"Medicine, Faculty of"@en . "Medicine, Department of"@en . "Experimental Medicine, Division of"@en . "DSpace"@en . "UBCV"@en . "Liu, Yitao"@en . "2010-01-18T20:45:48Z"@en . "2006"@en . "Doctor of Philosophy - PhD"@en . "University of British Columbia"@en . "Ischemic brain damage is largely due to excitotoxicity mediated by glutamate receptors, notably the N-methyl-D-aspartate-type receptors (NMDARs); however, to date none of the NMDAR antagonists have shown therapeutic benefits in treating stroke. Moreover, recent studies indicate blockade of NMDARs may even cause neuronal death. An explanation of the molecular mechanisms underlying this paradox is urgently required in order to develop new, effective stroke therapeutics. In this doctoral thesis project, we hypothesized that different NMDAR subtypes have opposing roles in neuronal fate and effective treatment of ischemic brain injury may be achieved through selective activation of the NMDAR subtype mediating neuronal survival and/or blockage of the NMDAR subtype mediating neuronal death. We first determined whether the two major NMDAR subtypes in mature cortical neurons, NR2A- and NR2B-containing NMDARs play differential roles in neuronal apoptosis. The results showed that NR2B-containing NMDARs are coupled to neuronal death whereas NR2A-containing receptors mediate neuronal survival. Further investigation revealed that the subcellular location (synaptic versus extrasynaptic) of NMDARs has little effect on their roles in neuronal fate. We then tested whether selective activation of NR2A-mediated neuronal survival signaling [i.e. neuronal survival signalling] or inhibition of NR2B-mediated neuronal death pathway is neuroprotective in stroke models. The data showed that blockade of NR2B is neuroprotective but has a relatively narrow therapeutic window. In contrast, selective activation of NR2A attenuates ischemic brain injury even when delivered 4.5 h post stoke onset. These findings suggest for the first time that selective stimulation of NR2A-containing NMDARs may constitute a promising therapy for stroke. Due to the critical role of NR2B in neuronal death and the narrow time window of NR2B antagonists, we examined whether disrupting the excitotoxic signaling [i.e. excitotoxic signalling] pathway downstream of NR2B activation was efficacious in treating ischemic damage. We found that post-ischemic administration of an interference peptide derived from the carboxyl tail of NR2B remarkably reduces stroke-induced brain injury. Thus, perturbing protein-protein interaction downstream of NR2B activation may represent another novel therapy for stroke. This research project provides a molecular basis for the dual roles of NMDARs in neuronal survival and demise and thereby suggests a number of clinically relevant new stroke therapies."@en . "https://circle.library.ubc.ca/rest/handle/2429/18582?expand=metadata"@en . "OPPOSING ROLES OF NMDA RECEPTOR SUBTYPES IN NEURONAL FATE AND NOVEL TREATMENTS FOR ISCHEMIC BRAIN INJURY by Y I T A O L I U A THESIS S U B M I T T E D I N P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R O F P H I L O S O P H Y in T H E F A C U L T Y O F G R A D U A T E S T U D I E S (Experimental Medicine) T H E U N I V E R S I T Y OF B R I T I S H C O L U M B I A August 2006 \u00C2\u00A9 Yitao L i u , 2006 ABSTRACT Ischemic brain damage is largely due to excitotoxicity mediated by glutamate receptors, notably the N-methyl-D-aspartate-type receptors (NMDARs); however, to date none of the N M D A R antagonists have shown therapeutic benefits in treating stroke. Moreover, recent studies indicate blockade of N M D A R s may even cause neuronal death. A n explanation of the molecular mechanisms underlying this paradox is urgently required in order to develop new, effective stroke therapeutics. In this doctoral thesis project, we hypothesized that different N M D A R subtypes have opposing roles in neuronal fate and effective treatment of ischemic brain injury may be achieved through selective activation of the N M D A R subtype mediating neuronal survival and/or blockage of the N M D A R subtype mediating neuronal death. We first determined whether the two major N M D A R subtypes in mature cortical neurons, N R 2 A - and NR2B-containing N M D A R s play differential roles in neuronal apoptosis. The results showed that NR2B-containing N M D A R s are coupled to neuronal death whereas NR2A-containing receptors mediate neuronal survival. Further investigation revealed that the subcellular location (synaptic versus extrasynaptic) of N M D A R s has little effect on their roles in neuronal fate. We then tested whether selective activation of NR2A-mediated neuronal survival signaling or inhibition of NR2B-mediated neuronal death pathway is neuroprotective in stroke models. The data showed that blockade of NR2B is neuroprotective but has a relatively narrow therapeutic window. In contrast, selective activation of N R 2 A attenuates ischemic brain injury even when delivered 4.5 h post stoke onset. These findings suggest for the first time that selective stimulation of NR2A-containing N M D A R s may constitute a promising therapy for stroke. Due to the critical role of N R 2 B in neuronal death ii and the narrow time window of N R 2 B antagonists, we examined whether disrupting the excitotoxic signaling pathway downstream of N R 2 B activation was efficacious in treating ischemic damage. We found that post-ischemic administration of an interference peptide derived from the carboxyl tail of N R 2 B remarkably reduces stroke-induced brain injury. Thus, perturbing protein-protein interaction downstream of N R 2 B activation may represent another novel therapy for stroke. This research project provides a molecular basis for the dual roles of N M D A R s in neuronal survival and demise and thereby suggests a number of clinically relevant new stroke therapies. i i i T A B L E O F C O N T E N T S ABSTRACT ii TABLE OF CONTENTS iv LIST OF TABLES viii LIST OF FIGURES ix LIST OF ABBREVIATIONS xi ACKNOWLEDGEMENTS xiii CHAPTER 1 : INTRODUCTION 1.1. Overview 1 1.2. Ischemic stroke and stroke models 3 1.3. Neuronal death following ischemic stroke 6 1.4. Mechanisms of ischemic neuronal death 7 1.5. Signaling pathways leading to neuronal apoptosis 12 1.6. N M D A R s as therapeutic targets for stroke , 15 1.6.1. N M D A R s ...15 1.6.2. Current understanding of the role of N M D A R s in ischemic neuronal injury 23 1.7. Current status of N M D A R antagonists in the treatment of stroke 25 1.8. Why N M D A R antagonists failed in stroke clinical trials 29 1.9. Rationales, hypotheses and specific aims 30 CHAPTER 2: MATERIALS AND METHODS 2.1. Drugs and solutions 34 2.2. Primary culture of cortical neurons 34 iv 2.3. In vitro stimulations 35 2.4. Assessment of neuronal death 36 2.5. Experimental stroke models 36 2.5.1. O G D in vitro 36 2.5.2. M C A o in rats 37 2.6. Electrophysiology 38 2.6.1. Recording of mEPSC 38 2.6.2. Recording of N M D A induced current mediated by extrasynaptic N M D A R s .. .39 2.6.3. Recording of fEPSCs in hippocampal slices 40 2.7. Peptide construction and delivery 40 2.8. Construction of cDNAs encoding N R 2 A and NR2B C-terminals 41 2.9. R N A i 42 2.10. Transfection 43 2.11. Western blot 43 2.12. Co-IP 44 2.13. Calcium Imaging 45 2.14. Data analysis 45 C H A P T E R 3: R E S U L T S Part I: N R 2 A - and NR2B-containing N M D A R s play opposing roles in excitotoxicity 3.1.1. Summary 46 3.1.2. N R 2 A - and NR2B-containing N M D A R s play differential roles in neuronal apoptosis 47 3.1.3. N R 2 A - and NR2B-containing N M D A R s are coupled to distinct signaling pathways 59 3.1.4. The subunit composition rather than the subcellular location determines the role of N M D A R s in excitotoxicity 62 Part II: Treatment of stroke by selective activation of NR2A-containing N M D A R s 3.2.1. Summary 83 3.2.2. Selective activation of NR2A-containing reduces hypoxic neuronal injury in vitro 83 3.2.3. Selective activation of NR2A-containing reduces ischemic brain damage in vivo 87 Part III: Treatment of stroke by disrupting NR2B-PSD-95 interaction 3.3.1. Summary 105 3.3.2. Perburbing NR2B-PSD-95 interaction diminishes ischemica brain damage in vivo 106 CHAPTER 4: DISCUSSION 4.1. The use of N V P A A M 0 7 7 as a selective antagonist of N R 2 A 116 4.2. Dual role of N M D A R s in neuronal fate and possible underlying mechanisms 117 4.3. Subunit composition of N M D A R s determines their role in excitotoxicity 122 4.4. Selective activation of NR2A-containing N M D A R s represents a novel therapy for stroke 125 4.5. Perturbing protein-protein interaction downstream of N R 2 B activation is an alternative to NR2B blockage 128 4.6. Promising NMDAR-based neuroprotective therapies for ischemic brain damage...130 4.7. Future directions : 134 R E F E R E N C E L I S T 136 A P P E N D I X 162 LIST O F T A B L E S Table Title Page Table 1 Ionotropic glutamate receptor subunits 16 Table 2 Differential properties of N M D A R subtypes 22 LIST O F FIGURES Figure Title Page Functional NR2A and NR2B-containing N M D A receptors are Fig. 1 present in cultured neurons and are preferentially blocked by their 50-51 respective antagonists. Fi 2a b Activation of NR2A- and NR2B-containing N M D A R s exerts g 4 5 ^ \u00E2\u0080\u00A2 differential effects on neuronal apoptosis Activation of NR2A- and NR2B-containing N M D A R s triggers Fig. 2c neuronal survival (Akt) and apoptotic (caspase-3) pathway, 60-61 respectively. _ . , Functional synaptic NR2B-containing N M D A R s are present in , . r i g . jSk . . B cultured cortical neurons p . Enhanced activation of synaptic NR2A- and NR2B-containing ^ ^ N M D A R s exerts opposing actions on neuronal fate Spontaneously activated synaptic NR2A- and NR2B-containing Fig. 3c-d N M D A R s have opposing roles in promoting neuronal survival and 70-72 death, respectively p . ^ a Functional NR2A-containing N M D A R s are present at extra- ^ \u00C2\u00B0 ' synaptic sites Activation of extra-synaptic NR2A-containing N M D A R s protects Fig. 4b against neuronal death mediated by extrasynaptic NR2B- 78-79 containing N M D A R s p . ^ c Activation of extra-synaptic NR2A-containing N M D A R s can ^ ^ 8 ' counteract NMDAR-independent apoptosis Pretreatments with NR2A- and NR2B-specific antagonists Fig. 5a respectively promote neuronal survival and death in stroke in vitro 85-86 (OGD) Fi 5b d Activation of N R 2 A - and NR2B-containing N M D A R s exerts ^ ^ ^ ' opposing effects on ischemic neuronal injuries in vivo r . Potentiated activation of NR2A-containing N M D A R s by N M D A R F l g- 6a . , . ^ . . . . 94-95 co-agonist glycine exerts neuroprotection in vitro p. ^ Glycine treatment reduces OGD-induced neuronal apoptosis in ^ ^ vitro Post-ischemic potentiation of NR2A-containing, but not blockade Fig. 6c-e of NR2B-containing N M D A R s reduces ischemic brain damage in 101-104 an in vivo focal ischemic stroke model Fig. 7 Neuroprotection by treatment with Tat-NR2B9c in vivo 110-114 p. g Schematic illustrates potential NMDAR-based therapies for stroke ^ 133 and other neurodegenerative diseases x LIST OF ABBREVIATIONS Abbreviation Definition ACSF Artificial cerebrospinal fluid AIF Apoptosis-inducing factor AMPA a-amino-3-hydroxyl-5-methyl-4-isoxazolepropionic acid AMPAR a-amino-3-hydroxyl-5-methyl-4-isoxazolepropionic acid type glutamate receptor ANOVA Analysis of variance AP-5/APV 2-amino-5-phosphonopentanoic acid Apaf-1 Apoptosis protease-activating factor 1 ASIC Acid-sensing ion channel Bcl-2 B-cell lymphoma-2 CaMKII Calcium-calmodulin kinase II CARD Caspase recruitment domain CNQX 6-cyano-7-nitroquinoxaline-2,3-dione CNS Central nervous system Co-IP Coimmunoprecipitation Cyt C Cytochrome C DED Death effector domain DIABLO Direct IAP binding protein with low p i DISC Death-inducing signaling complex DIV (cultures) Day in vitro DNQX 6,7-dinitroquinoxaline-2,3-dione ECS Extracellular solution Endo G Endonuclease G EPSC Excitatory postsynaptic current ER Endoplasmic reticulum Erk Extracellular signal-regulated kinase FADD Fas-associated adaptor protein with death domain fEPSC Field excitatory postsynaptic current GABA y-aminobutyric acid H&E Hematoxylin and eosin staining ICA Internal cerebral artery iGluR Ionotropic glutamate receptor KA Kainic acid MCA Middle cerebral artery MCAo Middle cerebral artery occlusion mEPSC Miniature excitatory postsynaptic current MOMP Mitochondrial outer membrane permeabilization mRFP Mutant red fluorescence protein NCX Na+/Ca2+ exchanger NMDA N-methyl-D-aspartate NMDAR N-methyl-D-aspartate-type glutamate receptors nNOS Neuronal nitric oxide synthase NO Nitric oxide OGD Oxygen-glucose deprivation PDZ PSD-95, Dig, and ZO-1 homology domain P I 3 K Phosphatidylinositol 3-kinase PSD Postsynaptic density PT Permeability transition PTEN Phosphatase and tensin homolog deleted on chromosome 10 RasGRF Ras guanine nucleotide-releasing factor RNAi R N A interference ROS Reactive oxygen species SAP Synapse associated protein SEM Standard error of mean siRNA Small interfering R N A Smac second mitochondria-derived activator of caspase SynGAP Synaptic GTPase activating protein tBid Truncated B id VGCC Voltage-gated calcium channel XOD Xanthine oxidase xi i ACKNOWLEDGEMENTS \"If I have seen further it is by standing on ye shoulders of Giants.\" -Isaac Newton No scientific endeavor can be accomplished without the support and assistance of many people. As such, I am deeply indebted to my supervisor Dr. Yu-Tian Wang for believing in me, giving me the freedom to explore unchartered scientific frontiers and providing guidance and encouragement throughout my study. Had it not been for him, my research project would not have been possible. I am also very thankful to Dr. Michael Tymianski, who was a member of my supervisory committee when I first started my Ph.D. study in the University of Toronto, for offering me the opportunity to study in his lab and to collaborate with his research team. I would like to thank my supervisory committee Drs. Max Cynader and Wolfram Tetzlaff for their constructive inputs and suggestions on my research. I also want to express my utmost gratitude to Drs. Tak Pan Wong and Lidong L iu for their help in the electrophysiological experiments, and Drs. Michelle Aarts and Dongchuan Wu for their assistance in the animal studies. This project would not have been finished without the efforts of those wonderful people in the labs of Dr. Yu-Tian Wang and Dr. Michael Tymianski, and particularly I want to thank Mr. Steve Van Iderstine for his excellent management, Ms. Yuping L i and Ms. Lucy Teves for their technical support, and Drs Yushan Wang and Jie L u for their insights. Finally, I would like to thank my dearest Ying for her unconditional and unwavering love, faith and patience, and also my parents who have always believed in and supported me. xi i i CHAPTER 1: INTRODUCTION 1.1. Overview Stroke is now a major cause of morbidity and mortality in most industrialized countries (Hacke et al., 1991;Dirnagl et al., 1999). In the United States alone, every 45 seconds there is a person suffering a stroke and every 3.1 minutes someone dies of a stroke (American Heart Association, 2004) . In addition to its devastating effect on the patient and his/her family, it also imposes an enormous burden on a nation's economy (Ter Horst and Postigo, 1997;Lin, 2002). According to the estimates, the annual cost of stroke-related expenses in the US accumulates up to approximately 71.8 billion dollars (Hinkle and Bowman, 2003;Murphy, 2003). These statistics suggest that a cure for stroke is urgently needed. Current therapies for ischemic stroke can be classified into two categories: neuroprotective therapy and thrombolytic therapy (Fisher and Bogousslavsky, 1998). The rationale of neuroprotective therapy lies in that brain ischemia triggers a series of pathological changes that result in neuronal death. Theoretically, it is possible to salvage the brain tissue i f the neurotoxic events can be promptly interrupted and/or an inherent neuroprotective ability can be activated in time. In accord, many neuroprotectants have been developed in hope of diminishing brain damage resulting from stroke (Lutsep and Clark, 2001;Hinkle and Bowman, 2003;Danton and Dietrich, 2004;Wahlgren and Ahmed, 2004;Beresford et al., 2003); however, none of these agents, including many glutamate receptor antagonists, have exhibited clinical efficacy. Thrombolytic therapy takes advantage of the fact that clinical ischemic 1 stroke is mostly due to disruption of cerebral blood flow by thrombosis. Therefore, thrombolysis may reconstitute the circulation of the ischemic brain regions and hence prevent further brain injury (Jenkinson, 2004;Cornu et al., 2001 ;del Zoppo, 2004). To date the thrombolytics that have been investigated include recombinant tissue-type plasminogen activator (rt-PA), streptokinase, urokinase and prourokinase (Cornu et al., 2001;Madden, 2002;Hawn and Baldwin, 1997). rt-PA is currently the only drug approved by F D A for the treatment of acute stroke; however, whether it is effective remains controversial (Hacke et al., 1998;Albers et al., 2002). Furthermore, rt-PA has a narrow therapeutic time window (Mielke et al., 2004;Schellinger and Warach, 2004;Tomsick, 2004;Madden, 2002), which allows no more than 7% of stroke patients to receive this therapy (Cocho et al., 2005). In other words, thrombolytic therapy for stroke is also far from satisfactory (Caplan, 2004). Given the clear role of glutamate receptors, especially the N M D A subtype in excitotoxicity (Lipton and Rosenberg, 1994), it is paradoxical that none of the N M D A R antagonists have succeeded in stroke clinical trials to date. Further understanding of the functioning of N M D A R s during ischemic stroke is critical. It has been known that N M D A R s have several subtypes depending on their subunit composition, and further investigation have shown that these subtypes possess distinct pharmacological and electrophysiological properties; however, the functional differentiation of these N M D A R subtypes are not well-studied. Most recently, it has been found that the two major N M D A R subpopulations, N R 2 A - and N R 2 B -containing N M D A R s , may govern the polarity of synaptic plasticity in the cortex and hippocampus (Liu et al., 2004a;Massey et al., 2004b), indicating N M D A R subtypes 2 may be functionally diverse. In this thesis project, I hypothesized that the roles of N M D A R s in excitotoxicity are also dependent on the N M D A R subtype activated. I first studied the effects of activation of the two major N M D A R subtypes in mature cortical neurons, N R 2 A - and NR2B-containing N M D A R s , on neuronal apoptosis. The results showed that N R 2 A -containing N M D A R s have anti-apoptotic activity while their NR2B-containing counterparts mainly mediate neuronal death. Based on these discoveries, two new therapies for stroke were developed: the first was to specifically activate N R 2 A -containing N M D A R s , and the second was to disrupt the neurotoxic signaling pathway downstream of activation of NR2B-containing N M D A R s (in collaboration with Dr. Michael Tymianski's lab). These two strategies were shown to have dramatic effects in attenuating ischemic brain damage in vivo either prior to or, most importantly, following stroke onset. These findings suggest specifically enhancing the survival-promoting action of NR2A-containing N M D A R s and/or perturbing the signaling pathway downstream of activation of the apoptosis-mediating NR2B-containing N M D A R s may avoid the drawbacks of current N M D A R antagonists in treating stroke and constitute promising therapies for stroke. 1.2. Ischemic stroke and stroke models Being one of the most active organs in the body, the brain has a huge energy demand. Under physiological conditions, the majority of energy is produced through oxidative phosphorylation of glucose, which requires not only glucose as a substrate but also an adequate supply of oxygen (O2). It is estimated that although it only represents 2% of 3 the total body weight, the brain consumes about 25% of the total body glucose utilization and 20% of the resting total body 0 2 (Clarke and Sokoloff L . , 1994;Brust, 2000); however, the 0 2 storage in the brain is miniscule. Thus, the brain is dependent upon continuous replenishment of 0 2 through cerebral circulation (Clarke and Sokoloff L . , 1994). Ischemic stroke is a condition in which the cerebral blood flow is abruptly cut off or reduced drastically, resulting in lack of 0 2 and glucose in the ischemic regions. Naturally-occurring cerebral ischemia falls into two types: focal brain ischemia and global ischemia (Neumar, 2000). Focal ischemia is more commonly seen in humans. It may happen when a cerebral artery, such as the lenticulostriate artery or middle cerebral artery ( M C A ) , is blocked by thrombosis or embolism. In contrast, global ischemia is usually caused by a systematic reduction in blood flow like a cardiac arrest. As aforementioned, stroke has a tremendous negative impact on the society but no effective cure for stroke has been discovered. To study the mechanisms of ischemic brain injury, many experimental stroke models have been developed over the past years. At present, two types of stroke models, i.e. in vitro and in vivo models, have been established. In vitro experimental strokes can be produced in either brain slices or cell cultures. One most widely used paradigm in vitro is the oxygen-glucose deprivation (OGD) model. The bathing solution for brain slices or cultured cells is rapidly changed from 0 2 / C 0 2 - to N 2/C0 2-equilibrated solution without glucose. The lack of blood flow in stroke models in vitro makes it easier to interpret the obtained results; however, cerebral blood flow is an important factor that affects the outcome 4 of stroke. In addition, the anatomical relationship between an in vivo brain and in vitro cultured cells is not the same, and hence the changes observed in vitro may not reflect the situations in vivo (Lipton, 1999a). Animal models have long been used to study stroke. In correspondence with the cerebral ischemia that occurs in humans, strokes models in vivo are divided into two types: global and focal ischemia (Ginsberg and Busto, 1989). Global ischemia is produced by occlusion of vessels that results in an ischemia affecting a large portion of the forebrain. Global ischemia usually leads to selective cell death in the C A I region of the hippocampus. Delayed cell death is a major feature of global ischemia (Pulsinelli and Brierley, 1979;Kirino, 1982;Pulsinelli et al., 1982). The more frequently used model in vivo is the focal ischemic model, which usually involves occlusion of the M C A (MCAo) (Longa et al., 1989;Ginsberg and Busto, 1989;Strong et al., 1983;Symon, 1975). Among the different techniques of M C A occlusion, one well-established and widely used method is to insert a nylon suture through internal carotid artery (ICA) until the tip of the suture reaches the point at which the M C A branches from ICA so that M C A is blocked at its origin. In this model, there is a gradient in the ischemic region. In the core area, the blood flow is reduced to less than 15% of normal value (Duverger and MacKenzie, 1988;Nedergaard et al., 1986;Sakatani et al., 1990;Tamura et al., 1981), while in the region surrounding the core, which is termed penumbra, the blood flow is about 15-40% of normal (Ginsberg and Busto, 1989;Hossmann, 1994;Back et al., 1995). There is also the peri-infarct region whereby the blood flow is reduced but remains above 40% of normal. Focal ischemia produces a contiguous mass of damaged brain tissue termed infarct instead 5 of selective lesion in certain vulnerable regions of the brain. 1.3. Neuronal death following ischemic stroke Post-ischemic neurons mainly exhibit characteristics of two distinct forms of cell death: necrosis and apoptosis, depending on the intensity of the insults (Chopp and L i , 1996;Majno and Joris, 1995;Snider et al., 1999;Yuan et al., 2003;Pettmann and Henderson, 1998). Necrosis progresses rapidly and it is morphologically characterized by cellular and organelle swelling followed by disruption of nuclear, organelle and plasma membranes, along with disintegration of nuclear structure and cytoplasmic organelles (Majno and Joris, 1995;Clarke, 1990). Later in necrosis, chromatin may disappear entirely. Cells are killed by lipolysis, proteolysis and loss of ion homeostasis (Dirnagl et al., 1999;Aarts et al., 2003b). Inflammation may ensue because of the release of cellular contents into the extracellular space (Dirnagl et al., 1999;Neumar, 2000). In contrast, apoptosis is a form of delayed and programmed cell death (Yuan and Yankner, 2000). Apoptotic cells display characteristics of compaction and margination of the nuclear chromatin, cytoplasmic shrinkage and condensation with preservation of organelles, and nuclear and cytoplasmic budding to form membrane-bound fragments, i.e. apoptotic bodies (Kerr et al., 1972;Yuan et al., 2003;Neumar, 2000). Under appropriate conditions, apoptotic bodies are engulfed by macrophages and hence cellular contents are usually not spilled into extracellular space. At the molecular levels, apoptosis can be distinguished by exposure of phosphatidylserine (PS) on the outer surface of the plasma membrane, cleavage and activation of caspases, and eventual D N A fragmentation (Danial and Korsmeyer, 6 2004;Yuan and Yankner, 2000;Thornberry and Lazebnik, 1998). Until the early 1990s, ischemic brain damage was generally thought to result exclusively from necrosis, but in recent years mounting evidences have shown that apoptosis is an important part of ischemic brain injury (Sims and Anderson, 2002;Chopp and L i , 1996;Choi, 1996;Love, 2003;Love, 2003). Whether ischemic neurons die through necrosis or apoptosis depends heavily on the extent and duration of ischemia (Lipton, 1999a). Necrosis occurs within the ischemic core, whereas apoptosis is typically seen in the penumbra region (Juurlink and Sweeney, 1997). In the ischemic core of a transient focal ischemia, the ratio of necrotic to apoptotic cells is estimated to be around 1:1, whereas in the penumbra, the ratio is about 1:9 (Charriaut-Marlangue, 2004). Given these mechanisms and the rapidity with which the process occurs, necrosis is usually regarded as an uncontrolled/unregulated process. On the contrary, apoptosis has been shown to be tightly controlled, and the signaling pathways leading to apoptosis have been delineated over the past years. 1.4. Mechanisms of ischemic neuronal death The precise mechanisms of post-ischemic neuronal death are yet to be determined albeit the extensive studies conducted. Advances in this research area over the past years, however, have greatly improved our understanding of how ischemic brain injury occurs. It is now believed that ischemic neuronal death may occur through an interplay between alteration of C a 2 + homeostasis, reduced A T P production, elevated generation of reactive oxygen species (ROS) and acidosis. Ca is a vital intracellular messenger governing cellular functions, such as synaptic 7 activity and membrane excitability. Neurons maintain a tight command of C a 2 + homeostasis, including both intracellular levels and distribution of C a 2 + , through an interplay between C a 2 + influx and efflux, C a 2 + buffering and internal C a 2 + storage (Arundine and Tymianski, 2003). At resting state, the concentration of free C a 2 + in the cytosol is kept at low levels (-100 nM). During stroke, intracellular C a 2 + concentration increases dramatically. C a 2 + influx through glutamate receptors, especially the N M D A subtype, has been found to be a major source of the intracellular Ca 2 +increase (Arundine and Tymianski, 2004a;Choi, 1985;Choi, 1995). C a 2 + released from intracellular C a 2 + stores such as endoplasmic reticulum (ER) and mitochondria may contribute to the disruption of C a 2 + homeostasis as well (Hayashi and Abe, 2004;Ganitkevich, 2003;Paschen, 2000;Paschen and Doutheil, 1999). Most recently, it is shown that the major plasma membrane C a 2 + extruding system, the N a + / C a 2 + exchanger (NCX), also plays a critical role in delayed C a 2 + elevation in neurons (Bano et al., 2005). N C X 3 is cleaved in neurons undergoing excitotoxicity, and this cleavage aggravates C a 2 + accumulation in neurons. Although voltage-gated C a 2 + channels (VGCCs) are an important site of calcium entry, it is still controversial whether this C a 2 + influx is toxic (Hardingham et al., 2002;Bading et al., 1993;Hardingham et al., 1999). Thus, the \"source specificity\" hypothesis proposes that C a 2 + toxicity occurs through Ca 2 +-signaling pathways linked to specific routes of C a 2 + influx (Sattler and Tymianski, 2000;Tymianski et al., 1993). Whatever the source is, in vivo studies have demonstrated cytosolic and mitochondrial C a 2 + levels are dramatically elevated during ischemia and early reperfusion. A brief global ischemia can trigger a -2,000-fold elevation in the selectively vulnerable 8 hippocampal C A I and cortical neurons (Silver and Erecinska, 1992;Erecinska and Silver, 1992). In focal ischemia, total tissue C a 2 + i n both ischemic core and penumbra increases and lasts up to 24 hours after reperfusion (Kristian et al., 1998). Mitochondrial C a 2 + elevation in the ischemic core is proportional to the duration of ischemia (Gutierrez-Diaz et al., 1985). C a 2 + disregulation is paramount to neuronal death (Arundine and Tymianski, 2003;Kristian and Siesjo, 1998). Intracellular C a 2 + overload can trigger activation of lipases, proteases, endonucleases and kinases/phosphatases which eventually leads to neuronal death; however, the exact mechanism by which C a 2 + mediates excitotoxicity is still unclear. Under physiological conditions, mitochondria are capable of sequestering large amounts of intracellular C a 2 + ; however, abnormal accumulation of C a 2 + during stroke can cause mitochondrial dysfunction. C a 2 + is sequestered into the mitochondrial matrix via a proton electrochemical gradient that is generated by the electron transport chain and depolarizes the mitochondrial potential (Akerman, 1978;Gunter and Pfeiffer, 1990;Loew et al., 1994). This influx of Ca 2 +results in a reduction in the electrochemical gradient, and consequently the reduction of A T P production. In the meantime, more A T P is consumed by cells in order to extrude the intracellular C a 2 + overload. It's estimated that a 10-minute global ischemia can lead to a drop of A T P levels to 10% or less of normal values (Wagner and Lanier, 1994). In focal ischemia, the loss of A T P is less drastic, with -25% and -50-70% of normal in the ischemic core and penumbra, respectively (Sun et al., 1995;Welsh et al., 1991). The concurrent accumulation of intramitochondrial C a 2 + , decreased A T P generation and increased A T P consumption are critical in mediating early ischemic cell death (Schinder et al., 9 1996). ROS, including oxygen free radicals, nitric oxide (NO) and peroxynitrite (ONOO\"), the reaction product of superoxide and N O , are important mediators of ischemic neuronal death. Neurons are exposed to a baseline of free radical-mediated oxidative stress that is presumably tolerated by the cells. When ROS production exceeds the normal neuronal buffering capacity, it can impinge on neuronal integrity. Impairment of mitochondria during ischemia results in deficits in mitochondrial electron transport chain, which in turn leads to excessive free radical production (Green and Reed, 1998;Antonsson et al., 1997;Kumar et al., 1990). Free radicals can be generated through several pathways. For example, the breakdown of adenine nucleotides leads to accumulation of hypoxanthine. In the meantime, xanthine oxidase (XOD) production is increased as well (Mishra and ivoria-Papadopoulos, 1999). X O D can metabolize hypoxanthine to produce free radicals (Sussman and Bulkley, 1990). In addition, arachadonic acid accumulates during cerebral ischemia (Zhang and Sun, 1995), and arachadonic acid metabolism by oxidases can generate free radicals as well. Studies in vivo have demonstrated that the production of free radicals is elevated during and after global and focal ischemia (Kumar et al., 1990). In neurons, N O is mainly synthesized from L-arginine through the catalytic activity of neuronal N O synthase (nNOS). Ischemia dramatically increases nNOS activation in a calcium/calmodulin-dependent manner (Zhang et al., 1994;Dawson et al., 1996). More recently, it is shown that N M D A R activation is specifically linked to N O production via the post synaptic density (PSD) scaffolding protein PSD-95 (Sattler et al., 1999). N O production rises to the low micromolar range during global and focal 10 ischemia and elevated after focal ischemia as well (Lipton, 1999a). N O has many roles in the central nervous system (CNS) as a messenger molecule; however, it can be neurotoxic when generated in excess (Dawson and Dawson, 1996;Lipton, 1999b). On the injurious side, N O can damage D N A , eventually leading to neuronal death (Li and Wogan, 2005;Martin et al., 2005). N O also inhibits mitochondrial respiratory chain enzymes (Murray et al., 2003), and reacts with superoxide anion to produce ONOO\" . Although there is no direct evidence showing that ONOO\" production is elevated, the increased generation of superoxide and N O should allow formation of ONOO\" (Beckman, 1994). ONOO\" is a potent oxidant that reacts with sulphydryls and with zinc-thiolate moieties. It can also react with nitrate and hydroxylate aromatic rings on amino acid residues to oxidize lipids, protein and D N A (Aarts et al., 2003b). A n array of studies over the past years has shown that ROS play a central role in the development of ischemia-induced neuronal damage (Chan, 2001;Kontos, 2001). Complete oxidation of glucose is required for the brain to fulfill its energy requirement. A lack of blood supply and hence a shortage of oxygen following ischemia inevitably perturbs oxidative phosphorylation and forces neurons to switch to anaerobic glycolysis. Consequently, excessive lactic acid is produced as a byproduct of glycolysis, and protons also accumulate due to A T P hydrolysis. These all contribute to the drop of pH value in the ischemic tissue (Siesjo et al., 1996;Rehncrona, 1985). Under normoglycemic conditions, tissue pH can fall to as low as 6.5-6.0 during ischemia; however, i f a hyperglycemia exists or an ischemia is extremely severe, pH can drop even lower (below 6.0) (Siesjo et al., 1996;Rehncrona, 1985;Nedergaard et al., 1991a;Nedergaard et al., 1991b). Acidosis has been shown to 11 greatly exacerbate ischemic brain damage (Huang and McNamara, 2004;Siesjo et al., 1996;Tombaugh and Sapolsky, 1993). Most recently, a novel mechanism by which acidosis induces neuronal death has been revealed (Yermolaieva et al., 2004;Xiong et al., 2004). It has been shown that acidosis activates acid-sensing ion channels (ASICs) and hence elicits C a 2 + influx independent of glutamate receptors. Furthermore, both ASIC blockers and knockout of ASIC gene can protects the brain from ischemic injury. 1.5. Signaling pathways leading to neuronal apoptosis As discussed, neuronal apoptosis is a tightly-controlled cell death pathway that develops relatively slowly when compared to necrosis. Therefore, many neuroprotective approaches target different intermediate steps leading to neuronal apoptosis following stroke. Two apoptotic pathways have been characterized so far: the intrinsic and extrinsic apoptotic pathways. In post-ischemic neurons, neurotoxic events such as A T P depletion, intracellular C a 2 + overload and generation of ROS may converge on mitochondria to induce mitochondrial outer membrane permeabilization (MOMP), and hence cytochrome C (Cyt C) release (Aarts and Tymianski, 2004;Zipfel et al., 2000;Neumar, 2000;Nicholls, 2004), which initiates intrinsic apoptotic pathway. Therefore, mitochondria are generally thought to play a central role in this signaling pathway (Zamzami and Kroemer, 2001;Yuan and Yankner, 2000;Green and Reed, 1998;Finkel, 2001 ;Cai et al., 1998;Mignotte and Vayssiere, 1998). The mechanisms responsible for M O M P during apoptosis are still at issue (Halestrap et al., 2002a). It is argued that 12 M O M P is closely related to the B-cell lymphoma-2 (Bcl-2) family proteins (Sharpe et al., 2004;Kuwana and Newmeyer, 2003;Scorrano and Korsmeyer, 2003;Wei et al., 2001;Letai et al., 2002). The Bcl-2 family of apoptosis-regulating proteins contains both anti- and pro-apoptotic members which reside immediately upstream of mitochondria (Chan and Y u , 2004;Harada and Grant, 2003;Tsujimoto and Shimizu, 2000;Tsujimoto, 1998). Anti-apoptotic Bcl-2 proteins such as Bcl-2 and B c l - X L function to block M O M P while the pro-apoptotic members promote it (Halestrap et al., 2002b;Wei et al., 2001;Letai et al., 2002;Degli and Dive, 2003). Another mechanism of M O M P may be the opening of the permeability transition (PT) pore, which results in loss of mitochondrial inner transmembrane potential (A*Pm) and swelling of the matrix. After CytC is released from mitochondria, it binds to apoptosis protease-activating factor 1 (Apaf-1) (Hill et al., 2003). Apaf-1 is a cytosolic protein that contains a caspase recruitment domain (CARD), a nucleotide-binding domain and several WD-40 domains (Lauber et al., 2001;Zou et al., 1999). Following the binding of Apaf-1 to CytC, the nucleotide dATP or A T P binds to the complex and triggers its oligomerization to form an apoptosome (Hill et al., 2003). The C A R D domain is then exposed in the apoptosome, which subsequently recruits procaspase 9. Recruitment of procaspase 9 leads to its autoactivation through cleavage into caspase 9. The active caspase 9 then cleaves procaspase 3 into caspase 3, the active form of the enzyme (Porter and Janicke, 1999). Caspase 3, along with other effector caspases such as caspase 6 and 7, is the major effector caspase in apoptosis (Slee et al., 2001;Van de et al., 1999;Stennicke and Salvesen, 1997). Caspases are aspartate-specific cysteine proteases which can cleave proteins such as 13 the DNA-repairing enzyme poly (ADP-ribose) polymerase (PARP) and the cytoskeleton protein, gelsolin, and result in neuronal disassembly (Love, 2003). M O M P also results in the release of other intermitochondrial membrane proteins such as second mitochondria-derived activator of caspase (Smac)/direct IAP binding protein with low pi (DIABLO), apoptosis-inducing factor (AIF) and endonuclease G (Endo G) (Du et al., 2000;Verhagen et al., 2000;Lu et al., 2003;Cande et al., 2002;Widlak and Garrard, 2005;Schafer et al., 2004;Arnoult et al., 2003;Li et al., 2001;Li et al., 2001). Smac/DIABLO can promote CytC/Apaf-1 -dependent caspase activation. AIF and Endo G, on the other hand, can mediate caspase-independent forms of apoptosis. Neuronal death that follows experimental brain injury has been shown to involve AIF translocation from the mitochondria to cell nuclei (Zhang et al., 2002). Endo G is a DNase that induces nuclear D N A cleavage and apoptosis, which may be translocated from mitochondria to nucleus following cerebral ischemia (Li et al., 2001;Lee et al., 2005). Death receptor-dependent extrinsic apoptotic pathway may be involved in ischemic neuronal death as well (Love, 2003;Felderhoff-Mueser et al., 2000;Carboni et al., 2005). Cerebral ischemia may upregulate death receptors such as Fas/CD95 (Felderhoff-Mueser et al., 2000;Carboni et al., 2005). The mechanism is still not fully understood but p53 levels can be increased by ischemic injury, and p53 is capable of upregulating Fas/CD95 (Rich et al., 2000). Upon binding of Fas ligand (FasL), Fas receptors trimerize and undergo a conformational change, which subsequently assembles on the cytoplasmic tail a signaling complex known as the death-inducing signaling complex (DISC) (Muzio et al., 1996). The Fas-associated adaptor protein 14 with death domain (FADD) binds to Fas via its D D domain and recruits procaspase 8 via its death effector domain (DED) (Kischkel et al., 1995). Procaspase 8 is then activated through autoproteolysis. Following activation of caspase 8, there are two possible pathways leading to apoptosis (Scaffidi et al., 1998). The first is through cleavage of caspases 3 and 7, and the other pathway involves truncation of Bid. Truncated B id (tBid) can be translocated to mitochondria and hence trigger the intrinsic apoptotic pathway (Li et al., 1998a;Luo et al., 1998). 1.6. NMDARs as therapeutic targets for stroke Glutamate is a major neurotransmitter in the mammalian central nervous system (CNS). Glutamate receptors are divided into two pharmacologically and functionally distinct families, i.e. metabotropic and ionotropic glutamate receptors (mGluRs and iGluRs). Given the main research aims in this project, here I wi l l only focus on one type of iGluRs, i.e. N M D A R s . 1.6.1. NMDARs Based on their affinity to three selective agonists, N-methyl-D-aspartate ( N M D A ) , a-amino-3-hydroxyl-5-methyl-4-isoxazolepropionic acid ( A M P A ) and kainic acid (KA) , iGluRs can be divided into three subfamilies, i.e. N M D A R s , A M P A receptors (AMPARs) and kainate receptors, respectively (Erreger et al., 2004;Dingledine et al., 1999). Each subfamily is further divided into distinct subunits that come together to form a homo- or heteromeric ion channels (Table 1). 15 Table 1. Ionotropic glutamate receptor subunits Receptor family Subunit Selective agonist NMDARs NR1 NR2A, NR2B, NR2C, NR2D NR3A, NR3B NMDA AMPARs GluRl GluR2 GluR3 GluR4 AMPA Kainate receptors GluR5 GluR6 GluR7 KA1 KA2 Kainic acid 16 N M D A R s were first cloned around 1990. Three types of N M D A R subunits, NR1 , NR2 and NR3 , are encoded by three distinct gene families (Dingledine et al., 1999). Transmembrane topology study indicates that each subunit has four hydrophobic membrane-spanning domains, three of which are transmembrane domains ( M l , M3 and M4) and one is a cytoplasmic-facing re-entrant membrane loop (M2) (Dingledine et al., 1999;Kuner et al., 1996). The N terminus of each subunit is located extracellularly and the C terminus intracellularly. The M 2 domain lines the pore of the ion channel (Kuner et al., 1996). In the homologous regions of the NR2 subunit resides the agonist (glutamate or N M D A ) binding site. Interestingly, activation of N M D A R s requires a co-agonist, glycine or D-serine (Kleckner and Dingledine, 1988;Fern et al., 1996;Mayer et al., 1989). The co-agonist binding site is formed by the region preceding segment M l (SI domain) and the loop region between M3 and M 4 (S2 domain) of the NR1 subunit (Laube et al., 1998;Kuryatov et al., 1994;Dingledine et al., 1999). Using a model based on random aggregation of N M D A R subunits, it was first proposed that N M D A R s be a pentameric structure (Premkumar and Auerbach, 1997). But later studies favored an assembly of four subunits (Rosenmund et al., 1998;Laube et al., 1998). Functional analyses in vitro have shown that only the recombinant heteromers, but not the homomer of either NR1 or N R 2 A - D , match the physiological and pharmacological responses of the native N M D A R s (Janssen et al., 2005). Therefore, it is now generally believed that N M D A R s are heterotetramers comprised of two copies of the NR1 subunit and two copies of N R 2 A - D subunit (complexes formed by NR1 and N R 3 A or 3B have been shown to be gated by glycine (Chatterton 17 et al., 2002) and hence not classical N M D A R s ) . Some native N M D A R s may be triheteromers which contain NR1 and two different types of NR2 subunit (Cull-Candy et al., 2001;Sheng et al., 1994), e.g., NR1/NR2A/NR2B; however, the ratio of triheteromeric vs. diheteromeric N M D A R s in the brain is still under debate (Balhos and Wenthold, 1996; Luo et al., 1997). The distribution of N M D A R subtypes in the mammalian nervous system undergoes a developmental change (Monyer et al., 1994;Akazawa et al., 1994;Sheng et al., 1994;Li et al., 1998b). NR1 subunit is ubiquitously expressed throughout development and adulthood. N R 2 A expression cannot be detected until soon after birth, but it increases gradually when approaching adulthood. Conversely, the levels of NR2B subunit are high during the prenatal period but gradually decrease with development. The expression N R 2 C first appears at the third postnatal week and increases steadily. Similar to NR2B, N R 2 D is highly expressed prenatally, but the levels of N R 2 D drop quickly after birth. In an adult CNS, N M D A R subtypes distribute differentially (Monyer et al., 1994). The NR2A-containing N M D A R s are widely distributed, with high levels in the forebrain, hippocampus, and cerebellum. NR2B-containing N M D A R s are highly represented in the forebrain, striatum and midbrain. The adult cerebellum has the greatest concentration of the NR2C-containing receptors, while NR2D-containing N M D A R s is weakly expressed in the brainstem and spinal cord. The distribution of N M D A R subtypes within a neuron may differ as well. N M D A R s are mostly localized to the postsynaptic sites. It has been shown in cortical and hippocampal neurons that NR2A-containing N M D A R s are preferentially located at excitatory synapses, whereas NR2B-containing 18 N M D A R s are predominant at extrasynaptic sites (Tovar and Westbrook, 1999;Stocca and Vicini , 1998). NR2D-containing N M D A R s are present extrasynaptically in dorsal horn spinal neurons (Momiyama, 2000) but there is no evidence for the presence of synaptic NR1/NR2D receptors (Cull-Candy et al., 2001). In contrast, NR1/NR2C receptors can be found at synapses in cerebellum (Cathala et al., 2000). Compared with A M P A R and kainate receptors, N M D A R s activate slowly and deactivate with a much slower time course. In the continued presence of glutamate, the responses of N M D A R s to the agonist are diminished, i.e., N M D A R s desensitize. Three forms of desensitization of N M D A R channels have been identified: glycine-dependent desensitization, glycine-independent desensitization and Ca2+-dependent desensitization (also referred to as Ca2+-dependent inactivation) (Yamakura and Shimoji, 1999). N M D A R s are ion-channels highly permeable to N a + , K + and most notably C a 2 + (Burnashev et al., 1995b;Schneggenburger, 1996;Garaschuk et al., 1996). In normal extracellular solution with a C a 2 + concentration of 1.8mM, fractional C a 2 + currents through the NR1/NR2A channels are around 11% (Burnashev et al., 1995a). The asparagine residue at the N site in segment M 2 of the NR1 subunit that contributes to voltage-dependent M g 2 + block also determines the C a 2 + permeability of the N M D A receptor channel (Burnashev et al., 1992); conversely, the NR2 subunit does not contribute to the C a 2 + permeability. Interestingly, C a 2 + ions not only flux through but also block N M D A R channels when the external C a 2 + is increased. C a 2 + can markedly reduce single channel conductance of N M D A R s (Jahr and Stevens, 1993). This C a 2 + block is independent of membrane potential, and mutation of the N site asparagines of 19 the NR1 or N R 2 A subunit increases the C a 2 + block (Ruppersberg et al., 1993). Besides the glutamate and glycine binding sites, there are several modulatory sites on N M D A R s . N M D A R s are blocked by M g 2 + in a voltage-dependent manner (Nowak et al., 1984;Jahr and Stevens, 1990). Therefore, activation of N M D A R s is also dependent on membrane potential. At resting membrane potentials, N M D A R channels are blocked by physiological concentrations of extracellular M g 2 + , but when neurons are depolarized by intense activation of postsynaptic A M P A R s , the voltage-dependent block by M g 2 + can be relieved. These properties render N M D A R s a co-incidence detector for synaptic release of glutamate and depolarization. In addition to M g 2 + , several endogenous allosteric modulators of N M D A R s have been identified. For example, N M D A R s are sensitive to inhibition by proton (FT\"), and Z n 2 + can serve as an allosteric modulator of N M D A R s as well. Other substances that exert allosteric modulations on N M D A R s include polyamines and redox agents. Although NR1 subunit is indispensable for functional N M D A R s (Forrest et al., 1994); (Fukaya et al., 2003), it is the NR2 subunit that confers distinct properties to the receptors (Yamakura and Shimoji, 1999;Cull-Candy et al., 2001). N M D A R s containing different NR2 subunit differ in their sensitivity to endogenous and exogenous ligands, permeation and blockage by divalent ions, kinetic properties, and interaction with intracellular proteins (Table 2). For example, the deactivation time course of NR1/NR2A is 6-fold faster than that of NR2B-containing N M D A R s . This suggests that NR1/NR2B receptors, once activated, may stay open for a much longer time than those receptors containing N R 2 A . It is also noteworthy that not many N M D A R antagonists are subunit-specific except highly selective antagonists for 20 NR2B such as Ifenprodil and Ro 25-6981. To date antagonists with high selectivity for N R 2 C and NR2D have not been identified. Only recently has a relatively selective antagonist for N R 2 A , N V P A A M 0 7 7 (NVP) been developed (Auberson et al., 2002; (Liu et al., 2004a), but its selectivity is still controversial (Berberich et al., 2005; Weitlaufetal, 2005). 21 Table 2. Differential properties of NMDAR subtypes * Properties NMDAR subtypes NR1/NR2A NR1/NR2B NR1/NR2C NR1/NR2D EC50 of endogenous agonists L-Glutamate (uM) 1.7 0.8 0.7 0.4 Glycine OxM) 2.1 0.3 0.2 0.1 Desensitization Glycine-independent form Prominent Less prominent Absent Absent Ca2+-dependent form Present Slight/absent Slight/absent Present Electrophysiological properties Single-channel conductance (pS) 40 or 50 40 or 50 22 or 36 16 or 35 Deactivation (T w , ms) 50 300 280 1700 Allosteric modulations Sensitivity to M g 2 + block High High Low Low IC50(pH) for proton 7.2 7.3 6.2 7.3 IC50forZn 2 + (uM) 0.005-0.08 0.5-10 14-38 14 Specific antagonists NVP AAM077 (?) Ifenprodil Ro 25-6981 CP 101,606 Unidentified Unidentified T w : Weighted deactivation time constant *: Adapted from (Cull-Candy et al., 2001;Yamakura and Shimoji, 1999). 22 1.6.2. Current understanding of the role of NMDARs in ischemic neuronal injury It is estimated that the normal extracellular glutamate in the brain is about 1-5 | i M (Wahl et al., 1994). Following the onset of stroke, glutamate levels increase dramatically (Nishizawa, 2001). Within 1-2 min after global ischemia, extracellular glutamate begins to increase and it can rise to 16-30 u M by 10-15 min. During focal ischemia, extracellular glutamate also increases quickly, with concentrations varying between 30-50 u M in the ischemic regions (Lipton, 1999a). Results from human stroke patients also support this observation (Castillo et al., 1997). The mechanisms of this glutamate accumulation include enhanced efflux of glutamate and reduced glutamate uptake. The excessive release of glutamate immediately after the onset of ischemia is triggered by activation of voltage-dependent calcium channels and ensuing C a 2 + influx (Silver and Erecinska, 1990). With the progress of ischemia, glutamate transporters may operate in reverse mode due to the imbalance of N a + across plasma membranes, which exacerbates the accumulation of extracellular glutamate (Barbour et al., 1988;Taylor et al., 1995); however, it is also demonstrated that glutamate accumulation can be cleared rapidly in global ischemia. In transient focal ischemia, the high glutamate levels do not last for a long time, either. Following reperfusion, glutamate concentrations decline to control (basal) levels quickly, usually less than 1 hour (Nishizawa, 2001 ;Obrenovitch and Richards, 1995;Davalos etal., 1997). The neurotoxic effects of glutamate were first discovered 5 decades ago by Lucas and Newhouse ( L U C A S and N E W H O U S E , 1957), and Olney confirmed this glutamate 23 toxicity (Olney and Sharpe, 1969;01ney, 1969) and coined the term \"excitotoxicity\". It is now thought that excitotoxicity during stroke is caused by excessive release of glutamate which in turn lead to over-stimulation of glutamate receptors, especially N M D A R s . Since the early studies showing that glutamate and N M D A R s play a crucial role in hypoxic/ischemic injury in vitro and in vivo (Choi and Rothman, 1990;Kass and Lipton, 1982;Rothman, 1983;Simon et al., 1984), NMDAR-mediated excitotoxicity has been widely regarded as the major mechanism by which ischemic brain damage occurs. The signaling pathways linking N M D A R overactivation to neuronal death have caught much attention over the past years. As discussed above, excessive C a 2 + enters into neurons through N M D A R s following stroke onset, which can activate calmodulin and then nNOS (Dawson and Dawson, 1996). At excitatory synapses of central neurons, N M D A R s interact with multiple scaffolding and signaling proteins, such as PSD-95, PSD-93 and SAP 102, within the PSD, a microscopic structure associated with the postsynaptic membrane (Sheng and Pak, 2000). For example, the cytoplasmic carboxyl terminals of N R 2 A and NR2B subunits have been shown to bind to the second PDZ (PSD-95, Dig and ZO-1 homology) domain of PSD-95 (Kornau et al., 1995). PSD-95 also interacts with nNOS through its second PDZ domain (Brenman et al., 1996;Christopherson et al., 1999); (Komiyama et al., 2002;Kim et al., 1998). Therefore, PSD-95 is thought to play a central role in coupling the over-activation of N M D A R to N O production, and hence neuronal death; however, the exact mechanisms of NMDAR-mediated neuronal death are still not fully understood. 24 Interestingly, the NR2 subunits of NMDARs may interact differentially with downstream signaling proteins. While calcium-calmodulin kinase II (CaMKII) are associated with both NR2A and NR2B, only its binding to NR2B but not NR2A locks CaMKII in an activated state (Bayer et al., 2001;Strack et al., 2000;Leonard et al., 1999;Gardoni et al., 1998;Strack and Colbran, 1998). NR2B can interact directly with RasGRFl, a Ca2+/calmodulin-dependent Ras-guanine-nucleotide-releasing factor whereas NR2A cannot (Krapivinsky et al., 2003). PSD-95 can bind to both NR2A and NR2B (Niethammer et al., 1996a;Aarts et al., 2002), but it seems that NR2B may preferentially bind to SAP 102 (van et a l , 2004). These subunit-specific interactions may lead to differential effects following NMDAR activation. In summary, the role of glutamate and NMDARs in ischemic neuronal death is clear, but further investigations are needed to elucidate the precise mechanisms. 1.7. C u r r e n t s ta tus o f N M D A R a n t a g o n i s t s i n the t r e a t m e n t o f s t r o k e In light of the importance of NMDARs in ischemia-induced neuronal injury, NMDAR antagonists have been considered to be promising neuroprotectants for stroke, and subsequently many drugs have been developed for this purpose. To date NMDAR antagonists that have been identified can be divided into four types according to their pharmacological properties. The first type is the competitive antagonists; the prototype compounds are 3-(2-carboxypiperazin-4-yl)-propyl-l-phosphonic acid (CPP) but its derivatives d-CPPene and CGS 19755 (Selfotel) and 2-amino-5-phosphonopentanoic acid (AP-5 or APV) and 2-amino-7-phosphonoheptanoic acid (AP-7) are more widely used (Dingledine et al., 1999;Wood 25 and Hawkinson, 1997). Other competitive N M D A antagonists developed include MDL-100,453, CGP-40116, W A Y 126090, N7, NPC17742 and SYM-2351 (De et al., 1999). The most recently developed N R 2 A specific antagonist N V P A A M 0 7 7 has also been shown to be a competitive antagonist (Auberson et al., 2002). A subgroup of competitive antagonists target glycine (co-agonist) binding site on N M D A R s , including Licostinel (ACEA-1021), Gavestinel (GV-150526), Harkoeride, M D L 29951, Z D 9379, M R Z 2/576 and L689,560 (Jansen and Dannhardt, 2003;Coyle and Tsai, 2004;Coyle et al., 2002).The second type of N M D A R antagonists is the noncompetitive antagonists. Most NR2B-specific N M D A R antagonists are noncompetitive. Ifenprodil and its analogs, including eliprodil and haloperidol, have drawn the most attention (Lynch and Gallagher, 1996;Gallagher et al., 1996). Derivatives of ifenprodil with higher selectivity such as Ro 25-6981, Ro 8-4304 and CP-101,606 have also been developed (Menniti et al., 1997;Fischer et al., 1997b;Kew et al., 1998). Other less commonly known noncompetitive antagonists include ethanol (at intoxicating concentrations) and dynorphin peptides (Dingledine et al., 1999). The uncompetitive antagonists are the third type of N M D A R antagonists. They act only on the activated (open) receptors but not the receptors at rest. The unique property of these antagonists is that they exert a state-dependent and use-dependent block, i.e., they wil l not reach the binding site unless the ion channel is open and the M g 2 + block is removed (Dzubay and Jahr, 1996;Jahr, 1992). For this reason, these drugs are also known as open channel blockers. Once they are bound, these blockers can be trapped by channel closure and recovery from the trapped blocked state is generally slow. Early agents of this type include dizocilpine (MK801), phencyclidine (PCP) and its 26 derivatives N-ethyl-l-phenylcyclohexamine (PCE) and l-[l-2(thienyl)cyclohexyl]-piperadine (TCP) (Dzubay and Jahr, 1996;Jahr, 1992;MacDonald et al., 1991;MacDonald et al., 1990). Other drugs that have also been well studied include aptiganel (CNS 1102, Cerestat), ketamine, amino-adamantane derivatives, such as memantine and amantadine, and MRZ2/579 (Hewitt, 2000;Sareen, 2002;Liu et al., 2000). The fourth type of N M D A R antagonists contains those blocking N M D A R s with unknown mechanisms, such as nitrous oxide and some mGluR agonists (Dingledine et al., 1999). Since the seminal work by Simon et al. showing that the competitive antagonist A P V protect the brain from ischemic damage (Simon et al., 1984), the effects of a variety of N M D A R antagonists on neuronal injury resulting from hypoxia/ischemia have been extensively studied in both the in vitro and in vivo systems. In cultured neurons, almost all types of N M D A R antagonists tested have shown neuroprotection to some extent. Goldberg et al first showed that A P V and several other N M D A R antagonists were potent neuroprotectants against OGD-induced cell death in cortical cultures (Goldberg et al., 1987). The authors also found that uncompetitive antagonists such as MK801 and PCP blocked neuronal injury triggered by N M D A stimulation and O G D challenge (Goldberg et al., 1988). Excitotoxic neuronal death can be ameliorated by NR2B subunit-specific antagonists such as ifenprodil and Ro 25-6981 as well (Fischer et al., 1997a;Graham et al., 1992). Glycine-site antagonists also exert neuroprotection in cultured cortical neurons (Boireau et al., 1996). The neuroprotective actions of N M D A R antagonists have also been determined using organotypic corticostriatal and hippocampal brain slices (Reyes et al., 1998;Calabresi 27 et al., 2003). The neuroprotection of N M D A R antagonists observed in the hypoxia models in vitro have been confirmed in many animal studies as well. In transient or even permanent focal ischemia model, there is consistent evidence showing reduction in infarct size and improvement in neurological function by all types of antagonists. When administered before or during a temporary ischemia, almost all of the N M D A R antagonists are very effective. Some studies also reported that administration of N M D A R antagonists following focal ischemia can still provide neuroprotection, albeit in a somewhat less dramatic way (Bar-Joseph et al., 1994;Warner et al., 1991). N M D A R antagonists seem to protect the penumbra only in focal ischemia, and the core of the lesion cannot be rescued by N M D A R blockade (Lipton, 1999a), however, these postischemic effects are still controversial (Nellgard and Wieloch, 1992). In global ischemia, the neuroprotective effects of N M D A antagonists are still not convincing. The open channel blocker MK801 and other glycine-site antagonists have been reported to show no effects (Warner et al., 1995;Buchan et al., 1991). Inspired by the huge success in animal studies, Albers et al. proposed that N M D A R antagonists might be ready for clinical trials (Albers et al., 1989). Over the past years, dozens of N M D A R antagonists have entered acute ischemic stroke clinical trials. This long list includes many types of N M D A R antagonists available (De et al., 1999;Bleich et al., 2003). Selfotel, a representative of competitive N M D A R antagonists, proceeded to phase III stroke trial but was halted because of risk/benefit ratio concern and potential neurotoxic effects accompanied with this drug (Davis et al., 2000;Davis et al., 1997). For similar reasons, the phase III clinical trial for 28 noncompetitive NR2B-specific antagonist eliprodil was terminated (Lees, 1997). Uncompetitive antagonists were considered to be good candidates for stroke therapy, but a carefully designed clinical trial demonstrated that aptiganel hydrochloride (Ceresat, CNS 1102) was not efficacious or may even be harmful (Lees, 1997;Albers et al., 2001). Glycine-site N M D A R antagonists were shown to have fewer side effects in animal studies, but again, to the disappointment of scientists and clinicians, gavestinel failed in clinical trial for stroke (Lees et al., 2000;Sacco et al., 2001). In conclusion, not a single N M D A R antagonist has shown positive results in human patients. 1.8. Why NMDAR antagonists failed in stroke clinical trials As discussed, none of the N M D A R antagonists has been successful in clinical trials involving stroke. Whether it is because of the defects of the antagonists per se, the inappropriate timing of treatment with these antagonists or some other unknown mechanisms remains to be determined (Cheng et al., 2004). Given the clear role of N M D A R s in excitotoxicity, the failure of such antagonists has long been perplexing. Ikonomidou and Turski (Ikonomidou and Turski, 2002b) recently argued that, since N M D A R s mediate the slow component of synaptic transmission, and synaptic transmission is essential for proper functioning of the brain, N M D A R antagonists actually may hinder normal brain function. Moreover, since glutamate levels in the brain may return to normal rapidly (within l h after stroke onset, (Nishizawa, 2001;Obrenovitch and Richards, 1995;Davalos et al., 1997)), glutamate-induced excitotoxicity may not contribute to brain injury at all after the acute phase of stroke 29 (within 6 h after stroke onset); however, in a c l in ical setting most patients w i l l not be able to receive treatment within 6 h. Together, these may explain why N M D A R antagonists failed in al l stroke cl in ical trials. In fact, despite the overwhelming reports on the neuroprotective effects o f N M D A R antagonists on ischemic brain damage, over the past years some studies d id indicate that N M D A R antagonists might be deleterious. It has been shown that blockade o f N M D A R s with M K 8 0 1 during normoxia can cause acute neuronal damage (Olney et al. , 1989;Allen and Iversen, 1990). M o r e recently, Ikonomidou et al demonstrated that the blockade o f N M D A R s during early development triggers extensive apoptosis in the brain including cortex, hippocampus and thalamus (Ikonomidou et al . , 1999b). Therefore, N M D A R s may also play a role in neuronal survival. This is further supported by the evidence that stimulation o f N M D A R s can lead to upregulation o f anti-apoptotic proteins o f the B c l - 2 family (Zhu et al . , 2005), whereas blockade o f N M D A R s in the developing brain results in impaired Erk activity, and hence apoptosis (Hansen et al. , 2004). Taken together, there is evidence suggesting the dual role o f N M D A R s in both neuronal survival and death. The key to answer why N M D A R antagonists are not successful in the treatment o f ischemic brain injury may lie in the understanding o f the mechanisms by wh ich N M D A R s function differentially in determining neuronal fate. 1.9. Rationales, hypotheses and specific aims Whi le the mechanisms o f the paradoxical role o f N M D A R s in excitotoxicity are still 30 elusive, Hardingdam et al. found recently that stimulation of N M D A R s at synapses activates cell-survival gene B D N F expression and prevents neurons from apoptosis, whereas activation of extrasynaptic N M D A R s triggers a pro-death signaling pathway that can be overwhelming (Hardingham et al., 2002). Thus, synaptic and extrasynaptic N M D A R s seem to play opposing roles in neuronal fate. Furthermore, these differential effects of N M D A R s are developmentally regulated (Hardingham and Bading, 2002). Consequently, the different anatomical locations of N M D A R s may account for the paradoxical roles of N M D A R s in neuronal fate. However, most recent evidence has shown that synaptic activity can be excitotoxic (Bellizzi et a l , 2005b), which makes the site-specific action of N M D A R s somewhat questionable. Furthermore, although it is being challenged (Thomas et al., 2006), mounting evidence has shown that synaptic and extrasynaptic N M D A R s have distinct subunit composition (Tovar and Westbrook, 1999). NR2A-containing N M D A R s may be predominant at synapses while NR2B-containing N M D A R s at extrasynaptic sites. Interestingly, recent studies suggest that the subunit composition of N M D A R s may dictate their functions. For example, N R 2 A - and NR2B-containing N M D A R subtypes may play differential roles in mediating synaptic plasticity (Liu et al., 2004a;Massey et al., 2004b), and different N M D A R subtypes may also be involved in different pathological conditions, including some neurodegenerative diseases (Lynch and Guttmann, 2002;Waxman and Lynch, 2005). In this study, I hypothesized that N R 2 A - and NR2B-containing N M D A R s may have differential roles in supporting neuronal survival and mediating neuronal death. The specific aims of this thesis project are: 31 Aim 1: Determine if NR2A- and NR2B-containing NMDARs play differential roles in neuronal apoptosis. I will induce NMDAR-mediated apoptosis in mature cortical neurons, which possess mainly NR2A- and NR2B-containing NMDAR subtypes, by relatively mild stimulation with NMDA, and then take advantage of NR2A- and NR2B-selective antagonists to study the individual role of these NMDAR subtypes. If I find these NMDAR subtypes have differential roles in neuronal apoptosis, I will then Aim 2: Determine if the subcellular (synaptic versus extrasynaptic) location or the subunit composition of NMDARs causes their functional differentiation. I will first determine the subunit composition of synaptic and extrasynaptic NMDARs by electrophysiological methods. Thereafter, the two NMDAR subtypes at either subcellular location will be isolated pharmacologically and the roles of these NMDAR subtypes in neuronal apoptosis be studied. If it is the subunit composition of NMDARs that determines their role in neuronal apoptosis, I will proceed to Aim 3: Determine if specifically stimulating the pro-survival NMDAR subtype and/or inhibiting the pro-death subtype reduces neuronal apoptosis following stroke in vitro. I will adopt the well-established in vitro OGD stroke model to achieve this goal. If the results show that OGD-induced neuronal apoptosis can be attenuated by enhancing the survival-promoting and/or antagonizing the pro-apoptotic NMDAR subtype, an in vivo study will follow to Aim 4: Determine if the treatments conducted in vitro (Aim 3) will be effective in vivo. A well-characterized focal cerebral ischemia model MCAo will be used, and neurological behavior and cerebral infarction will be compared between the treated 32 and non-treated groups to assess the efficacy of the treatments. The ultimate goal of this research project is to cast light on the mechanisms of the NMDAR-dependent ischemic brain damage, hence to develop effective N M D A R -based treatment of ischemic brain injury. 33 CHAPTER 2: MATERIALS AND METHODS 2 . 1 . Drugs and solutions A l l chemicals and drugs used were purchased from Sigma unless specifically indicated. The extracellular solution (ECS) were composed of (in mM): 25 HEPES acid, 140 NaCl , 33 glucose, 5.4 K C l 1.3 CaCl 2 and 1 M g C l 2 (normal ECS). Mg 2 +-free ECS did not contain M g C l 2 . Osmolarity of ECS was adjusted to 320-330 mosM and pH to 7.35. Modified RIPA buffer contained: 150 m M NaCl , 50 m M Tris (pH 7.4), 0.1% SDS, 1% NP-40, 0.5% sodium deoxycholate (DOC), 1 m M E D T A and 1 m M N a 3 V 0 4 . Protease inhibitors, including 10 pg/ml each of aprotinin, leupeptin (Peptides International) and I m M phenylmethylsulfonyl fluoride (PMSF), were added before use. PBS contained (in mM): 138 NaCl , 2.67 K C l , 8.1 N a 2 H P 0 4 , 1.47 K H 2 P 0 4 , 0.9 C a C l 2 and 0.5 M g C l 2 . pH was adjusted to 7.4 with HCI. M g 2 + and C a 2 + free PBS did not contain M g C l 2 or C a C l 2 (Invitrogen). The glucose-free bicarbonate-buffered solution for O G D experiment contained (in mM): 121 NaCl , 5 K C l , 1 Na-pyruvate, 1.8 C a C l 2 , 25 N a H C 0 3 , 0.01 glycine. Solutions used in electrophysiological experiments were detailed below (See Electrophysiology). 2 . 2 . Primary culture of cortical neurons Dissociated cultures of cortical neurons were prepared from 18-day Sprague-Dawley rat embryos (Mielke and Wang, 2005). The cortices were dissected from the brain and collected in a 35mm dish with M g 2 + and C a 2 + free PBS. The cortices were then digested with 0.5% trypsin in a 5% C 0 2 incubator at 37\u00C2\u00B0C for 10 min. After digestion, the neurons 34 were further dissociated in Neurobasal medium by trituration with a glass pipette. Neurons were then seeded onto 12-well plates or dishes (35mm or 10cm) at a density of 2.5-3.0 x 105/ml and grown in Neurobasal medium containing 2% B-27 supplement and 0.5 m M glutamine (all from Invitrogen) at 37\u00C2\u00B0C in a humidified 5% C O 2 atmosphere. To obtain mixed cortical cultures enriched with neurons, uridine (10 uM) and 5-Fluor-2'-deoxyuridine (10 uM) were added to the culture medium at 3 days in vitro (DIV) and maintained for 48 h to inhibit non-neuronal cell proliferation, and then the cultures were shifted back to the regular culture medium. The medium was changed every 4 days. Mature neurons (11-14 DIV) were used for experiments. 2.3. In vitro stimulations N M D A (50 uM) and glycine (10 uM) were bath applied to neurons for 20 min to facilitate induction of apoptosis. To induce apoptosis by staurosporine (STS), neurons were treated with STS (100 nM) for 1 h by bath application. Specific blockade of synaptic N M D A R s was achieved by treatment with MK801 (10 uM) in the presence of bicuculline (50 uM) for 10-15 min, followed by thorough wash with normal ECS to remove any trace of MK801 . Wherever NR2A-specific N M D A R antagonist N V P A A M 0 7 7 (NVP, 0.4 u M ; generous gift of Dr. Yves P. Auberson, Novartis Pharma A G , Basel, Switzerland) or NR2B-specific N M D A R antagonist Ro 25-6981 (Ro, 0.5 uM) was involved, the antagonist was pre-incubated with neurons for 10 min and maintained at the same concentration throughout the treatments. In vitro stimulations were conducted in 2+ M g free ECS. Termination of stimulations was achieved by washing 2x with normal ECS. Neurons were then changed back to normal culturing conditions until further assay. 35 2.4. Assessment of neuronal apoptosis To visualize apoptotic neurons, Hoechst-33342 (10 ug/ml) was added to the culture medium 20 h after treatments and incubated at 37\u00C2\u00B0C for 45 min. Images of different treatment groups were taken with a fluorescence microscope (Leica DMIRE2, 40x). To quantify neuronal apoptosis, 20 h following treatments, neurons were lysed, centrifuged (200 x g, 10 min) and then the supernatant (20 ul) was transferred into a strepatavidin precoated microplate. Neuronal apoptosis was determined by measuring intranucleosomal D N A fragmentation using a Cell Death Detection E L I S A P L U S kit (Roche Applied Science). Data analysis was carried out according to the manufacturer's instructions for the kit. 2.5. Experimental stroke models 2.5.1. O G D in vitro Cortical cultures were transferred to an anaerobic chamber containing a 5% CO2, 10% H2, and 85% N 2 (<0.2% 0 2 ) atmosphere (Goldberg and Choi, 1993), and then washed 3* with the glucose-free bicarbonate-buffed solution (deoxygenated in the anaerobic chamber for 30 min before use) and maintained anoxic for 1 h at 37\u00C2\u00B0C in the incubator inside the anaerobic chamber. O G D was terminated by washing the cultures 2x with ECS, and then the neurons were switched back to the original culturing conditions until assaying for cell death. 36 2.5.2. M C A o in rats Transient focal cerebral ischemia was produced by M C A o , as previously described (Liu et al., 2000;Longa et al., 1989). Briefly, male Sprague-Dawley rats (Charles River Laboratories) weighing ~300g were fasted overnight but allowed free access to water. Anesthesia was induced and maintained with 4.0% and 1.5% Isoflurane, respectively. M C A o was achieved by introducing a 3-0 monofilament suture to the M C A via internal carotid artery, as previously described (Longa et al., 1989). Body temperature was controlled at 37 \u00C2\u00B1 0.5\u00C2\u00B0C, and blood pressures and gases were monitored during the experiment. To study the effects of pretreatment with drugs, the right femoral vein of the rats was cannulized following anesthetization, and a single bolus of vehicle (saline) or drug was given via intravenous injection (i.v.). 30 min (for N V P and Ro) or 1 hour (for Tat-NR2B-A A and Tat-NR2B9c) following injection, the animals were subject to a 1 hour (for N V P and Ro) or 90 min (for Tat-NR2B-AA and Tat-NR2B9c) M C A o . For post-treatment experiments, animals were first subject to a 90-min cerebral ischemia produced by M C A o . To determine the effects of activation of NR2A-containing N M D A R s , the animals were then treated 4.5 hour after onset of M C A o with vehicle (saline) or drug(s) (glycine and/or N V P and Ro) through intraperitoneal injection (i.p.). To examine the effects of NR2B C-tail peptides, vehicle (saline) or peptide (Tat-NR2B-AA or Tat-NR2B9c) was given 1 hour after M C A o onset. Neurological function testing was performed to grade neurological function on a scale of 0 to 12 (normal = 0; worst = 12) 60 min following M C A o onset and/or 10 min before the animals were sacrificed. Specifically, a battery of two tests that have been used 37 previously to evaluate various aspects of neurological function (De et al., 1989;Bederson et al., 1986) were carried out: (1) the postural reflex test to examine upper body posture, and (2) the forelimb placing test to examine sensorimotor integration in forelimb placing responses to visual, tactile, and proprioceptive stimuli. 24 hours following M C A o , brains were perfusion-fixed with 4% paraformaldehyde, and brain blocks were embedded in paraffin. Coronal brain sections (5 um) were cut and stained with hematoxylin and eosin (H & E). 8 coronal levels throughout the brain were selected and images were taken with a microscope (Zeiss, Axiovert 200, 1.5x). Cerebral infarct area (S) of 8 selected sections (between Bregma +3.2mm and -5.8mm) was traced using the ImageJ software, and then total infarct volume (V) was calculated using the following formula: n ZVj = (Si.,+Si)*dM/2, i=l where n=9, Si represents the infarct area of a selected brain section i , V i denotes the infarct volume between section Sui and Si, d is the distance between two adjacent brain sections, So=0 and S9=0. 2.6. Electrophysiology 2.6.1. Recording of miniature postsynaptic currents (mEPSCs) Recording of mEPSCs was done on 11-day old cultured cortical neurons. The neurons on coverslips were transferred to a recording chamber that was continuously perfused with ECS. In addition, bicuculline (10 uM) and tetrodotoxin (0.5 u M ; Alomone) were added to the ECS to block gama-aminobutyric acid A type ( G A B A A ) receptors and voltage-gated sodium channels, respectively, to isolate action potential-independent 38 mEPSCs. Patch pipettes were pulled from borosilicate glass capillaries (Sutter Instrument) and filled with an intracellular solution (pH 7.2; 300-310 mOsm) composed of (mM): 140 CsCl gluconate, 0.1 CaCl 2 , 10 HEPES, 2 M g C l 2 , 10 B A P T A , and 4 A T P . A MultiClamp 700A amplifier (Axon Instruments) was used for the recording. The series resistance was monitored throughout each recording so that recordings with series resistance varied by more than 10% were rejected. No electronic compensation for series resistance was employed. Whole-cell patch-clamp recordings were performed in voltage-clamp mode while maintaining the membrane potential at -60 mV. Recordings were low-pass filtered at 2 kHz, sampled at 10 kHz, and stored in a PC using Clampex 8.0 (Axon). Synaptic events were analyzed offline using the Min i Analysis Program 6.0 (Synaptosoft). The removal of extracellular M g 2 + eradicates the Mg 2 +-mediated blockade of N M D A R s so that mEPSCs comprising both A M P A and N M D A receptor-mediated components can be measured. Antagonist for N M D A R s (Ro or A P V (Tocris)) was bath applied for at least 10 minutes to obtain sufficient length of recording for analysis after achieving a stable level of N M D A R blockade. Synaptic events before and after application of N M D A R antagonists were automatically detected from computer stored recordings using the same detection parameters in Min i Analysis Program. Subtraction of averaged traces was done in Excel (Microsoft). 2.6.2. Recording of NMDA induced current mediated by extrasynaptic NMDARs Extrasynaptic N M D A R s were isolated by specifically blocking synaptic N M D A R s using open channel blocker MK801 (10 u M , Tocris). To ensure all synaptic N M D A R s are activated during the MK801 treatment, cortical neurons were treated with high 39 concentration of bicuculline (50 uM) to enhance the excitatory synaptic inputs for 15 minutes before and during the MK801 treatment (10 min). After extensive wash by normal ECS to remove any trace of MK801 that is not trapped in opened N M D A R s , the coverslip with treated cortical neurons was transferred to a recording chamber for whole-cell patch clamp recording. Extrasynaptic N M D A R s in voltage-clamped cortical neurons were activated by N M D A (200 uM) using a fast perfusion system (Warner). 2.6.3. Recording of field EPSCs (fEPSCs) in hippocampal slices Acute hippocampal slices (400 um) were prepared from SD rats of 20-36 days old using a vibratome. Hippocampal slices were perfused at room temperature with artificial cerebrospinal fluid (ACSF) containing (in mM) 126 NaCl , 3 K C l , 2 M g C l 2 , 2 C a C l 2 , 1.2 K H 2 P 0 4 , 26 NaHCC-3 and 10 glucose and bubbled with 95% 0 2 and 5% C 0 2 . Field potentials were recorded with glass micropipettes (2-4 MQ) filled with A C S F placed in the striatum radiatum 60-80 um from the cell body layer. Synaptic responses were evoked by stimulation (0.05 ms) of the Schaffer collateral-commissural pathway with a bipolar tungsten electrode in the presence of bicuculline methiotide (10 uM). 2.7. Peptide construction and delivery A l l the peptides used were rendered cell-permeant by fusing each to the cell-membrane transduction domain of the human immunodeficiency virus-type 1 (HIV-1) Tat protein ( Y G R K K R R Q R R R ) (Schwarze et al., 1999). The wild-type NR2B peptide (Tat-NR2B9c) is a fusion protein comprising Tat and the last 9 residues of the N R 2 B subunit (KLSSIESDV). The control peptides for NR2B9c include: Tat38-48 (comprising HIV-1 40 Tat residues 38-48 outside the transduction domain), Tat-AA (comprising Tat and two alanine residues), and Tat-NR2B-AA (containing the nine amino acids of the C O O H -terminal of NR2B but with a double point mutation in the PSD-95 binding motif, K L S S I E A D A ) . The pTat-PDZl-2 and pTat-GK fusion proteins were generated by insertion of PSD-95 residues 65-248 encoding the first and second P D Z domains (PDZ1-2), residues 534-724 encoding the guanylate kinase-like domain, respectively, into pTat-H A plasmids. Fusion proteins contained a 6X His-tag, Tat and a hemaglutinin-tag N -terminal to the insert. Plasmids were transformed into BL21 (DE3) LysS bacteria (Invitrogen) and recombinant proteins were isolated under denaturing conditions on a Nickel-His column. To visualize the delivery of Tat-NR2B9c into the brain, fluorescent Tat-NR2B9c-dansyl and cell-impermeant Tat38-48-dansyl were injected (i.p.) into C57BL/6 mice (25 g) (3 nmol/g). 1 hour later, the animals were perfused with fixative solution (3% paraformaldehyde, 0.25% glutaradehyde, 10% sucrose, 10 U/ml heparin in saline). Brains were removed, frozen in 2-methylbutaneat (-42 \u00C2\u00B0C) and 40 p M sections were cut with a cryostat. Coronal sections were examined for dansyl fluorescence by UV-laser confocal microscopy. 2.8. Construction of cDNAs of N R 2 A and N R 2 B C-terminals Two pairs of primers, ( 5 ' ) C G C G G A T C C G A G C A C C T C T T C T A C T G G A A G ( 3 ' ) , (5') G G C G G A T C C T T A A A C A T C A G A T T C G A T A C ( 3 ' ) and ( 5 ' ) C G C G G A T C C G A G -C A T C T G T T C T A T T G G C A G ( 3 ' ) , ( 5 ' ) C G C G G A T C C T C A G A C A T C A G A C T C A A T A C (3') were used to amplify the C-terminals of N R 2 A (residues of 837 to 1465) and NR2B 41 (residues 838 to 1482), respectively. The PCR amplified C-terminal of N R 2 A or NR2B was then inserted into the BamHI sites of m R F P - C l expression vector. Sequencing was performed to confirm the constructs. 2.9. R N A i (RNA interference) The siRNAs for R N A i comprised 19 nucleotides with a 2-nucleotide 3' overhang. The duplex for silencing N R 2 A m R N A (nucleotides 3182-3200) was: sense: (5') C U C U C A A U G A G U C C A A C C C A A (3'), with a phosphate group at the 5' end; anti-sense: (5') G G G U U G G A C U C A U U G A G A G U G (3'). The duplex designed to target NR2B m R N A (nucleotides 2983-3001) was: sense: (5') A G G A G C G C C A A U C C G U G A U C U (3'), with a phosphate group at the 5' end; anti-sense: (5') A U C A C G G A U U G G C G C U C C U C U (3'). siRNAs were synthesized by Qiagen (Ontario, Canada). Neurons (DIV 9) cultured on 35mm dishes were transfected with either N R 2 A or NR2B siRNAs using the Transmessenger Transfection kit (Qiagen) and following the manufacturer's suggestions. Briefly, diluted Enhancer R in the appropriate volume of Buffer E C - R and then added s iRNA (the ratio of nucleic acids to Enhancer R was 1:8). The siRNA-Enhancer R mixture was incubated briefly (2-5min) at room temperature. TransMessenger Transfection Reagent was added to the mixture and incubated for another 10 min at room temperature to allow formation of the nucleic acid-TransMessenger Reagent complexes. Diluted the complexes with equal volume of culture medium and added drop-wise to the neurons. After incubation for 3 hours under normal growth conditions, neurons were washed l x with PBS, and maintained under original culture conditions for 48 h before further analysis. 42 2.10. Transfection cDNAs of N R 2 A or NR2B carboxyl tail were transfected into neurons (DIV 9) cultured on 1.8 cm coverslips using the CalPhos mammalian transfection kit (Clontech). Briefly, for each coverslip, 2 pg of c D N A was first mixed with C a C l 2 solution, and then added to HEPES-buffered saline. This transfection solution was added drop-wise to the neurons. After incubation for 1-3 hours in a CO2 incubator at 37 \u00C2\u00B0C, the transfection was terminated by further incubation with culture medium (pre-equilibrated in a 10% CO2 incubator) for 20 min. The coverslips were then transferred back to the original culture medium and maintained under normal growth conditions. 48 h later, the transfection efficiency for neurons were examined. 2.11. Western blot For detection of Akt phosphorylation and caspase 3 activation, neurons were lysed using modified RIPA buffer in the presence of protease inhibitors 12 hours after treatments, and proteins were extracted. 40 pg of total proteins were subjected to 10% and 15% S D S -P A G E for probing of phospho-Akt and cleaved caspase 3, respectively, and then immunoblotted with respective antibodies according to the manufacturer's instruction. Except for the P-tubulin antibody (Sigma), antibodies for phospho-Akt (Ser473), Akt and cleaved caspase 3 (Aspl75) were purchased from Cell Signaling. For sequential re-probing of the same blots, the membranes were stripped of the initial primary and secondary antibodies and subjected to immunoblotting with another antibody. To detect N R 2 A and NR2B expressions after R N A i , proteins were extracted for western blotting 48 hours after s iRNA transfection, and 20 pg of total protein were separated on 8% S D S -43 P A G E gels and probed with anti-NR2A or -2B antibody (US Biological). Detection of proteins was achieved using horseradish peroxide (HRP)-conjugated secondary antibodies and enhanced chemiluminescence (Amersham). 2.12. Co-immunoprecipitation (Co-IP) Rat brain proteins were prepared under weakly denaturing conditions known to permit NMDAR/PSD-95 interaction. Adult Wistar rat forebrains were removed and homogenized in ice-cold buffer (0.32 M sucrose, O. lmM Na3V04 , 0.02 M p-nitrophenyl phosphate, 0.02 M glycerol phosphate, 0.1 m M PMSF, and 5 ug/ml each of antipain, aprotinin, and leupeptin). Homogenates were centrifuged at 800 x g for 10 min at 4 \u00C2\u00B0C. The supernatants were centrifuged at 11,000 x g at 4 \u00C2\u00B0C for 20 min and the pellets (P2) were resuspended in homogenization buffer. P2 membranes, a fraction enriched with synaptic structures, were adjusted to 200 \ig protein/90 ul with homogenization buffer, and D O C and Triton X-100 were added to final concentrations of 1% and 0.1% respectively. After 30 min at 37 \u00C2\u00B0C, suspensions were centrifuged at 100,000 x g a v for 10 min and the supernatants were used for co-IP. Tat peptides (100 uM) were incubated with 200 ug of rat forebrain lysate for 1 hour at 37 \u00C2\u00B0C. Following the protocol described above, PSD-95 was precipitated from rat forebrain extracts using anti-NR2A and NR2B antibodies, which were generated against the amino acid residues 934-1203 of the N R 2 A protein and residues 935-1455 of the NR2B protein, respectively. Proteins were then separated on 8% SDS-PAGE gels and probed with the appropriate antibodies (See Western blot). 44 2.13. Calcium imaging Ratiometric measurements of free intracellular calcium concentration [Ca 2 +]i were performed in primary cultured neurons loaded with the fluorescent C a 2 + indicator fura-2 / A M . N M D A (250 p M dissolved in Mg 2 +-free solution) was applied by pressure injection ( I s , every 90 s) from a micropipette placed near the cell soma. Evoked calcium responses were recorded before and after bath application of Tat-NR2B9c (50 n M , 20 min). Rate of [Ca 2 +]i increase was calculated from the time taken to rise from 5% to 95% of peak values. For each cell, peak ratio and rate of [Ca 2 + ]i was averaged from 4 responses obtained before and 4 after Tat-NR2J39c application. Data were expressed as before :after ratios of responses obtained before and after peptide application. 2.14. Data Analysis Data were expressed as Mean \u00C2\u00B1 S E M where appropriate. Relative difference in apoptosis was calculated by normalizing the acquired absorbance readings to control (set as 100%) and then subtracting control from all the normalized values. Analysis of Variance ( A N O V A ) was used for comparison among multiple groups, followed by Holm-Sidak test for comparison between two groups. Statistical significance was defined as p<0.05. 45 C H A P T E R 3: R E S U L T S Part I: NR2A- and NR2B-containing NMDAR subtypes have opposing roles in excitotoxicity 3.1.1. Summary N R 2 A - and NR2B-containing N M D A R subtypes may dictate the polarity of synaptic plasticity (Liu et al., 2004a;Massey et al., 2004b); however, whether they play differential roles in excitotoxicity is unknown. In this study, I investigated the roles of these two major N M D A R subtypes in mature cortical neurons in neuronal apoptosis. I found that, while NR2B-containing N M D A R subtype mediates neuronal apoptosis, N M D A R s containing N R 2 A play an opposing role, i.e. activation of N R 2 A is anti-apoptotic. The distinct roles of these two N M D A R subtypes are not affected by their subcellular (synaptic versus extrasynaptic) location. Taken together, N R 2 A - and NR2B-containing N M D A R subtypes may play opposing roles in excitotoxicity. These discoveries suggest that general N M D A R antagonism may hinder the pro-survival action of N R 2 A -containing N M D A R s and cause neuronal death, and hence may explain, at least partially, the failure of N M D A R antagonists in the treatment of stroke-related brain injury in clinical studies and lay the foundation for developing novel NMDAR-based therapies for stroke. 46 3.1.2. NR2A- and NR2B-containing NMDARs play differential roles in neuronal apoptosis 3.1.2.1. NVP AAM077 shows a relatively high selectivity for NR2A-containing NMDARs in the present study In the adult forebrain, the two major N M D A R subtypes are N R 2 A - and NR2B-containing N M D A R s . To distinguish the roles of these N M D A R subpopulations in neuronal apoptosis in mature cortical neurons, we took advantage of two subunit-specific N M D A R antagonists, N V P A A M 0 7 7 (NVP) and Ro 25-6981 (Ro) for N R 2 A - and NR2B-containing N M D A R s , respectively. Ro is highly selective for NR2B-containing N M D A R s and has been widely used in many studies to identify the unique properties of NR2B-containing N M D A R s . The specificity of the newly developed N R 2 A antagonist N V P , however, has been under debate since its discovery. On one hand, Auberson and colleagues (Auberson et al., 2002) showed that N V P has a 130-fold preference for N R 2 A relative to N R 2 B . This relatively high selectivity was also observed in other studies (Liu et al., 2004a;Massey et al., 2004b). On the other hand, Feng et al. found that N V P not only has a poor selectivity for N R 2 A (only 13-fold over NR2B), but also acts as a potent antagonist for NR2C- and NR2D-containing receptors (Feng et al., 2004). More recent studies also suggested that N V P might also inhibit NR2B-containing N M D A R s substantially, especially when applied before agonists (Berberich et al., 2005a;Weitlauf et al., 2005b). Due to the questionable selectivity of N V P and the developmental change in N R 2 A and NR2B expression, we first determined i f both subtypes of N M D A R s exist in the neurons used in the present study (cortical cultures of 11-14 days in vitro (DIV)) and i f N V P functions as a N R 2 A specific antagonist. 47 To do this, we examined the ability of these antagonists to inhibit whole-cell currents evoked with a rapid and brief application of N M D A (50 p M N M D A , 10 p M glycine, 5 p M strychnine). As shown in Fig. la , bath application of either N V P (0.4 pM) or Ro (0.5 pM) alone produced a partial, but significant, blockade of the NMDA-induced currents. To determine the potentially non-selective, overlapping blockade of these antagonists, we applied the two antagonists sequentially and compared the degree of blockade produced by each antagonist when it was applied alone and following the blockade by the other antagonist. As shown in Fig. l a and b, N V P produced similar blockage when applied either alone (42.9% \u00C2\u00B1 5.9%) or after blockade with Ro (43.4% \u00C2\u00B1 12.4%), confirming that at the respective concentrations used here, Ro is a very specific antagonist to N R 2 B subunit with little effect on NR2A-containing receptors (Mutel et al., 1998;Fischer et al., 1997a) and N V P can effectively block NR2A-containing receptor-mediated current (Liu et al., 2004a;Massey et al., 2004b;Tigaret et al., 2006a). However, following N V P blockade, the percentage of N M D A current inhibition by Ro was reduced proximately by 5.6% (34.6% \u00C2\u00B1 1.8% when applied alone vs 29.0% \u00C2\u00B1 3.3% after N V P , P>0.05). Although the reduction is not significant, it may reflect a small degree of cross-inhibition of NR2B receptors by N V P in these neurons under our experimental conditions, as suggested by several recent studies (Weitlauf et al., 2005a;Berberich et al., 2005b;Tigaret et al., 2006b). However, several recent studies have demonstrated that such a small percentage of contaminant NR2B inhibition does not significantly affect the utility of N V P as an NR2A-subunit preferential antagonist under experimental conditions such as in the present study (Liu et al., 2004b;Massey et al., 2004a;Tigaret et al., 2006c). Together, our results indicate that both N R 2 A - and NR2B-containing receptor subtypes 48 are expressed in these neurons and under our experimental conditions, the two antagonists selectively block the expected receptor subtypes with little cross-receptor subtype antagonism. 49 a 0.5 MM RO25-6981 + 0.4 MM NVP-AAM077 0.4 |JM N V P - A A M 0 7 7 + 0.5 MM RO25-6981 Control 500 pA 500 ms > ~ 6 0 \u00C2\u00B0 / o 1 .n n TJ \u00C2\u00A3 50%-.2 \" 40%-\u00E2\u0080\u00A2o <2 p S 30% C Q . oo o o o o o O/ H 4 * i\u00E2\u0080\u0094i * * * * * * Fig . 2a-b. Activation of NR2A- or NR2B-containing NMDARs exerts differential effects on neuronal apoptosis. a) Representative images illustrate the differential roles of NR2A- and NR2B-containing NMDARs in NMDA-induced neuronal apoptosis. Mature neurons (DIV 11-14) were treated with 50 pM NMDA plus 10 pM glycine for 20 min in the presence of NR2A-specific antagonist NVP-AAM077 (NVP; 0.4 uM) or NR2B-specific antagonist Ro 25-6981 (Ro; 0.5 uM), and stained with Hoechst-33342 20 h following indicated treatments. NMDA stimulation-produced neuronal apoptosis characterized by chromatin condensation and/or fragmentation was aggravated in the presence of NVP (NVP+NMDA), but eliminated in the presence of Ro (Ro+NMDA). b). Quantitative measurement confirmed the differential effects shown in a). Cell death ELISA assay for apoptosis was performed 20 h after the indicated treatments. Data are presented as the difference in apoptosis levels as a percentage of control. *** denotes p < 0.001 compared with non-treated control; # and ### denote p < 0.05 and p < 0.001, respectively, compared with NMDA treatment alone; n = 18 tissue culture wells from three separate experiments for each group. 56 3.1.2.2. Examination of the roles of NR2A- and NR2B-containing NMDARs in neuronal apoptosis by RNAi and over-expression of NR2 subunit C-terminal Pharmacological tools may have limitations when used alone to characterize the function of NMDARs (Neyton and Paoletti, 2006). To confirm that results obtained by using subunit-specific NMDAR antagonists, I attempted to use other methods such as RNA interference (RNAi) and over-expression of the NR2A or NR2B carboxyl tail in neurons. RNAi can silence proteins without knocking out the genes that encode such proteins. Thus, I tried to acutely knock down NR2A or NR2B in cortical cultures using small interfering RNAs (siRNAs) derived from the sequence of NR2A or NR2B mRNA. Unfortunately, in my hands this strategy was not able to specically and efficiently knock down the expression of NR2A or NR2B (data not shown). The C-terminal of NR2 subunits is very important for the function of NMDARs (Sprengel et al., 1998) since it contains the domains that allow interactions with other intracellular proteins critical for NMDAR signaling (Sheng and Pak, 2000;Kohr et al., 2003;Niethammer et al., 1996b). Therefore, disrupting these receptors from their downstream signaling pathway may be an alternative to pharmacological approaches to examine the functions of NR2A- and NR2B-containing receptors. To do this, I attempted to over-express the C-terminal of NR2A or NR2B in the cortical cultures. The cDNA encoding the full length of NR2A or NR2B C-tail was inserted into the mRFP-Cl expression vector and transfected into neurons (DIV 9-11) using calcium phosphate method (Jiang et al., 2004). Unfortunately, the transfection efficiency of the cDNA of NR2A or NR2B C-tail was near zero (data not shown), rendering it impossible to carry out further experiments. 57 Because of time limitation, I was not able to optimize the experimental conditions. Nonetheless, both R N A i and over-expression of N R 2 subunit C-tail remain potentially good methods to address the issues about the role of N M D A R subtypes in excitotoxicity. 58 3.1.3. NR2A- and NR2B-containing N M D A R s are coupled to distinct signaling pathway As discussed, cell survival and death are mediated by distinct mechanisms. Thus, we next examined the mechanisms underlying the opposite effects of N R 2 A - and N R 2 B -containing N M D A R s on neuronal fate. Since phosphatidylinositol 3-kinase (PI3K)-Akt pathway has been shown to promote survival in manifold types of cells including neurons (Brunet et al., 2001;Sutton and Chandler, 2002) whereas caspase 3-dependent apoptotic pathway is activated in NMDAR-mediated neuronal apoptosis (Allen et al., 1999;Tenneti and Lipton, 2000), I probed the levels of phosphorylated Akt and cleaved (activated) caspase-3 12 h after the same treatments as in Fig. 2a and b by Western blotting. Similar to previous reports (Allen et al., 1999;Tenneti and Lipton, 2000;Puka-Sundvall et al., 2000;Sutton and Chandler, 2002;Zhu et al., 2002;Gines et al., 2003), we found that N M D A treatment (50 p M , 20 min) decreased Akt phosphorylation (Ser473) but increased caspase 3 (Asp 175) cleavage. When NR2B-containing N M D A R s were selectively blocked by Ro (0.5 pM), however, N M D A treatment enhanced Akt and lessened caspase 3 activation, confirming that stimulation of NR2A-containing N M D A R s facilitates neuronal survival. Consistent with the effects on neuronal apoptosis, specific activation of N M D A R s containing NR2B by blocking N R 2 A with N V P (0.4 uM) during N M D A treatment further diminished NMDA-induced reduction in Akt phosphorylation while elevated the level of cleaved caspase 3 (Fig. 2c). These findings suggest that the distinct roles of N R 2 A - and NR2B-containing N M D A R subtypes are probably due to differential downstream signaling. The exact intermediate steps of these signal transduction pathways are yet to be studied. 59 P-Akt Akt Casp3 P-Tubulin Fig . 2c. Activation of NR2A- and NR2B-containing NMDARs triggers neuronal survival (Akt) and apoptotic (caspase-3) pathways, respectively. Analysis of phosphorylated Akt (P-Akt; Ser473) and cleaved caspase 3 (Casp 3; Aspl75) levels by Western blotting indicates that Akt phosphorylation was increased following stimulation of NR2A-containing NMDARs (Ro+NMDA) but decreased after stimulation of NMDARs containing NR2B (NVP+NMDA). In contrast, caspase-3 cleavage shows a changing pattern opposite to that of Akt phosphorylation. 61 3.1.4. The subunit composition rather than the subcellular location determines the role of NMDARs in excitotoxicity In a recent study, Hardingham et al. reported that selective activation of synaptic and extrasynaptic N M D A R s mediates neuronal survival and death, respectively (Hardingham et al., 2002). However, most recent evidence indicated that activation of synaptic N M D A R s can lead to neuronal death (Bellizzi et al., 2005a). Furthermore, available immunochemical and electrophysiological evidence suggests that at least in the mature neurons of rat hippocampus and cortex, N R 2 A - and NR2B-containing receptors are differentially expressed at synaptic and extrasynaptic sites (Tovar and Westbrook, 1999;Stocca and Vicini , 1998). These data raise the possibility that subunit composition, rather than the subcellular (synaptic versus extrasynaptic) location, of N M D A R s mediates the opposing actions observed in the present study. To verify such a possibility, we functionally mapped the expressions of N R 2 A and NR2B-containing N M D A R s at synaptic and extrasynaptic sites and investigated their roles in promoting cell survival or death in cultured cortical neurons. 3.1.4.1. Synatpic NR2A- and NR2B-containing NMDARs mediate opposing effects on neuronal fate 3.1.4.1.1. A significant number of NR2B-containing NMDARs exist at excitotary glutamatergic synapses Although the vast majority of synaptic N M D A R s are NR2A-containing, in C A I neurons in hippocampal slices prepared from rats aged from 3 to 4 weeks, electrophysiological evidence clearly indicates that small proportion of functional NR2B-containing receptors 62 are expressed at synaptic sites (Liu et al., 2004a). Therefore, we first examined i f functional NR2B-containing receptors are also expressed at the synaptic sites of the cultured cortical neurons used in the present study using whole-cell recording of spontaneous mEPSCs. As shown in Fig. 3a, under this recording condition, mEPSCs comprise both fast, AMPAR-mediated component which was completely blocked by non-NMDAR antagonist 6,7-dinitroquinoxaline-2,3-dione (DNQX) (data not shown), and the slow, NMDAR-mediated component which was fully blocked by N M D A R antagonist A P V . Consistent with the presence of certain proportion of functional synaptic NR2B-containing receptors, the N M D A component was significantly reduced by bath application of the specific NR2B-containing N M D A R antagonist Ro (0.5 p M ; Fig. 3a-3 and 4). As mEPSCs are primarily mediated by synaptically localized receptors that were activated by glutamate spontaneously released from presynaptic terminals, the sensitivity to N R 2 B antagonist demonstrates that functional NR2B-containing N M D A R s are present within the glutamatergic synapses of the neurons under study. On average, the NR2B-containing receptor-mediated component accounts for 32.4 \u00C2\u00B1 3.6% of the synaptic N M D A currents (Fig. 3a-3 and 4) and the rest was primarily mediated by N R 2 A -containing receptors as it was largely eliminated in the presence of N R 2 A specific antagonist N V P (0.4 pM). Thus, similar to hippocampal C A I neurons prepared from brain slices (Liu et al., 2004a), both functional N R 2 A - and NR2B-containing subpopulations of N M D A R s , albeit the former being predominant, are expressed at the synapses of the cultured neurons used herein. 63 3a Fig. 3 a. Functional synaptic NR2B-containing NMDARs are present in cultured cortical neurons. Spontaneous mEPSCs were recorded in whole-cell voltage-clamp mode at a holding membrane potential of -60 mV in the presence of tetrodotoxin (0.5 uM) and bicuculline (10 uM) with zero added Mg 2 + . a-1. Examples of mEPSC traces (averaged from 100 individual events) obtained in the absence (Control) and presence of Ro (0.5 uM) or the broad spectrum NMDAR antagonist APV (APV; 50 uM). a-2. Total NMDAR-mediated component of mEPSCs was obtained by subtracting the averaged mEPSC recorded in the presence of APV from the averaged control mEPSC (Control-APV; shaded region), a-3. The NR2B-containing receptor component was obtained by subtracting the averaged mEPSC recorded in the presence of Ro from the averaged control mEPSC (Control-Ro; shaded region). a-4. Bar graph summarizes data obtained from five individual neurons (n=5). Charge transfer is equivalent to the area of the shaded regions. 65 3.1.4.1.2. The pro-apoptotic effect of synaptic NR2B-containing NMDARs is unmasked by blockade of NR2A-containing counterparts After establishing the presence of both N R 2 A - and NR2B-containing receptors at the synaptic sites, I examined the function of the two subpopulations of synaptic N M D A R s in mediating neuronal survival or death. I reasoned that i f the location of the receptors is critical, activation of either receptor population should produce effects of promoting neuronal survival. But, i f the subunit composition is the determinant, one would expect that the two populations, although both being synaptically localized, should have opposing actions. To increase activation of synaptic N M D A R s by synaptically released glutamate, neurons were incubated with G A B A A receptor antagonist bicuculline (50uM) for 4 hours. Bicuculline increases neuronal excitation by blocking G A B A A receptor-mediated synaptic inhibition and thereby enhances action potential-dependent synchronized release of glutamate from presynaptic terminals (Hardingham et al., 2002). Neuronal apoptosis was quantified 20 h following the treatments. I found that stimulation of synaptic N M D A R s by application of bicuculline alone or in the presence of NR2B antagonist Ro (0.5 pM) did not cause apoptotic cell death (Fig. 3b). In contrast, blocking synaptic NR2A-containing receptors by co-application of N V P (0.4 pM) with bicuculline significantly increased neuronal apoptosis (p<0.001, Fig. 3b). The N R 2 A blockade-induced neuronal apoptosis was mediated by synaptic NR2B-containing receptors as it was prevented in the presence of Ro (0.5 u M ; Fig. 3b). 66 o/ Difference in Apoptosis (% relative to control) ho o CO o o cn o ''c * * * *H1 9 as Fig . 3 b . Enhanced activation of synaptic NR2A- and NR2B-containing NMDARs exerts opposing actions on neuronal fate. Potentiation of synaptic NMDAR activation was achieved by increasing the presynaptic release of glutamate through incubation of cultured neurons with bicuculline (Bic; 50 pM) for 4 h in the absence or presence of NR2-specific antagonists. Blockade of NR2A- (Bic+NVP), but not NR2B- (Bic+Ro) containing NMDARs increased neuronal apoptosis. The NR2A blockade-induced apoptosis was prevented by a further blockade of NR2B-containing receptors (Bic+NVP+Ro). *** p < 0.001 compared with control; n = 14-16 for each group from three separate experiments. 68 Since the increased action potential-dependent synaptic release of glutamate during bicuculline incubation may also activate extrasynaptic N M D A R s through glutamate spillover, I next examined the effects of blocking synaptic N M D A R activation by glutamate spontaneously released from terminal under basal, non-stimulated conditions. As shown in Fig. 3c, incubating neurons with N V P (0.4 pM) for 4 h did not cause significant neuronal apoptosis when compared with control cells. These results suggested that N V P has little toxic effect on the cortical cultures and as well confirmed that cell death induced by bath application of N R 2 A antagonist N V P for 20 min (Fig. 2a and b) is not due to the toxicity of N V P . While a short-term incubation of N V P had little effect on neuronal death, an extended incubation period of 48 h did produce a remarkable increase in neuronal apoptosis (Fig. 3d, pO.Ol ) , which was comparable with the level of apoptosis following N V P treatment under bicuculline-stimulated conditions shown in Fig. 3b. In contrast, blockade of synaptic NR2B alone up-to 48 h did not induce any neuronal apoptosis (Fig. 3d). Similarly, the synaptic N R 2 A antagonist-induced apoptosis was also prevented by the blockade of synaptic NR2B receptors with Ro (0.5 pM). Together, the data strongly suggest that synaptic NR2B-containing N M D A R s may mediate neuronal apoptosis, but under physiological conditions, this effect is overwhelmed by the anti-apoptotic activity of synaptic NR2A-containing counterparts. Therefore, our study indicated that at synapses, activation of N R 2 A - and NR2B-containing receptors have opposing roles in promoting cell survival and death, respectively. 69 Difference in Apoptosis (% relative to control) Difference in Apoptosis (% relative to control) \u00E2\u0080\u00A2 i cn o 0/ o cn o cn K> ro o cn Fig . 3c-d. Spontaneously activated synaptic NR2A- and NR2B-containing NMDARs have opposing roles in promoting neuronal survival and death, respectively. Although a relatively short incubation of neurons with NR2A-specific antagonist NVP (0.4 uM) or NR2B-specific antagonist Ro (0.5 uM) does not have effects on neuronal fate (c), an extended duration (48 h) of NVP but not Ro incubation (in the absence of bicuculline) was sufficient to produce an increase in apoptosis (d). The NVP-induced apoptosis was prevented by addition of Ro (NVP+Ro). These results indicated that both synaptic NR2A- and NR2B-containing subpopulations of NMDARs are spontaneously activated by presynaptically released glutamate, exerting counteracting effects on cell survival and death, but synaptic NR2A-containing receptor activation is predominant and required for maintaining normal neuronal survival. ** p < 0.01 compared with control; n=12-18 and n = 15-17 for each group from three separate experiments in Fig. 3c and 3d, respectively. 72 3.1.4.2. Extrasynaptically located NR2A- and NR2B-containing NMDARs also have disparate roles in neuronal fate. 3.1.4.2.1. NR2A-containing NMDARs exist at extrasynaptic sites albeit the predominance of NMDARs containing NR2B. In contrast to the predominant expression of NR2A-containing receptors at synapses, NR2B-containing receptors are thought to be primarily expressed at extrasynaptic sites in mature neurons (Massey et al., 2004b;Tovar and Westbrook, 1999). To determine i f some of NR2A-containing N M D A R s are also expressed at extrasynaptic sites in the neurons under study, we first pharmacologically blocked all N M D A R s expressed at synapses and then examined i f currents specifically gated through extrasynaptic N M D A R s are sensitive to N R 2 A subunit-specific antagonist. The selective blockade of synaptic N M D A R s was achieved by co-application of bicuculline (50 pM) and MK-801 for lOmin. As described above, bicuculline enhances synaptic release of glutamate and thereby selectively activates synaptic N M D A R s . MK801 , as an irreversible blocker of open N M D A R channels (Huettner and Bean, 1988), only blocks bicuculline-activated synaptic N M D A R s , and does not block extrasynaptic channels that are not activated during bicuculline stimulation. The complete blockade of synaptic N M D A R s by co-application of bicuculline and MK-801 was achieved within 10 min of drug applications as indicated by the virtual elimination of the slow and APV-sensitive component of mEPSCs (Fig. 4a-1 and 2). Different from what was observed by Tovar et al. (Tovar and Westbrook, 2002), we found little recovery of mEPSCs within one hour following wash out of the drugs. The currents gated through extrasynaptic N M D A R s were then induced by fast perfusion of N M D A (200pM) to the MK801 pretreated cells immediately after 73 washing out bicuculline and MK-801. The extrasynaptic NMDAR-mediated currents could be largely reduced by N R 2 B antagonist Ro (0.5 u M ; Fig. 4a-3 and 4), consistent with the idea that predominant extrasynaptic N M D A R s are NR2B-containing (Stocca and Vicini , 1998;Tovar and Westbrook, 1999). The small component of extracellular NMDAR-mediated currents that were resistant with NR2B antagonist were almost completely abolished by N R 2 A antagonist N V P (0.4 p M ; Fig. 4a-3 and 4), indicating the non-NR2B-containing extrasynaptic N M D A R s are exclusively NR2A-containing receptors. On average, about 26.6 \u00C2\u00B1 2.3% (n=5) of total currents gated by extrasynaptic N M D A R s was mediated by NR2A-containing receptors (Fig. 4a-4). Another experiment designed to confirm the results obtained involved reversing the order of N V P and Ro applications as shown in Fig. 4a-3, and then determining the percentage of extrasynaptic NR2A-containing N M D A R S . However, since the respective concentrations of N V P and Ro used in the present study show little cross-receptor blockage (Fig. l a and b), this experiment might be redundant. Overall, these results provided further evidence for the existence of functional N R 2 A -containing N M D A R s at extrasynpatic sties in the mature cultured cortical neurons. 74 4a 75 Fig. 4a. Functional NR2A-containing NMDARs are present at extrasynaptic sites. Whole-cell recordings were performed at a holding membrane potential of -60 mV. a-1. Averaged traces of mEPSCs showing an APV-sensitive (50 uM) NMDAR-mediated component (Control-APV). a-2. Averaged traces of mEPSCs showing the blockade of synaptic NMDARs by the open channel blocker MK-801 (10 uM plus 50 uM bicuculline, 10 min), as demonstrated by the elimination of the NMDAR-mediated component of the mEPSCs (Control-APV). a-3. Example traces of whole-cell currents evoked by NMDA (200 pM) following the blockade of synaptic NMDARs with MK-801 in the absence (control; A) or presence of Ro (0.5 uM; B) or Ro and NVP (0.4 uM; Ro+NVP; C). NMDAR-mediated currents were evoked by fast application of NMDA within 10 min of washing out MK-801 and bicuculline. Currents remaining following the blockade of extrasynaptic NR2B-containing receptors were virtually abolished by the addition of NVP, suggesting the presence of functional extrasynaptic NR2A-containing NMDARs in these neurons, a-4. Histogram shows the summarized data from 5 individual neurons. 76 3.1.4.2.2. Extrasynaptic NMDARs containing NR2A have pro-survival action which can protect neurons against NMDAR- and non-NMDAR-dependent apoptosis. Since we found that functional NR2A-containing N M D A R s are localized to extrasynaptic sites, the next question we wished to address was the role of extrasynaptic N R 2 A - and NR2B-containing receptors in mediating NMDA-induced cell survival and death. After specific blockade of synaptic N M D A R s and washing out bicuculline and MK-801, the neurons were treated with N M D A (50 pM) for 20min in the absence or presence of N V P (0.4 pM) or Ro (0.5 pM). Quantitative neuronal apoptosis assays performed 20h after the treatments showed that N M D A application alone (non-selective activation of extrasynaptic N M D A R s ) elicited significant apoptosis (pO.OOl, Fig. 4b), which could be prevented by selective blockade of NR2B-containing extrasynaptic N M D A R s with Ro. In sharp contrast, selective blockade of NR2A-containing receptors with N V P potentiated NMDA-mediated apoptosis (Fig. 4b, p<0.05, compared with N M D A ) . Thus, like synaptic N M D A R s , activation of extrasynaptic NR2A-containing receptors has a role in promoting neuronal survival that can counteract NR2B-mediated neuronal apoptosis. Taken together, the data illustrated in Fig. 3 and 4 strongly indicate that, regardless of their anatomical (synaptic versus extrasynaptic) locations, N R 2 A - and NR2B-containing receptors have opposing roles in mediating NMDA-elici ted neuronal survival and apoptosis and that these opposing roles are dictated by their subunit compositions, but not their anatomical localizations. 77 Difference in Apoptosis (% relative to control) 4^ ro o ro o 4^ CD CXI o o o o o to o O/ X * * * CO o T3 \u00E2\u0080\u0094s 0 i\u00E2\u0080\u0094*-<3 03 3 a> 00 Fig . 4b. Activation of extrasynaptic NR2A-containing NMDARs protects against neuronal death mediated by extrasynaptic NR2B-containing NMDARs. Excitotoxic neuronal death was induced in cortical neurons by bath application of NMDA (50 pM, 20 min) after the blockade of synaptic NMDARs with MK-801 plus bicuculline, and cell death was assayed 20 h later. NMDA elicited neuronal apoptosis, which was exacerbated when the NR2B-containing component was selectively stimulated (NVP+NMDA), but eradicated when the NR2A-containing component was specifically activated (Ro+NMDA). ** p < 0.01, *** p<0.001 compared with control, n = 11-12 from two separate experiments for each group. 79 The finding that N R 2 A exerts a neuronal survival effect that counteracts NR2B-mediated neuronal apoptosis prompted us to further examine whether the NR2A-mediated neuronal survival effect may also be protective against neuronal damage caused by factors other than NMDAR-mediated excitotoxicity. To selectively stimulate extrasynaptic N R 2 A -containing receptors with bath application of N M D A , I first irreversibly blocked all synaptic N M D A R s with co-application of bicuculline and MK-801 as described previously and then the extrasynaptic NR2B-containing receptors by incubating the neurons in the presence of Ro (0.5 pM) throughout the entire course of the experiments. Under these conditions, bath application of N M D A alone did not increase neuronal apoptosis, confirming the effective blockade of NR2B-mediated pro-apoptotic actions. NMDA-independent neuronal death was then induced by incubation of the cultured neurons with staurosporine (STS), a potent apoptosis inducer (Budd et al., 2000). As shown in Fig. 4c, treatment with STS (100 n M , lh) alone triggered tremendous neuronal apoptosis. The STS-induced apoptosis was significantly reduced by a brief bath application of N M D A (200 u M ; 5 min) prior to STS treatment (p<0.001, compared with STS alone, Fig. 4c). This neuronal protection provided by N M D A pretreatment is mediated through extrasynaptic N R 2 A receptors as it was prevented by co-application of N V P (0.4 p M ; pO.OOl , compared with STS alone). Therefore, extrasynaptic NR2A-containing N M D A R s indeed possess neuronal survival-promoting capability that, once activated, is not only against NMDAR-dependent (NR2B-mediated apoptosis), but also NMDAR-independent neuronal damage. 80 Difference in Apoptosis (% relative to control) o o o o o o o o o o O/ - ! I L. H * * * * * * * * *= Fig . 4c. Activation of extrasynaptic NR2A-containing NMDARs can counteract NMDAR-independent apoptosis. Bath application of staurosporine (STS, 100 nM, 1 h), after blockade of synaptic NMDARs with pretreatment of MK-801 plus bicuculline and of extrasynaptic NR2B receptors in the presence of Ro (0.5 pM ), induced significant increase in neuronal apoptosis (Ro+STS). Brief application of NMDA (200 pM, 5 min) did not produce neuronal aopotosis on its own (Ro+NMDA), but significantly reduced the STS-induced neuronal apoptosis (Ro+NMDA+STS) and the NMDA-induced neuroprotective action was abrogated by co-application of NVP (0.4 pM; Ro+NVP+NMDA+STS). *** p < 0.001 compared with Ro treatment. ### p < 0.001 compared with Ro+STS treatment, n = 8-12 for each group from three separate experiments. 82 CHAPTER 3: RESULTS Part II: Treatment of stroke by selective activation of NR2A-containing NMDARs 3.2.1. Summary Since NR2A-con ta in ing N M D A R s has anti-apoptotic activity that can protect neurons against N M D A R - and non-NMDAR-dependent apoptosis, in this study I investigated the effects o f selective activation o f NR2A-con ta in ing N M D A R s on hypoxic/ischemic neuronal death in vitro ( O G D ) and in vivo ( M C A o ) . The data showed that, although pretreatment with NR2B-spec i f i c N M D A R antagonist can protect against ischemic injury, post-ischemic inhibition o f N R 2 B has a narrow therapeutic window (ineffective when administered 4.5 h after stroke onset). In striking contrast, specific stimulation o f N R 2 A -containing N M D A R s can remarkably reduce ischemic brain damage when carried out either prior to or, most importantly, 4.5 h fol lowing ischemic insults. Therefore, treating stroke by selectively activating NR2A-con ta in ing N M D A R s is a strategy fundamentally different from general N M D A R antagonism, and may represent a novel therapy for stroke. 3.2.2. Selective activation of NR2A-containing reduces hypoxic neuronal injury in vitro The results shown in my previous experiments clearly indicate that activation o f N R 2 A -containing N M D A R s may play a critical role in protecting against neuronal damage during brain insults such as stroke in which both N M D A - and n o n - N M D A dependent neuronal injuries have been implicated. Therefore, I reasoned that selective activation 83 of NR2A-containing N M D A R s may represent a novel therapy for stroke that is completely different from general N M D A R antagonism. To investigate whether activation of NR2A-containing N M D A R s can reduce brain damage resulting from ischemic insults, I first compared the effects of selective blockade of N R 2 A - and NR2B-containing N M D A R s on neuronal apoptosis in a well characterized in vitro stroke model, O G D (Goldberg and Choi, 1993). Mature cortical cultures of 11-14 DIV were subject to an anaerobic atmosphere for 1 hour in a glucose-free solution in the absence or the presence of either N V P (0.4 uM) or Ro (0.5 uM). Neuronal apoptosis was quantitatively determined 20 h after O G D . As shown in Fig. 5a, 1 h O G D was able to produce a pronounced increase in neuronal apoptosis. As I expected, selective inhibition of NR2A-containing N M D A R s with N V P significantly enhanced OGD-induced neuronal apoptosis (p<0.05, compared with OGD), but in contrast, specific blockade of the NR2B-containing N M D A R s by Ro markedly attenuated neuronal apoptosis (p<0.001, compared with O G D , Fig. 5a). As discussed previously, blockade of NR2A-containing N M D A R s virtually results in activation of the NR2B-containing counterparts in these mature cortical cultures, and vice versa; therefore, the data suggested that stimulation of the pro-survival action of N R 2 A does offer neuroprotection during stroke in vitro. 84 Difference in Apoptosis (% relative to control) o M -f^ CT) OO O O O O o ro o o O/ Q Q * * * _i i i_ * * * Fig. 5a. Pretreatments with NR2A- and NR2B-specific antagonists respectively promote neuronal survival and death in stroke in vitro. Cortical cultures were challenged with a 1-h OGD and apoptosis was assayed 23 h after the challenge. OGD resulted in a significant increase in neuronal apoptosis compared with non-challenged controls (Control) and the OGD-induced apoptosis was potentiated by the NR2A specific antagonist NVP (0.4 pM; NVP+OGD) and inhibited by the NR2B antagonist Ro (0.5 pM; Ro+OGD) when bath applied 10 min prior to, and during, the OGD challenge, indicating NR2A- and NR2B-containing receptors exert opposing effects in anoxic neuronal death in vitro. *** p < 0.001 compared with control. # p < 0.05, ### p < 0.001 compared with OGD. n = 17-18 for each group from three separate experiments. 86 3.2.3. Selective activation of NR2A reduces ischemic brain damage in vivo 3.2.3.1. Pretreatment with NR2A-specific antagonist has deleterious effects on ischemic brain damage in rats Following the effects observed in the in vitro stroke model, I determined whether these findings could be reproduced in vivo using a rat focal ischemic stroke model - middle cerebral artery occlusion (MCAo) (Longa et al., 1989). The animals were infused with N R 2 A specific antagonist N V P (2.4mg/kg, personal communication with Auberson Y.P.) or N R 2 B specific antagonist Ro (6mg/kg (Loschmann et al., 2004)) or vehicle (saline) through intravenous injection (i.v.) 30 min prior to stroke (MCAo) onset and then subjected to a 1-h transient ischemic stroke induced by M C A o . This relatively short duration of ischemia was chosen to unmask the potential contribution of NR2A-mediated neuroprotective effects. Neurological score and cerebral infarction was examined 24 h after M C A o onset (see Methods and Materials). We found that the infarct areas and the total infarct volume were significantly increased by blockade of NR2A-containing N M D A R s through pre-treatment with N V P , but, in sharp contrast, remarkably reduced by N R 2 B antagonism (Fig. 5b and c). Specifically, when compared with saline-treated animals, N V P pre-treatment gave rise to a 67.0 \u00C2\u00B1 17.9% increase in total infarct volume (n=5; p<0.05), while Ro treatment decreased the total infarct volume by 67.8 \u00C2\u00B1 4.3% (n=6; p O . O l ; Fig. 5b and c). Neurological behavioral test also showed that the N V P -treated animals exhibited a trend toward aggravation while Ro treatment produced a protective effect (Fig. 5d). Taken together, these observations indicate that N R 2 A - and NR2B-containing N M D A R subtypes play opposing roles in stroke-induced brain damage in vivo and blockade of NR2A-containing N M D A R s may have negative effects on 87 excitotoxic brain injury. These findings also suggested that activation of NR2A-containing N M D A R s may counteract brain damage resulting from stroke. It is noteworthy that the anesthetic isoflurane used in this experiment (and the following in vivo experiments as well) may reduce NMDAR-induced excitotoxicity (Harada et al., 1999). To make the results obtained comparable, we anesthetized all the animals in the same manner during the surgery. Therefore, the effects observed should only be attributable to the specific treatment in each experimental group. 88 Infuse drug 30min Start MCAo Start reperfusion Neurol, testing T I A Sacrifice 5b o o Fig . 5b-d. Activation of NR2A- and NR2B-containing receptors exerts opposing effects on ischemic neuronal injuries in vivo. Adult rats were subjected to a 1-h focal cerebral ischemia produced by middle cerebral artery occlusion (MCAo), and cerebral infarction was assessed 24 h after MCAo onset. Intravenous infusion 30 min before MCAo onset of NVP (2.4 mg/kg; NVP+MCAo; n = 5) and Ro (6 mg/kg; Ro+MCAo; n = 6) respectively increased and decreased both infarct area (b) and total infarct volume (c). * p < 0.05, ** p < 0.01 compared with MCAo. d. Neurological scores assessed 24 h after stroke onset in the same groups of animals shown in (b) and (c) indicate that blockade of the NR2A-containing NMDARs resulted in a trend toward worsening neurological function, whereas blockade of NMDARs containing NR2B markedly improved neurological behavior. ** p < 0.01 compared with MCAo. 91 3.2.3.2. Postischemic potentiation of NR2A-containing NMDARs is neuroprotective 3.2.3.2.1. Glycine can exert neuroprotection through activation of NR2A-containing NMDARs A practical therapy for stroke would be one that can be implemented following the onset of stroke. Since this study indicates that NR2A-containing N M D A R subtype has a role in promoting neuronal survival, I reasoned that selective activation of such N M D A R subtype after stroke attack would reduce brain damage. The simplest way to activate NR2A-containing N M D A R s specifically is to use a highly selective N R 2 A agonist. However, to date no NR2A-specific agonists have been identified. To circumvent this, an alternative strategy was designed. The N M D A R co-agonist glycine has been shown previously to potentiate the activation of N M D A R s by endogenously released glutamate from presynaptic terminals in vivo (De et al., 2000). Moreover, our lab has demonstrated that exogenous application of a suprasaturating concentration of glycine (200 pM), can selectively activate synaptic N M D A R s and induce long-term potentiation (LTP) in cultured hippocampal neurons (Lu et al., 2001). Given that NR2A-containing N M D A R s are predominant at synapses, and that N R 2 A -containing N M D A R s may be critical for mediation of L T P (Liu et al., 2004a), the effects of glycine on L T P induction must be primarily mediated by potentiation of synaptic N M D A R s containing N R 2 A . Therefore, glycine may be a good agent to enhance activation of NR2A-containing N M D A R s and hence may boost neuronal survival. Indeed, our study indicated that, following a brief (10 min) pre-stimulation with glycine (300 pM) and strychnine (10 pM) in M g 2 + free ECS, cortical cultures challenged with N M D A (50 p M ; 20 min) exhibited significantly less neuronal apoptosis (p<0.05, 92 compare with N M D A alone; Fig. 6a). Strychnine is a highly specific glycine receptor antagonist to which the glycine-binding site on N M D A R s is insensitive (Jansen and Dannhardt, 2003). As a competitive antagonist, strychnine at 10 p M is more than sufficient to completely block the activation of glycine receptors by 300 p M glycine (Zhang et al., 2006). Thus, co-application of glycine ensured exclusive enhanced stimulation of the glycine site on N M D A R s . The survival-promoting effect of glycine was indeed mediated by activation of the NR2A-containing N M D A R subpopulation at synapses because, in the presence of NR2A-specific antagonist N V P (0.4 pM), glycine pretreatment no longer exhibited neuroprotective effects, whereas in the presence of NR2B-specific antagonist Ro (0.5 pM), glycine still exerted significant neuroprotection (pO.Ol , compared with N M D A alone; Fig. 6a). A n alternative explanation for the neuroprotective effect of glycine is that pretreatment by glycine leads to ensuing reduction in membrane N M D A R s (Nong et al., 2003), and hence NMDAR-mediated excitotoxicity. If the glycine effect was largely due to the decreased membrane N M D A R number, then co-application of N V P , which binds to N R 2 but not NR1 (containing glycine binding site), with glycine would not interfere with the neuroprotection by glycine. However, the data shown in Fig. 6a indicated that co-application of N V P virtually eliminated the neuroprotection by glycine pretreatment. Therefore, the endocytosis of N M D A R s primed by glycine may contribute minimally to the neuroprotective effect provided by glycine pretreatment in our study. 93 Difference in Apoptosis (% relative to control) ro ^ a> co o o o o Fig. 6a. Potentiated activation of NR2A-containing NMDARs by NMDAR co-agonist glycine exerts neuroprotection in vitro. NMDA-induced neuronal apoptosis was significantly reduced by a brief pre-treatment with glycine (300 uM) in the presence of strychnine (10 uM). However, this neuprotective effect was abolished when NR2A-specific antagonist NVP (0.4 uM) was co-applied (NVP+Gly-NMDA), indicating NR2A-containing NMDARs mediated the neuroprotection. In contrast, NR2B-specific antagonist Ro (0.5 uM) was not able to block the effect of glycine. * p<0.05, *** p < 0.001 compared with control. # p < 0.05, ## p < 0.01 compared with NMDA. n = 17-18 for each group from three separate experiments. 95 3.2.3.2.2. Pretreatment with glycine reduces OGD-induced neuronal apoptosis After establishing that bath application of glycine can protect neurons from N M D A -induced neuronal death in cortical cultures, I examined i f it was neuroprotective in the in vitro stroke (OGD) model. Prior to OGD, the neurons were treated briefly (10 min) with glycine (300 pM) in the presence of strychnine (10 pM). Quantification of neuronal apoptosis 20 h following O G D showed that glycine markedly, albeit only partially, decreased OGD-induced cell death (p<0.001, compared with O G D ; Fig. 6b). Thus, glycine can serve as a neuroprotectant for hypoxia-induced neuronal injury. 96 6b 70 i H i= 60 50 40 i 30 i 20 10 to ^ O P Q- \u00C2\u00A3= s \u00C2\u00B0 < o c ~ \u00E2\u0080\u0094 C CO 0 ID CD Control PO .001 * * * OGD Gly+OGD 97 Fig. 6b. G l y c i n e t r e a t m e n t r e d u c e s O G D - i n d u c e d n e u r o n a l a p o p t o s i s in vitro. N e u r o n a l a p o p t o s i s i n d u c e d b y O G D (1 h ) w a s s i g n i f i c a n t l y a m e l i o r a t e d b y a b r i e f p r e - t r e a t m e n t (10 m i n ) w i t h g l y c i n e (300 u M ) p l u s s t r y c h n i n e (10 u M ) , s u g g e s t i n g g l y c i n e c a n e x e r t n e u r o p o r t e c t i o n f r o m h y p o x i c n e u r o n a l i n j u r y i n v i t r o . * * * p < 0.001 c o m p a r e d w i t h c o n t r o l , n = 22-24 f o r e a c h g r o u p f r o m f o u r s e p a r a t e e x p e r i m e n t s . 98 3.2.3.2.3. Stimulation of NR2A-containing NMDARs following stroke onset protects the brain from ischemic injury in rats In most clinical situations, a stroke can only be treated after its onset. Thus, a therapy for stroke is not practical unless it can be carried out following stroke onset. Since we have shown selective activation of N R 2 A is effective in reducing ischemic brain damage before stroke onset, we wanted to determine whether post-ischemic potentiation of N R 2 A is also neuroprotective. Given that exogenous application of glycine can primarily activate NR2A-containing N M D A R s and exert neuroprotection (Fig. 6a and b), we attempted to use glycine to enhance activation of NR2A-containing N M D A R s . Although glycine is an endogenous co-agonist of N M D A R s and present in the rat brain at a concentration in the low micromolar range (Danysz and Parsons, 1998), the glycine site on N M D A R s may not be saturated in the brain (Danysz and Parsons, 1998;Javitt, 2006;Wood, 1995). Indeed, intraperitoneal injection of a high dose glycine (800 mg/Kg) can enhance the activity of N M D A R s in vivo (De et al., 2000). Therefore, we attempted to administer glycine to enhance N R 2 A activation in rats. To exclude the possibility that a large amount of NR2B-containing N M D A R s may be simultaneously activated by glycine, NR2B specific antagonist Ro was co-applied during glycine treatment. A l l the animals were subject to a 90-min transient M C A o . 3 h after reperfusion (4.5 h after stroke onset), the animals were treated with either vehicle (saline) or drug(s) through intraperitoneal injection (i.p.). As presented in Fig. 6c and d, while Ro alone (Ro, 6 mg/Kg, n=10) did not provide neuroprotection when compared with M C A o alone (control, n=10), co-administration of glycine (800mg/kg) with Ro (Gly+Ro, n=9) resulted in remarkable reduction in total 99 infarct volume assessed 24 h after M C A o onset (49\u00C2\u00B1 8%, p<0.001). In line with the lesser infarction, neurological testing 24 h following stroke onset also indicated that glycine improves neurological function (p<0.001; Fig. 6e). The efficacy of glycine is mediated through activation of NR2A-containing N M D A R s as co-administration of N V P (2.4mg/kg; Gly+Ro+NVP, n=10) was able to reverse both the decrease in infarction and the betterment of neurological behavior to the control (MCAo) levels (Fig. 6c, d and e). These results suggested that post-ischemic potentiation of the pro-survival action of NR2A-containing N M D A R s is beneficial for the recovery of brain injury and glycine is a potential neuroprotectant. 100 Infuse drug @ 3h T Neurol, testing T Start MCAo Start reperfusion Sacrifice 6d \u00C2\u00A9 Fig . 6 c - C Post-ischemic potentiation of NR2A-containing, but not blockade of NR2B-containing NMDARs reduces ischemic brain damage in an in vivo focal ischemic stroke model. Adult rats received intraperitoneal injection of either drug or saline 4.5 h after the onset of a 1.5-h MCAo challenge, c. Post-stroke treatment with NR2B antagonist Ro (6 mg/kg; n = 10) was no longer neuroprotective, whereas a post-stroke treatment with NMDAR co-agonist glycine (800 mg/kg) in the presence of Ro (6 mg/kg) significantly reduced total infarct volume (Gly+Ro; n = 9, p < 0.001) compared with saline controls (Control; n = 10). The glycine-induced neuroprotection was mediated by enhancement of NR2A-containing receptor activation as it was prevented by co-administration of NR2A antagonist NVP (Gly+Ro+NVP; n = 10). d. Representative brain sections from each treatment group stained with hematoxylin and eosin (H & E). Pale staining indicates infarct, e. Neurological score assessment indicated a significant improvement in neurobehaviour following glycine treatment in the same treatment group. *** p < 0.001 compared with Control. 104 CHAPTER 3: RESULTS Part III: Treatment of stroke via disruption of the neurotoxic pathway downstream of activation of NR2B-containing NMDARs The results presented herein are essentially the same as appeared in: Michelle Aarts*, Yitao Liu*, Lidong Liu, Shintaro Besshoh, Mark Arundine, James W. Gurd, Yu-Tian Wang, Michael W. Salter, Michael Tymianski (2002). Treatment of ischemic brain damage by perturbing NMDAR- PSD-95 protein interactions. Science 298:846-50 (see Appendix). This project was a collaborative study involving several laboratories of Canadian Stroke Network, but primarily carried out in Dr. Michael Tymianski's lab and ours. I was the primary investigator examining the effects of Tat-NR2B9c peptide on ischemic injury in M C A o stroke model in vivo. Due to my contribution to this study, I was assigned one of the two equal first authors. 3.3.1. Summary Our data indicate that NR2B-containing N M D A R s may be the major mediator of excitotoxicity and hence an ideal therapeutic target for stroke. However, the results obtained from in vivo study showed that blockade of NR2B has a relatively narrow therapeutic window, and furthermore, blockade of the normal activity of NR2B may lead to untoward side effects. A l l of these render direct inhibition of NR2B-containing N M D A R s an impractical treatment for stroke in clinic. Results from Dr. Tymianski's lab suggested that a post-synaptic density protein PSD-95 specifically couples N M D A R activation to excitotoxicity and that a small peptide derived from the last nine amino 105 acids of the carboxyal tail of NR2B subunit (KLSSIESDV) can selectively disrupt the interaction between NR2B-containing NMDARs and PSD-95 and reduce NMDA-induced neuronal death in vitro. Since perturbing the events downstream of NR2B-containing NMDAR activation may have a broader time window and fewer side effects (due to the avoidance of blocking NR2B-containing receptor functioning), this peptide may be used as an alternative to NR2B blockade to treat stroke. In this study we examined whether this peptide has neuroprotective effects in vivo. We found that this peptide significantly attenuates ischemic brain injury when delivered either before or, more importantly, after stroke onset. These findings suggested that perturbing protein-protein interaction downstream of NR2B-containing NMDAR activation may constitute a promising therapy for stroke as well. 3.3.2. Perturbing NR2B-PSD-95 interaction diminishes ischemic brain damage in vivo 3.3.2.1. Tat-NR2B9c can be delivered into the brain and perturb the interaction between PSD-95 and NR2B- but not NR2A-containing NMDARs (these experiments were done by Dr. Tymianski's lab) The NR2B C-tail peptide (NR2B9c) was made cell permeable by fusing to Tat protein (Tat-NR2B9c, See Methods and Materials). Tat-NR2B9c can transduce into cultured cortical neurons readily when bath applied and remain in the neurons for up to 5 h (Appendix, Fig. 1C). To test its feasibility for treatment of ischemic brain injury, they further examined whether this peptide could be delivered into the brain in intact animals when systematically administered (i.p.). C57BL/6 mice (-25 g) were given a 500-umol 106 dose of fluorescent peptide Tat-NR2B9c-dansyl or Tat38-48-dansyl (as a cell-impermeant control). Coronal brain sections were taken 1 h after injection and examined by confocal microscopy for fluorescent peptide uptake. They found that the animals injected with Tat-NR2B9c, but not Tat38-48-dansyl, exhibited strong fluorescence in the cortex (Appendix, Supplemental Figure 2A). Similar results were obtained when Tat-NR2B9c-dansyl was infused (i.v.) to rats, confirming that Tat-NR2B9c enters the brain upon peripheral administration. Next they tested whether NR2B9c could selectively dissociate the interaction between NR2B and PSD-95. The P2 membrane protein fraction of rat forebrain tissue, which is enriched in synaptic structures, was incubated with Tat-NR2B9c or with one of the three controls, Tat38-48, Tat-AA, or Tat-NR2B-AA, each lacking an intact PDZ binding motif (Kornau et al., 1995). Then they co-immunoprecipitated N R 2 A or N R 2 B protein with PSD-95 antibody. Western blot analysis showed that Tat-NR2B9c reduced the co-immunoprecipitation of PSD-95 with NR2B but not N R 2 A (Appendix, Fig. ID and E). This suggested that the NR2B9c peptide can selectively disrupt the binding of NR2B to PSD-95 and leave the association of N R 2 A with PSD-95 alone. 3.3.2.2. Pretreatment with Tat-NR2B9c decreases ischemic brain insults Based on the results from the Tymianski group, Tat-NR2B9c peptide attenuates N M D A -induced excitotoxicity in cultured cortical neurons in vitro. Given that it can be effectively delivered into the brain in the rats and selectively perturb N2B-PSD95 complexes in the brain tissue, we wished to determine whether NR2B9c could be used to treat stroke in vivo. 107 First, we determined whether pretreatment with this peptide effectively in reduced ischemic damage. Rats were pretreated 45 min before stroke (MCAo) by a single bolus injection (i.v.) with saline, the Tat-NR2B-AA control, or Tat-NR2B9c (3 nmol/g; n=6 for each group), and the extent of cerebral infarction was measured 24 h thereafter. Neuronal function testing was performed 60 min during M C A o and 10 min before animals were sacrificed. As shown in Appendix, Supplemental Figure 2C, pretreatment with Tat-NR2B9c reduced the total cerebral infarction volume by 54.6 \u00C2\u00B1 11.3% (pO.Ol) was largely accounted for by a 70.7 \u00C2\u00B1 11.2% reduction (p<0.01) in cortical infarction thought to be caused largely by NMDAR-dependent mechanisms. Moreover, this treatment also produced a trend toward improved 24-h neuorological scores (Appendix, Supplemental Figure 2B). In contrast, the control peptide Tat-NR2B-AA had effects on neither neurological function nor cerebral infarction. These results suggested that pre-stroke administration of Tat-NR2B9c ameliorates ensuing brain injury. 3.3.2.3. Post-treatment with Tat-NR2B9c also attenuates ischemic brain injury and improves neurological function A stroke therapy would be most therapeutically valueable i f effective when given after the onset of ischemia. Thus, we next evaluated whether treatment with Tat-NR2B9c could decrease ischemic damage in vivo when applied following stroke onset. Rats were subjected to transient M C A o for 90 min, and saline or Tat peptide (Tat-NR2B9c or Tat-N R 2 B - A A , 3 nmol/g) was injected (i.v.) 1 h after M C A o onset. Infarction volume and neurological outcome measurements were performed identical to the pretreatment study. Surprisingly, post-ischemic treatment with Tat-NR2B9c caused an even greater reduction 108 in the volume of total cerebral infarction by 67.0 \u00C2\u00B1 3.7% (Fig. 7a, pO.OOl , n=9) when compared with control group (saline, n=10). Similar to the pretreatment study, this reduction was accounted for by an 87.0 \u00C2\u00B1 4.4% reduction in cortical infarction volume (Fig. 7b, pO.OOl). These Tat-NR2B9c-treated animals also exhibited a significant improvement in 24-h neurological scores (Fig. 7c, p<0.001). In contrast, treatment with Tat-NR2B-AA (n=8) did not have any effects on cerebral infarction or neurological function (Fig. 7c). Thus, Tat-NR2B9c also boosts stroke recovery when given post-ischemically. 109 Stereotactic Coordinates (mm) 110 7b Cortical Infarct Stereotactic Coordinates (mm) i n 7c N e u r o l o g i c a l S c o r e i i C o n t r o l D u r i n g M C A o A f t e r M C A o ( 6 0 m i n ) (24hr ) 112 7d SALINE Tat-NR2B-AA Tat-NR2B9c \u00E2\u0080\u00A2 181 ^ A n ^ f * AHH Fig.\" 7. Neuroprotection by treatment with Tat-NR2B9c in vivo, a, b) Treatment with Tat-NR2B9c (3 nmol/g, n=9) but not mutated Tat-NR2B-AA (n=8) or saline controls (n=10) significantly reduced (a) total infarct area and volume (inset) and (b) cortical infarct area and volume (inset), measured 24 hours after transient MCAO. *** denote p<0.001, compared with control, c) Composite neurological scores during and 24 hours after MCAO. *** p< 0.001, compared with control, d) Representative appearance of hematoxylin and eosin-stained rat brain sections from each treatment group. 114 C H A P T E R 4: DISCUSSION In this thesis research project, the roles of the two major N M D A R subtypes in mature cortical neurons, N R 2 A - and NR2B-containing N M D A R s , in excitotoxicity were investigated. The results showed that, while NR2B-containing N M D A R s mediate neuronal apoptosis, the NR2A-containing counterparts have an opposing role\u00E2\u0080\u0094activation of N M D A R s containing N R 2 A can protect against neuronal death. It was also demonstrated that coupling to distinct downstream signaling pathways may underlie the mechanisms by which these two N M D A R subtypes function differentially. Based on these findings, two novel therapies for stroke were designed: the first was to selectively activate the NR2A-containing N M D A R s , and the second was to disrupt the signaling pathway downstream of activation of NR2B-containing N M D A R s . In in vitro and in vivo stroke models, both therapies were substantiated to be efficacious in reducing ischemic brain damage even when implemented following stroke onset. Thus, this study provides a molecular basis for the paradoxical roles of N M D A R s in promoting neuronal survival and mediating neuronal damage, and as well, at least partially, an explanation for the failure of N M D A R antagonists in previous clinical trials of stroke. Furthermore, the data suggest that NMDAR-based therapies can still be efficacious; however, rather than general N M D A R antagonism, a practical treatment should aim at selective enhancement of NR2A-containing N M D A R activation or perturbing the signaling pathway downstream of activation of N M D A R s containing NR2B. These novel therapies may represent a new direction in the treatment of stroke. 115 4.1 . The use of N V P A A M 0 7 7 as a selective antagonist for N R 2 A Auberson et al first showed that N V P has a 130-fold preference for recombinant human NR1/NR2A over NR1/NR2B receptors expressed in oocytes (Auberson et al., 2002). The selectivity of N V P was also confirmed by Liu et al. Using hippocampal cells, these authors found that at a concentration of 0.4 p M , N V P could effectively inhibit NR2A-containing N M D A R s while essentially sparing the NR2B-containing subtype (Liu et al., 2004a). On the other hand, Feng et al. found that N V P not only has a poor selectivity for recombinant rodent N M D A R s (only 13-fold over NR2B), but also acts as a potent antagonist for NR2C- and NR2D-containing receptors, which suggested, at least in rodents, N V P is not a subunit-specific antagonist whatsoever (Feng et al., 2004). Bererich et al. tested the effects of N V P at the same concentration used in our study (0.4 pM) and found that it blocked around 67% of NR1/NR2B receptors expressed in H E K cells (Berberich et al., 2005a). These studies seem to suggest that N V P may be a poor NR2A specific antagonist. More recently, Weitlauf et al. demonstrated that N V P displays time-dependent blockade kinetics: while simultaneous application of N V P with N M D A does not result in significant inhibition of recombinant NR1/NR2B receptors, pre-appliaction of N V P may produce substantial inhibition of peak currents mediated by these NR2B-containing receptors (Weitlauf et al., 2005b). Although the conflicting results may be partially due to the differential experimental conditions in these studies, the reasons for the distinction remain to be determined. We found that at the concentrations used in our study, N V P only has a negligible cross-receptor inhibition of NR2B-containing receptors. Under the conditions in our experiments, N V P is sensitive enough to distinguish the roles of NR2A- and NR2B-116 containing N M D A R s in neuronal fate. Our results clearly indicated that the neuronal survival action can be fully blocked by N V P . Although N V P may have a small contaminant inhibition of NR2B- receptors, the lack of blockade of the N M D A receptor-mediated cell survival action by the NR2B antagonist Ro essentially rule out the contribution of this subunit. On the other hand, the efficient blockade of N M D A receptor-dependent cell death by Ro, but not by N V P , strongly suggested that it is the N R 2 B - but not NR2A-containing N M D A receptor subpopulation that plays a primary role in triggering intracellular cascades that leading to N M D A - or ischemia-induced neuronal apoptosis. Whether N V P blocks other N M D A R subtypes (NR2C- and/or NR2D-containing N M D A R s ) has not been considered in this study due to their low expression in mature cortical neurons. 4.2. Dual role of NMDARs in neuronal fate and the possible underlying mechanisms Although it has been well known that the NR2 subunits contribute to the distinct pharmacological and electrophysiological properties of N M D A R s , the functional diversity of N M D A R s conferred by NR2 subunits remains largely unknown. Compelling evidence seems to suggest that most N M D A R subtypes, such as those containing NR2A- , N R 2 B - and NR2C subunits, are all involved in mediating excitotoxicity during stroke (Morikawa et al., 1998;Wang and Shuaib, 2005;Kadotani et al., 1998). In this thesis project, we have established for the first time that, as opposed to stimulation of N R 2 B -containing N M D A R s , activation of the NR2A-containing counterparts has pro-survival action in mature cortical cultures and the brain of adult rats. Therefore, N M D A R activation can produce either neuronal survival or death promoting action. 117 This work has provided a molecular basis to explain the paradoxical roles of N M D A R antagonism in producing extensive neuronal apoptosis in developing animals (Ikonomidou et al., 1999a) and neuroprotection against ischemic brain damage in animal studies (Lee et al., 1999a;Arundine and Tymianski, 2004b). For instance, under normal conditions spontaneously released glutamate preferentially stimulates synaptic N M D A R s which comprise mainly NR2A-containing N M D A R s , and hence activates predominantly the NR2A-containing receptor-dependent cell survival signaling, thereby playing an essential role in maintaining the physiological survival of the neurons. As demonstrated in this project, NR2A-containing receptor activation, in addition to counteracting NR2B-containing receptor-mediated cell death, has the ability to guard against non-NMDAR-mediated apoptotic processes which may be particularly active as part of the process of eliminating excess numbers of neurons in the developing animal. Thus, blocking these synaptically activated N M D A R s may lead to extensive neuronal apoptosis. However, under some pathological conditions, such as stroke and brain trauma, there is usually a transient and rapid increase in extracellular glutamate concentrations, and consequently extrasynaptic receptors, which are predominantly NR2B-containing and not usually activated by synaptically released glutamate under normal synaptic transmission, become activated, resulting in a predominant activation of the NR2B-containing receptor-dependent cell death pathway. Thus, blocking these receptors can be neuroprotective under these pathological conditions. The pro-survival role of NR2A-containing N M D A R s agrees with the observation that a selective reduction in N R 2 A subunit renders the brain more vulnerable to glutamate excitotoxicity in young animals (Gurd et al., 2002;McDonald and Johnston, 1990). On 118 the other hand, that NMDAR-dependent excitotoxity is found to be largely attributable to NR2B-containing N M D A R s is in accord with previous reports showing that N R 2 B -selective N M D A R antagonists exhibit excellent neuroprotective effects during stroke (Wang and Shuaib, 2005), and that the C A I area of hippocampus, which is particularly sensitive to ischemia, expresses greater levels of NR2B subunit (Coultrap et al., 2005). The pro-death activity of NR2B-containing N M D A R s is also consistent with that specific disruption of the neurotoxic signaling pathway downstream of NR2B dramatically can attenuate excitotoxicity without affecting N M D A R activity (Aarts et al., 2002). Furthermore, it was demonstrated that in Huntington's disease, striatal neurons selectively damaged via apoptosis show a high density of the NR2B-containing N M D A R s , suggesting high levels of NR2B may sensitize these neurons to apoptosis (Li et al., 2003;Zeron et al., 2002). Interestingly, the amount of cell death induced by selective activation of extrasynaptic NR2B-containing N M D A R s (Fig. 4b) was comparable to that shown in Fig. 2b whereby whole cells were stimulated, but much greater than that evoked by specific stimulation of synaptic N M D A R s containing NR2B indicated in Fig. 3d. This difference could be due to the different protocols used to stimulate N M D A R s in the experiments. We took advantage of the spontaneously released glutamate to examine the role of synaptic N M D A R s on neuronal fate. In contrast, exogenous N M D A of 50 p M was bath applied continuously for 20 min to stimulate N M D A R s on whole neurons or extrasynaptic sites. Alternatively, the amount of neuronal death may be largely dependent on the amount of activated NR2B-containing receptors. The pro-survival action of NR2A-containing receptors may only counteract the NR2B-mediated neuronal death to a certain extent. 119 Once the majority of NR2B receptors are activated, the NR2B-mediated death-promoting action wil l be overwhelming. The findings presented in our study are seemingly contradictory to those observations using transgenic mice. It has been shown that NR2A-knock-out mice are viable and more resistant to ischemic insults in adulthood (Morikawa et al., 1998), whereas NR2B-knock-out mice die shortly after birth (Kutsuwada et al., 1996). However, knocking out of an N M D A R subunit gene may change the normal subunit composition, distribution and functioning of N M D R s in adulthood. Thus, the data obtained from knockout studies may not reflect the real situations in adult animals that developed normally. Further investigations wi l l be needed to explain these discrepancies. The mechanisms underlying the distinction of N R 2 A - and NR2B-containing N M D A R s in excitotoxicity remain to be studied. Recent evidence suggests that different N R 2 subunits may couple to disparate postsynaptic signaling pathways, most likely via distinct protein interactions with their cytoplasmic carboxyl tails (Sheng and Pak, 2000;Kohr et al., 2003). In the present study we examined the classical Akt cell survival signaling pathways, and found that activation of NR2A-containing N M D A R s is specifically coupled to Akt phosphorylation. Why activation of NR2A-containing can selectively lead to Akt activation remains to be determined. It has been shown that stimulation of N R 2 A -containing N M D A R s can preferentially activate Ras-Erk pathway (Kim et al., 2005). Since Ras is also an upstream activator of Akt (Sutton and Chandler, 2002), Akt may be concomitantly and selectively activated upon activation of NR2A-containing N M D A R s . Indeed, it has been shown N M D A R activation can trigger both Akt (Sutton and Chandler, 2002) and Erk phosphorylation (Chandler et al., 2001). Recently, a PI3K-dependent pro-120 neuronal survival signaling has been linked with NR2A-containing N M D A R activation (Lee et al., 2002), suggesting the PI3K-Akt pathway may be specifically coupled to N R 2 A activation as well. Here we also investigated the caspase 3-mediated apoptotic pathway. The data suggest that NR2B-dependent neuronal apoptosis may be mediated by activation of caspase 3 and inhibition of Akt phosphorylation. The mechanisms underlying this specificity are unclear, either. However, recent evidence indicates that Akt can be negatively regulated by the tumor suppressor P T E N (phosphatase and tensin homolog deleted on chromosome 10), a lipid phosphatase that dephosphorylates PI3K. P T E N forms a complex with NR2B- but not NR2A-containing N M D A R s in vivo (Ning et al., 2004), suggesting activation of NR2B-containing N M D A R s may facilitate activation of P T E N , and hence inactivation of Akt. Moreover, SynGAP, a PSD protein that inhibits the activity of Ras, preferentially associates with NRZB-containing N M D A R s in the brain, and stimulation of NR2B-containing N M D A R s has been reported to specifically inhibit Ras-ERK pathway (Kim et al., 2005). Therefore, through preferential interaction between NR2B and SynGAP, activation of NR2B-containing N M D A R s may result in the specific inhibition of Ras, which in turn decreases Akt phosphorylation. Bad is a pro-apoptotic Bcl-2 family protein lying upstream of caspase 3 activation in the intrinsic apoptotic signaling pathway (Wang et al., 1999). However, only when Bad is dephosphorylated and translocated to mitochondria can it lead to activation of caspase 3. Given the critical role of Akt in the phosphorylation of Bad (Datta et al., 1997), NR2B-mediated Akt dephosphorylation may facilitate Bad dephosphorylation, and hence the activation of the caspase-3. Mounting evidence has demonstrated the involvement of caspase-3 in NMDAR-dependent neuronal apoptosis. The results shown here indicate that N M D A R s 121 containing NR2B may be primarily responsible for the activation of caspase-3 in mature cortical neurons. How activation of N R 2 A - and NR2B-containing receptors initiates distinct signal transduction pathways is largely unknown. The distinct gating kinetics of these two receptor subpopulations (Erreger et al., 2005;Cull-Candy and Leszkiewicz, 2004) may lead to cell-survival or -death signaling by providing differentially required levels and kinetics of rising [Ca 2 +]j in the postsynaptic neurons (Lipton and Nicotera, 1998;Choi, 1994). In addition, it has been shown that the N-terminal domains of NR2 subunits may be involved in the regulation of N M D A R s (Erreger and Traynelis, 2005;Krupp et al., 1998). Thus, the difference between the N-terminals of N R 2 A - and NR2B-containing receptors may lead to distinct signaling pathways as well. Whatever the precise mechanisms are, this study has established that N R 2 A - and N R 2 B -containing subpopulations of N M D A R s have differential roles in mediating neuronal survival and death, and hence provides a molecular basis for the paradoxical actions of N M D A R antagonism in producing both pro- and anti-apoptotic effects under different conditions. 4.3. Subunit composition of NMDARs determines their role in excitotoxicity The results from in this study demonstrate that, despite the predominance of N R 2 A -containing N M D A R s , a significant amount of N M D A R s containing NR2B is also localized to excitatory synapses, and these NR2B-containing N M D A R s are functional. NR2 subunits of N M D A R s undergo a diverse temporal and regional distribution change during development. Prenatally, N R 2 B is predominant, whereas N R 2 A expression increases quickly after birth (Monyer et al., 1994). In developing neurons, N R 2 A -122 containing N M D A R s are inserted into the synaptic sites and replace the existent synaptic NR2B-containing N M D A R s with the formation of excitatory synapses, rendering N R 2 A -containing N M D A R s the predominant N M D A R subtype at glutamatergic synapses (Tovar and Westbrook, 1999). Intriguingly, it is also argued that the switch in synaptic N M D A R subunit compostion is not due to the displacement of NR2B by N R 2 A ; instead, the predominance of NR2A-containing N M D A R s at synapses may reflect the formation of new synapses from which NR2B is lacking (Liu et al., 2004c). N M D A R s can move laterally on the cell surface from extrasynaptic to synaptic localization (Groc et al., 2004), but it remains unclear whether the NR2A-containing N M D A R s inserted into synapses are from extrasynaptic sites or directly from the N R 2 A pool in the cytosol. It is commonly believed that NR2B-containing N M D A R s accounts for a small fraction at synapses. This may be due to the more robust endocytosis of NR2B (Lavezzari et al., 2004) and/ or the constant removal of NR2B from synapses even when there is no synaptic activity (Barria and Malinow, 2002). However, our results demonstrated that in the majority of mature cortical neurons, functional NR2B-containing N M D A R s make up around one third of synaptic N M D A R s . Our findings may provide new insights into the N M D A R subunit composition at excitatory synapses. It was also found that at extrasynaptic sites, functional NR2A-containing N M D A R s , albeit a small portion, co-localize with their NR2B-containing counterparts, which are generally considered to be predominant at these sites at least in mature cortical and hippocampal neurons. Similar to synaptic N M D A R s , the N R 2 A component has been shown to increasingly contribute to extrasynaptic N M D A R s during development although the NR2B component is still predominant (Mohrmann et al., 2000). Thus, 123 extrasynatpic receptors seem to undergo a change in subunit composition as well. Interestingly, N M D A R s containing N R 2 A with truncated C-tail are located at extrasynaptic but not synaptic sites (Steigerwald et al., 2000), suggesting the localization of extrasynaptic NR2A-containing N M D A R s does not require tethering proteins interacting with N R 2 A C-tail. In contrast with previous data, a recent study suggested that N R 2 A - and N R 2 B -containing N M D A R s might be evenly distributed at synaptic and extrasynaptic sites in hippocampal neurons (Thomas et al., 2006). Further evidence wi l l be needed to confirm these findings. It is also noteworthy that triheteromeric N M D A R s which contain NR1 and two different types of NR2 subunit (Cull-Candy et al., 2001;Sheng et al., 1994), e.g., NR1/NR2A/NR2B may also exist in cortical neurons. Using co-immunoprecipitation method, Luo et al. even found that most N M D A R s in the cortex in rats contain both N R 2 A and N R 2 B . Data from Tovar et al. also suggested that at hippocampal synapses, NR1/NR2A/NR2B receptors are predominant (Tovar and Westbrook, 1999). However, the portion of these triheteromers in the brain is controversial (Blahos and Wenthold, 1996;Liu et al., 2004c). Whether these triheteromeric receptors possess the properties of both NR1/NR2A and NR1/NR2B receptors is largely unknown. Hatton and Paoletti showed that N M D A R s containing a single copy of N R 2 A or N R 2 B are still sensitive to subunit-specific agents (zinc or ifenprodil) although the maximal level of inhibition was greatly reduced when compared with those diheteromers containing two copies of N R 2 A or NR2B (Hatton and Paoletti, 2005). Due to the unknown properties of these triheteromeric receptors, their contribution to neuronal survival or death remains to be investigated. 124 Although N M D A R s are localized to both synaptic and extrasynaptic sites, I have shown here that the dual action of N M D A R s on neuronal fate is dictated by their subunit composition and not subcellular (synaptic vs. extrasynaptic) localization. Activation of NR2B-containing N M D A R s , synaptic or extrasynaptic, initiates apoptotic signaling cascades and promotes neuronal death. Conversely, selective activation of N R 2 A -containing N M D A R s stimulates pro-survival signaling, thereby exerting a neuroprotective action against both N M D A R - and non-NMDAR-dependent neuronal injuries. Thus, the net impact of N M D A R activation on neuronal survival and death is dictated by the balance between the activation of N R 2 A - and NR2B-containing N M D A R subpopulations but not by the subcellular locations. However, because of the preferential expression of the two subunits at synaptic versus extrasynaptic sites, selective activation of synaptic and extrasynaptic N M D A R s could be expected to respectively result in the activation of predominantly NR2A-containing receptor-dependent cell survival and NR2B-containing receptor-mediated apoptotic pathways. Thus, our results may not be in conflict with recently reported differential actions on cell survival and death following selective activation of synaptic and extrasynaptic N M D A R s (Hardingham et al., 2002). Furthermore, most recently it has been shown that synaptic N M D A R s can mediate neuronal death as well, confirming that the subcellular location of N M D A R s does not decide the role of N M D A R s in excitotoxicity. 4.4. Selective activation of NR2A-containing NMDARs represents a novel therapy for stroke Current strategies for stroke treatment fall into two categories: thrombolysis and 125 neuroprotection. Despite extensive research, only one thrombolytic, t-PA, has been successful in the treatment of acute stroke (NINDS, 1995). However, the efficacy of t-PA is still controversial (Hacke et al., 1998;Albers et al., 2002). Although t-PA may effectively reduce stroke-induced brain injury, it must be administered within 3-6 h after the onset of ischemic stroke. It is estimated that in the US only no more than 7% patients can receive this therapy. While a many neuroprotective agents have been studied, the future of neuroprotection is even more dismal. To date none of the neuroprotectants tested clinically have proved to be effective, including a variety of N M D A R antagonists (Lutsep and Clark, 2001). The data shown here indicate that selective activation of NR2A-containing N M D A R s can effectively ameliorate ischemic brain damage. Treatment of stroke by specifically activating NR2A-containing N M D A R s may have several obvious advantages over previously proposed N M D A R antagonism-based stroke therapies: first, as demonstrated in the present work, it possesses a broader therapeutic window than NR2B-containing receptor blockade. In fact, theoretically it should not have any therapeutic window limitation as it protects neurons against brain damage by promoting neuronal survival rather than blocking the activation of a death signal initiated by the stroke insult; second, it is significant to point out that NR2A-containing receptor activation is not only effective against NMDAR-mediated cell death (primary neuronal injuries), but can also guard against non-NMDAR-mediated cell death (secondary neuronal injuries). Increasing evidence supports the fact that some of the non-NMDAR-mediated mechanisms, while secondary to N M D A R activation, may contribute significantly to brain damage, particularly following severe stroke insults (Xiong et al., 2004;Aarts et al., 2003a). Thus, 126 activation of NR2A-containing receptors, rather than blockade of N M D A R s , appears to be a more effective post-stroke neuroprotective therapy. In addition to the neuronal injuries caused by acute brain insults such as stroke and brain trauma, activation of NR2A-containing receptor-dependent pro-survival signaling may also prove to be a potential neuroprotective therapy for a number of chronic neurodegenerative disorders in which a \"slow\" NMDAR-mediated excitotoxicity is implicated, such as Parkinson's disease, Huntington's disease, amyotrophic lateral sclerosis and Alzheimer's disease (Ikonomidou and Turski, 2002b;Lipton and Rosenberg, 1994). According to the observations in this study, an NR2A-containing receptor activation-based neuroprotective strategy that can diminish both acute brain injury following stroke or brain trauma and chronic brain damage associated with a large number of neurodegenerative diseases may represent a novel therapy for these diseases. Thus, this work calls for the urgent development of highly specific agonists for the N R 2 A -containing N M D A R subtype. In the absence of such subunit specific agonists, a specific enhancement of NR2A-containing N M D A R function, as demonstrated in the present work, may be achieved by the combination of a non-subunit specific N M D A R enhancer, such as glycine, and an NR2B specific antagonist. Glycine is an endogenous co-agonist of N M D A R s . Probably because of its relatively high concentration in the brain under physiological conditions (Danysz and Parsons, 1998), it is commonly assumed that glycine is freely available for N M D A R s and the glycine site on N M D A R s is always saturated. However, recent evidence suggests that the glycine site may be unsaturated in vivo (Wood, 1995); (Javitt, 2006;Danysz and Parsons, 1998). This may explain why in our study the activity of N M D A R s , NR2A-containing receptors in particular, can be 127 enhanced by a high dose of exogenous glycine. It is interesting to note, in this regard, that both N M D A R glycine site agonists, such as D-cycloserine (Posey et a l , 2004), and NR2B specific antagonists(Chazot, 2000) are now available for clinical trials. Thus, our study may have an immediate impact on the treatment of stroke. 4.5. Perturbing protein-protein interaction downstream of NR2B activation may be an alternative to NR2B blockage Previous stroke clinical trials with N M D A R antagonists, including NR2B-selective blockers, have not seen success (Lee et al., 1999b;Kemp and McKernan, 2002;Ikonomidou and Turski, 2002b). In the present study, we also demonstrated that in the rat focal cerebral ischemia model, although NMDAR-mediated excitotoxic neuronal injuries following stroke are primarily mediated by NR2B-containing receptors, as NR2B-containing NMDAR-specific antagonist applied prior to the stroke onset significantly reduced ischemic brain damage, NR2B antagonist offers little protection when administered 4.5 h after the stroke onset. The reasons for this failure are still controversial. The rapid recovery of extracellular glutamate concentrations to pre-stroke levels (Benveniste et al., 1984;Ikonomidou and Turski, 2002a), which are not sufficient to activate extrasynaptic N M D A R s , may partially account for this narrow therapeutic window. In most clinical settings, due to the time required to transport a patient to the hospital and obtain a definitive diagnosis, treatment is not usually possible until several hours after the stroke onset, which is most likely outside the window of efficacy for any N M D A R blockers including NR2B-selective antagonists. Thus, consistent with previous suspicions, our results suggest that the failure to administer N M D A R antagonists within 128 their efficacy window might be one of the major reasons for their failure in the clinical trials of stroke and neurotrauma conducted thus far. Although the rapid recovery of glutamate levels following stroke makes it unfeasible for NR2B-specific antagonists to treat stroke, it takes time to transduce the pro-death signals following activation of NR2B-containing N M D A R s . Therefore, an alternative that is able to disrupt the downstream signaling may provide an opportunity to prevent neuronal death. The NR2B9c peptide used in this study was derived from the last nine amino acids of the C-tail of NR2B subunit. Since both N R 2 A and NR2B C-tails contain the same binding motif (ESDV) in the very last four amino acids for the second PDZ domain on PSD-95, NR2B9c peptide is very likely to dissociate the binding of PSD-95 to N R 2 A as well. However, data from the Tymianski group suggest this peptide only selectively perturbs the interaction between NR2B and PSD-95. Comparison of the sequences of last nine residues of the N R 2 A C-tail and the NR2B9c peptide indicates there are two different residues between them. This incomplete homology of NR2B9c for N R 2 A C-tail may account for its ineffectiveness in perturbing the binding of PSD-95 to N R 2 A . Alternatively, N R 2 A may be more tightly bound to PSD-95 because it may preferentially associate with PSD-95 (van et al., 2004). Given that the NR2B9c peptide contains only nine amino acids and that it shares a high homology with N R 2 A C-tail but still does not interfere with the NR2A-PSD-95 interaction, this peptide may have few non-specific binding partners besides PSD-95. The high selectivity of NR2B9c is also indirectly confirmed by the lack of observable untoward effects in the experiments in vivo. Consistent with the effects of the NR2B9c peptide on excitotoxicity in vitro (Aarts et al., 2002), our results suggested that it also has dramatic effects on ischemic brain damage in 129 vivo. This therapy is fundamentally different from blockade of NR2B-containing N M D A R s in that it may occur without affecting N M D A R activity, and hence the N R 2 A -mediated cell survival signaling is intact. In addition, it is efficacious when carried out after the insult onset. A l l these suggest targeting of the NR2B-PSD-95 interaction is a practical strategy for treating stroke. Thus, this may represent a new direction in stroke therapeutics. Lastly, it is also probable that a similar approach could be used to modulate signals mediated by protein-protein interactions that lead to other human diseases. 4.6. Promising NMDAR-based neuroprotective therapies for ischemic brain damage In this research project, we have shown that N M D A R s are still on the center stage of excitotoxic injuries but also play a crucial role in neuronal survival. This functional differentiation is dictated by their subunit composition, i.e., N M D A R subtypes. According to these findings, here we proposed several NMDAR-based strategies to achieve neuroprotection from stroke-induced brain injury (Fig. 8). If a stroke is predictable, blocking NR2B-containing N M D A R s before stroke would constitute an ideal neuroprotective therapy. Through administration of a highly selective N R 2 B antagonist just before stroke attack, the overactivation of N M D A R s containing N R 2 B during stroke can be inhibited, and hence the associated cell death signaling wil l not be initiated. In the meantime, the ensuing excessive release of glutamate wil l stimulate NR2A-containing N M D A R s , which wil l in turn trigger the anti-apoptotic signaling pathway, and guard neurons against NMDAR-independent injuries during stroke. Regretfully, at present it is almost impossible to know when a stroke wil l occur; therefore, albeit appealing, this therapy may not be clinically useful. However, i f a stroke 130 patient can receive medical attention within the therapeutic time window (usually less than 3 h), this therapy wil l , still be beneficial. Unfortunately, in most clinical settings, stroke patients have missed out the optimal time window. This makes an effective post-ishemic therapy much more desirable. Theoretically, neuroprotection offered by selective activation of NR2A-containing N M D A R s has no time limit and could counteract not only NMDAR-dependent but also NMDAR-independent neuronal death. Accordingly, stimulation N M D A R s containing N R 2 A by highly specific N R 2 A agonists may be an optimal therapy for stroke patients at this stage. Alternatively, disruption of the signaling pathway downstream of N R 2 B activation with agents such as the NR2B C-tail interference peptide may also be efficacious. Since these two approaches are not occlusive, a comprehensive treatment that integrates N R 2 A activation and dissociation of NR2B-mediated cell death signaling could be more protective. 131 Fig. 8 132 Fig . 8. Schematic illustrates potential NMDAR-based therapies for stroke. Before stroke occurs, selective blockage of NR2B may be an ideal approach to prevent ischemic injury. NR2B-specific blockers can still be efficacious if administered timely during stroke. Following stroke onset, dissociation of the cell death signaling pathway downstream with interfering agents such as NR2B peptide (NR2B pep.) used in this study and/or post-stroke stimulation of NR2A with highly specific agonists may constitute feasible therapies for ischemic brain damage. These strategies may also apply to other neurodegenerative diseases in which excitotoxicity is involved. 133 4 .7 . Future directions We have shown here that N M D A R subtypes may play disparate roles in neuronal fate, and also demonstrated for the first time that selective activation of NR2A-containing N M D A R s may exert neuroprotection. However, because of the limitation of time, there is still so much left to be studied. In the future, I wish to elucidate three major issues. The first question I would like to address in the near future is the exact signaling pathways of NR2A- and NR2B-containing N M D A R s . Although we have found that stimulation of these two N M D A R subtypes may activate opposing signaling pathways leading to cell survival or death, the mechanisms underlying this differentiation are still far from clear. However, only when these downstream pathways are fully understood can we develop suitable neuroprotective agents to promote cell survival and/or prevent cell death. There are several ways to achieve this goal, for example, imaging potential downstream mediators of these signaling pathways in live neurons following specific activation of a N M D A R subtype may provide insights into this issue. Since we have shown in animals that selective activation of NR2A-containing N M D A R s can greatly reduce brain injury, I would like to collaborate with industry and develop some highly specific NR2A agonists. The efficacy of such NR2A-selective agonists wil l then be examined in animals. If found to be able to reduce ischemic injury, they would move onto stroke clinical trials. In addition, as the NR2B C-tail peptide used in this study has shown amazing effects in attenuating brain injury, a clinical trial for this peptide to treat stroke should be under consideration. Hopefully, some new lines of drugs for treating stroke can be invented. Finally, I want to determine the roles of other N M D A R subtypes such as NR2C- and 134 NR2D-containing N M D A R s in excitotoxicity. In humans stroke mainly occurs in the forebrain; however, other regions in the central nervous system, such as brain stem, cerebellum and spinal cord, are also frequently attacked by stroke. The subunit composition of N M D A R s in these regions is different from that in forebrain. Thus, determining the functions of the N M D A R subtypes other than N R 2 A - and NR2B-containing N M D A R s is also critical in stroke therapeutics. 135 Reference List Aarts M , Iihara K , Wei W L , Xiong Z G , Arundine M , Cerwinski W, MacDonald JF, Tymianski M (2003a) A key role for TRPM7 channels in anoxic neuronal death. Cell 115:863-877. Aarts M , L iu Y , L i u L , Besshoh S, Arundine M , Gurd JW, Wang Y T , Salter M W , Tymianski M (2002) Treatment of ischemic brain damage by perturbing N M D A receptor- PSD-95 protein interactions. Science 298:846-850. Aarts M M , Arundine M , Tymianski M (2003b) Novel concepts in excitotoxic neurodegeneration after stroke. Expert Rev M o l Med 2003:1-22. Aarts M M , Tymianski M (2004) Molecular mechanisms underlying specificity of excitotoxic signaling in neurons. Curr M o l Med 4:137-147. Akazawa C, Shigemoto R, Bessho Y , Nakanishi S, Mizuno N (1994) Differential expression of five N-methyl-D-aspartate receptor subunit mRNAs in the cerebellum of developing and adult rats. J Comp Neurol 347:150-160. Akerman K E (1978) Changes in membrane potential during calcium ion influx and efflux across the mitochondrial membrane. Biochim Biophys Acta 502:359-366. Albers GW, Clark W M , Madden K P , Hamilton SA (2002) A T L A N T I S trial: results for patients treated within 3 hours of stroke onset. Alteplase Thrombolysis for Acute Noninterventional Therapy in Ischemic Stroke. Stroke 33:493-495. Albers G W , Goldberg M P , Choi D W (1989) N-methyl-D-aspartate antagonists: ready for clinical trial in brain ischemia? Ann Neurol 25:398-403. Albers GW, Goldstein L B , Hall D, Lesko L M (2001) Aptiganel hydrochloride in acute ischemic stroke: a randomized controlled trial. J A M A 286:2673-2682. Allen H L , Iversen L L (1990) Phencyclidine, dizocilpine, and cerebrocortical neurons. Science 247:221. Allen JW, Knoblach S M , Faden A I (1999) Combined mechanical trauma and metabolic impairment in vitro induces N M D A receptor-dependent neuronal cell death and caspase-3-dependent apoptosis. F A S E B J 13:1875-1882. American Heart Association (2004) Heart Diseases and Stroke Statistics-2005 update. Dallas,Tex.: American Heart Association. Antonsson B , Conti F, Ciavatta A , Montessuit S, Lewis S, Martinou I, Bernasconi L , 136 Bernard A , Mermod JJ, Mazzei G, Maundrell K , Gambale F, Sadoul R, Martinou JC (1997) Inhibition of Bax channel-forming activity by Bcl-2. Science 277:370-372. Arnoult D , Gaume B , Karbowski M , Sharpe JC, Cecconi F, Youle RJ (2003) Mitochondrial release of AIF and EndoG requires caspase activation downstream of Bax/Bak-mediated permeabilization. E M B O J 22:4385-4399. Arundine M , Tymianski M (2004b) Molecular mechanisms of glutamate-dependent neurodegeneration in ischemia and traumatic brain injury. Cell M o l Life Sci 61:657-668. Arundine M , Tymianski M (2003) Molecular mechanisms of calcium-dependent neurodegeneration in excitotoxicity. Cell Calcium 34:325-337. Arundine M , Tymianski M (2004a) Molecular mechanisms of glutamate-dependent neurodegeneration in ischemia and traumatic brain injury. Cell M o l Life Sci 61:657-668. Auberson Y P , Allgeier H , Bischoff S, Lingenhoehl K , Moretti R, Schmutz M (2002) 5-Phosphonomethylquinoxalinediones as competitive N M D A receptor antagonists with a preference for the human 1A/2A, rather than 1A/2B receptor composition. Bioorg Med Chem Lett 12:1099-1102. Back T, Zhao W, Ginsberg M D (1995) Three-dimensional image analysis of brain glucose metabolism-blood flow uncoupling and its electrophysiological correlates in the acute ischemic penumbra following middle cerebral artery occlusion. J Cereb Blood Flow Metab 15:566-577. Bading H , Ginty D D , Greenberg M E (1993) Regulation of gene expression in hippocampal neurons by distinct calcium signaling pathways. Science 260:181-186. Bano D, Young K W , Guerin CJ, Lefeuvre R, Rothwell NJ , Naldini L , Rizzuto R, Carafoli E, Nicotera P (2005) Cleavage of the plasma membrane Na+/Ca2+ exchanger in excitotoxicity. Cell 120:275-285. Bar-Joseph A , Berkovitch Y , Adamchik J, Biegon A (1994) Neuroprotective activity of HU-211, a novel N M D A antagonist, in global ischemia in gerbils. M o l Chem Neuropathol 23:125-135. Barbour B , Brew H , Attwell D (1988) Electrogenic glutamate uptake in glial cells is activated by intracellular potassium. Nature 335:433-435. Barria A , Malinow R (2002) Subunit-specific N M D A receptor trafficking to synapses. Neuron 35:345-353. Bayer K U , De K P , Leonard A S , Hell JW, Schulman H (2001) Interaction with the N M D A receptor locks CaMKII in an active conformation. Nature 411:801-805. Beckman JS (1994) Peroxynitrite versus hydroxyl radical: the role of nitric oxide in superoxide-dependent cerebral injury. Ann N Y Acad Sci 738:69-75. 137 Bederson JB, Pitts L H , Tsuji M , Nishimura M C , Davis R L , Bartkowski H (1986) Rat middle cerebral artery occlusion: evaluation of the model and development of a neurologic examination. Stroke 17:472-476. Bellizzi M J , L u S M , Masliah E, Gelbard H A (2005b) Synaptic activity becomes excitotoxic in neurons exposed to elevated levels of platelet-activating factor. J Clin Invest 115:3185-3192. Bellizzi M J , L u S M , Masliah E, Gelbard H A (2005a) Synaptic activity becomes excitotoxic in neurons exposed to elevated levels of platelet-activating factor. J Clin Invest 115:3185-3192. Benveniste H , Drejer J, Schousboe A , Diemer N H (1984) Elevation of the extracellular concentrations of glutamate and aspartate in rat hippocampus during transient cerebral ischemia monitored by intracerebral microdialysis. J Neurochem 43:1369-1374. Berberich S, Punnakkal P, Jensen V , Pawlak V , Seeburg P H , Hvalby O, Kohr G (2005b) Lack of N M D A receptor subtype selectivity for hippocampal long-term potentiation. J Neurosci 25:6907-6910. Berberich S, Punnakkal P, Jensen V , Pawlak V , Seeburg P H , Hvalby O, Kohr G (2005a) Lack of N M D A receptor subtype selectivity for hippocampal long-term potentiation. J Neurosci 25:6907-6910. Beresford IJ, Parsons A A , Hunter A J (2003) Treatments for stroke. Expert Opin Emerg Drugs 8:103-122. Blahos J, Wenthold RJ (1996) Relationship between N-methyl-D-aspartate receptor NR1 splice variants and NR2 subunits. J Biol Chem 271:15669-15674. Bleich S, Romer K , Wiltfang J, Kornhuber J (2003) Glutamate and the glutamate receptor system: a target for drug action. Int J Geriatr Psychiatry 18:S33-S40. Boireau A , Malgouris C, Burgevin M C , Peny C, Durand G, Bordier F, Meunier M , Miquet J M , Daniel M , Chevet T, Jimonet P, Mignani S, Blanchard JC, Doble A (1996) Neuroprotective effects of RPR 104632, a novel antagonist at the glycine site of the N M D A receptor, in vitro. Eur J Pharmacol 300:237-246. Brenman JE, Chao DS, Gee SH, McGee A W , Craven SE, Santillano DR, Wu Z , Huang F, X i a H , Peters M F , Froehner SC, Bredt DS (1996) Interaction of nitric oxide synthase with the postsynaptic density protein PSD-95 and alpha 1-syntrophin mediated by PDZ domains. Cell 84:757-767. Brunet A , Datta SR, Greenberg M E (2001) Transcription-dependent and -independent control of neuronal survival by the PI3K-Akt signaling pathway. Curr Opin Neurobiol 11:297-305. Brust JC (2000) Circulation of the brain. In: Principles of Neural Science pp 1302-1306. 138 New York: McGraw-Hil l . Buchan A , L i H , Pulsinelli W A (1991) The N-methyl-D-aspartate antagonist, MK-801, fails to protect against neuronal damage caused by transient, severe forebrain ischemia in adult rats. J Neurosci 11:1049-1056. Budd SL, Lipton SA (1999) Signaling events in N M D A receptor-induced apoptosis in cerebrocortical cultures. Ann N Y Acad Sci 893:261-264. Budd SL, Tenneti L , Lishnak T, Lipton SA (2000) Mitochondrial and extramitochondrial apoptotic signaling pathways in cerebrocortical neurons. Proc Natl Acad Sci U S A 97:6161-6166. Burnashev N , Schoepfer R, Monyer H , Ruppersberg JP, Gunther W, Seeburg P H , Sakmann B (1992) Control by asparagine residues of calcium permeability and magnesium blockade in the N M D A receptor. Science 257:1415-1419. Burnashev N , Zhou Z, Neher E, Sakmann B (1995a) Fractional calcium currents through recombinant GluR channels of the N M D A , A M P A and kainate receptor subtypes. J Physiol 485 (Pt 2):403-418. Burnashev N , Zhou Z, Neher E, Sakmann B (1995b) Fractional calcium currents through recombinant GluR channels of the N M D A , A M P A and kainate receptor subtypes. J Physiol 485 (Pt 2):403-418. Cai J, Yang J, Jones DP (1998) Mitochondrial control of apoptosis: the role of cytochrome c. Biochim Biophys Acta 1366:139-149. Calabresi P, Centonze D, Cupini L M , Costa C, Pisani F, Bernardi G (2003) Ionotropic glutamate receptors: still a target for neuroprotection in brain ischemia? Insights from in vitro studies. Neurobiol Dis 12:82-88. Cande C, Cohen I, Daugas E, Ravagnan L , Larochette N , Zamzami N , Kroemer G (2002) Apoptosis-inducing factor (AIF): a novel caspase-independent death effector released from mitochondria. Biochimie 84:215-222. Caplan L R (2004) Treatment of acute stroke: still struggling. J A M A 292:1883-1885. Carboni S, Antonsson B , Gaillard P, Gotteland JP, Gillon JY, Vitte P A (2005) Control of death receptor and mitochondrial-dependent apoptosis by c-Jun N-terminal kinase in hippocampal C A I neurones following global transient ischaemia. JNeurochem 92:1054-1060. Castillo J, Davalos A , Noya M (1997) Progression of ischaemic stroke and excitotoxic aminoacids. Lancet 349:79-83. Cathala L , Misra C, Cull-Candy S (2000) Developmental profile of the changing properties of N M D A receptors at cerebellar mossy fiber-granule cell synapses. J Neurosci 139 20:5899-5905. Chan P H (2001) Reactive oxygen radicals in signaling and damage in the ischemic brain. J Cereb Blood Flow Metab 21:2-14. Chan SL, Y u V C (2004) Proteins of the bcl-2 family in apoptosis signalling: from mechanistic insights to therapeutic opportunities. Clin Exp Pharmacol Physiol 31:119-128. Chandler L J , Sutton G, Dorairaj N R , Norwood D (2001) N-methyl D-aspartate receptor-mediated bidirectional control of extracellular signal-regulated kinase activity in cortical neuronal cultures. J Bio l Chem 276:2627-2636. Charriaut-Marlangue C (2004) Apoptosis: a target for neuroprotection. Therapie 59:185-190. Chatterton JE, Awobuluyi M , Premkumar L S , Takahashi H , Talantova M , Shin Y , Cui J, Tu S, Sevarino K A , Nakanishi N , Tong G, Lipton SA, Zhang D (2002) Excitatory glycine receptors containing the NR3 family of N M D A receptor subunits. Nature 415:793-798. Chazot P L (2000) CP-101606 Pfizer Inc. Curr Opin Investig Drugs 1:370-374. Cheng Y D , Al-Khoury L , Zivin JA (2004) Neuroprotection for Ischemic Stroke: Two Decades of Success and Failure. Neurorx 1:36-45. Choi D W (1996) Ischemia-induced neuronal apoptosis. Curr Opin Neurobiol 6:667-672. Choi D W (1994) Calcium and excitotoxic neuronal injury. [Review]. Annals of the New York Academy of Sciences 747:162-171. Choi D W (1995) Calcium: still center-stage in hypoxic-ischemic neuronal death. Trends Neurosci 18:58-60. Choi D W (1985) Glutamate neurotoxicity in cortical cell culture is calcium dependent. Neurosci Lett 58:293-297. Choi D W , Rothman S M (1990) The role of glutamate neurotoxicity in hypoxic-ischemic neuronal death. Annu Rev Neurosci 13:171 -82.: 171 -182. Chopp M , L i Y (1996) Apoptosis in focal cerebral ischemia. Acta Neurochir Suppl 66:21-26. Christopherson K S , Hillier BJ , L im W A , Bredt DS (1999) PSD-95 assembles a ternary complex with the N-methyl-D-aspartic acid receptor and a bivalent neuronal N O synthase P D Z domain. J Bio l Chem 274:27467-27473. Clarke D D , Sokoloff L . (1994) Cellular Neurochemistry: Circulation and Energy 140 Metabolism of the Brain. In: Basic Neurochemistry (Siegel G.J., Agranoff B.W., Albers R.W., Molinoff P.B., eds), pp 645-680. New York: Raven Press. Clarke P G (1990) Developmental cell death: morphological diversity and multiple mechanisms. Anat Embryol (Berl) 181:195-213. Cocho D, Belvis R, Marti-Fabregas J, Molina-Porcel L , az-Manera J, Aleu A , Pagonabarraga J, Garcia-Bargo D, Mauri A , Marti-Vilalta JL (2005) Reasons for exclusion from thrombolytic therapy following acute ischemic stroke. Neurology 64:719-720. Cornu C, Amsallem E, Serradj-Jaillard A A (2001) Thrombolytic therapy for acute ischemic stroke. A m J Cardiovasc Drugs 1:281-292. Coultrap SJ, Nixon K M , Alvestad R M , Fernando V C , Browning M D (2005) Differential expression of N M D A receptor subunits and splice variants among the C A I , CA3 and dentate gyrus of the adult rat. Brain Res M o l Brain Res 135:104-111. Coyle JT, Tsai G (2004) The N M D A receptor glycine modulatory site: a therapeutic target for improving cognition and reducing negative symptoms in schizophrenia. Psychopharmacology (Berl) 174:32-38. Coyle JT, Tsai G, Goff D C (2002) Ionotropic glutamate receptors as therapeutic targets in schizophrenia. Curr Drug Targets CNS Neurol Disord 1:183-189. Cull-Candy S, Brickley S, Farrant M (2001) N M D A receptor subunits: diversity, development and disease. Curr Opin Neurobiol 11:327-335. Cull-Candy SG, Leszkiewicz D N (2004) Role of distinct N M D A receptor subtypes at central synapses. Sci S T K E 2004:rel6. Danial N N , Korsmeyer SJ (2004) Cell death: critical control points. Cell 116:205-219. Danton G H , Dietrich W D (2004) The search for neuroprotective strategies in stroke. A J N R A m J Neuroradiol 25:181 -194. Danysz W, Parsons A C (1998) Glycine and N-methyl-D-aspartate receptors: physiological significance and possible therapeutic applications. Pharmacol Rev 50:597-664. Datta SR, Dudek H , Tao X , Masters S, Fu H , Gotoh Y , Greenberg M E (1997) Akt phosphorylation of B A D couples survival signals to the cell-intrinsic death machinery. Cell 91:231-241. Davalos A , Castillo J, Serena J, Noya M (1997) Duration of glutamate release after acute ischemic stroke. Stroke 28:708-710. Davis S M , Albers GW, Diener H C , Lees K R , Norris J (1997) Termination of Acute 141 Stroke Studies Involving Selfotel Treatment. ASSIST Steering Committed. Lancet 349:32. Davis S M , Lees K R , Albers GW, Diener H C , Markabi S, Karlsson G, Norris J (2000) Selfotel in acute ischemic stroke : possible neurotoxic effects of an N M D A antagonist. Stroke 31:347-354. Dawson V L , Dawson T M (1996) Nitric oxide neurotoxicity. J Chem Neuroanat 10:179-190. Dawson V L , Kizushi V M , Huang PL, Snyder SH, Dawson T M (1996) Resistance to neurotoxicity in cortical cultures from neuronal nitric oxide synthase-deficient mice. J Neurosci 16:2479-2487. De K J , Suiter G, Luiten P G (1999) Clinical trials with neuroprotective drugs in acute ischaemic stroke: are we doing the right thing? Trends Neurosci 22:535-540. De R M , Van RJ, Borgers M , Wauquier A , Janssen P A (1989) Photochemical stroke model: flunarizine prevents sensorimotor deficits after neocortical infarcts in rats. Stroke 20:1383-1390. De SG, Siniscalchi A , Ferreri G, Gallelli L , De S A (2000) N M D A and AMPA/kainate receptors are involved in the anticonvulsant activity of riluzole in D B A / 2 mice. Eur J Pharmacol 408:25-34. Degli E M , Dive C (2003) Mitochondrial membrane permeabilisation by Bax/Bak. Biochem Biophys Res Commun 304:455-461. del Zoppo GJ (2004) Thrombolysis: from the experimental findings to the clinical practice. Cerebrovasc Dis 17 Suppl 1:144-152. Dingledine R, Borges K , Bowie D, Traynelis SF (1999) The glutamate receptor ion channels. Pharmacol Rev 51:7-61. Dirnagl U , Iadecola C, Moskowitz M A (1999) Pathobiology of ischaemic stroke: an integrated view. Trends Neurosci 22:391-397. Du C, Fang M , L i Y , L i L , Wang X (2000) Smac, a mitochondrial protein that promotes cytochrome c-dependent caspase activation by eliminating IAP inhibition. Cell 102:33-42. Duverger D, MacKenzie ET (1988) The quantification of cerebral infarction following focal ischemia in the rat: influence of strain, arterial pressure, blood glucose concentration, and age. J Cereb Blood Flow Metab 8:449-461. Dzubay JA, Jahr C E (1996) Kinetics of N M D A channel opening. J Neurosci 16:4129-4134. Erecinska M , Silver IA (1992) Relationship between ions and energy metabolism: 142 cerebral calcium movements during ischaemia and subsequent recovery. Can J Physiol Pharmacol 70 Suppl:S190-S193. Erreger K , Chen PE, Wyllie DJ, Traynelis SF (2004) Glutamate receptor gating. Crit Rev Neurobiol 16:187-224. Erreger K , Dravid S M , Banke TG, Wyllie DJ, Traynelis SF (2005) Subunit-specific gating controls rat NR1/NR2A and NR1/NR2B N M D A channel kinetics and synaptic signalling profiles. J Physiol 563:345-358. Erreger K , Traynelis SF (2005) Allosteric interaction between zinc and glutamate binding domains on N R 2 A causes desensitization of N M D A receptors. J Physiol 569:381-393. Felderhoff-Mueser U , Taylor D L , Greenwood K , Kozma M , Stibenz D, Joashi U C , Edwards A D , Mehmet H (2000) Fas/CD95/APO-l can function as a death receptor for neuronal cells in vitro and in vivo and is upregulated following cerebral hypoxic-ischemic injury to the developing rat brain. Brain Pathol 10:17-29. Feng B , Tse H W , Skifter D A , Morley R, Jane D E , Monaghan DT (2004) Structure-activity analysis of a novel NR2C/NR2D-preferring N M D A receptor antagonist: 1-(phenanthrene-2-carbonyl) piperazine-2,3-dicarboxylic acid. Br J Pharmacol 141:508-516. Fern R, Connolly GP, Harrison PJ (1996) Evidence for functional co-activation of N -methyl-D-aspartate receptors by glycine. Neuroreport 7:1953-1956. Finkel E (2001) The mitochondrion: is it central to apoptosis? Science 292:624-626. Fischer G, Mutel V , Trube G, Malherbe P, Kew JN, Mohacsi E, Heitz M P , Kemp JA (1997a) Ro 25-6981, a highly potent and selective blocker of N-methyl-D-aspartate receptors containing the NR2B subunit. Characterization in vitro. J Pharmacol Exp Ther 283:1285-1292. Fischer G, Mutel V , Trube G, Malherbe P, Kew JN, Mohacsi E, Heitz M P , Kemp JA (1997b) Ro 25-6981, a highly potent and selective blocker of N-methyl-D-aspartate receptors containing the N R 2 B subunit. Characterization in vitro. J Pharmacol Exp Ther 283:1285-1292. Fisher M , Bogousslavsky J (1998) Further evolution toward effective therapy for acute ischemic stroke. J A M A 279:1298-1303. Forrest D, Yuzaki M , Soares H D , N g L , Luk D C , Sheng M , Stewart C L , Morgan JI, Connor JA, Curran T (1994) Targeted disruption of N M D A receptor 1 gene abolishes N M D A response and results in neonatal death. Neuron 13:325-338. Fukaya M , Kato A , Lovett C, Tonegawa S, Watanabe M (2003) Retention of N M D A receptor N R 2 subunits in the lumen of endoplasmic reticulum in targeted NR1 knockout mice. Proc Natl Acad Sci U S A 100:4855-4860. 143 Gallagher M J , Huang H , Pritchett D B , Lynch D R (1996) Interactions between ifenprodil and the N R 2 B subunit of the N-methyl-D-aspartate receptor. J B io l Chem 271:9603-9611. Ganitkevich V Y (2003) The role of mitochondria in cytoplasmic Ca2+ cycling. Exp Physiol 88:91-97. Garaschuk O, Schneggenburger R, Schirra C, Tempia F, Konnerth A (1996) Fractional Ca2+ currents through somatic and dendritic glutamate receptor channels of rat hippocampal C A I pyramidal neurones. J Physiol 491 ( Pt 3):757-772. Gardoni F, Caputi A , Cimino M , Pastorino L , Cattabeni F, D i L M (1998) Calcium/calmodulin-dependent protein kinase II is associated with N R 2 A / B subunits of N M D A receptor in postsynaptic densities. J Neurochem 71:1733-1741. Gines S, Ivanova E, Seong IS, Saura C A , MacDonald M E (2003) Enhanced Akt signaling is an early pro-survival response that reflects N-methyl-D-aspartate receptor activation in Huntington's disease knock-in striatal cells. J B io l Chem 278:50514-50522. Ginsberg M D , Busto R (1989) Rodent models of cerebral ischemia. Stroke 20:1627-1642. Goldberg M P , Choi D W (1993) Combined oxygen and glucose deprivation in cortical cell culture: calcium-dependent and calcium-independent mechanisms of neuronal injury. J Neurosci 13:3510-3524. Goldberg M P , Viseskul V , Choi D W (1988) Phencyclidine receptor ligands attenuate cortical neuronal injury after N-methyl-D-aspartate exposure or hypoxia. J Pharmacol Exp Ther 245:1081-1087. Goldberg M P , Weiss JH, Pham PC, Choi D W (1987) N-methyl-D-aspartate receptors mediate hypoxic neuronal injury in cortical culture. J Pharmacol Exp Ther 243:784-791. Graham D, Darles G , Langer SZ (1992) The neuroprotective properties of ifenprodil, a novel N M D A receptor antagonist, in neuronal cell culture toxicity studies. Eur J Pharmacol 226:373-376. Green DR, Reed JC (1998) Mitochondria and apoptosis. Science 281:1309-1312. Groc L , Heine M , Cognet L , Brickley K , Stephenson F A , Lounis B , Choquet D (2004) Differential activity-dependent regulation of the lateral mobilities of A M P A and N M D A receptors. Nat Neurosci 7:695-696. Gunter TE , Pfeiffer D R (1990) Mechanisms by which mitochondria transport calcium. A m J Physiol 258:C755-C786. Gurd JW, Bissoon N , Beesley PW, Nakazawa T, Yamamoto T, Vannucci SJ (2002) Differential effects of hypoxia-ischemia on subunit expression and tyrosine phosphorylation of the N M D A receptor in 7- and 21-day-old rats. J Neurochem 82:848-856. 144 Gutierrez-Diaz JA, Cuevas P, Reimers D, Dujovny M , Diaz F G , Ausman JI (1985) Quantitative electron microscopic study of calcium accumulation in cerebral ischemia mitochondria. Surg Neurol 24:67-72. Hacke w, Hennerici M , Gelmers HJ , Kramer G (1991) Epidemiology and classification of strokes. In: Cerebral Ischemia pp 31-52. New York: Springer-Verlag. Hacke w, Kaste M , Fieschi C, von K R , Davalos A , Meier D , Larrue V , Bluhmki E , Davis S, Donnan G, Schneider D, ez-Tejedor E, Trouillas P (1998) Randomised double-blind placebo-controlled trial of thrombolytic therapy with intravenous alteplase in acute ischaemic stroke (ECASS II). Second European-Australasian Acute Stroke Study Investigators. Lancet 352:1245-1251. Halestrap A P , McStay GP , Clarke SJ (2002a) The permeability transition pore complex: another view. Biochimie 84:153-166. Halestrap A P , McStay GP, Clarke SJ (2002b) The permeability transition pore complex: another view. Biochimie 84:153-166. Hansen H H , Briem T, Dzietko M , Sifringer M , Voss A , Rzeski W, Zdzisinska B , Thor F, Heumann R, Stepulak A , Bittigau P, Ikonomidou C (2004) Mechanisms leading to disseminated apoptosis following N M D A receptor blockade in the developing rat brain. Neurobiol Dis 16:440-453. Harada H , Grant S (2003) Apoptosis regulators. Rev Cl in Exp Hematol 7:117-138. Harada H , Kel ly PJ, Cole DJ , Drummond JC, Patel P M (1999) Isoflurane reduces N -methyl-D-aspartate toxicity in vivo in the rat cerebral cortex. Anesth Analg 89:1442-1447. Hardingham G E , Bading H (2002) Coupling of extrasynaptic N M D A receptors to a C R E B shut-off pathway is developmentally regulated. Biochim Biophys Acta 1600:148-153. Hardingham G E , Chawla S, Cruzalegui F H , Bading H (1999) Control of recruitment and transcription-activating function of C B P determines gene regulation by N M D A receptors and L-type calcium channels. Neuron 22:789-798. Hardingham G E , Fukunaga Y , Bading H (2002) Extrasynaptic N M D A R s oppose synaptic N M D A R s by triggering C R E B shut-off and cell death pathways. Nat Neurosci 5:405-414. Hatton CJ , Paoletti P (2005) Modulation of triheteromeric N M D A receptors by N -terminal domain ligands. Neuron 46:261-274. Hawn A M , Baldwin L M (1997) Thrombolytic therapy for acute ischemic stroke. J Fam Pract 45:470. Hayashi T, Abe K (2004) Ischemic neuronal cell death and organellae damage. Neurol 145 Res 26:827-834. Hewitt DJ (2000) The use of NMDA-receptor antagonists in the treatment of chronic pain. Cl in J Pain 16:S73-S79. H i l l M M , Adrain C, Martin SJ (2003) Portrait of a killer: the mitochondrial apoptosome emerges from the shadows. M o l Interv 3:19-26. Hinkle JL, Bowman L (2003) Neuroprotection for ischemic stroke. J Neurosci Nurs 35:114-118. Hossmann K A (1994) Viability thresholds and the penumbra of focal ischemia. Ann Neurol 36:557-565. Huang Y , McNamara JO (2004) Ischemic stroke: \"acidotoxicity\" is a perpetrator. Cel l 118:665-666. Huettner JE, Bean BP (1988) Block of N-methyl-D-aspartate-activated current by the anticonvulsant MK-801 : selective binding to open channels. Proc Natl Acad Sci U S A 85:1307-1311. Ikonomidou C, Bosch F, Miksa M , Bittigau P, Vockler J, Dikranian K , Tenkova TI, Stefovska V , Turski L , Olney JW (1999a) Blockade of N M D A receptors and apoptotic neurodegeneration in the developing brain. Science 283:70-74. Ikonomidou C, Bosch F, Miksa M , Bittigau P, Vockler J, Dikranian K , Tenkova TI, Stefovska V , Turski L , Olney JW (1999b) Blockade of N M D A receptors and apoptotic neurodegeneration in the developing brain. Science 283:70-74. Ikonomidou C, Turski L (2002b) Why did N M D A receptor antagonists fail clinical trials for stroke and traumatic brain injury? Lancet Neurol 1:383-386. Ikonomidou C, Turski L (2002a) Why did N M D A receptor antagonists fail clinical trials for stroke and traumatic brain injury? Lancet Neurol 1:383-386. Jahr C E (1992) High probability opening of N M D A receptor channels by L-glutamate. Science 255:470-472. Jahr C E , Stevens C F (1990) Voltage dependence of NMDA-activated macroscopic conductances predicted by single-channel kinetics. J Neurosci 10:3178-3182. Jahr C E , Stevens C F (1993) Calcium permeability of the N-methyl-D-aspartate receptor channel in hippocampal neurons in culture. Proc Natl Acad Sci U S A 90:11573-11577. Jansen M , Dannhardt G (2003) Antagonists and agonists at the glycine site of the N M D A receptor for therapeutic interventions. Eur J M e d Chem 38:661-670. Janssen W G , Vissavajjhala P, Andrews G , Moran T, Hof PR, Morrison J H (2005) 146 Cellular and synaptic distribution of N R 2 A and N R 2 B in macaque monkey and rat hippocampus as visualized with subunit-specific monoclonal antibodies. Exp Neurol 191 Suppl 1.S28-S44. Javitt D C (2006) Is the glycine site half saturated or half unsaturated? Effects of glutamatergic drugs in schizophrenia patients. Curr Opin Psychiatry 19:151-157. Jenkinson D (2004) Thrombolytic therapy for acute ischaemic stroke. Hosp Med 65:164-169. Jiang M , Deng L , Chen G (2004) High Ca(2+)-phosphate transfection efficiency enables single neuron gene analysis. Gene Ther 11:1303-1311. Juurlink B H , Sweeney M I (1997) Mechanisms that result in damage during and following cerebral ischemia. Neurosci Biobehav Rev 21:121-128. Kadotani H , Namura S, Katsuura G , Terashima T, Kikuchi H (1998) Attenuation of focal cerebral infarct in mice lacking N M D A receptor subunit NR2C. Neuroreport 9:471-475. Kass IS, Lipton P (1982) Mechanisms involved in irreversible anoxic damage to the in vitro rat hippocampal slice. J Physiol 332:459-72.:459-472. Kemp JA, McKernan R M (2002) N M D A receptor pathways as drug targets. Nat Neurosci 5 Suppl: 1039-1042. Kerr JF, Wyllie A H , Currie A R (1972) Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics. Br J Cancer 26:239-257. Kew JN, Trube G , Kemp JA (1998) State-dependent N M D A receptor antagonism by Ro 8-4304, a novel N R 2 B selective, non-competitive, voltage-independent antagonist. Br J Pharmacol 123:463-472. K i m JH, Liao D, Lau L F , Huganir R L (1998) SynGAP: a synaptic RasGAP that associates with the PSD-95/SAP90 protein family. Neuron 20:683-691. K i m M J , Dunah A W , Wang Y T , Sheng M (2005) Differential roles of N R 2 A - and NR2B-containing N M D A receptors in Ras-ERK signaling and A M P A receptor trafficking. Neuron 46:745-760. Kirino T (1982) Delayed neuronal death in the gerbil hippocampus following ischemia. Brain Res 239:57-69. Kischkel F C , Hellbardt S, Behrmann I, Germer M , Pawlita M , Krammer P H , Peter M E (1995) Cytotoxicity-dependent APO-1 (Fas/CD95)-associated proteins form a death-inducing signaling complex (DISC) with the receptor. E M B O J 14:5579-5588. Kleckner N W , Dingledine R (1988) Requirement for glycine in activation of N M D A -receptors expressed in Xenopus oocytes. Science 241:835-837. 147 Kohr G, Jensen V , Koester HJ , Mihaljevic A L , Utvik JK, Kvello A , Ottersen OP, Seeburg P H , Sprengel R, Hvalby O (2003) Intracellular domains of N M D A receptor subtypes are determinants for long-term potentiation induction. J Neurosci 23:10791-10799. Komiyama N H , Watabe A M , Carlisle HJ , Porter K , Charlesworth P, Monti J, Strathdee DJ, O'Carroll C M , Martin SJ, Morris R G , O'Dell TJ, Grant S G (2002) SynGAP regulates E R K / M A P K signaling, synaptic plasticity, and learning in the complex with postsynaptic density 95 and N M D A receptor. J Neurosci 22:9721-9732. Kontos H A (2001) Oxygen radicals in cerebral ischemia: the 2001 Will is lecture. Stroke 32:2712-2716. Kornau H C , Schenker L T , Kennedy M B , Seeburg P H (1995) Domain interaction between N M D A receptor subunits and the postsynaptic density protein PSD-95. Science 269:1737-1740. Krapivinsky G, Krapivinsky L , Manasian Y , Ivanov A , Tyzio R, Pellegrino C, Ben-Ari Y , Clapham D E , Medina I (2003) The N M D A receptor is coupled to the E R K pathway by a direct interaction between N R 2 B and R a s G R F l . Neuron 40:775-784. Kristian T, Gido G , Kuroda S, Schutz A , Siesjo B K (1998) Calcium metabolism of focal and penumbral tissues in rats subjected to transient middle cerebral artery occlusion. Exp Brain Res 120:503-509. Kristian T, Siesjo B K (1998) Calcium in ischemic cell death. Stroke 29:705-718. Krupp JJ, Vissel B , Heinemann SF, Westbrook G L (1998) N-terminal domains in the NR2 subunit control desensitization of N M D A receptors. Neuron 20:317-327. Kumar C, Okuda M , Ikai I, Chance B (1990) Luminol enhanced chemiluminescence of the perfused rat heart during ischemia and reperfusion. F E B S Lett 272:121-124. Kuner T, Wollmuth L P , Karlin A , Seeburg P H , Sakmann B (1996) Structure of the N M D A receptor channel M 2 segment inferred from the accessibility of substituted cysteines. Neuron 17:343-352. Kuryatov A , Laube B , Betz H , Kuhse J (1994) Mutational analysis of the glycine-binding site of the N M D A receptor: structural similarity with bacterial amino acid-binding proteins. Neuron 12:1291-1300. Kutsuwada T, Sakimura K , Manabe T, Takayama C, Katakura N , Kushiya E , Natsume R, Watanabe M , Inoue Y , Yagi T, Aizawa S, Arakawa M , Takahashi T, Nakamura Y , Mor i H , Mishina M (1996) Impairment of suckling response, trigeminal neuronal pattern formation, and hippocampal L T D in N M D A receptor epsilon 2 subunit mutant mice. Neuron 16:333-344. Kuwana T, Newmeyer D D (2003) Bcl-2-family proteins and the role of mitochondria in 148 apoptosis. Curr Opin Cell B io l 15:691-699. Laube B , Kuhse J, Betz H (1998) Evidence for a tetrameric structure of recombinant N M D A receptors. J Neurosci 18:2954-2961. Lauber K , Appel H A , Schlosser SF, Gregor M , Schulze-Osthoff K , Wesselborg S (2001) The adapter protein apoptotic protease-activating factor-1 (Apaf-1) is proteolytically processed during apoptosis. J B io l Chem 276:29772-29781. Lavezzari G , McCallum J, Dewey C M , Roche K W (2004) Subunit-specific regulation of N M D A receptor endocytosis. J Neurosci 24:6383-6391. Lee BI, Lee DJ , Cho K J , K i m G W (2005) Early nuclear translocation of endonuclease G and subsequent D N A fragmentation after transient focal cerebral ischemia in mice. Neurosci Lett. Lee FJ, Xue S, Pei L , Vukusic B , Chery N , Wang Y , Wang Y T , Niznik H B , Y u X M , L i u F (2002) Dual regulation of N M D A receptor functions by direct protein-protein interactions with the dopamine DI receptor. Cell 111:219-230. Lee J M , Zipfel GJ , Choi D W (1999b) The changing landscape of ischaemic brain injury mechanisms. Nature 399:A7-14. Lee J M , Zipfel GJ , Choi D W (1999a) The changing landscape of ischaemic brain injury mechanisms. Nature 399.A7-14. Lees K R (1997) Cerestat and other N M D A antagonists in ischemic stroke. Neurology 49.S66-S69. Lees K R , Asplund K , Carolei A , Davis S M , Diener H C , Kaste M , Orgogozo J M , Whitehead J (2000) Glycine antagonist (gavestinel) in neuroprotection ( G A I N International) in patients with acute stroke: a randomised controlled trial. G A I N International Investigators. Lancet 355:1949-1954. Leonard A S , L i m IA, Hemsworth D E , Home M C , Hell JW (1999) Calcium/calmodulin-dependent protein kinase II is associated with the N-methyl-D-aspartate receptor. Proc Natl Acad Sci U S A 96:3239-3244. Letai A , Bassik M C , Walensky L D , Sorcinelli M D , Weiler S, Korsmeyer SJ (2002) Distinct BH3 domains either sensitize or activate mitochondrial apoptosis, serving as prototype cancer therapeutics. Cancer Cel l 2:183-192. L i C Q , Wogan G N (2005) Nitric oxide as a modulator of apoptosis. Cancer Lett 226:1-15. L i H , Zhu H , X u CJ, Yuan J (1998a) Cleavage of BID by caspase 8 mediates the mitochondrial damage in the Fas pathway of apoptosis. Cel l 94:491-501. L i JH, Wang Y H , Wolfe B B , Krueger K E , Corsi L , Stocca G , Vic in i S (1998b) 149 Developmental changes in localization of N M D A receptor subunits in primary cultures of cortical neurons. Eur J Neurosci 10:1704-1715. L i L , Fan M , Icton C D , Chen N , Leavitt BR, Hayden M R , Murphy T H , Raymond L A (2003) Role of NR2B-type N M D A receptors in selective neurodegeneration in Huntington disease. Neurobiol Aging 24:1113-1121. L i L Y , Luo X , Wang X (2001) Endonuclease G is an apoptotic DNase when released from mitochondria. Nature 412:95-99. L i n R (2002) Preface. In: New Concepts in Cerebral Ischemia pp vi i -v i i i . New York: C R C Press. Lipton P (1999a) Ischemic cell death in brain neurons. Physiol Rev 79:1431-1568. Lipton S A (1999b) Neuronal protection and destruction by N O . Cel l Death Differ 6:943-951. Lipton SA, Nicotera P (1998) Calcium, free radicals and excitotoxins in neuronal apoptosis. Cell Calcium 23:165-171. Lipton SA, Rosenberg P A (1994) Excitatory amino acids as a final common pathway for neurologic disorders. N Engl J Med 330:613-622. L i u L , Wong TP, Pozza M F , Lingenhoehl K , Wang Y , Sheng M , Auberson Y P , Wang Y T (2004a) Role of N M D A receptor subtypes in governing the direction of hippocampal synaptic plasticity. Science 304:1021 -1024. L i u L , Wong TP, Pozza M F , Lingenhoehl K , Wang Y , Sheng M , Auberson Y P , Wang Y T (2004b) Role of N M D A receptor subtypes in governing the direction of hippocampal synaptic plasticity. Science 304:1021-1024. L i u X B , Murray K D , Jones E G (2004c) Switching of N M D A receptor 2 A and 2B subunits at thalamic and cortical synapses during early postnatal development. J Neurosci 24:8885-8895. L i u Y , Belayev L , Zhao W, Busto R, Ginsberg M D (2000) M R Z 2/579, a novel uncompetitive N-methyl-D-aspartate antagonist, reduces infarct volume and brain swelling and improves neurological deficit after focal cerebral ischemia in rats. Brain Res 862:111-119. Loew L M , Carrington W, Tuft R A , Fay FS (1994) Physiological cytosolic Ca2+ transients evoke concurrent mitochondrial depolarizations. Proc Natl Acad Sci U S A 91:12579-12583. Longa EZ , Weinstein PR, Carlson S, Cummins R (1989) Reversible middle cerebral artery occlusion without craniectomy in rats. Stroke 20:84-91. 150 Loschmann P A , De G C , Smith L , Wullner U , Fischer G, Kemp JA, Jenner P, Klockgether T (2004) Antiparkinsonian activity of Ro 25-6981, a NR2B subunit specific N M D A receptor antagonist, in animal models of Parkinson's disease. Exp Neurol 187:86-93. Love S (2003) Apoptosis and brain ischaemia. Prog Neuropsychopharmacol Bio l Psychiatry 27:267-282. L u C X , Fan TJ, Hu G B , Cong RS (2003) Apoptosis-inducing factor and apoptosis. Sheng Wu Hua Xue Y u Sheng Wu Wu L i Xue Bao (Shanghai) 35:881-885. L u W, Man H , Ju W, Trimble WS, MacDonald JF, Wang Y T (2001) Activation of synaptic N M D A receptors induces membrane insertion of new A M P A receptors and L T P in cultured hippocampal neurons. Neuron 29:243-254. L U C A S DR, N E W H O U S E JP (1957) The toxic effect of sodium L-glutamate on the inner layers of the retina. A M A Arch Ophthalmol 58:193-201. Luo X , Budihardjo I, Zou H , Slaughter C, Wang X (1998) Bid , a Bcl2 interacting protein, mediates cytochrome c release from mitochondria in response to activation of cell surface death receptors. Cel l 94:481-490. Lutsep H L , Clark W M (2001) Current status of neuroprotective agents in the treatment of acute ischemic stroke. Curr Neurol Neurosci Rep 1:13-18. Lynch DR, Gallagher M J (1996) Inhibition of N-methyl-D-aspartate receptors by haloperidol: developmental and pharmacological characterization in native and recombinant receptors. J Pharmacol Exp Ther 279:154-161. Lynch DR, Guttmann RP (2002) Excitotoxicity: perspectives based on N-methyl-D-aspartate receptor subtypes. J Pharmacol Exp Ther 300:717-723. MacDonald JF, Bartlett M C , Mody I, Pahapill P, Reynolds JN, Salter M W , Schneiderman JH, Pennefather PS (1991) Actions of ketamine, phencyclidine and M K -801 on N M D A receptor currents in cultured mouse hippocampal neurones. J Physiol 432:483-508. MacDonald JF, Bartlett M C , Mody I, Reynolds JN, Salter M W (1990) The PCP site of the N M D A receptor complex. Adv Exp Med B io l 268:27-34. Madden K (2002) Optimal timing of thrombolytic therapy in acute ischaemic stroke. C N S Drugs 16:213-218. Majno G , Joris I (1995) Apoptosis, oncosis, and necrosis. A n overview of cell death. A m J Pathol 146:3-15. Martin L J , Chen K , L i u Z (2005) Adult motor neuron apoptosis is mediated by nitric oxide and Fas death receptor linked by D N A damage and p53 activation. J Neurosci 151 25:6449-6459. Massey P V , Johnson B E , Moult PR, Auberson Y P , Brown M W , Molnar E, Collingridge G L , Bashir ZI (2004b) Differential roles of N R 2 A and NR2B-containing N M D A receptors in cortical long-term potentiation and long-term depression. J Neurosci 24:7821-7828. Massey P V , Johnson B E , Moult PR, Auberson Y P , Brown M W , Molnar E, Collingridge G L , Bashir ZI (2004a) Differential roles of N R 2 A and NR2B-containing N M D A receptors in cortical long-term potentiation and long-term depression. J Neurosci 24:7821-7828. Mayer M L , Vykl icky L , Jr., Clements J (1989) Regulation of N M D A receptor desensitization in mouse hippocampal neurons by glycine. Nature 338:425-427. McDonald JW, Johnston M V (1990) Physiological and pathophysiological roles of excitatory amino acids during central nervous system development. Brain Res Brain Res Rev 15:41-70. Menniti F, Chenard B , Collins M , Ducat M , Shalaby I, White F (1997) CP-101,606, a potent neuroprotectant selective for forebrain neurons. Eur J Pharmacol 331:117-126. Mielke JG, Wang Y T (2005) Insulin exerts neuroprotection by counteracting the decrease in cell-surface G A B A receptors following oxygen-glucose deprivation in cultured cortical neurons. J Neurochem 92:103-113. Mielke O, Wardlaw J, L i u M (2004) Thrombolysis (different doses, routes of administration and agents) for acute ischaemic stroke. Cochrane Database Syst RevCD000514. Mignotte B , Vayssiere JL (1998) Mitochondria and apoptosis. Eur J Biochem 252:1-15. Mishra OP, ivoria-Papadopoulos M (1999) Cellular mechanisms of hypoxic injury in the developing brain. Brain Res Bul l 48:233-238. Mohrmann R, Hart H , Gottmann K (2000) Developmental regulation of subunit composition of extrasynaptic N M D A receptors in neocortical neurones. Neuroreport 11:1203-1208. Momiyama A (2000) Distinct synaptic and extrasynaptic N M D A receptors identified in dorsal horn neurones of the adult rat spinal cord. J Physiol 523 Pt 3:621-628. Monyer H , Burnashev N , Laurie DJ, Sakmann B , Seeburg P H (1994) Developmental and regional expression in the rat brain and functional properties of four N M D A receptors. Neuron 12:529-540. Morikawa E , Mor i H , Kiyama Y , Mishina M , Asano T, Kirino T (1998) Attenuation of focal ischemic brain injury in mice deficient in the epsilonl (NR2A) subunit of N M D A 152 receptor. J Neurosci 18:9727-9732. Murphy J (2003) Pharmacological treatment of acute ischemic stroke. Crit Care Nurs Q 26:276-282. Murray J, Taylor SW, Zhang B , Ghosh SS, Capaldi R A (2003) Oxidative damage to mitochondrial complex I due to peroxynitrite: identification of reactive tyrosines by mass spectrometry. J B i o l Chem 278:37223-37230. Mutel V , Buchy D, Klingelschmidt A , Messer J, Bleuel Z , Kemp JA, Richards JG (1998) In vitro binding properties in rat brain of [3H]Ro 25-6981, a potent and selective antagonist of N M D A receptors containing N R 2 B subunits. J Neurochem 70:2147-2155. Nedergaard M , Gjedde A , Diemer N H (1986) Focal ischemia of the rat brain: autoradiographic determination of cerebral glucose utilization, glucose content, and blood flow. J Cereb Blood Flow Metab 6:414-424. Nedergaard M , Goldman SA, Desai S, Pulsinelli W A (1991a) Acid-induced death in neurons and glia. J Neurosci 11:2489-2497. Nedergaard M , Kraig RP, Tanabe J, Pulsinelli W A (1991b) Dynamics of interstitial and intracellular pH in evolving brain infarct. A m J Physiol 260:R581-R588. Nellgard B , Wieloch T (1992) Postischemic blockade of A M P A but not N M D A receptors mitigates neuronal damage in the rat brain following transient severe cerebral ischemia. J Cereb Blood Flow Metab 12:2-11. Neumar R W (2000) Molecular mechanisms of ischemic neuronal injury. Ann Emerg Med 36:483-506. Neyton J, Paoletti P (2006) Relating N M D A receptor function to receptor subunit composition: limitations of the pharmacological approach. J Neurosci 26:1331-1333. Nicholls D G (2004) Mitochondrial dysfunction and glutamate excitotoxicity studied in primary neuronal cultures. Curr M o l Med 4:149-177. Niethammer M , K i m E , Sheng M (1996a) Interaction between the C terminus of N M D A receptor subunits and multiple members of the PSD-95 family of membrane-associated guanylate kinases. J Neurosci 16:2157-2163. Niethammer M , K i m E, Sheng M (1996b) Interaction between the C terminus of N M D A receptor subunits and multiple members of the PSD-95 family of membrane-associated guanylate kinases. J Neurosci 16:2157-2163. NINDS (1995) Tissue plasminogen activator for acute ischemic stroke. The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group. N Engl J Med 333:1581-1587. 153 Ning K , Pei L , Liao M , L i u B , Zhang Y , Jiang W, Mielke JG, L i L , Chen Y , El-Hayek Y H , Fehlings M G , Zhang X , L i u F, Eubanks J, Wan Q (2004) Dual neuroprotective signaling mediated by downregulating two distinct phosphatase activities of P T E N . J Neurosci 24:4052-4060. Nishizawa Y (2001) Glutamate release and neuronal damage in ischemia. Life Sci 69:369-381. Nong Y , Huang Y Q , Ju W, Kalia L V , Ahmadian G , Wang Y T , Salter M W (2003) Glycine binding primes N M D A receptor internalization. Nature 422:302-307. Nowak L , Bregestovski P, Ascher P, Herbet A , Prochiantz A (1984) Magnesium gates glutamate-activated channels in mouse central neurones. Nature 307:462-465. Obrenovitch TP, Richards D A (1995) Extracellular neurotransmitter changes in cerebral ischaemia. Cerebrovasc Brain Metab Rev 7:1-54. Olney JW (1969) Brain lesions, obesity, and other disturbances in mice treated with monosodium glutamate. Science 164:719-721. Olney JW, Labruyere J, Price M T (1989) Pathological changes induced in cerebrocortical neurons by phencyclidine and related drugs. Science 244:1360-1362. Olney JW, Sharpe L G (1969) Brain lesions in an infant rhesus monkey treated with monsodium glutamate. Science 166:386-388. Paschen W (2000) Role of calcium in neuronal cell injury: which subcellular compartment is involved? Brain Res Bul l 53:409-413. Paschen W, Doutheil J (1999) Disturbances of the functioning of endoplasmic reticulum: a key mechanism underlying neuronal cell injury? J Cereb Blood Flow Metab 19:1-18. Pettmann B , Henderson C E (1998) Neuronal cell death. Neuron 20:633-647. Porter A G , Janicke R U (1999) Emerging roles of caspase-3 in apoptosis. Cell Death Differ 6:99-104. Posey DJ , Kern D L , Swiezy N B , Sweeten T L , Wiegand R E , McDougle C J (2004) A pilot study of D-cycloserine in subjects with autistic disorder. A m J Psychiatry 161:2115-2117. Premkumar L S , Auerbach A (1997) Stoichiometry of recombinant N-methyl-D-aspartate receptor channels inferred from single-channel current patterns. J Gen Physiol 110:485-502. Puka-Sundvall M , Hallin U , Zhu C, Wang X , Karlsson JO, Blomgren K , Hagberg H (2000) N M D A blockade attenuates caspase-3 activation and D N A fragmentation after neonatal hypoxia-ischemia. Neuroreport 11:2833-2836. 154 Pulsinelli W A , Brierley JB (1979) A new model of bilateral hemispheric ischemia in the unanesthetized rat. Stroke 10:267-272. Pulsinelli W A , Brierley JB, Plum F (1982) Temporal profile of neuronal damage in a model of transient forebrain ischemia. Ann Neurol 11:491-498. Rehncrona S (1985) Brain acidosis. Ann Emerg Med 14:770-776. Reyes M , Reyes A , Opitz T, Kapin M A , Stanton P K (1998) Eliprodil, a non-competitive, NR2B-selective N M D A antagonist, protects pyramidal neurons in hippocampal slices from hypoxic/ischemic damage. Brain Res 782:212-218. Rich T, Allen R L , Wyllie A H (2000) Defying death after D N A damage. Nature 407:777-783. Rosenmund C, Stern-Bach Y , Stevens CF (1998) The tetrameric structure of a glutamate receptor channel. Science 280:1596-1599. Rothman S M (1983) Synaptic activity mediates death of hypoxic neurons. Science 220:536-537. Ruppersberg JP, Mosbacher J, Gunther W, Schoepfer R, Fakler B (1993) Studying block in cloned N-methyl-D-aspartate ( N M D A ) receptors. Biochem Pharmacol 46:1877-1885. Sacco R L , DeRosa JT, Haley E C , Jr., Levin B , Ordronneau P, Phillips SJ, Rundek T, Snipes R G , Thompson JL (2001) Glycine antagonist in neuroprotection for patients with acute stroke: G A I N Americas: a randomized controlled trial. J A M A 285:1719-1728. Sakatani K , Iizuka H , Young W (1990) Somatosensory evoked potentials in rat cerebral cortex before and after middle cerebral artery occlusion. Stroke 21:124-132. Sareen D (2002) Neuroprotective agents in acute ischemic stroke. J Assoc Physicians India 50:250-258. Sattler R, Tymianski M (2000) Molecular mechanisms of calcium-dependent excitotoxicity. J M o l Med 78:3-13. Sattler R, Xiong Z , L u W Y , Hafner M , MacDonald JF, Tymianski M (1999) Specific coupling of N M D A receptor activation to nitric oxide neurotoxicity by PSD-95 protein. Science 284:1845-1848. Scaffidi C , Fulda S, Srinivasan A , Friesen C, L i F, Tomaselli K J , Debatin K M , Krammer P H , Peter M E (1998) Two CD95 (APO-l/Fas) signaling pathways. E M B O J 17:1675-1687. Schafer P, Scholz SR, Gimadutdinow O, Cymerman IA, Bujnicki J M , Ruiz-Carrillo A , Pingoud A , Meiss G (2004) Structural and functional characterization of mitochondrial EndoG, a sugar non-specific nuclease which plays an important role during apoptosis. J 155 M o l B io l 338:217-228. Schellinger PD, Warach S (2004) Therapeutic time window of thrombolytic therapy following stroke. Curr Atheroscler Rep 6:288-294. Schinder A F , Olson E C , Spitzer N C , Montal M (1996) Mitochondrial dysfunction is a primary event in glutamate neurotoxicity. J Neurosci 16:6125-6133. Schneggenburger R (1996) Simultaneous measurement of Ca2+ influx and reversal potentials in recombinant N-methyl-D-aspartate receptor channels. Biophys J 70:2165-2174. Schwarze SR, Ho A , Vocero-Akbani A , Dowdy SF (1999) In vivo protein transduction: delivery of a biologically active protein into the mouse. Science 285:1569-1572. Scorrano L , Korsmeyer SJ (2003) Mechanisms of cytochrome c release by proapoptotic B C L - 2 family members. Biochem Biophys Res Commun 304:437-444. Sharpe JC, Arnoult D , Youle R J (2004) Control of mitochondrial permeability by Bcl-2 family members. Biochim Biophys Acta 1644:107-113. Sheng M , Cummings J, Roldan L A , Jan Y N , Jan L Y (1994) Changing subunit composition of heteromeric N M D A receptors during development of rat cortex. Nature 368:144-147. Sheng M , Pak D T (2000) Ligand-gated ion channel interactions with cytoskeletal and signaling proteins. Annu Rev Physiol 62:755-778. Siesjo B K , Katsura K , Kristian T (1996) Acidosis-related damage. Adv Neurol 71:209-233. Silver IA, Erecinska M (1992) Ion homeostasis in rat brain in vivo: intra- and extracellular [Ca2+] and [H+] in the hippocampus during recovery from short-term, transient ischemia. J Cereb Blood Flow Metab 12:759-772. Silver IA, Erecinska M (1990) Intracellular and extracellular changes of [Ca2+] in hypoxia and ischemia in rat brain in vivo. J Gen Physiol 95:837-866. Simon RP, Swan JH, Griffiths T, Meldrum BS (1984) Blockade of N-methyl-D-aspartate receptors may protect against ischemic damage in the brain. Science 226:850-852. Sims N R , Anderson M F (2002) Mitochondrial contributions to tissue damage in stroke. Neurochem Int 40:511-526. Slee E A , Adrain C, Martin SJ (2001) Executioner caspase-3, -6, and -7 perform distinct, non-redundant roles during the demolition phase of apoptosis. J B io l Chem 276:7320-7326. 156 Snider BJ , Gottron FJ, Choi D W (1999) Apoptosis and necrosis in cerebrovascular disease. Ann N Y Acad Sci 893:243-253. Sprengel R, Suchanek B , Amico C, Brusa R, Burnashev N , Rozov A , Hvalby O, Jensen V , Paulsen O, Andersen P, K i m JJ, Thompson RF, Sun W, Webster L C , Grant SG, Eilers J, Konnerth A , L i J, McNamara JO, Seeburg P H (1998) Importance of the intracellular domain of NR2 subunits for N M D A receptor function in vivo. Cel l 92:279-289. Steigerwald F, Schulz TW, Schenker L T , Kennedy M B , Seeburg P H , Kohr G (2000) C-Terminal truncation of N R 2 A subunits impairs synaptic but not extrasynaptic localization of N M D A receptors. J Neurosci 20:4573-4581. Stennicke HR, Salvesen GS (1997) Biochemical characteristics of caspases-3, -6, -7, and -8. J B io l Chem 272:25719-25723. Stocca G , Vic in i S (1998) Increased contribution of N R 2 A subunit to synaptic N M D A receptors in developing rat cortical neurons. J Physiol 507 ( Pt 1): 13-24. Strack S, Colbran RJ (1998) Autophosphorylation-dependent targeting of calcium/ calmodulin-dependent protein kinase II by the N R 2 B subunit of the N-methyl- D -aspartate receptor. J B io l Chem 273:20689-20692. Strack S, McNei l l R B , Colbran RJ (2000) Mechanism and regulation of calcium/calmodulin-dependent protein kinase II targeting to the N R 2 B subunit of the N -methyl-D-aspartate receptor. J B io l Chem 275:23798-23806. Strong A J , Venables GS, Gibson G (1983) The cortical ischaemic penumbra associated with occlusion of the middle cerebral artery in the cat: 1. Topography of changes in blood flow, potassium ion activity, and E E G . J Cereb Blood Flow Metab 3:86-96. Sun G Y , Zhang JP, L i n T A , L i n T N , He Y Y , Hsu C Y (1995) Inositol trisphosphate, polyphosphoinositide turnover, and high-energy metabolites in focal cerebral ischemia and reperfusion. Stroke 26:1893-1900. Sussman M S , Bulkley G B (1990) Oxygen-derived free radicals in reperfusion injury. Methods Enzymol 186:711-723. Sutton G , Chandler L J (2002) Activity-dependent N M D A receptor-mediated activation of protein kinase B/Akt in cortical neuronal cultures. J Neurochem 82:1097-1105. Symon L (1975) Experimental model of stroke in the baboon. Adv Neurol 10:199-212. Tamura A , Graham DI, McCulloch J, Teasdale G M (1981) Focal cerebral ischaemia in the rat: 2. Regional cerebral blood flow determined by [14C]iodoantipyrine autoradiography following middle cerebral artery occlusion. J Cereb Blood Flow Metab 1:61-69. Taylor CP, Burke SP, Weber M L (1995) Hippocampal slices: glutamate overflow and 157 cellular damage from ischemia are reduced by sodium-channel blockade. J Neurosci Methods 59:121-128. Tenneti L , Lipton S A (2000) Involvement of activated caspase-3-like proteases in N -methyl-D-aspartate-induced apoptosis in cerebrocortical neurons. J Neurochem 74:134-142. Ter Horst GJ , Postigo A (1997) Stroke: prevalence and mechanisms of cell death. In: Clinical Pharmacology of Cerebral Ischemia pp 1-30. New Jersey: Humana. Thomas C G , Mil ler A J , Westbrook G L (2006) Synaptic and extrasynaptic N M D A receptor NR2 subunits in cultured hippocampal neurons. J Neurophysiol 95:1727-1734. Thornberry N A , Lazebnik Y (1998) Caspases: enemies within. Science 281:1312-1316. Tigaret C M , Thalhammer A , Rast G F , Specht C G , Auberson Y P , Stewart M G , Schoepfer R (2006a) Subunit dependencies of N-methyl-D-aspartate ( N M D A ) receptor-induced alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid ( A M P A ) receptor internalization. M o l Pharmacol 69:1251-1259. Tigaret C M , Thalhammer A , Rast GF, Specht C G , Auberson Y P , Stewart M G , Schoepfer R (2006b) Subunit dependencies of N M D A receptor-induced A M P A receptor internalization. M o l Pharmacol. Tigaret C M , Thalhammer A , Rast GF, Specht C G , Auberson Y P , Stewart M G , Schoepfer R (2006c) Subunit dependencies of N M D A receptor-induced A M P A receptor internalization. M o l Pharmacol. Tombaugh G C , Sapolsky R M (1993) Evolving concepts about the role of acidosis in ischemic neuropathology. J Neurochem 61:793-803. Tomsick T A (2004) Intravenous thrombolysis for acute ischemic stroke. J Vase Interv Radiol 15:S67-S76. Tovar K R , Westbrook G L (1999) The incorporation of N M D A receptors with a distinct subunit composition at nascent hippocampal synapses in vitro. J Neurosci 19:4180-4188. Tovar K R , Westbrook G L (2002) Mobile N M D A receptors at hippocampal synapses. Neuron 34:255-264. Tsujimoto Y (1998) Role of Bcl-2 family proteins in apoptosis: apoptosomes or mitochondria? Genes Cells 3:697-707. Tsujimoto Y , Shimizu S (2000) Bcl-2 family: life-or-death switch. F E B S Lett 466:6-10. Tymianski M , Charlton M P , Carlen P L , Tator C H (1993) Source specificity of early calcium neurotoxicity in cultured embryonic spinal neurons. J Neurosci 13:2085-2104. 158 Van de C M , Declercq W, Van db, I, Fiers W, Vandenabeele P (1999) The proteolytic procaspase activation network: an in vitro analysis. Cell Death Differ 6:1117-1124. van Z B , Yoshii A , Constantine-Paton M (2004) Receptor compartmentalization and trafficking at glutamate synapses: a developmental proposal. Trends Neurosci 27:428-437. Verhagen A M , Ekert P G , Pakusch M , Silke J, Connolly L M , Reid G E , Moritz R L , Simpson RJ, Vaux D L (2000) Identification of D I A B L O , a mammalian protein that promotes apoptosis by binding to and antagonizing TAP proteins. Cell 102:43-53. Wagner SR, Lanier W L (1994) Metabolism of glucose, glycogen, and high-energy phosphates during complete cerebral ischemia. A comparison of normoglycemic, chronically hyperglycemic diabetic, and acutely hyperglycemic nondiabetic rats. Anesthesiology 81:1516-1526. Wahl F, Obrenovitch TP, Hardy A M , Plotkine M , Boulu R, Symon L (1994) Extracellular glutamate during focal cerebral ischaemia in rats: time course and calcium dependency. J Neurochem 63:1003-1011. Wahlgren N G , Ahmed N (2004) Neuroprotection in cerebral ischaemia: facts and fancies\u00E2\u0080\u0094the need for new approaches. Cerebrovasc Dis 17 Suppl 1:153-166. Wang C X , Shuaib A (2005) N M D A / N R 2 B selective antagonists in the treatment of ischemic brain injury. Curr Drug Targets C N S Neurol Disord 4:143-151. Wang H G , Pathan N , Ethell I M , Krajewski S, Yamaguchi Y , Shibasaki F, McKeon F, Bobo T, Franke TF, Reed JC (1999) Ca2+-induced apoptosis through calcineurin dephosphorylation of B A D . Science 284:339-343. Wang Y , Ju W, L iu L , Fam S, D'Souza S, Taghibiglou C, Salter M , Wang Y T (2004) alpha-Amino-3-hydroxy-5-methylisoxazole-4-propionic acid subtype glutamate receptor ( A M P A R ) endocytosis is essential for N-methyl-D-aspartate-induced neuronal apoptosis. J B io l Chem 279:41267-41270. Warner DS, Martin H , Ludwig P, McAllister A , Keana JF, Weber E (1995) In vivo models of cerebral ischemia: effects of parenterally administered N M D A receptor glycine site antagonists. J Cereb Blood Flow Metab 15:188-196. Warner M A , Nei l l K H , Nadler JV, Crain BJ (1991) Regionally selective effects of N M D A receptor antagonists against ischemic brain damage in the gerbil. J Cereb Blood Flow Metab 11:600-610. Waxman E A , Lynch D R (2005) N-methyl-D-aspartate receptor subtypes: multiple roles in excitotoxicity and neurological disease. Neuroscientist 11:37-49. Wei M C , Zong W X , Cheng E H , Lindsten T, Panoutsakopoulou V , Ross A J , Roth K A , MacGregor GR, Thompson C B , Korsmeyer SJ (2001) Proapoptotic B A X and B A K : a requisite gateway to mitochondrial dysfunction and death. Science 292:727-730. 159 Weitlauf C, Honse Y , Auberson Y P , Mishina M , Lovinger D M , Winder D G (2005b) Activation of NR2A-containing N M D A receptors is not obligatory for N M D A receptor-dependent long-term potentiation. J Neurosci 25:8386-8390. Weitlauf C, Honse Y , Auberson Y P , Mishina M , Lovinger D M , Winder D G (2005a) Activation of NR2A-containing N M D A receptors is not obligatory for N M D A receptor-dependent long-term potentiation. J Neurosci 25:8386-8390. Welsh F A , Marcy V R , Sims R E (1991) N A D H fluorescence and regional energy metabolites during focal ischemia and reperfusion of rat brain. J Cereb Blood Flow Metab 11:459-465. Widlak P, Garrard WT (2005) Discovery, regulation, and action of the major apoptotic nucleases DFF40 /CAD and endonuclease G . J Cell Biochem 94:1078-1087. Wood P L (1995) The co-agonist concept: is the NMDA-associated glycine receptor saturated in vivo? Life Sci 57:301-310. Wood P L , Hawkinson JE (1997) N-methyl-D-aspartate antagonists for stroke and head trauma. Expert Opin Investig Drugs 6:389-397. Xiong 2 G , Zhu X M , Chu X P , Minami M , Hey J, Wei W L , MacDonald JF, Wemmie JA, Price M P , Welsh M J , Simon RP (2004) Neuroprotection in ischemia: blocking calcium-permeable acid-sensing ion channels. Cell 118:687-698. Yamakura T, Shimoji K (1999) Subunit- and site-specific pharmacology of the N M D A receptor channel. Prog Neurobiol 59:279-298. Yermolaieva O, Leonard A S , Schnizler M K , Abboud F M , Welsh M J (2004) Extracellular acidosis increases neuronal cell calcium by activating acid-sensing ion channel la . Proc Natl Acad Sci U S A 101:6752-6757. Yuan J, Lipinski M , Degterev A (2003) Diversity in the mechanisms of neuronal cell death. Neuron 40:401-413. Yuan J, Yankner B A (2000) Apoptosis in the nervous system. Nature 407:802-809. Zamzami N , Kroemer G (2001) The mitochondrion in apoptosis: how Pandora's box opens. Nat Rev M o l Cell B io l 2:67-71. Zeron M M , Hansson O, Chen N , Wellington C L , Leavitt BR, Brundin P, Hayden M R , Raymond L A (2002) Increased sensitivity to N-methyl-D-aspartate receptor-mediated excitotoxicity in a mouse model of Huntington's disease. Neuron 33:849-860. Zhang JP, Sun G Y (1995) Free fatty acids, neutral glycerides, and phosphoglycerides in transient focal cerebral ischemia. J Neurochem 64:1688-1695. Zhang L H , X u L , X u T L (2006) Glycine receptor activation regulates short-term 160 plasticity in CAI area of hippocampal slices of rats. Biochem Biophys Res Commun 344:721-726. Zhang X , Chen J, Graham SH, Du L, Kochanek P M , Draviam R, Guo F, Nathaniel PD, Szabo C, Watkins SC, Clark RS (2002) Intranuclear localization of apoptosis-inducing factor (AIF) and large scale DNA fragmentation after traumatic brain injury in rats and in neuronal cultures exposed to peroxynitrite. J Neurochem 82:181-191. Zhang ZG, Chopp M , Gautam S, Zaloga C, Zhang RL, Schmidt HH, Pollock JS, Forstermann U (1994) Upregulation of neuronal nitric oxide synthase and mRNA, and selective sparing of nitric oxide synthase-containing neurons after focal cerebral ischemia in rat. Brain Res 654:85-95. Zhu D, Lipsky RH, Marini A M (2002) Co-activation of the phosphatidylinositol-3-kinase/Akt signaling pathway by N-methyl-D-aspartate and TrkB receptors in cerebellar granule cell neurons. Amino Acids 23:11-17. Zhu D, Wu X , Strauss KI, Lipsky RH, Qureshi Z, Terhakopian A, Novelli A , Banaudha K, Marini A M (2005) N-methyl-D-aspartate and TrkB receptors protect neurons against glutamate excitotoxicity through an extracellular signal-regulated kinase pathway. J Neurosci Res 80:104-113. Zipfel GJ, Babcock DJ, Lee JM, Choi DW (2000) Neuronal apoptosis after CNS injury: the roles of glutamate and calcium. J Neurotrauma 17:857-869. Zou H, L i Y , Liu X , Wang X (1999) An APAF-1.cytochrome c multimeric complex is a functional apoptosome that activates procaspase-9. J Biol Chem 274:11549-11556. 161 APPENDIX Part of the work presented in this thesis has been published in: A a r t s M* , L i u Y * , L i u L , B e s s h o h S, A r u n d i n e M , G u r d J W , W a n g Y T , S a l t e r M W , T y m i a n s k i M (2002) . T r e a t m e n t o f i s c h e m i c b r a i n d a m a g e b y p e r t u r b i n g N M D A r e c e p t o r - P S D - 9 5 p r o t e i n i n t e r a c t i o n s . S c i e n c e 2 9 8 ( 5 5 9 4 ) : 8 4 6 - 5 0 . 162 R E P O R T S with Tat38-48-dansyl (10 pM), devoid of the transduction domain, exhibited only back-ground signal indicating no peptide uptake (Fig. IB, right). Tat-NR2B9c-dansyl accumulation was detectable in neurons within 10 min of application, peaked during the next 20 min, and remained detectable for 5 hours after washing the peptide from the bath (Fig. 1C). Next we determined whether Tat-NR2B9c could perturb preformed NMDAR-PSD-95 complexes by examining its effects on the coimmunoprecipitation of PSD-95 with NR2 subunits (13). The P2 membrane protein frac-tion of rat forebrain tissue, which is enriched in synaptic structures, was incubated with Tat-NR2B9c or with one of three controls: Tat38-48, the Tat transduction sequence con-jugated to two alanine residues (Tat-AA), or a Tat-NR2B9c peptide in which the COOH-terminal tSXV motif contained a double point mutation (Lys-Leu-Ser-Ser-lle-Glu-Ala-Asp-Ala; Tat-NR2B-AA) rendering it incapable of binding PSD-95 (J). None of these controls, each lacking an intact PDZ binding motif, reduced the coimmunoprecipitation of PSD-95 with NR2B. However, Tat-NR2B9c, in which the Ile-Glu-Ser-Asp-Val sequence is preceded by residues unique to the NR2B COOH-terrninus, selectively reduced the co-immunoprecipitation of PSD-95 with NR2B (Fig. ID), but not with NR2A (Fig. IE). Thus, NR2A may be more tightly bound to PSD-95. Alternatively, the incomplete ho-mology of Tat-NR2B9c for the NR2A COOH-terrninus may make it less effective in perturbing PSD-95-NR2A binding (13). Because synaptic and whole-cell NMDAR currents are unaffected when PSD-95 is lacking (6, 7), we examined NMDAR currents and C a 2 + fluxes when PSD-95 is dissociated. Bath application of Tat-NR2B9c (50 nM) to acute rat hippocampal slices had no effect on synaptic responses of CAI neurons evoked by stimula-tion of the Schaffer collateral-commissural pathway (Fig. 2A). Tat-NR2B9c also had no effect on patch recordings in CAI neurons of the primarily AMPAR (AMPA receptor)-me-diated total excitatory postsynaptic current (EPSC) (Fig. 2B) (IS), nor on the isolated NMDA component of the EPSC (Fig. 2C) (13). Moreover, pretreating cultured cortical neurons with Tat-NR2B9c or with pTat-PDZl-2 (each at 50 nM) did not alter 4 5 C a 2 + uptake produced by applying NMDA (Fig. 2D). The rate of rise and peak levels of free intracellular calcium concentration ([Ca2+]j) in response to N M D A were also unaffected by Tat-NR2B9c (13. 16). CNQX (6-cyano-7-nitroquinoxaline-2,3-dione; 10 p.M) and nimodipine (2 |xM) were present in the extracellular solution in these and all further experiments in cultures, so as to isolate signal-ing to NMDARs and prevent secondary activa-tion of AMPARs or of voltage-gated C a 2 + channels, respectively (7, 17, IS). We next determined whether Tat-NR2B9c affected signaling events downstream of NMDAR activation. NMDA-evoked changes in guanosine 3',5'-monophosphate (cGMP) level were measured as a surrogate measure of NO production by nNOS (7, 19). We focused on nNOS activity because it mediates neurotox-ic sequelae of NMDAR activation (5) and, along with other signaling molecules bound to * \u00E2\u0080\u00A2 * aidansy l^ .* ' *\u00C2\u00BB \u00E2\u0080\u00A2 j d p ' \u00E2\u0080\u00A2 \" A \u00E2\u0080\u00A2' 1 <\u00E2\u0080\u00A2* \" w M Tat-38-48-dansyl Application of 10|iM Tat-NR2B9c-dansyl 120 150 180 210 270 330 390 Time (min) r 510 nm; representative of five experiments). Fluorescence of cultures treated with Tat38-48-dansyl was similar to background. (C) Time course of Tat-NR2B9c-dansyl (10 JJL.M) fluorescence after application to cortical cultures at room temperature (symbols: mean \u00C2\u00B1 SE of four experiments). Inset: fluores-cence images from a representative experiment. (D and E) Coimmunoprecipitation of PSD-95 with NR2 subunits in rat forebrain P2 membrane fractions treated with Tat peptides. (D) Tat-NR2B9c reduced the optical density (O.D.) ratio of PSD-95:NR2B by 37.6 \u00C2\u00B1 8.2% relative to controls. ANOVA, f = 6.086, *P = 0.0041. (E) No significant effect on O.D. ratio of PSD-95:NR2A while reducing O.D. ratio of PSD-95:NR2B (ANOVA, *P < 0.01) in same tissue extract. Top: Represen-tative gels. Bottom: Means \u00C2\u00B1 SE of four to eight experiments. www.sciencemag.org SCIENCE VOL 298 25 OCTOBER 2002 R E P O R T S PSD-95, should be dissociated from NMDARs by Tat-NR2B9c. Cultured cortical neurons were pretreated with Tat-NR2B9c (50 nM), the noninteracting Tat-NR2B-AA (50 nM), or sham washes for 1 hour and then challenged with N M D A (0 to 1000 u.M). N M D A produced a concentration-dependent increase in cGMP that was significantly sup-pressed (average of 39.5 \u00C2\u00B1 6.7%) by pretreat-ing the cultures with Tat-NR2B9c, but not with Tat-NR2B-AA (Fig. 2E). Although Tat-NR2B9c treatment did not affect NMDAR-mediated currents or C a 2 + fluxes, it interfered with NMDAR-PSD-95 binding and suppressed downstream NO sig-naling. Thus, we examined whether such treatment enhances neurons' resilience to NMDA toxicity. Cortical neuronal cultures were pretreated with sham washes, Tat-NR2B9c, or the noninteracting control Tat-NR2B-AA (each at 50 nM) for 1 hour, then exposed to NMDA (0 to 100 u.M) for 1 hour followed by a 20-hour observation period (fig. SI, inset). Cell death at all NMDA con-centrations was significantly reduced by Tat-NR2B9c pretreatment, whereas Tat-NR2B-A A was ineffective (fig. SI). Pretreatment with pTat-PDZl-2 also attenuated N M D A neurotoxicity to a similar degree (fig. SI) (13), which suggested that targeting either side of the NMDAR-PSD-95 interaction (Fig. IA) reduces excitotoxic damage. Agents that block NMDAR activity are del-eterious or ineffective in treating stroke in ani-mals and humans (9-11). Because Tat-NR2B9c attenuates NMDA toxicity without blocking NMDARs, we reasoned that its appli-cation in the treatment of stroke would consti-tute an improvement over NMDAR blockers. We first determined whether Tat-NR2B9c could be delivered into the brain in the intact animal. C57BL/6 mice (25 g) were injected intraperitoneally with a 500-p.mol dose of ei-ther Tat-NR2B9c-dansyl or Tat38-48-dansyl as a cell-impermeant control. Coronal brain sec-tions taken 1 hour after injection were exam-ined by confocal microscopy for fluorescent peptide uptake (13). Brains from animals inject-ed with Tat-NR2B9c, but not Tat38^8-dansyl, exhibited strong fluorescence in the cortex (fig. S2A) (20). Similar results were obtained with intravenous injection in rats (21), confirming that Tat-NR2B9c enters the brain upon periph-eral administration. Next, we examined whether pretreatment with Tat-NR2B9c would reduce stroke dam-age. Adult male Sprague-Dawley rats were subjected to transient middle cerebral artery occlusion (MCAO) for 90 min by the intralu-minal suture method (13, 22, 23). Animals were pretreated by a single intravenous bolus injection with saline, the Tat-NR2B-AA con-trol, or Tat-NR2B9c 45 min before M C A O (3 nmol/g). Body temperature, blood pressure, and blood gases were monitored and main-A (L PS 120 UJ KB o 80 B sucrose, lOU/mL heparin in Saline) 1 hour after peptide injection. Brains were removed, frozen in 2-methylbutane at -42\u00C2\u00B0C and 40pm sections were cut with a cryostat. Coronal sections were examined for dansyl fluorescence by UV-laser confocal microscopy. Focal cerebral ischemia. Animals were fasted overnight and injected with atropine sulfate (0.5 mg/kg IP). After 10 minutes anesthesia was induced with 3.5% halothane in a mixture of nitrous oxide and oxygen (Vol. 2:1) and maintained with 0.8% halothane. Rats were orally intubated, mechanically ventilated, and paralyzed with pancuronium bromide (0.6 mg/kg IV). Body temperature was naintained at 36.5-37.5\u00C2\u00B0C with a heating lamp. Polyethylene catheters in the femoral artery and vein were used to continuously record blood pressure and to sample blood for gas and pH measurements. Transient M C A O was achieved for 90 min by introducing a poly-L-rysine-coated 3-0 monofilament nylon suture into the circle of Willis via the internal carotid artery, effectively occluding the middle cerebral artery. This produces an extensive infarction encompassing the cerebral cortex and basal ganglia. A l l experimental manipulations and analyses of data were performed by individuals blinded to the treatment groups. Supplemental Online Text. The Tat-NR2B9c peptide affected the interaction of PSD-95 with NR2B, but not NR2A subunits (Fig. IE). Thus, amino acids upstream from the IESDV sequence common to the NR2A and NR2B c-terminus may indicate the binding efficiency of these subunits to PSD-95 3 . The factors governing these differential interactions remain to be elucidated. However, our findings indicate that perturbing PSD-95/NR2B binding is sufficient to treat excitotoxic and ischemic neuronal damage. This may indicate that NMDARs that mediate excitotoxicity preferentially contain NR2B over NR2A subunits, or that perturbing the interaction of PSD-95 with NR2B subunits is sufficient to suppress downstream excitotoxic signaling. If Tat-NR2B9c suppresses N M D A excitotoxicity by interfering with the binding of NR2B to PSD-95 then interfering with this binding by an alternative means should also suppress the toxicity. We therefore also tested pTat-PDZl-2, predicted to interfere with PSD-95 binding to NR2B and to permeate into the cells, though without effect on NMDA-evoked C a 2 + accumulation (Fig. 2D). Pre-treating the cultures with pTat-PDZl-2 attenuated the neurotoxicity of N M D A to a similar degree as Tat-NR2B9c (Fig. SI). As a control we made and used pTat-GK, a Tat fusion protein containing residues 534-724 of PSD-95 comprising the carboxyl- terminal guanylate-kinase homology domain that lacks enzymatic activity4. pTat-GK, which is devoid of the necessary domains to bind NR2B, had no effect on the N M D A -evoked cell death (Fig. SI). Thus, interfering with the NMDAR/PSD-95 interaction using peptides that target either side of the interaction reduces in vitro excitotoxicity produced by N M D A R activation. Pre-treatment Study 0 20 40 100 NMDA Concentration ( M) Cells at 24h Supplemental Figure 1 Decreased excitotoxicity at 20h at a l l N M D A concentrations in cultured cortical neurons pre-treated with 50 n M Tat-NR2B9c or p T a t - P D Z l - 2 for l h . Asterisk: differences from control, T a t - N R 2 B - A A and pTat-G K at each N M D A concentration (Bonferroni t-test, p < 0.005). Right panels: Representative phase contrast and propidium iodide fluorescence images o f cultures 20h after challenge with 100 M N M D A with and without Tat -NR2B9c treatment. Bars indicate the mean \u00C2\u00B1 S.E. for 12 cultures in 3 separate experiments. (92. B Neurological Score 12 10 cu o 8 o 00 \"cn 6 o CT) o 4 o 0 z 2 Control Tat-NR2B-AA Tat-NR2B9c During MCAo After MCAo (60min) (24hr) Pre-treatment Study Y , , \u00E2\u0080\u00A2 ^45~ r r y As 4\u00C2\u00AB <*0 \u00E2\u0080\u00A2 22.5 h E 600 E Total Infarct Control Tat-NR2B-AA \u00E2\u0080\u00A2Tat-NR2B9c 11 Cortical Infarct 150 100 E 50 o ro 1 2 3 4 5 6 7 8 Stereotactic Coordinates (mm) 1 2 3 4 5 6 7 8 Stereotactic Coordinates (mm) Supplementa l F igu re 2 Neuroprotection by Tat-NR2B9c pretreatment in-vivo. (A) Detection of Tat-NR2B9c-dansyl but not Tat38-48-dansyl in the cortex of C57BL/6 mouse brain lh after intraperitoneal injection (0.5 pmole total dose). Fluorescence of brains from animals treated with Tat-38-48-dansyl was similar to background. (B) Composite neurological scores (see text) during and 24h after MCAo. (C) Pre-treatment with 3 nmole/g Tat-NR2B9c but not mutated Tat-NR2B-A A or saline (control) significantly reduced (i) total infarct area and volume (inset), ANOVA; F= 7.3, p<0.005 and (ii) cortical infarct area and volume (inset), ANOVA; F= 8.35, p<0.005 measured 24h after transient MCAo. (n = 6 animals per group; symbols and bars indicate mean \u00C2\u00B1 S.E). Infarct volume was calculated by analyzing the infarct area in 8 stereotactic coordinates of the brain as shown at right inset. m Supplemental Table 1: Physiological Variables in Pre-Treatment M C A O Study Physiological Variables Control TAT-NR2BAA TAT-NR2B9c (n=6) (n=6) (n=6) Before anesthesia Body weight, g 269 \u00C2\u00B1 6 273 \u00C2\u00B1 7 271 \u00C2\u00B1 5 Before MCAo(45min) Body Temperature, \u00C2\u00B0C 36.7 \u00C2\u00B10 .07 36.7 \u00C2\u00B10 .17 36.6 \u00C2\u00B10 .21 MABP, mmHg 119 \u00C2\u00B1 4 115 \u00C2\u00B1 5 120 \u00C2\u00B1 9 Before MCAo(30min) Body Temperature, \u00C2\u00B0C 36.8 \u00C2\u00B1 0.08 36.5 \u00C2\u00B10 .12 36.7 \u00C2\u00B10 .19 M A B P , mmHg 107 \u00C2\u00B1 3 110 \u00C2\u00B1 4 76 \u00C2\u00B1 5 * Blood gases PH 7.44 \u00C2\u00B10 .02 7.44 \u00C2\u00B10 .02 7.44 \u00C2\u00B1 0.02 P02,mmHg 104 \u00C2\u00B1 3 110 \u00C2\u00B1 7 123 \u00C2\u00B1 8 PC02,mmH g 39.6 \u00C2\u00B1 1.3 39.1 \u00C2\u00B1 1.4 38.1 \u00C2\u00B1 1 . 4 Before MCAo(15min) Body Temperature, \u00C2\u00B0C 36.9 \u00C2\u00B10 .11 36.6 \u00C2\u00B10 .15 36.7 \u00C2\u00B10 .20 MABP, mmHg 1 1 1 \u00C2\u00B1 6 1 1 5 \u00C2\u00B1 5 9 0 \u00C2\u00B1 6 * During MCAo (5min) Body Temperature, \u00C2\u00B0C 36.9 \u00C2\u00B10 .03 36.6 \u00C2\u00B10 .17 36.7 \u00C2\u00B10 .16 M A B P , mmHg 132 \u00C2\u00B1 6 135 \u00C2\u00B1 7 112 \u00C2\u00B1 9 Blood gases PH 7.44 \u00C2\u00B10 .02 7.44 \u00C2\u00B10 .02 7.44 \u00C2\u00B1 0.02 P02,mmHg 118 \u00C2\u00B1 3 109 \u00C2\u00B1 4 112 \u00C2\u00B1 6 PC02,mmHg 39.2 \u00C2\u00B1 0.6 39.6 \u00C2\u00B1 0.5 41.0 \u00C2\u00B1 1 . 3 During MCAo (15min) Body Temperature, \u00C2\u00B0C 36.9 \u00C2\u00B1 0.09 36.7 \u00C2\u00B10 .15 36.8 \u00C2\u00B1 0.23 MABP, rnmHg 116 \u00C2\u00B1 9 1 1 1 \u00C2\u00B1 6 9 8 \u00C2\u00B1 6 After MCAo (15min) Body Temperature, \u00C2\u00B0C 36.9 \u00C2\u00B10 .09 36.8 \u00C2\u00B10 .08 36.8 \u00C2\u00B10 .12 After MCAo (24hr) Body Temperature, \u00C2\u00B0C 36.6 \u00C2\u00B1 0 . 1 4 37.0 \u00C2\u00B1 0 . 2 5 36.5 \u00C2\u00B1 0 . 1 4 Body weight, g 238 \u00C2\u00B1 6 244 \u00C2\u00B1 6 250 \u00C2\u00B1 5 MABP: Mean arterial blood pressure * : PO.05, Student's t-test Supplemental Table 2: Physiological Variables in Post- Treatment MCAO Study Physiological Variables Control TAT-NR2BAA TAT-NR2B9c (n=10) (n=8) (n=9) Before anesthesia Body weight, g 314 \u00C2\u00B1 4 301 \u00C2\u00B1 5 306 \u00C2\u00B1 7 Before MCAo(15min) Body Temperature, \u00C2\u00B0C 36.9 \u00C2\u00B10 .07 36.7 \u00C2\u00B10 .07 36.6 \u00C2\u00B1 0 . 0 7 MABP, mmHg 103 \u00C2\u00B1 4 103 \u00C2\u00B1 6 103 \u00C2\u00B1 5 Blood gases PH 7.43 \u00C2\u00B10 .01 7.45 \u00C2\u00B10 .01 7.43 \u00C2\u00B1 0.02 P02,mmHg 113 \u00C2\u00B1 4 113 \u00C2\u00B1 4 105 \u00C2\u00B1 4 PC02,mmHg 39.4 \u00C2\u00B1 1.0 37.9 \u00C2\u00B11 .1 40.1 \u00C2\u00B1 1.0 During MCAo (15min) Body Temperature, \u00C2\u00B0C 36.9 \u00C2\u00B10 .07 36.7 \u00C2\u00B10 .11 37.0 \u00C2\u00B10 .07 MABP, mmHg 120 \u00C2\u00B1 5 121 \u00C2\u00B1 5 119 \u00C2\u00B1 8 Blood gases PH 7.44 \u00C2\u00B10 .01 7.46 \u00C2\u00B10 .01 7.43 \u00C2\u00B10 .01 P02,mmHg 113 \u00C2\u00B1 3 108 \u00C2\u00B1 2 111 \u00C2\u00B1 4 PC02,mmHg 39.3 \u00C2\u00B1 0.7 48.0 \u00C2\u00B1 1.2 39.8 \u00C2\u00B1 0 . 9 During MCAo (60min) Body Temperature, \u00C2\u00B0C 37.1 \u00C2\u00B10 .21 37.0 \u00C2\u00B10 .31 36.7 \u00C2\u00B10 .11 MABP, mmHg 146 \u00C2\u00B1 5 149 \u00C2\u00B1 4 143 \u00C2\u00B1 5 During MCAo (65min) Body Temperature, \u00C2\u00B0C 37.1 \u00C2\u00B10 .16 37.0 \u00C2\u00B10 .29 36.9 \u00C2\u00B10 .08 MABP, mmHg 134 \u00C2\u00B1 6 136 \u00C2\u00B1 5 137 \u00C2\u00B1 4 After MCAo (15min) Body Temperature, \u00C2\u00B0C 37.0 \u00C2\u00B1 0.09 36.9 \u00C2\u00B1 0.23 36.8 \u00C2\u00B1 0.08 MABP, mmHg 128 \u00C2\u00B1 6 116 \u00C2\u00B1 4 119 \u00C2\u00B1 4 After MCAo (24hr) Body Temperature, \u00C2\u00B0C 36.6 \u00C2\u00B10 .14 36.7 \u00C2\u00B1 0 . 2 7 36.4 \u00C2\u00B10 .24 Body weight, g 276 \u00C2\u00B1 3 276 \u00C2\u00B1 6 279 \u00C2\u00B1 8 MCAo: Middle cerebral artery occlusion; MABP: Mean arterial blood pressure Reference List 1. N. Takagi, R. Logan, L. Teves, M. C. Wallace, J. W. Gurd, J.Neurochem. 74,169 (2000). 2. S. R. Fam, C. J. Gallagher, M . W. Salter, J.Neurosci. 20, 2800 (2000). 3. P. Bassand, A. Bernard, A. Rafiki, D. Gayet, M . Khrestchatisky, Eur.J.Neurosci. 11, 2031 (1999). 4. U. Kistner, C. C. Garner, M. Linial, FEBS Lett. 359, 159 (1995). "@en . "Thesis/Dissertation"@en . "10.14288/1.0092964"@en . "eng"@en . "Experimental Medicine"@en . "Vancouver : University of British Columbia Library"@en . "University of British Columbia"@en . "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en . "Graduate"@en . "Opposing roles of NMDA receptor subtypes in neuronal fate and novel treatments for ischemic brain injury"@en . "Text"@en . "http://hdl.handle.net/2429/18582"@en .