3–2 and GST from solubilized rat hippocampal lysates andimmunoblotted with primary antibody against GAPDH. H-J: Using an in vitro binding assay, [35 S]-GAPDH probe bound with GST-GluR2NT1 (H), GST-GluR2NT1-3 (I) and GST-GluR2NT1-3–2 (J), but not with other GST fusion proteins or GST alone.Wang et al. Molecular Brain 2012, 5:13 Page 3 of 12http://www.molecularbrain.com/content/5/1/13GluR2/GAPDH complex formation by 75 ± 18% and58 ± 11% (mean ± SEM, n = 3), respectively (Figure 2D, E;top panels). In each Co-IP experiment, 500 μg of proteinwere incubated in the presence of primary antibodiesanti-GluR2 or rabbit IgG, and 50 μg of extracted proteinwas used as positive control. The level of directly immu-noprecipitated GluR2 subunit was not significantlyaltered by the agonist stimulation (Figure 2D, E; bottompanels). If the GluR2NT1-3–2 region is essential for GluR2to interact with GAPDH, application of the peptide en-coding GluR2NT1-3–2 would disrupt the GluR2/GAPDHinteraction by competing with GluR2 for GAPDH. Asexpected, pre-incubation of the GluR2NT1-3–2 peptide(10 μM, 1 hour), but not the scrambled GluR2NT1-3–2peptide (GluR2NT1-3-2Scram), significantly inhibited theagonist-induced increase of the GluR2/GAPDH complexformation in transfected HEK-293 T cells (Figure 2D,65 ± 8% decrease; mean ± SE, n = 3) and in hippocampalneurons (Figure 2E, 46 ± 6% decrease; mean ± SE, n = 3).The fact that extracellular application of the interferingGluR2NT1-3–2 peptide was able to disrupt the GluR2/GAPDH interaction further supports the notion that theGluR2/GAPDH complex formation occurs extracellularly.Disruption of GluR2/GAPDH interaction inhibitsAMPAR-mediated excitotoxicityBoth AMPAR and GAPDH have been independently shownto be involved in cell toxicity [38-42]. The observation thatAMPAR activation promoted GluR2/GAPDH complex for-mation suggested that the GluR2/GAPDH interactionmight be involved in AMPAR-mediated excitotoxicity. Be-fore conducting further experiments, we first confirmed theA IP: GluR2Anti-GAPDHAnti-GluR235130C CMAnti-GAPDHAnti-α tubulin 5535BIP: GAPDHWB: anti-GAPDH35DAnti-GAPDHIP:GluR2HEK293T CellsAnti-GluR235130IP:GluR2Anti-GAPDHHippocampal NeuronsAnti-GluR235130EFigure 2 GluR2/GAPDH interaction occurs extracellularly. A, Rat hippocampal neurons were incubated with sulfo-NHS-LC biotin to label cellsurface proteins. GAPDH that co-immunoprecipitated with GluR2 antibody was examined in both non-biotinylated (NB) and biotinylated (B)proteins. B, Using a rabbit anti-GAPDH antibody, GAPDH was immunoprecipitated from the conditioned medium (CM; medium incubated withneurons/cells for 24 hours) of primary cultures of rat hippocampus but not from fresh medium. A mouse GAPDH antibody was used for Westernblotting and rabbit IgG was used as negative control. C, Western blot analysis of GAPDH and α-tubulin in concentrated conditioned medium ofnon-transfected HEK-293 T cells (non-T) and HEK-293 T cells transfected with GluR1/2 subunits (AMPAR), in the presence or absence of glutamate(AMPAR+Glut). Cell lysates were used as controls. D-E: Coimmunoprecipitation of GAPDH by primary antibody against GluR2 subunit (with orwithout glutamate treatment) from HEK-293 T cells expressing GluR1/2 subunits (D) and hippocampal neurons (E) pre-treated with GluR2NT1-3–2 orGluR2NT1-3–2-scram peptides (top panels). Each Coimmunoprecipitation was in parallel with Western blot analysis of the directly immunoprecipitatedproteins (bottom panels). All western blot analysis and co-immunoprecipitation assays in this figure are representative of at least 3 independentexperiments.Wang et al. Molecular Brain 2012, 5:13 Page 4 of 12http://www.molecularbrain.com/content/5/1/13ability of glutamate (300 μM, 24 hour; plus 25 μMcyclothiazide to prevent AMPAR desensitization) to inducecell death in HEK-293 T cells expressing GluR1/2 (Fig-ure 3A), which is consistent with previous studies[43,44]. To investigate the role of the GluR2/GAPDHinteraction in AMPAR-mediated cell death, HEK-293 Tcells expressing GluR1/2 were pre-treated with theGluR2NT1-3–2 peptide (10 μM, 1 hour), which is able todisrupt the GluR2/GAPDH association (confirmed inFigure 2D). As shown in Figure 3B, pre-incubationwith the GluR2NT1-3–2 peptide significantly attenuatedglutamate-induced (300 μM, 500 μM) cell death. TheGluR2NT1-3–2 peptide itself showed no effect in eitherthe absence of glutamate treatment (Figure 3B) or innon-transfected cells regardless of glutamate treat-ment (Figure 3C). The specificity of the GluR2NT1-3–2peptide was also confirmed in HEK-293 T cellsexpressing GluR1/3, GluR1/4 or GluR3/4 subunits,where pre-incubation with the GluR2NT1-3–2 peptidefailed to inhibit AMPAR-mediated cell death(Figure 3D).To study the GluR2/GAPDH interaction in a relevantcellular milieu, rat hippocampal neurons were utilized inparallel experiments. We previously confirmed in Figure 2EC+GluR1:020406080100ControlGlutamateCell Death(Percentage of Total)GluR2NT1-3-2GluR2:- -+--A020406080100120Cell Death(Percentage of Total)***GluR1/2B GlutamateGluR2NT1-3-2+GlutamateControl 300 M 500 M 1mM020406080Cell Death(Percentage of Total)*******####100120DGluR1/3020406080100120GluR1/4 GluR3/4GlutamateGluR2NT1-3-2+glutamateFraction Dead(Relative to Glutamate of GluR1/3)Figure 3 Regulation of the AMPAR-mediated cell death in transfected cells. A, Bar graph summarizing the quantitative measurements of PIfluorescence from HEK-293 T cells expressing GluR1/2 subunits with/without glutamate treatment (300 μM glutamate, 25 μM CTZ, 24 hr).***Significantly different from control group (P< 0.001, n = 9 per group), t-test. B, Bar graph summarizing the quantitative measurements of PIfluorescence from HEK-293 T cells expressing GluR1/2 subunits with/without glutamate treatment at various doses in the presence/absence ofGluR2NT1-3–2 peptide (10 μM, 1 hr). **, *** Significantly different from control group (P< 0.01, 0.001), ANOVA followed by post-hoc SNK test; ##,significant from the corresponding glutamate group (P< 0.01, n = 9 per group), t-test. C, Bar graph summarizing the quantitative measurements of PIfluorescence from non-transfected HEK-293 T cells or HEK-293 T cells expressing GluR1/2 subunits. Cells were pre-treated with the GluR2NT1-3–2 peptidewith/without glutamate treatment (n= 9 per group). D, Bar graph summarizing the quantitative measurements of PI fluorescence from HEK-293 T cellsexpressing GluR1/3, GluR1/4 or GluR3/4 subunits with glutamate treatment in the presence/absence of the GluR2NT1-3–2 peptide (n= 9 per group). AllPI fluorescence measurement assays were performed 3 times independently.Wang et al. Molecular Brain 2012, 5:13 Page 5 of 12http://www.molecularbrain.com/content/5/1/13that pre-incubating hippocampal neurons with theGluR2NT1-3–2 peptide interrupted the GluR2/GAPDHinteraction promoted by the AMPAR activation. Thus, weexamined whether the disruption of this interaction inhippocampal neurons by applying the GluR2NT1-3–2peptide would rescue neurons from AMPAR-mediatedexcitotoxicity. AMPAR-mediated cell death was inducedby treating neurons with KA (100 μM, 1 hour) in thepresence of NMDAR and Ca2+ channel antagonists(10 μM MK-801 and 2 μM nimodipine). As shown inFigure 4A, pretreatment with the GluR2NT1-3–2 peptidesignificantly inhibited AMPAR-mediated cell death.AMPAR-mediated toxicity is often considered acontributing, if not an underlying, causative factor inischemia, which deprives brain cells of glucose andoxygen, causing irreversible brain damage withinminutes. Cells in ischemic brain tissue undergo anumber of changes: they rapidly lose their energysupplies, their membranes become depolarized, calciumloads are increased, reactive oxygen types are producedand excitotoxic effects are found. These biochemicalchanges are followed by irreversible changes to cellularstructures and cell death. The oxygen glucosedeprivation (OGD) cell lesion model represents a validsimulation of the conditions in brain ischemia [45,46].Therefore, we assessed the effectiveness of theGluR2NT1-3–2 peptide to rescue cells from neurotoxicstress in the OGD model to verify the implication of theGluR2/GAPDH interaction in ischemia. As shown inFigure 4B, the GluR2NT1-3–2 peptide pretreatment(10 μM, 1 hour) was able to significantly attenuateOGD-induced cell death (30.4% ± 9.5%) in the presenceof 10 μM MK-801 and 2 μM nimodipine.In order to further confirm the role of GAPDH in theAMPAR-mediated cell death, GAPDH siRNA wastransfected into HEK-293 T cells to block the expressionof GAPDH, but not the expression of GluR2 (Figure 4C).As shown in Figure 4D-E, AMPAR-mediated cell deathwas significantly attenuated in the presence of GAPDHsiRNA. Together, these data suggest that the GluR2/GAPDH interaction may play a critical role in theGluR2-contaning AMPAR-mediated cell death.Activation of AMPAR induces AMPAR/GAPDH complexinternalization through the GluR2/GAPDH interactionPrevious studies demonstrated that agonist stimulationcould induce AMPAR endocytosis [47-49]. Thus, weexamined whether the extracellular GAPDH wouldinternalize along with AMPAR through the GluR2/GAPDHinteraction upon the activation of AMPAR. To quantifyGluR2 and GAPDH cell surface levels in HEK-293 T cellsexpressing GluR1/2, a cell-based ELISA assay was appliedas previously described [49,50]. We first confirmed theresults from previous studies that the glutamate stimulation(100 μM, 30 minutes) induced a significant decrease inplasma membrane GluR2 (Figure 5A). We then testedwhether the cell surface-associated GAPDH is alsodecreased upon agonist stimulation of AMPAR. Asshown in Figure 5B, activation of AMPAR signifi-cantly decreased the cell surface-associated GAPDHin HEK-293 T cells expressing GluR1/2, a phenomenathat can be abolished by the pre-treatment ofGluR2NT1-3–2 peptide. These data, together with theinability of glutamate stimulation to internalize thecell surface-associated GAPDH in the non-transfectedHEK-293 T cells (Figure 5C) or HEK-293 T cellstransfected with GluR1/3 subunits (Figure 5D),suggest that GAPDH internalization may be a passiveprocess enabled by the GluR2/GAPDH interaction.To further investigate whether the observed GAPDHinternalization is dependent on the GluR2 internalization,we tested whether blockade of GluR2 endocytosis willinhibit GAPDH internalization. Previous studies demon-strated that GluR2 endocytosis is dynamin-dependent andthat the expression of the dominant-negative dynaminmutant (K44E) was able to block the GluR2 internalization[47,49]. Thus, after confirming the ability of the K44Emutant to block the GluR2 internalization (Figure 5E), weexamined whether the K44E mutant affected cell surface-associated GAPDH internalization in HEK-293 T cellsexpressing GluR1/2 subunits. As shown in Figure 5F, theK44E mutant significantly inhibited glutamate-induced cellsurface-associated GAPDH internalization, indicating thatGAPDH internalized through a dynamin-dependentpathway and further confirmed that GAPDH was co-internalize with the GluR2 subunit. Moreover, the K44Emutant also attenuated glutamate-induced cell death inHEK-293 T cells expressing GluR1/2 subunits (Figure 5G),indicating that GluR2/GAPDH complex internalizationmay play an important role in the GluR2-containingAMPAR-mediated cell death.DiscussionAMPAR-mediated excitotoxicity has been implicated inthe pathogenesis of neuronal loss associated with anumber of brain disorders, including transient forebrainischemia [8-20]. However, the underlying mechanismsremain unclear. An uncontrollable rise in intracellular Ca2+and Zn2+, with subsequent activation of diverse downstreamcell death signals has been one of the most prominenthypotheses to explain excitotoxic neuronal death[19,20,51-55]. Although GluR2-containing AMPARs arecalcium impermeable, recent studies have suggestedthat selective reductions in the expression of GluR2,resulting in an increase in Ca2+−permeable AMPA recep-tors, have been associated with an increased vulnerability ofneurons to ischemic injury [16,56-61]. Although themechanisms involved are not fully understood, it has beenWang et al. Molecular Brain 2012, 5:13 Page 6 of 12http://www.molecularbrain.com/content/5/1/13suggested that GluR2 internalization may enhance the Ca2+-influx that results in neurotoxicity, either through newlysynthesized Ca2+-permeable AMPARs [57] or by activationof a caspase-dependent apoptotic pathway [62]. Consistentwith previous studies, our data has shown that agoniststimulation of AMPAR results in the internalization ofGluR2 and promotes extracellular GAPDH internalizationvia a GluR2/GAPDH coupling-dependent process. This isthe first evidence showing that the N-terminal of theGluR2 subunit plays an important role in AMPA receptor-mediated excitotoxicity through regulating AMPARtrafficking. Many studies have shown that agonist-induced GluR2 internalization is a dynamin-dependentprocess [47,49]. The observations of our study that mutantdynamin abolishes both GluR2 and GAPDH internalizationand the inability of GAPDH to internalize in cells lackingGluR2 suggest that GAPDH internalization is a passiveprocess facilitated by the GluR2/GAPDH interaction andmediated by GluR2 internalization.Given the fact that GAPDH interacts with the extracellu-lar NT of GluR2, it is likely that the GluR2/GAPDH proteincomplex may be in an endocytosed vesicle following theagonist-induced internalization. On this basis it would belogical to further ask how the GluR2/GAPDH complex getsDCellDeath020406080100(Percentage of Total)NeuronEA020406080100*GluR1/2GAPDH siRNA ++-CellDeath(Percentage of Total)BC0255075100OGDCellDeath(Percentage of Total)GluR1/GluR2+GAPDH siRNA Propidium iodideStaining of HEK 293T CellsGluR1/GluR2Anti-GAPDHHEK293TCellsGluR1/2GAPDH siRNA +-Anti-GluR2+++Figure 4 Regulation of the AMPAR-mediated cell death in cultured neurons and OGD model. A, Bar graph summarizing the quantitativemeasurements of PI fluorescence from rat hippocampal primary culture with KA treatment (100 μM, 1 hr) in the presence/absence of theGluR2NT1-3–2 peptide. ***Significantly different from KA group (P< 0.001, n = 9 per group), t-test. B, Bar graph summarizing the quantitativemeasurements of PI fluorescence from rat hippocampal primary culture with OGD protocol in the presence/absence of the GluR2NT1-3–2 peptide.*Significantly different from OGD group (P< 0.05, n = 9 per group), t-test. C, Western blot analysis of GAPDH (upper panel) and GluR2 (lowerpanel) expression in HEK 293 T cells expressing GluR1/2 subunits in the presence/absence of the GAPDH siRNA. D, Bar graph summarizing thequantitative measurements of PI fluorescence from HEK-293 T cells expressing GluR1/2 subunits with glutamate treatment in the presence/absence of the GAPDH siRNA (P< 0.05, n = 9 per group), t-test. E, Propidim iodide positive cells (red) from HEK-293 T expressing GluR1/2 subunitswith glutamate treatment in the presence/absence of the GAPDH siRNA, scale bar 50 μm. All PI fluorescence measurement assays wereperformed 3 times independently.Wang et al. Molecular Brain 2012, 5:13 Page 7 of 12http://www.molecularbrain.com/content/5/1/13out of the vesicle and promotes excitotoxic neuronal death.There are many possibilities for this question. First, thecomplex may be transported to the nucleus via a retrogradevesicle transport mechanism leading to the fusion of thevesicle with ER or nuclear membranes or via mechanismsrecently proposed for the nuclear translocation of anotherplasma membrane receptor, the EGF receptor [63,64].Second, the GluR2/GAPDH complex formation in theA020406080100120Cell Surface GluR2(Percent Control)B DC020406080100120Cell Surface-associated GAPDH(Percent Control)020406080100120Cell Surface-associated GAPDH(Percent Control)GluR1/3020406080100120Cell Surface-associated GAPDH(Percent Control)Non-transfectedHEK-293TEWT K44E020406080100120ControlGlutamateDynaminCell Surface GluR2(Percent WT Control)FWT K44E020406080100120ControlGlutamateDynaminCell Surface-associated GAPDH(Percent WT Control)WT K44E020406080100##Cell Death(Percentage of Total)ControlGlutamateGDynaminFigure 5 Activation of AMPAR induces GluR2/GAPDH co-internalization. A, Quantification of GluR2 expression at the plasma membranewith/without glutamate treatment (100 μM, 30 minutes) in HEK-293 T cells expressing GluR1/2 subunits. *Significantly different from control group(P< 0.05, n = 9 per group), t-test. B, Quantification of cell surface-associated GAPDH with/without glutamate treatment in the presence/absence ofthe GluR2NT1-3–2 peptide in HEK-293 T cells expressing GluR1/2 subunits. *Significantly different from control group; #, significantly different fromglutamate group (P< 0.05, n = 9 per group), ANOVA followed by post-hoc SNK test. C, Quantification of cell surface-associated GAPDH innon-transfected HEK-293 T cells with/without glutamate treatment (n = 9 per group). D, Quantification of cell surface-associated GAPDHwith/without glutamate treatment in HEK-293 T cells expressing GluR1/3 subunits (n = 9 per group). Quantification of plasma membrane GluR2 (E)or cell surface-associated GAPDH (F) expression at the plasma membrane with/without glutamate treatment in HEK-293 T cells expressingGluR1/2 subunits with wild type dynamin (WT) or mutant K44E dynamin (K44E). *Significantly different from the corresponding control group(P< 0.05, n = 9 per group), t-test. G, Bar graph summarizing the quantitative measurements of PI fluorescence from HEK-293 T cells expressingGluR1/2 subunits with wild type dynamin (WT) or mutant K44E dynamin (K44E) with/without glutamate treatment. ***Significantly different fromcontrol WT group (P< 0.001, n = 9 per group); ##significantly different from control K44E group (P< 0.01, n = 9 per group), t-test. All assays inthis figure were performed 3 times independently.Wang et al. Molecular Brain 2012, 5:13 Page 8 of 12http://www.molecularbrain.com/content/5/1/13vesicle may lead to the activation of lysosome in thevesicle that breaks the vesicle and release the GluR2/GAPDH into the cytoplasm.The possible mechanisms that underlie this GluR2/GAPDH related cell death is particularly interesting. It issomewhat surprising to find that the AMPAR-mediatedcell death involves GAPDH, a key enzyme involved inglycolysis with a ubiquitous intracellular distribution.However, additional roles for GAPDH have beendiscovered recently, including membrane fusion/transport,binding to low molecular weight G proteins, regulation ofthe cytoskeleton, accumulation of glutamate intopresynaptic vesicles, and apoptosis [65-71]. Recent studieshave shown that GAPDH binds to Siah1 and triggersapoptosis [39]. Moreover GAPDH has also been reportedto interact with p53 [72], a tumor suppressor andtranscription factor that has been implicated in glutamate-mediated excitotoxicity [73-75]. Numerous evidence showthat activation of p53 can trigger apoptosis (for reviews,see [76]) under conditions of cellular stress mediated byphosphorylation or acetylation of p53 [77]. Whether Siah1,p53 or other molecules are involved in GluR2/GAPDH-related cell death pathway requires much more additionalwork for a better understanding of the detailed molecularmechanisms.Stroke is the second leading cause of death worldwideyet there are very few effective pharmacologicaltreatments for patients suffering ischemic stroke.Thrombolytics such as alteplase and tenecteplase havebeen a significant advance in the treatment of ischemicstroke. However, thrombolytics must be given soon aftera stroke to be effective (within 3 hours of ischemicepisode). This short time frame has limited their use inmany situations. There continues to be a significantunmet need for acute pharmacological treatmentsbeyond thrombolytics. Advances in recent years includehypothermia [78-80], oxygen therapy [81], stem celltransplantation [82] and cerebral plasticity stimulation(trophic factor) strategies [83]. These novel techniquesare intriguing, but will require further well-designedprospective trials to assess clinical feasibility, safety, andefficacy [84]. Another approach that has received consider-able attention is agents that inhibit ischemia-inducedexcitotoxicity though directly blocking glutamate receptors.However, all have failed at various stages of developmentfor a variety of reasons. One of the main drawbacks of theglutamate receptor antagonists is that they block normalexcitatory neurotransmission necessary for maintainingbasic brain functions. For this reason, much research hasbeen directed at identifying drugs and peptides that may beable to selectively target protein-protein interactions thathave more narrow function than a certain neurotransmitterreceptor. In the present study, we have shown that adminis-tration of the interfering GluR2NT1-3–2 peptide to interruptthe GluR2/GAPDH interaction significantly mitigatesneuronal cell death in a cell model of ischemia, revealing apreviously unappreciated signaling pathway underlyingAMPAR-mediated excitotoxicity and it may provide a newavenue for the development of a complementarytherapeutics in the treatment of neuropathologicaldisorders, such as stroke and epilepsy.Materials and methodsCell culture and transient transfectionHEK293 T cells were cultured in α-MEM (Invitrogen,Carlsbad, CA) supplemented with 10% fetal bovine serum(Invitrogen) and maintained in incubators at 37°C, 5%CO2. HEK293T cells were transiently transfected withplasmid constructs and/or siRNA using lipofectamine2000 reagents (Invitrogen). Cells were harvested 48 hourspost transfection.Primary hippocampal neuron culture and OGD treatmentPrimary cultures from hippocampus were prepared fromfetal Wistar rats (embryonic day 17–19) on Cell + (Sarstedt)culture dishes as previously described [85-87]. The cultureswere used for experiments on 12–15 days after plating.Hippocampal cultures were pretreated GluR2NT1-3–2peptides prior to kainic acid treatment. OGD treatment wasperformed in the presence of MK-801 and nimodipine aspreviously described [57].GST fusion proteinsTo construct GST-fusion proteins encoding truncatedGluR2 and GAPDH, cDNA fragments were amplified byusing PCR method with specific primers. Except wherespecified, all 5′ and 3′ oligonucleotides incorporatedBamH1 site (GGATCC) and Xho1 sites (CTCGAG),respectively, to facilitate subcloning into vector pGEX-4T3(for GST-fusion protein construction). GST-fusion proteinswere prepared from bacterial lysates with GlutathioneSepharose 4B beads as described by the manufacturer(Amersham). To confirm appropriate splice fusion and theabsence of spurious PCR generated nucleotide errors, allconstructs were resequenced.Protein affinity purification, in vitro binding,co-immunoprecipitation and western blotProtein affinity purification, in vitro binding, co-immuno-precipitation and Western blot analyses were performed aspreviously described [85-87]. Antibodies used for immuno-precipitation, Western blots and cell surface ELISA assaysinclude GAPDH (polyclonal from Abcam, monoclonalfrom Chemicon), GluR2 (Western blots: Chemicon;immunoprecipitation: Upstate), and α-tubulin (monoclonal,Sigma-Aldrich).Wang et al. Molecular Brain 2012, 5:13 Page 9 of 12http://www.molecularbrain.com/content/5/1/13Cell-ELISA assaysHEK-293 T cells transfected with plasmid constructs weretreated with 100 μM glutamate or extracellular solution(ECS) before fixing in 4% (W/V) paraformaldehyde for 10minutes in the absence (non-permeabilized conditions) orpresence (permeabilized conditions) of 1% (V/V) TritonX-100. Cells were incubated in 1% (W/V) glycine for 10minutes at 4°C to recover from the fixing. Cells werethen incubated with specific primary antibodies for thepurpose of labeling the receptors or proteins on the cellsurface under non-permeabilized conditions or theentire receptor pool under permeabilized conditions.After incubation with corresponding HRP-conjugatedsecondary antibodies (Sigma-Aldrich), the HRP substrateo-phenylenediamine (Sigma-Aldrich) was added to pro-duce a color reaction that was stopped with the equal vol-ume of 3 N HCl. Fluorescence intensity in each well wasmeasured with a plate reader (Victor3; PerkinElmer). Thecell surface expression of HA-GluR2 after pre-treatmentwith glutamate was presented as the ratio of colorimetricreadings under non-permeabilized conditions to thoseunder permeabilized conditions, and then normalized totheir respective control groups (pretreated with ECS).Afterwards, cells were scrapped from the dishes, and theprotein concentration of each dish was measured. Theresults of cell surface expression of receptors or proteinswere calibrated by the protein concentration of each well.Analysis was done using at least 9 separate wells in eachgroup. Cell ELISA using primary hippocampal neuronswas performed identically with assays using HEK-293 Tcells, with the exception that the anti-GluR2 antibody(MAB397; Chemicon) was used as primary antibody in-stead of anti-HA.Quantification of AMPAR-mediated excitotoxicityHEK-293 T cells transfected with GluR1/2 subunits wereexposed to 300 μM glutamate/25 μM cyclothiazide at 37°Cfor 24 hour. Cells were allowed to recover for 24 hours at37°C. To quantify AMPAR-mediated cell death, culturemedium was replaced by extracellular solution containing50 μg/ml of propidium iodide (PI) (Invitrogen, Carlsbad,CA). After 30 minutes incubation at 37°C, fluorescenceintensity in each well was measured with a plate reader(Victor3; PerkinElmer, Waltham, MA). The fraction ofdead cells was normalized to the total cell number.Primary hippocampal neurons were exposed to 100 μMKA/25 μM cyclothiazide in the presence of NMDAR andCa2+ channel antagonists (10 μM MK-801 and 2 μMnimodipine, respectively) at 37°C for 1 hour.Cell biotinylationFor cell surface biotinylation, cells were rinsed four timeswith ice-cold PBS2+ (PBS containing 0.1 mM CaCl2 and1.0 mM MgCl2) after treatment, and incubated twicewith 1.0 mg/ml sulfo-NHS-LC-biotin (Pierce, Rockford,IL) for 20 minutes at 4°C. Non-reactive biotin wasquenched by 20 minutes incubation at 4°C in ice-coldPBS2+ and 0.1 M glycine. Cells were solubilized in RIPAbuffer (10 mM Tris, ph7.4, 150 mM NaCl, 1.0 mMEDTA, 0.1% (W/V) SDS, 1.0% (V/V) Triton X-100 and1.0% (V/V) Sodium deoxycholate) containing proteaseinhibitors (1.0 mM PMSF and 1.0 μg/ml proteasecocktail). Biotinylated and non-biotinylated proteins wereseparated from equal amounts of cellular protein byincubation with 50 μl of 50% slurry of immobilizedstreptavidin-conjugated beads (Pierce, Rockford, IL)overnight with constant mixing at 4°C. Unboundproteins (supernatant) were saved for later co-immuno-precipitation experiment. Proteins bound to streptavidinbeads were eluted in biotin elution buffer. Biotinylatedand non-biotinylated samples were applied to protein A/GPLUS-agarose (Santa Cruz Biotechnology, Santa Cruz, CA)for co-immunoprecipitation.Competing InterestThe authors declare that they have no competing interests.AcknowledgmentsWe gratefully acknowledge Mr. Brian Vukusic for technical assistance. Wethank Dr. Paul Talyor for help with the MS experiments and analysis. Wethank Dr. Qi Wan for providing the OGD cell model. F.L. is a recipient ofCareer Investigator Award of the Heart and Stroke Foundation of Canada. S.P.L. is a recipient of post-doctoral fellowship of the Heart and StrokeFoundation of Canada. The work is supported by operating grant from Heartand Stroke Foundation of Canada (F. L.).Author details1Department of Neuroscience, Centre for Addiction and Mental Health,Toronto, Canada. 2Brain Research Center, University of British Columbia,Vancouver, Canada. 3Department of Psychiatry, University of Toronto,Toronto, Canada. 4Department of Neuroscience, Centre for Addiction andMental Health, Clarke Division, 250 College Street, Toronto, ON M5T 1R8,Canada.Authors’ contributionsMW carried out all experiments, with the help of SL for constructing GST-fusion proteins, HZ for AMAPR-mediated excitotoxicity assays, LP for the co-immunoprecipitation and SZ for the GST-pull down assays. FJK and YTWhelped to edit the manuscript. FL supervised the study and wrote themanuscript. All authors read and approved the final manuscript.Received: 10 March 2012 Accepted: 26 April 2012Published: 26 April 2012References1. Bliss TV, Collingridge GL: A synaptic model of memory: long-termpotentiation in the hippocampus. Nature 1993, 361:31–39.2. Choi DW: Glutamate neurotoxicity and diseases of the nervous system.Neuron 1988, 1:623–634.3. Hollmann M, Heinemann S: Cloned glutamate receptors. Annu Rev Neurosci1994, 17:31–108.4. Keinanen K, Wisden W, Sommer B, Werner P, Herb A, Verdoorn TA, SakmannB, Seeburg PH: A family of AMPA-selective glutamate receptors. Science1990, 249:556–560.5. Boulter J, Hollmann M, O’Shea-Greenfield A, Hartley M, Deneris E, Maron C,Heinemann S: Molecular cloning and functional expression of glutamatereceptor subunit genes. Science 1990, 249:1033–1037.6. Choi DW: Calcium: still center-stage in hypoxic-ischemic neuronal death.Trends Neurosci 1995, 18:58–60.Wang et al. Molecular Brain 2012, 5:13 Page 10 of 12http://www.molecularbrain.com/content/5/1/137. Simon RP, Swan JH, Griffiths T, Meldrum BS: Blockade of N-methyl-D-aspartate receptors may protect against ischemic damage in the brain.Science 1984, 226:850–852.8. Dos-Anjos S, Martinez-Villayandre B, Montori S, Regueiro-Purrinos MM,Gonzalo-Orden JM, Fernandez-Lopez A: Global ischemia-inducedmodifications in the expression of AMPA receptors and inflammation inrat brain. Brain Res 2009, 1287:20–27.9. Elger B, Gieseler M, Schmuecker O, Schumann I, Seltz A, Huth A: Extendedtherapeutic time window after focal cerebral ischemia by non-competitive inhibition of AMPA receptors. Brain Res 2006, 1085:189–194.10. Gill R, Lodge D: Pharmacology of AMPA antagonists and their role inneuroprotection. Int Rev Neurobiol 1997, 40:197–232.11. Gressens P, Spedding M, Gigler G, Kertesz S, Villa P, Medja F, Williamson T, Kapus G,Levay G, Szenasi G, et al: The effects of AMPA receptor antagonists in models ofstroke and neurodegeneration. Eur J Pharmacol 2005, 519:58–67.12. Matucz E, Moricz K, Gigler G, Simo A, Barkoczy J, Levay G, Harsing LG Jr,Szenasi G: Reduction of cerebral infarct size by non-competitive AMPAantagonists in rats subjected to permanent and transient focal ischemia.Brain Res 2004, 1019:210–216.13. McCarran WJ, Goldberg MP: White matter axon vulnerability to AMPA/kainate receptor-mediated ischemic injury is developmentally regulated.J Neurosci 2007, 27:4220–4229.14. Oguro K, Oguro N, Kojima T, Grooms SY, Calderone A, Zheng X, Bennett MV,Zukin RS: Knockdown of AMPA receptor GluR2 expression causes delayedneurodegeneration and increases damage by sublethal ischemia inhippocampal CA1 and CA3 neurons. J Neurosci 1999, 19:9218–9227.15. Pellegrini-Giampietro DE, Zukin RS, Bennett MV, Cho S, Pulsinelli WA: Switchin glutamate receptor subunit gene expression in CA1 subfield ofhippocampus following global ischemia in rats. Proc Natl Acad Sci USA1992, 89:10499–10503.16. Peng PL, Zhong X, Tu W, Soundarapandian MM, Molner P, Zhu D, Lau L, Liu S,Liu F, Lu Y: ADAR2-dependent RNA editing of AMPA receptor subunit GluR2determines vulnerability of neurons in forebrain ischemia. Neuron 2006,49:719–733.17. Pulsinelli WA, Levy DE, Duffy TE: Regional cerebral blood flow and glucosemetabolism following transient forebrain ischemia. Ann Neurol 1982,11:499–502.18. Schmidt-Kastner R, Freund TF: Selective vulnerability of the hippocampusin brain ischemia. Neuroscience 1991, 40:599–636.19. Weiss JH, Sensi SL: Ca2+−Zn2+ permeable AMPA or kainate receptors:possible key factors in selective neurodegeneration. Trends Neurosci 2000,23:365–371.20. Yin HZ, Sensi SL, Ogoshi F, Weiss JH: Blockade of Ca2 +−permeableAMPA/kainate channels decreases oxygen-glucose deprivation-inducedZn2+ accumulation and neuronal loss in hippocampal pyramidalneurons. J Neurosci 2002, 22:1273–1279.21. Tanaka H, Grooms SY, Bennett MV, Zukin RS: The AMPAR subunit GluR2:still front and center-stage. Brain Res 2000, 886:190–207.22. Geiger JR, Melcher T, Koh DS, Sakmann B, Seeburg PH, Jonas P, Monyer H:Relative abundance of subunit mRNAs determines gating and Ca2+permeability of AMPA receptors in principal neurons and interneurons inrat CNS. Neuron 1995, 15:193–204.23. Greger IH, Khatri L, Ziff EB: RNA editing at arg607 controls AMPA receptorexit from the endoplasmic reticulum. Neuron 2002, 34:759–772.24. Jonas P, Racca C, Sakmann B, Seeburg PH, Monyer H: Differences in Ca2+permeability of AMPA-type glutamate receptor channels in neocorticalneurons caused by differential GluR-B subunit expression. Neuron 1994,12:1281–1289.25. Lu W, Shi Y, Jackson AC, Bjorgan K, During MJ, Sprengel R, Seeburg PH, Nicoll RA:Subunit composition of synaptic AMPA receptors revealed by a single-cellgenetic approach. Neuron 2009, 62:254–268.26. Wenthold RJ, Petralia RS, Blahos J II, Niedzielski AS: Evidence for multiple AMPAreceptor complexes in hippocampal CA1/CA2 neurons. J Neurosci 1996,16:1982–1989.27. Dong H, O’Brien RJ, Fung ET, Lanahan AA, Worley PF, Huganir RL: GRIP: asynaptic PDZ domain-containing protein that interacts with AMPA receptors.Nature 1997, 386:279–284.28. Dong H, Zhang P, Song I, Petralia RS, Liao D, Huganir RL: Characterization of theglutamate receptor-interacting proteins GRIP1 and GRIP2. J Neurosci 1999,19:6930–6941.29. Lee SH, Liu L, Wang YT, Sheng M: Clathrin adaptor AP2 and NSF interactwith overlapping sites of GluR2 and play distinct roles in AMPA receptortrafficking and hippocampal LTD. Neuron 2002, 36:661–674.30. Nishimune A, Isaac JT, Molnar E, Noel J, Nash SR, Tagaya M, Collingridge GL,Nakanishi S, Henley JM: NSF binding to GluR2 regulates synaptictransmission. Neuron 1998, 21:87–97.31. Osten P, Srivastava S, Inman GJ, Vilim FS, Khatri L, Lee LM, States BA,Einheber S, Milner TA, Hanson PI, Ziff EB: The AMPA receptor GluR2 Cterminus can mediate a reversible, ATP-dependent interaction with NSFand alpha- and beta-SNAPs. Neuron 1998, 21:99–110.32. Srivastava S, Osten P, Vilim FS, Khatri L, Inman G, States B, Daly C, DeSouza S,Abagyan R, Valtschanoff JG, et al: Novel anchorage of GluR2/3 to thepostsynaptic density by the AMPA receptor-binding protein ABP. Neuron1998, 21:581–591.33. Wyszynski M, Valtschanoff JG, Naisbitt S, Dunah AW, Kim E, Standaert DG,Weinberg R, Sheng M: Association of AMPA receptors with a subset ofglutamate receptor-interacting protein in vivo. J Neurosci 1999, 19:6528–6537.34. Xia J, Zhang X, Staudinger J, Huganir RL: Clustering of AMPA receptors by thesynaptic PDZ domain-containing protein PICK1. Neuron 1999, 22:179–187.35. O’Brien RJ, Xu D, Petralia RS, Steward O, Huganir RL, Worley P: Synapticclustering of AMPA receptors by the extracellular immediate-early geneproduct Narp. Neuron 1999, 23:309–323.36. Saglietti L, Dequidt C, Kamieniarz K, Rousset MC, Valnegri P, Thoumine O, Beretta F,Fagni L, Choquet D, Sala C, et al: Extracellular interactions between GluR2 andN-cadherin in spine regulation. Neuron 2007, 54:461–477.37. Yamaji R, Chatani E, Harada N, Sugimoto K, Inui H, Nakano Y:Glyceraldehyde-3-phosphate dehydrogenase in the extracellular spaceinhibits cell spreading. Biochim Biophys Acta 2005, 1726:261–271.38. Chuang DM, Hough C, Senatorov VV: Glyceraldehyde-3-phosphatedehydrogenase, apoptosis, and neurodegenerative diseases. Annu RevPharmacol Toxicol 2005, 45:269–290.39. Hara MR, Agrawal N, Kim SF, Cascio MB, Fujimuro M, Ozeki Y, Takahashi M,Cheah JH, Tankou SK, Hester LD, et al: S-nitrosylated GAPDH initiatesapoptotic cell death by nuclear translocation following Siah1 binding.Nat Cell Biol 2005, 7:665–674.40. Ishitani R, Chuang DM: Glyceraldehyde-3-phosphate dehydrogenaseantisense oligodeoxynucleotides protect against cytosinearabinonucleoside-induced apoptosis in cultured cerebellar neurons. ProcNatl Acad Sci USA 1996, 93:9937–9941.41. Ishitani R, Tanaka M, Sunaga K, Katsube N, Chuang DM: Nuclear localizationof overexpressed glyceraldehyde-3-phosphate dehydrogenase incultured cerebellar neurons undergoing apoptosis. Mol Pharmacol 1998,53:701–707.42. Sawa A, Khan AA, Hester LD, Snyder SH: Glyceraldehyde-3-phosphatedehydrogenase: nuclear translocation participates in neuronal andnonneuronal cell death. Proc Natl Acad Sci USA 1997, 94:11669–11674.43. Iihara K, Joo DT, Henderson J, Sattler R, Taverna FA, Lourensen S, Orser BA,Roder JC, Tymianski M: The influence of glutamate receptor 2 expressionon excitotoxicity in Glur2 null mutant mice. J Neurosci 2001, 21:2224–2239.44. Zeevalk GD, Schoepp D, Nicklas WJ: Excitotoxicity at both NMDA and non-NMDA glutamate receptors is antagonized by aurintricarboxylic acid:evidence for differing mechanisms of action. J Neurochem 1995, 64:1749–1758.45. Goldberg MP, Choi DW: Combined oxygen and glucose deprivation incortical cell culture: calcium-dependent and calcium-independentmechanisms of neuronal injury. J Neurosci 1993, 13:3510–3524.46. Ying HS, Weishaupt JH, Grabb M, Canzoniero LM, Sensi SL, Sheline CT,Monyer H, Choi DW: Sublethal oxygen-glucose deprivation altershippocampal neuronal AMPA receptor expression and vulnerability tokainate-induced death. J Neurosci 1997, 17:9536–9544.47. Carroll RC, Beattie EC, Xia H, Luscher C, Altschuler Y, Nicoll RA, Malenka RC,von Zastrow M: Dynamin-dependent endocytosis of ionotropic glutamatereceptors. Proc Natl Acad Sci USA 1999, 96:14112–14117.48. Lin JW, Ju W, Foster K, Lee SH, Ahmadian G, Wyszynski M, Wang YT, ShengM: Distinct molecular mechanisms and divergent endocytotic pathwaysof AMPA receptor internalization. Nat Neurosci 2000, 3:1282–1290.49. Man HY, Lin JW, Ju WH, Ahmadian G, Liu L, Becker LE, Sheng M, WangYT: Regulation of AMPA receptor-mediated synaptic transmission byclathrin-dependent receptor internalization. Neuron 2000, 25:649–662.50. Pei L, Lee FJ, Moszczynska A, Vukusic B, Liu F: Regulation of dopamine D1receptor function by physical interaction with the NMDA receptors. JNeurosci 2004, 24:1149–1158.Wang et al. Molecular Brain 2012, 5:13 Page 11 of 12http://www.molecularbrain.com/content/5/1/1351. Sensi SL, Yin HZ, Carriedo SG, Rao SS, Weiss JH: Preferential Zn2+ influxthrough Ca2+−permeable AMPA/kainate channels triggers prolongedmitochondrial superoxide production. Proc Natl Acad Sci USA 1999,96:2414–2419.52. Sensi SL, Ton-That D, Sullivan PG, Jonas EA, Gee KR, Kaczmarek LK, Weiss JH:Modulation of mitochondrial function by endogenous Zn2+ pools. ProcNatl Acad Sci USA 2003, 100:6157–6162.53. Medvedeva YV, Lin B, Shuttleworth CW, Weiss JH: Intracellular Zn2+accumulation contributes to synaptic failure, mitochondrialdepolarization, and cell death in an acute slice oxygen-glucosedeprivation model of ischemia. J Neurosci 2009, 29:1105–1114.54. Jiang D, Sullivan PG, Sensi SL, Steward O, Weiss JH: Zn(2+) inducespermeability transition pore opening and release of pro-apoptoticpeptides from neuronal mitochondria. J Biol Chem 2001, 276:47524–47529.55. Bossy-Wetzel E, Talantova MV, Lee WD, Scholzke MN, Harrop A, Mathews E,Gotz T, Han J, Ellisman MH, Perkins GA, Lipton SA: Crosstalk between nitricoxide and zinc pathways to neuronal cell death involving mitochondrialdysfunction and p38-activated K + channels. Neuron 2004, 41:351–365.56. Kwak S, Weiss JH: Calcium-permeable AMPA channels inneurodegenerative disease and ischemia. Curr Opin Neurobiol 2006,16:281–287.57. Liu B, Liao M, Mielke JG, Ning K, Chen Y, Li L, El-Hayek YH, Gomez E, ZukinRS, Fehlings MG, Wan Q: Ischemic insults direct glutamate receptorsubunit 2-lacking AMPA receptors to synaptic sites. J Neurosci 2006,26:5309–5319.58. Soundarapandian MM, Tu WH, Peng PL, Zervos AS, Lu Y: AMPA receptorsubunit GluR2 gates injurious signals in ischemic stroke. Mol Neurobiol2005, 32:145–155.59. Spaethling JM, Klein DM, Singh P, Meaney DF: Calcium-permeable AMPAreceptors appear in cortical neurons after traumatic mechanical injuryand contribute to neuronal fate. J Neurotrauma 2008, 25:1207–1216.60. Anzai T, Tsuzuki K, Yamada N, Hayashi T, Iwakuma M, Inada K, Kameyama K,Hoka S, Saji M: Overexpression of Ca2 +−permeable AMPA receptorpromotes delayed cell death of hippocampal CA1 neurons followingtransient forebrain ischemia. Neurosci Res 2003, 46:41–51.61. Liu S, Lau L, Wei J, Zhu D, Zou S, Sun HS, Fu Y, Liu F, Lu Y: Expression of Ca(2+)-permeable AMPA receptor channels primes cell death in transientforebrain ischemia. Neuron 2004, 43:43–55.62. Li SY, Ni JH, Xu DS, Jia HT: Down-regulation of GluR2 is associated withCa2+−dependent protease activities in kainate-induced apoptotic celldeath in cultured [correction of culturd] rat hippocampal neurons.Neurosci Lett 2003, 352:105–108.63. Dittmann K, Mayer C, Kehlbach R, Rodemann HP: Radiation-inducedcaveolin-1 associated EGFR internalization is linked with nuclear EGFRtransport and activation of DNA-PK. Mol Cancer 2008, 7:69.64. Liao HJ, Carpenter G: Role of the Sec61 translocon in EGF receptortrafficking to the nucleus and gene expression. Mol Biol Cell 2007,18:1064–1072.65. Doucet JP, Tuana BS: Identification of low molecular weight GTP-bindingproteins and their sites of interaction in subcellular fractions fromskeletal muscle. J Biol Chem 1991, 266:17613–17620.66. Fuchtbauer A, Jockusch BM, Leberer E, Pette D: Actin-severing activitycopurifies with phosphofructokinase. Proc Natl Acad Sci USA 1986,83:9502–9506.67. Glaser PE, Gross RW: Rapid plasmenylethanolamine-selective fusion ofmembrane bilayers catalyzed by an isoform of glyceraldehyde-3-phosphate dehydrogenase: discrimination between glycolytic andfusogenic roles of individual isoforms. Biochemistry 1995, 34:12193–12203.68. Huitorel P, Pantaloni D: Bundling of microtubules by glyceraldehyde-3-phosphate dehydrogenase and its modulation by ATP. Eur J Biochem1985, 150:265–269.69. Ikemoto A, Bole DG, Ueda T: Glycolysis and glutamate accumulation intosynaptic vesicles. Role of glyceraldehyde phosphate dehydrogenase and3-phosphoglycerate kinase. J Biol Chem 2003, 278:5929–5940.70. Robbins AR, Ward RD, Oliver C: A mutation in glyceraldehyde 3-phosphatedehydrogenase alters endocytosis in CHO cells. J Cell Biol 1995, 130:1093–1104.71. Tisdale EJ: Glyceraldehyde-3-phosphate dehydrogenase is required forvesicular transport in the early secretory pathway. J Biol Chem 2001,276:2480–2486.72. Chen RW, Saunders PA, Wei H, Li Z, Seth P, Chuang DM: Involvement ofglyceraldehyde-3-phosphate dehydrogenase (GAPDH) and p53 inneuronal apoptosis: evidence that GAPDH is upregulated by p53.J Neurosci 1999, 19:9654–9662.73. Lakkaraju A, Dubinsky JM, Low WC, Rahman YE: Neurons are protectedfrom excitotoxic death by p53 antisense oligonucleotides delivered inanionic liposomes. J Biol Chem 2001, 276:32000–32007.74. Sakhi S, Bruce A, Sun N, Tocco G, Baudry M, Schreiber SS: p53 induction isassociated with neuronal damage in the central nervous system. ProcNatl Acad Sci USA 1994, 91:7525–7529.75. Uberti D, Belloni M, Grilli M, Spano P, Memo M: Induction of tumour-suppressor phosphoprotein p53 in the apoptosis of cultured ratcerebellar neurones triggered by excitatory amino acids. Eur J Neurosci1998, 10:246–254.76. Miller FD, Pozniak CD, Walsh GS: Neuronal life and death: an essential rolefor the p53 family. Cell Death Differ 2000, 7:880–888.77. Brooks CL, Gu W: Ubiquitination, phosphorylation and acetylation: themolecular basis for p53 regulation. Curr Opin Cell Biol 2003, 15:164–171.78. Colbourne F, Grooms SY, Zukin RS, Buchan AM, Bennett MV: Hypothermiarescues hippocampal CA1 neurons and attenuates down-regulation ofthe AMPA receptor GluR2 subunit after forebrain ischemia. Proc Natl AcadSci USA 2003, 100:2906–2910.79. Hemmen TM, Lyden PD: Induced hypothermia for acute stroke. Stroke2007, 38:794–799.80. Linares G, Mayer SA: Hypothermia for the treatment of ischemic andhemorrhagic stroke. Crit Care Med 2009, 37:S243–S249.81. Poli S, Veltkamp R: Oxygen therapy in acute ischemic stroke -experimental efficacy and molecular mechanisms. Curr Mol Med 2009,9:227–241.82. Burns TC, Verfaillie CM, Low WC: Stem cells for ischemic brain injury: acritical review. J Comp Neurol 2009, 515:125–144.83. Ohab JJ, Fleming S, Blesch A, Carmichael ST: A neurovascular niche forneurogenesis after stroke. J Neurosci 2006, 26:13007–13016.84. Gutierrez M, Merino JJ, de Lecinana MA, Diez-Tejedor E: Cerebralprotection, brain repair, plasticity and cell therapy in ischemic stroke.Cerebrovasc Dis 2009, 27(Suppl 1):177–186.85. Lee FJ, Xue S, Pei L, Vukusic B, Chery N, Wang Y, Wang YT, Niznik HB, Yu XM, Liu F:Dual regulation of NMDA receptor functions by direct protein-proteininteractions with the dopamine D1 receptor. Cell 2002, 111:219–230.86. Liu F, Wan Q, Pristupa ZB, Yu XM, Wang YT, Niznik HB: Direct protein-protein coupling enables cross-talk between dopamine D5 and gamma-aminobutyric acid A receptors. Nature 2000, 403:274–280.87. Zou S, Li L, Pei L, Vukusic B, Van Tol HH, Lee FJ, Wan Q, Liu F: Protein-protein coupling/uncoupling enables dopamine D2 receptor regulationof AMPA receptor-mediated excitotoxicity. J Neurosci 2005, 25:4385–4395.88. Hara MR, Thomas B, Cascio MB, Bae BI, Hester LD, Dawson VL, Dawson TM,Sawa A, Snyder SH: Neuroprotection by pharmacologic blockade of theGAPDH death cascade. Proc Natl Acad Sci USA 2006, 103:3887–3889.doi:10.1186/1756-6606-5-13Cite this article as: Wang et al.: Direct interaction between GluR2 andGAPDH regulates AMPAR-mediated excitotoxicity. Molecular Brain 20125:13.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionSubmit your manuscript at www.biomedcentral.com/submitWang et al. Molecular Brain 2012, 5:13 Page 12 of 12http://www.molecularbrain.com/content/5/1/13"@en ; edm:hasType "Article"@en ; edm:isShownAt "10.14288/1.0223275"@en ; dcterms:language "eng"@en ; ns0:peerReviewStatus "Reviewed"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "BioMed Central"@en ; ns0:publisherDOI "10.1186/1756-6606-5-13"@en ; dcterms:rights "Attribution 4.0 International (CC BY 4.0)"@en ; ns0:rightsURI "http://creativecommons.org/licenses/by/4.0/"@en ; ns0:scholarLevel "Faculty"@en ; dcterms:title "Direct interaction between GluR2 and GAPDH regulates AMPAR-mediated excitotoxicity"@en ; dcterms:type "Text"@en ; ns0:identifierURI "http://hdl.handle.net/2429/56402"@en .