UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Correlation of biophysical properties and cytotoxic potential of recombinant glutamate receptors Moshaver, Ali 1996

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1996-0356.pdf [ 5.55MB ]
Metadata
JSON: 831-1.0099031.json
JSON-LD: 831-1.0099031-ld.json
RDF/XML (Pretty): 831-1.0099031-rdf.xml
RDF/JSON: 831-1.0099031-rdf.json
Turtle: 831-1.0099031-turtle.txt
N-Triples: 831-1.0099031-rdf-ntriples.txt
Original Record: 831-1.0099031-source.json
Full Text
831-1.0099031-fulltext.txt
Citation
831-1.0099031.ris

Full Text

COPvRELATION OF BIOPHYSICAL PROPERTIES AND CYTOTOXIC POTENTIAL OF RECOMBINANT GLUTAMATE RECEPTORS by ALI MOSHAVER B.Sc. (Hon), University of British Columbia, 1994 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENT FOR THE DEGREE OF Master of Science in THE FACULTY OF GRADUATE STUDIES Graduate program in neuroscience We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH C O L U M B I A June 1996 © A l i Moshaver, 1996 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada DE-6 (2/88) ABSTRACT Data from both in vivo and in vitro studies indicate that prolonged exposure of neurons to glutamate results in over-activation of N-methyl-D-aspartate (NMDA)-type receptors, allowing an increase in [Ca 2 +]i and resulting in cell death. However, details of the molecular mechanisms, particularly the role of the N M D A receptor's structure, function, and subunit composition that underlie these events have been difficult to study in neuronal systems. We have co-expressed recombinant N M D A receptors with the marker protein (3-galactosidase in human embryonic kidney cells and determined the amount of viable transfected cells by immunostaining with NR1 specific antibodies and / or assaying for p-gal expression. Incubation of NR1 / NR2A/ P-gal transfected HEK-293 cells in bicarbonate-buffered physiological salt solution for 6 hours resulted in loss of transfected cells when N M D A was included. This effect was dose-dependent with an EC50 between 150-300 | l M N M D A . When these experiments were repeated with cells expressing Ca2+-impermeable mutant NR1 (N598R) / NR2A receptors, cell death was diminished by -75%. On the other hand, in experiments with wild-type NR1/NR2A-transfected cells, substitution of N-methyl-glucamine for Na + in the salt solution decreased cell death by 50%, suggesting that cytotoxicity was mediated in part by Na + flux through N M D A receptors. Using this system, it was shown that the rapid desensitization of non-NMDA receptors is protective against agonist-induced cell death. In addition, the effect of zinc on N M D A toxicity was examined. In NR1 / NR2A-transfected cells, zinc reduced the cyotoxicity of N M D A with an IC50 of -500 nM. Furthermore, it was shown that zinc inhibits the peak glutamate-evoked current responses ii and accelerates desensitization in whole-cell patch clamp recordings from NR1 / NR2A and NR1 / NR2B transfected cells. However, NR1 / NR2A was ~20-fold more sensitive to zinc inhibition than NR1 / NR2B, with IC50S of~500 nM and -10 uM, respectively. Finally, the effect of serum albumin on N M D A toxicity was analyzed. It was shown that serum albumin does not potentiate N M D A toxicity in transfected HEK-293 cells. However, the inhibitory effect of zinc on N M D A toxicity was reduced in the presence of serum albumin. Hence, increased concentrations of serum albumin in the brain observed during pathological conditions, could disrupt the regulatory function of Z n 2 + on N M D A receptors. iii TABLE OF CONTENTS Abstract i Table of Content iv List of Tables viii List of Figures ix List of Abbreviations xi Acknowledgments xv Chapter 1 Introduction 1 1.1 Historical perspective 1 1.2 The physiological and pharmacological properties of glutamate receptors 2 1.2.1 Kainate / A M P A receptors 3 1.2.2 N M D A receptors 4 1.3 The molecular biology of the glutamate receptors 6 1.3.1 Structure of ionotropic glutamate receptors 7 1.3.2 A M P A receptor subtypes 9 1.3.3 Localization of GluR 1-4 subunits 12 1.3.4 Kainate receptor subunits 13 1.3.5 Localization of kainate receptor subunits 15 1.3.6 Orphan receptors 16 iv 1.3.7 N M D A receptors 16 1.3.8 Localization of NR1 and NR2 subunits 20 1.4 Zinc in the central nervous system 21 1.4.1 The physiological role of zinc in glutamatergic transmission 25 1.5 Excitotoxicity 28 1.5.1 The mechanism of glutamate toxicity 29 1.5.2 The role of calcium in excitotoxicity 32 1.6 Research hypothesis 35 Chapter 2 Materials and Methods 37 2.1 Cell culture 37 2.2 PlasmidcDNA 37 2.3 Preparation of plasmid D N A 38 2.3.1 Transformation 38 2.3.2 Growth of transformant for plasmid preparation 38 2.3.3 Maxiprep of plasmid D N A 39 2.4 Transfections 39 2.5 Cytotoxicity experiments 40 2.6 Immunocytochemistry 41 2.7 (3-galactosidase staining 42 2.8 Assessment of cell death 43 2.9 Solution assay for (3-gal activity 43 v 2.10 Electrophysiology 44 2.10.1 Recordings 44 2.10.2 Recording solutions 45 2.10.3 Data analysis 46 Chapter 3 N M D A over-exposure causes selective death of transfected HEK-293 cells 47 3.1 Expression of functional N M D A receptors in HEK-293 cells 47 3.2 N M D A causes dose-dependent cell death in transfected HEK-293 cells 48 3.3 Using (3-gal as a marker of transfected cells 53 3.4 Glutamate did not cause cell death in transfected HEK-293 cells 58 3.5 Calcium and sodium influx contribute to NMDA-induced cytotoxicity 61 Chapter 4 Rapid desensitization of non-NMDA receptors is protective against agonist induced cytotoxicity 65 Chapter 5 Zinc protects against N M D A cytotoxicity in transfected HEK-293 cells 72 5.1 (3-galactosidase assay 72 5.2 Zinc protects against NMDA-induced toxicity 78 vi 5.3 Electrophysiological properties of recombinant N M D A receptors 84 5.4 Currents mediated by NR1 / NR2A are more sensitive to zinc than those of NR1 / NR2B 87 Chapter 6 Serum albumin inhibits the protective effect of zinc against N M D A -induced toxicity 105 6.1 Serum albumin did not potentiate N M D A toxicity 105 6.2 Serum albumin reduced the protective effect of zinc on N M D A -induced toxicity 106 Chapter 7 Discussion 112 7.1 Characterization of N M D A toxicity in transfected HEK-293 cells 112 7.2 Both sodium and calcium play a role in inducing cell death 115 7.3 Rapid desensitization of non-NMDA receptors is protective against agonist toxicity 116 7.4 Electrophysiological properties of NR1 / NR2A and NR1 / NR2B 118 7.5 Differential sensitivity of N M D A receptor subunits to zinc 119 7.6 Serum albumin inhibits the protective effects of zinc against N M D A toxicity 122 7.7 Conclusions 123 Literature cited 125 vii LIST OF T A B L E S Table 1. Splice variants of NR1 subunit of N M D A receptor. 18 Table 2. Characterization of glutamate-evoked current responses mediated 86 by NR1 / NR2A and NR1 / NR2B. viii LIST OF FIGURES Figure 1. Immunocytochemical staining of HEK-293 cells expressing NR1 and NR2A subunits of the N M D A receptors. Figure 2. Dose-response for NMDA-induced toxicity in transfected HEK-293 cells. Figure 3. (3-galactosidase staining of HEK-293 cells expressing NR1, NR2A, and P-gal protein. Figure 4. Dose-response for NMDA-induced toxicity in NR1 / NR2A / (3-gal transfected HEK-293 cells generated by staining surviving cells for [3-gal expression. Figure 5. The role of extracellular calcium in glutamate-induced cell death. Figure 6. The role of extracellular sodium in glutamate-induced cell death. Figure 7. AMPA-induced toxicity is only evident in the presence of cyclothiazide. Figure 8. Kainate-induced toxicity is only evident in cells pre-treated with concanavalin A. Figure 9. P-galalactosidase activity is linearly related to O . D .420. Figure 10. Dose-response for N M D A toxicity generated by assaying for p-gal activity. Figure 11. Zinc protects against N M D A toxicity in a non-competitive manner. Figure 12. Zinc protects against N M D A toxicity in a dose-dependent manner. ix Figure 13. The effect of extracellular zinc on glutamate-evoked whole-cell currents in cells expressing NR1 / NR2A and (3-gal. 89 Figure 14. Current-voltage relationship for NR1 / NR2A in control condition as well as in the presence of Z n 2 + . 92 Figure 15. Dose-response for zinc inhibition of peak current in cells transfected with NR1 / NR2A and (3-gal. 94 Figure 16. Zinc increases the rate of fast desensitization (TDF ) of current-responses from cells transfected with NR1 / NR2A. 96 Figure 17. The effect of extracellular zinc on glutamate-evoked whole-cell currents in cells expressing NR1 / NR2B and (3-gal. 98 Figure 18. Current-voltage relationship for NR1 / NR2B in control condition as well as in the presence o fZn 2 + . 100 Figure 19. Dose-response for zinc inhibition of peak current in cells transfected with NR1 / NR2B and (3-gal. 102 Figure 20. Zinc increases the rate of desensitization (tD) of current-responses from cells transfected with NR1 / NR2B. 104 Figure 21. Serum albumin inhibits the protective effects of zinc against N M D A toxicity. 109 Figure 22. Albumin decreased the inhibitory effects of zinc in a dose-dependent manner. I l l x LIST OF ABREVIATIONS A M P A (+)a-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid A P V (±)-2-amino-5-phosphonopentanoic acid BES N,N-bis[2-hydroxyethyl]-2-aminoethanesulfonic acid (3-gal p-galactosidase B S A Bovine serum albumin °C Degrees centigrade C a 2 + Calcium CaCb Calcium chloride Ca(OH)2 Calcium hydroxide cDNA complementary D N A CI" Chloride CNS Central nervous system CTZ Cyclothiazide C u 2 + Copper D N A Deoxyribonucleic acid EC Effective concentration FBS Fetal bovine serum Glu Glutamate GluR Glutamate receptor HEK-293 Human embryonic kidney-293 cells hr. Hour xi I Current IC Inhibitory concentration K + Potassium K A Kainate receptor KC1 Potassium chloride kHz kilohertz K O H Potassium hydroxide K 2 S 0 4 Potassium sulphate LTP Long term potentiation M molar M g 2 + Magnesium M g C l 2 Magnesium chloride ml milliliter pj microliter mm millimeter mM millimolar p M micromolar mOsm milliosmoles ms milliseconds mV millivolts nM nanomolar Na + Sodium xii nACh Nicotinic acetylcholine receptor NaCl Sodium chloride N a 2 C 0 3 Sodium carbonate NaH 2P04 Sodium phosphate, monobasic N a H C 0 3 Sodium bicarbonate NaOH Sodium hydroxide N a 2 S 0 4 Sodium sulfate N M D A N-methyl-D-aspartate NO Nitric oxide NOS Nitric oxide synthase NR1 N M D A receptor-1 NR2 N M D A receptor-2 O.D. Optical density ONPG Orthonitrophenyl galactoside pA pico amps PBS Phosphate-buffered saline pS pico siemen R P M Revolutions per minute R N A Ribonucleic acid TB Terrific broth x D Desensitization time constant TE Tris-EDTA xiii T M Transmembrane T O F F Decay time constant for glutamate dissociation V H Holding membrane potential Z n 2 + Zinc ZnCh Zinc chloride % Per cent xiv ACKNOWLEDGEMENTS 0 am indebted to my Mfaenviun "Dt, Jt. Raymond fan &&i continual auidance, advice, and Mfefront t6nocty&o«<£ my iiudy, *i¥ei enthuaia&m fan science 6een an Ut4fiOiatco*t fan me and CA wdat 0 cvili tememtfei t£e moot. s4 ifiecoxl tAatdU to "Di4, S, "Du^y, @. "Pnice, and 1. "TJtunfi&y fan tnein, Aelfrfad diAcuteion and totcfteitiani. I&ank to "P. JLee and ty. "Kenn&i fan technical ateiatance, aftecial ac&nowledament to my aood fauettdt "^o44ien S&ayan,, "Duncan "%o-, and "Kamxan "Khevuwi fan fnoofaieadwa t&Cb t&eaiA and /Ida, el-din &l-6a44ieni fan kelfciny me t&e loot two yeani,. 76an&& to my 6eat fatiend "Wtanyam /l/&jdani fan <safefi<nt and encomatfement. 'pwitAen, t&anfai, aoe& to my committee mem6e%&, "Dn,. fy. @/tunc& and T>i. "K. Sainrfnidae, fan tneOi uiefad comments, "penally, 1 coould U6e to t&ank my fr<vte*ttA and Oaten, fan t&ein. endlete, tufifunt and love. xv CHAPTER 1 Introduction Glutamate is believed to be the most ubiquitous neurotransmitter in the adult central nervous system and has been shown to play an important role in synaptogenesis, motor control, learning and memory as well as a number of pathological conditions such as stroke, epilepsy, and some neurodegenerative diseases (Albin and Greenamyre, 1992; Meldrum and Garthwaite, 1990). As a result, in the past few decades, an enormous amount of information about glutamate receptors has been accumulated. In this chapter, I intend to review the current findings on the physiology, pharmacology as well as the molecular biology of glutamate receptors relevant to this study. 1.1 Historical perspective The idea that excitatory amino acids might be involved in modulation of neuronal activity was first proposed by Hayashi (1954) who described the convulsant properties of L-glutamate in the cerebral cortex. Ionophoretic studies later confirmed the excitatory actions of L-glutamate and L-aspartate on spinal and virtually all other neurons in the central nervous system (Curtis et al, 1959, 1960). The role of amino acids as neurotransmitters was thought to be very unlikely because of their involvement in cellular metabolism. Also similar action of the L - and D-isomers seemed to argue against the enzymatic inactivation process in the synapse. Since relatively high concentrations of the amino acids were required to induce excitation of neurons, it was believed that these amino acids had non-specific effects on all neurons (Albin and Greenamyre, 1992). The 1 first model for the excitatory amino acid receptor proposed that could explain the binding of both isomers was the "three point receptor", which contained two positively charged and one negatively charged residue and was thought to be large enough to accommodate the substituent groups in both configurations (Curtis and Watkins, 1960). The essential substituent of the amino acids which conferred activity was shown to be the presence of an amino group optimally situated alpha to the carboxyl group and spaced two or three carbon atoms distant from the second acidic site (Watkins, 1962). In the three decades that have passed since these original investigations, two principle types of experimental approaches have been used to confirm the role of excitatory amino acids as mediators of synaptic activity. The first approach involved neurochemical analysis to determine the tissue concentration, uptake, and release of excitatory amino acids, and the second was the application of electrophysiological recording techniques to determine the postsynaptic changes elicited by application of exogenous L-glutamate or its structural analogues. Over the years, with increased knowledge about glutamate and the electrophysiological and pharmacological properties of its receptor, its role as a neurotransmitter has been well established. 1.2 The physiological and pharmacological properties of glutamate receptors It is now well accepted that glutamate. mediates most excitatory neurotransmission in the mammalian brain. Glutamate receptors have been divided into two categories: the ionotropic and the metabotropic receptors. The metabotropic glutamate receptors are G-protein linked and trigger second messenger pathways, such as the breakdown of PIP2 into 2 diacylglycerol and inositol triphosphate (JP3) or regulation of c A M P levels (Tsuji et al, 1995). Since this study involves the ionotropic glutamate receptors, the physiology and pharmacology of metabotropic receptors will not be discussed. For a review of metabotropic glutamate receptors refer to Schoepp and Conn (1993). Glutamate receptors are activated upon binding of glutamate to its binding site on the receptor; activation of ionotropic glutamate receptors produces changes in membrane conductance, suggesting that these receptors include an ion channel in their structure. These receptors have different physiology and pharmacology and are further subdivided into N-methyl-D-aspartate (NMDA), kainate, and a-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) (Mayer and Westbrook, 1987). Excitatory transmission appears to involve actions of one or more combinations of these receptors even at a single synapse. 1.2.1 Kainate/AMPA Because of the highly distinctive characteristic of the N M D A receptor, it has become an acceptable convention to refer to other types of ionotropic excitatory amino acid receptors as "non-NMDA" receptors. A clear agent that separates the A M P A and kainate receptors has not been found, but a variety of data suggest the presence of two different classes. The high-affinity binding site for [3H] kainate has a different anatomical distribution from the [3H] A M P A binding site (Monaghan et al, 1989). Several types of neurons are only responsive to one of the agonists, such as C-fiber afferent neurons of spinal cord which only respond to kainate (Evans et al, 1987). A M P A and kainate receptors are thought to mediate most fast excitatory neurotransmission in the brain. The 3 majority of these receptors are selectively permeable to monovalent cations; however, a minor subset show moderate C a 2 + permeability as well (lino et al, 1990). The glutamate-evoked currents at both A M P A and kainate receptors desensitize rapidly, with a decay time constant on the order of 5-15 ms (Mayer and Westbrook, 1987; Ascher and Nowak, 1988). The most potent kainate and A M P A antagonist currently available is C N Q X (Monaghan et al, 1990). The current-voltage relation in response to A M P A and kainate receptors, recorded from neurons under the voltage-clamp conditions shows a reversal potential close to 0 mV and is linear for the majority of the neurons, although a minority show an inward rectification ( Ascher and Nowak, 1988; lino et al, 1990). Single-channel analysis of A M P A and kainate evoked currents revealed the presence of multiple conductance states (Sansom and Usherwood, 1990). These substates differed in their amplitudes and opening duration. A M P A was shown to activate a low-conductance channel of ~8 pS, while kainate activated -20 pS channels (Jahr and Stevens, 1987; Cull-Candy and Usowicz, 1987) as well as a conductance state on the order of 140 fS (Cull-Candy et al, 1993). It is speculated that different properties of non-NMDA receptors arise from different subunit compositions of the receptor (see below, molecular biology of glutamate receptors). 1.2.2 NMDA receptors The N M D A receptor is the most complex of the glutamate receptors having a number of special characteristics which are thought to play an important role in the normal functioning of the central nervous system. Unlike non-NMDA receptor-mediated 4 responses, the NMDA-induced responses are voltage-dependent (Mayer and Westbrook, 1987). At negative membrane potentials, the receptor has a negative-slope conductance with the inward current being greatest at -20 to -30 mV with reduced amplitude at both more negative and more positive potentials (Mayer and Westbrook, 1987). This voltage-dependency is attributed to voltage-dependent block by extracellular M g 2 + , which allows channel opening only at more depolarized membrane potentials (Nowak et al, 1984). The N M D A receptor requires glycine as a co-agonist in order to be activated (Johnson and Ascher, 1987). It has been shown that glutamate and glycine have different binding sites on the receptor, and it is speculated that the binding of the two agonists is analogous to two ACh and two G A B A recognition sites in ACh or G A B A receptors (Johnson and Ascher, 1987). Another important feature of the N M D A receptor is their high permeability to C a 2 + as well as Na + and K + (Ascher and Nowak, 1986). The increase in intracellular C a 2 + ions observed after activation of N M D A receptors is thought to initiate various biochemical processes responsible for NMDA-induced plasticity observed in developing and adult CNS as well as NMDA-mediated excitotoxic cell death (Choi, 1988; Schulmen, 1995). Single-channel analysis of N M D A receptors revealed that both L-glutamate and N M D A activate channel opening to more than one conductance level, similar to non-N M D A receptors. However, the substate levels of the N M D A receptors have a larger amplitude, ranging from 20-50 pS, with 50 pS being the main conductance level (Sansom and Usherwood, 1990). 5 Currently, five distinct binding sites within this receptor have been suggested, and they include: 1) a transmitter binding site, 2) a glycine binding site, 3) a binding site for phencyclidine and related compounds, 4) a voltage-dependent M g 2 + binding site, and 5) a binding site for zinc. A number of other endogenous factors have also been shown to modulate the N M D A receptor activity. For example, certain subtypes of the receptor are sensitive to changes in extracellular pH, having an EC50 of pH 7.3 (Traynelis et al, 1995). Hence, at physiological pH, these subtypes are only half activated. Other subtypes are potentiated by micromolar concentrations of polyamines (Ransom and Stec, 1988). It is speculated that different subunit compositions of the receptor produce channels with different properties (see below). Further studies on the regulation of subunit composition of N M D A receptors is required for a better understanding of N M D A receptor function. 1 . 3 The molecular biology of the glutamate receptors With the application of molecular cloning technology and mutagenesis studies tremendous advances in understanding the structural features, expression, and the function of glutamate receptors have been made. Since the cloning of the first functional glutamate receptor, G l u R l , by Hollmann et al (1989), twenty two recombinant glutamate receptor cDNAs along with a considerable number of splice variants have been identified and isolated, and this number is growing as time goes on (Seeburg, 1993; Hollmann and Heinemann, 1994). These 22 subunits include 16 members of the ionotropic glutamate receptors and 6 members of the metabotropic family. 6 The first glutamate receptor subunit was identified using the expression cloning system. This technique involved the construction of a rat brain cDNA library, and its expression in an artificial system. The Xenopus oocyte is one of the more common expression systems used, which efficiently expresses foreign receptors and ion channels after injection of mammalian messenger RNA. The appropriate receptor expression was then verified using electrophysiological techniques. Due to the discovery of a large number of glutamate receptor subunits by different laboratories a clear and concise nomenclature for the different subunits has not been determined. In this thesis, I have tried to use the original names suggested for each of the subunits in order to avoid inconsistency. In this section, the current information available on the molecular biology of glutamate receptors will be discussed. For a more complete review of glutamate receptor molecular biology and the techniques used for isolation of these receptors refer to Hollmann and Heinemann (1994). 1.3.1 Structure of ionotropic glutamate receptors Initial examination of glutamate receptor structure led to the hypothesis that these receptors belong to a large family of ligand-gated ion channels which also includes nicotinic acetylcholine (nACh), and G A B A receptors. It was assumed that glutamate receptors may have a similar structure to the other members of this family, however, more recent data indicate that there are fundamental differences between the most conserved regions of glutamate receptors and other ligand-gated ion channels (Hollmann et al, 1994; Wood et al, 1995). The precise structure of the proteins can not be completely determined 7 until crystal structure analysis of the receptor proteins is available. The information currently available regarding the structural features of these receptors is derived mostly from hydrophobicity plots, mutagenesis studies, and studies determining the potential N -glycosylation and phosphorylation sites of the receptor. The majority of glutamate receptor subunits are of similar size (-900 amino acids; except for NR2A-D which have a slightly larger C-terminal) and contain a hydrophobic domain at the N-terminal that represents the signal peptide required for membrane insertion (Hollmann and Heinemann, 1994). Glutamate receptor subunits were shown to contain four hydrophobic domains with three of the hydrophobic regions near the middle of the protein and the fourth near the C-terminal of the protein, similar to nACh receptor. These hydrophobic regions were initially believed to be transmembrane portions of the protein and were termed T M I-IV. However, several recent reports indicate that not all these hydrophobic domains cross the membrane. Specifically, TM-II, which has been proposed to line the pore of the channel, forms a hairpin loop in the membrane without transversing it (Hollmann et al, 1994). This domain contains a number of negatively charged residues which line the channel pore and is believed to act as the selectivity filter for the receptor. Work by Zarei and Dani ( 1994, 1995) on the permeability and blocking characteristics of organic cations indicate that the N M D A receptor pore is asymmetric, and the narrowest cross section of the pore is short and rectangular which serves as the binding site for permeant ions. It is believed that other glutamate receptors have a similar structure to the N M D A receptor. Mutagenesis studies have revealed the importance of one amino acid in TM-II in the calcium permeability of the glutamate receptors (for reviews see 8 Nakanishi, 1992; Seeburg, 1993; Hollmann and Heinemann, 1994). This site, termed the Q / R site (due to the presence of either glutamine or arginine in non-NMDA receptor subunits), has been shown to be the key factor for determining the calcium permeability of the glutamate channels (see below). In general, it is assumed that amino acid sequences containing N-glycosylation sites are located extracellularly and those containing phosphorylation sites are intracellular. However, contradictory results have been observed for glutamate receptors. The loop between the TM-III and TM-IV has been shown to contain both phosphorylation and N-glycosylation sites making it difficult to localize the position of this portion (Raymond et al, 1993; Wang et al, 1993; Roche et al, 1994). According to the latest model proposed for glutamate receptor structure, the N-terminal is believed to be extracellular and to contain at least part of the agonist binding site(s), while the C-terminal has been shown to be intracellular (reviewed by Dani and Mayer, 1995). The currently accepted model places the region between TM-III and TM-IV on the extracellular side (Dani and Mayer, 1995). One possible explanation for this discrepancy is that this region may be dynamic; it may be located extracellularly in the unbound (closed) state and upon activation of the receptor, it may cross the membrane to place phosphorylation sites on the cytoplasmic sides, as proposed by Nakazawa and colleagues (1995). Further studies are required to resolve this controversy. 1.3.2 A M P A receptor subtypes 9 After the isolation of the first glutamate receptor, G l u R l , other glutamate receptor genes were isolated based on their sequence homology to this receptor. Using standard homology screening and polymerase chain reaction (PCR)-mediated screening for related sequences, three other genes were identified and termed GluR2, GluR3, and GluR4 (Gasic and Hollmann, 1992; Hollmann and Heinemann, 1994). . Expression of these subunits alone or in combination in HEK-293 cells or Xenopus oocytes results in high affinity A M P A binding sites, suggesting that these subunits belong to the A M P A receptor family of glutamate receptors. These subunits have been shown to have a 68-73 % amino acid homology, but were only 20 % homologous to other ligand-gated ion channels and -40% homologous to kainate receptors, indicating that they belong to a separate family of glutamate receptors (Hollmann and Heinemann, 1994). The receptor subunits were shown to be highly conserved between mammals with the rat, mouse and human GluRl sharing -97% homology at the amino acid level (Verdoorn et al, 1991). Electrophysiological studies have shown that functional receptors are formed by homomeric expression of these subunits, and these receptors show similar agonist potencies as found in [3H] A M P A ligand binding experiments in rat brain (Verdoorn et al, 1991). Agonist potencies follow the order domoic acid > A M P A > Glu > kainate, which led to the designation of these receptors as AMPA-preferring receptors. Glutamate and A M P A activate these receptors producing fast desensitizing currents, while kainate produces non-desensitizing currents (Keinanen et al, 1990). Expression of each of the different combinations of GluRl-4 produce receptors with different properties, suggesting an explanation for the large diversity of channels 10 observed in neuronal systems. Further diversity in the channels is obtained due to the presence of an exon that is alternatively spliced in each of the GluRl-4 subunits. A sequence of 115 base pairs preceding TM-IV can exist in two different forms termed flip (e.g. GluRli) and flop ( e.g. GluRl 0 ) creating receptors with different properties (Sommer et al, 1990). The two splice variants differ in their affinity for glutamate with the flip channels having a higher affinity than the flop channels. The pharmacological properties of the splice variants are otherwise similar. The current amplitude of both peak and steady state evoked by glutamate are four to five fold larger at flip receptors than flop receptors (Sommer et al, 1990). Hence, the flip receptors are called the "high gain" receptors as oppose to the "low gain" flop receptors (Nakanishi, 1992). Another interesting feature of the GluRl-4 subunits is the dominant effect of the GluR2 subunit on channel properties. Homomeric expression of G l u R l , 3, and 4 subunits has been shown to result in receptors of moderate calcium and magnesium permeability (Hollmann et al, 1991). The current-voltage relationship of the receptors formed with any of the subunits G l u R l , 3, or 4 is complex and is double rectifying. However, homomeric expression of GluR2 subunits results in receptors of very low calcium permeability and linear current voltage relationship much like neuronal receptors. Co-expression of G luRl , 3, or 4 subunits with GluR2 produces a current with similar properties to GluR2 homomeric channels, suggesting that the GluR2 subunit has a dominant effect (Burnashev et al, 1992b). A more careful examination of the amino acid sequence of the subunits revealed that the calcium permeable A M P A receptors have a glutamine in the Q / R site of the TM-II domain, while the GluR2 subunit contains an arginine residue at this site 11 (Burnashev et al, 1992b; Hollmann et al, 1991). Examination of the genomic sequence revealed that the Q / R site encodes for glutamine, and since only one gene exists for each of the four receptors and no alternative exons are present for T M II, a RNA-editing mechanism has been proposed for this splice variant (Seeburg, 1996). This mechanism allows the cell to regulate the amount of calcium entry and hence, regulate gene expression at different times. 1.3.3 Localization of GluRl-4 subunits In situ hybridization and GluRl-4 immunoreactivity both indicate that there is differential expression of GluRl-4 subunits in the adult central nervous system and during development. The distribution of GluRl-4 mRNA in the brain largely matches the distribution of [3H] A M P A binding sites. In the adult cortex, GluRl and GluR3 mRNA expression has been shown to be low in layers III and IV, whereas GluR4 mRNA is high and GluR2 evenly distributed (Martin et al, 1993; Molnar et al, 1993). GluR4 mRNA has been shown to be abundantly expressed in astrocytes, suggesting that these cells might have a role in brain signaling. In the hippocampus, there is a differential expression of GluR4 with high levels in CA1 and dentate gyrus and low levels in the CA3 region, while GluRl-3 are evenly distributed (Lambolez et al, 1992). In the striatum, GluRl and GluR4 are thought to be expressed mainly in aspiny neurons while GluR2 and GluR3 mRNA are found mainly in medium sized spiny neurons (Siegel et al, 1995). Also in cerebellum GluRl mRNA has been shown to be present in Purkinje cells and Bergmann glial cells but not in the granular cell layer (Brorson et al, 1994). In contrast, GluR2 mRNA is present in 12 Purkinje cells and granule cells but not in glial cells (Brorson et al, 1994; Hack and Balazs, 1995). GluRl-4 expression is observed as early as embryonic day 10 in rats. However, there is a developmental switch in the rats from predominantly flip variants before birth to flop after birth (Monyer et al, 1991). The functional role of the different splice variants is not quite clear, but it is speculated that insertion of the flip module into the receptor produces synapses operating with increased gain and might play an important role in synaptogenesis or underlying receptor modification observed in LTP. Both the flip and the flop form of the receptors are expressed in most regions of the adult brain. However, there is a slight differential expression observed in the hippocampal formation. Pyramidal CA3 region does not express any flop module of the A M P A receptors, while the pyramidal CA1 region as well as the granule cells of the dentate gyrus express them widely (Sommer et al, 1990). Examination of the subunit composition of calcium permeable A M P A receptors found in cerebellar Bergmann glial cells and retinal bipolar cells confirmed the absence of GluR2 subunits (Burnachev et al, 1992c). Expression of GluR2 subunits seems to be developmentally regulated with low levels of GluR2 in the fetal brain and increasing amounts postnatally (Hollmann and Heinemann, 1994). It is speculated that the increased calcium permeability of embryonic and early postnatal A M P A receptors might play an important role in brain development. 1.3.4 Kainate receptor subunits 13 Kainate receptor subunits were isolated using lower stringency hybridization screening with GluRl-4 probes and PCR-mediated D N A amplification with primers made to regions of high homology between GluRl-4 (Egebjerg et al, 1991). Five subunits were isolated and termed GluR5, GluR6, GluR7, K A - 1 , and K A 2 . GluR5-7 share 75-80% amino acid homology with each other (Nakanishi, 1992). KA1 and K A 2 have high homology with each other, but have only -37% homology with GluR5-7, suggesting that they form a different subfamily of glutamate receptors. Both groups have about 40% homology with GluRl-4. Structurally they are thought to form channels similar to other ligand gated ion channels. As for A M P A receptors, a RNA-editing mechanism has been shown for the TM-II Q / R site in GluR5 and GluR6; however, the editing of these receptors appears to be incomplete, since both the edited (R) and the unedited (Q) form of the GluR5 and GluR6 have been found in adult CNS (Kohler et al, 1993; Muller et al, 1992). Several splice variants have been found for GluR5, having different exons in the N - and the C-terminal domains of the receptor; their functional significance is still unclear (Hollmann and Heinemann, 1994). No splice variants for K A 1 , K A 2 , and GluR7 have been found. Expression of GluR5-7 in transfected cells or Xenopus oocytes produces currents in response to agonist with the same potency order of domoic acid > K A > Glu, as obtained in [3H] K A binding studies (Bettler et al, 1991, 1992). KA1 and 2 have been shown to have a much higher affinity in [3Ff] K A binding studies than GluR5-7, which led to the idea of high affinity kainate receptors (Sakimura et al, 1992). Homomeric expression of K A 1 , K A 2 , and GluR7 does not seem to form functional channels 14 (Sakimura et al, 1992). Homomeric GluR5 or GluR6, or heteromeric GluR5-7 subunits all produce functional channels with rapidly desensitizing current responses to glutamate or kainate (Hollmann and Heinemann, 1994). Co-expression of GluR5 or 6 with K A 2 has been shown to produce functional channels; however, no interaction between GluR7 and either KA1 or K A 2 has been demonstrated. Also, co-expression of GluR5-7 with GluRl -4 did not lead to any differences in channel properties, suggesting that heteromeric complexes containing both A M P A and kainate subunits do not form in vivo (Patin et al, 1993). It is postulated that GluR7, K A 1 , and K A 2 subunits may act to modulate the pharmacological and electrophysiological properties of kainate receptors (Hollmann and Heinemann, 1994). 1.3.5 Localization of kainate receptor subunits Using a radiolabeled ligand-binding approach, high affinity kainate binding sites have been shown to have distribution distinct from A M P A receptors (Barnard and Henley, 1990). In situ hybridization studies have shown that levels of expression of kainate subunit mRNA were in general lower than for GluRl-4 mRNA, and that there was a differential distribution of kainate subunits mRNAs in CA1 and CA2 pyramidal cells layer of the hippocampus, the caudate-putamen, and the septal region (Herb et al, 1992). Hippocampal cells mainly express GluR6 with low levels of GluR5. GluR6 mRNA is most abundant in the CA3 hippocampal region and cerebellar granule cells. GluR7 mRNA is pronounced in caudate and putamen (Bettler et al, 1992). Expression of these receptors are also developmentally regulated. As early as embryonic day 10, expression of GluR5 starts and gets more pronounced during development (Bettler et al, 1990). 15 Postnatally its expression decreases until adulthood. K A 2 mRNA is widely expressed in most parts of the brain, however, KA1 is largely restricted to CA3 pyramidal cells and dentate cells of hippocampus (Sakimura et al, 1992). 1.3.6 Orphan receptors Two other glutamate receptor subunits have been isolated based on sequence homology to GluRl-7. These genes, named 81 and 82, have similar structural features to other ligand-gated ion channels with proposed transmembrane domains (Yamazaki et al, 1992; Lomeli et al, 1993). TM-II contains a glutamine in the Q / R site as well as several sites for protein kinases. However, expression of these subunits as either homomers or heteromers does not seem to lead to functional receptors (Wisden and Seeburg, 1993). It is believed that either these receptors lack ligand binding sites or are not expressed in the cells tested. The electrophysiological and pharmacological properties of these receptors are very much unclear. They are highly expressed in the cerebellar Purkinje cells and are developmentally regulated, suggesting that they may have a role during development (Hollmann and Heinemann, 1994). Recent knockout experiments showed that the 82 subunit plays an important role in motor coordination, as well as formation of cerebellar Purkinje cell-parallel fiber and climbing fiber synapses (Kashiwabuchi et al, 1995). 1.3.7 N M D A receptors Homomeric NR1 channels 16 Similar techniques used for identifying the first GluRl receptor were used by Nakanishi's group to isolate a cDNA for N M D A receptors (Moriyoshi et al, 1991). The cDNA termed N M D A R 1 (or NR1) has -25-30 % homology with the other GluR cDNAs, and when expressed in Xenopus oocytes, produces a current with similar properties to neuronal N M D A receptors. The structure appears quite similar to GluRl-7 with the exception of a larger extracellular N-terminal. The current is activated by glutamate, N M D A , or L-homocysteic acid (HCA). The current-voltage relationship is linear with a 2 + reversal potential of -0 mV in the absence of Mg , similar to the neuronal N M D A receptors (Moriyoshi et al, 1991). The Q / R region in the TM-II has an asparagine residue giving it high calcium permeability (Yamazaki et al, 1992). A l l the competitive antagonists of the N M D A receptor, including D-2-amino-5-phosphonovalerate (D-APV), 3-(2-carboxypiperazine-4-yl) propylphosphonoate (CPP), and cis-4-(phosphonomethyl)piperidine-2-carboxylate (CGS 19755), as well as non-competitive antagonists such as MK-801, inhibited the current in a similar manner to that described for neuronal N M D A receptors (Le Bourdelles et al, 1994). In addition, external M g 2 + resulted in a voltage-dependent block, indicating that homomeric NR1 receptors contain all the interaction sites present in neuronal N M D A receptors (Hollman and Heinemann, 1994). The gene for NR1 has also been isolated and has been shown to be composed of 22 exons (Zukin and Bennett, 1995). Three of the exons , one in the N-terminal (exon 5) termed NI and two in the C-terminal (exon 21 and 22) termed CI and C2, are alternatively spliced, theoretically generating 8 possible splice variants of the NR1 subunit. These 17 splice variants are termed NR1A-G as shown in table 1 (an alternative name suggested for each splice variant is also included). The NI sequence is highly charged, having six positively charged residues arranged in two clusters at either end separated by three negatively charged residues (Durand et al, 1993). Exons 21 and 22 are 37 and 38 amino acid long, respectively. Use of alternative exon 22 causes removal of the original stop codon and creates a new one, giving a smaller C-terminal (Sugihara et al, 1993). Only seven of the eight possible splice variants have been found in cDNA libraries; the NR1 subunit including NI and CI but without C2 has not been found yet. Expression of each NR1 splice variant generates functional receptors with different physiological and pharmacological properties, suggesting that alternative splicing could be responsible for generation of high diversity in neuronal N M D A receptors. Table 1. Splice variants of NR1 subunit of N M D A receptor Splice variant NI insert CI C2 other names NR1A - + + NRlon NR1B + + + N R l m NR1C - - + NRIOOI NR1D - + - NRloio NR1E - - - NRlooo NR1F + - + NRlio i NR1G + - - NRlioo + + - NRl i io These splice variants have been shown to differ in agonist affinity (Durand at al, 1993, 1992; Hollmann et al, 1993; Tingley et al, 1993). The NI exon has been shown to play a 18 role in sensitivity of N M D A receptors to protons, and the CI exon plays a role in subcellular compartmentalization of these receptors (Ehlers et al, 1995; Traynelis et al, 1995). NR1 subunits which lack the NI exons show significant potentiation by micromolar concentration of zinc (Hollmann et al, 1993). However, caution should be exercised regarding the physiological relevance of these results, since homomeric receptors may not contribute to N M D A receptor currents in neurons. Heteromeric N M D A receptors Using primers made against the conserved region of NR1 and GluRl-7 and low stringency hybridization screening led to the isolation of four other subunits of N M D A receptors. These subunits termed NR2A-D have 45-56% sequence homology but differ from the other subunits, having <20% homology to NR1 subunits, suggesting that they are from a new family of glutamate receptors (Ikeda et al, 1992; Monyer et al, 1992). Structurally, these subunits are similar to the other glutamate receptor subunits except that they possess a larger C-terminal domain. Similar to the NR1 subunit, the Q / R site is occupied by asparagine (Burnashev et al, 1992a,b). A sequence related to a zinc-finger has also been shown to be present in the C-terminal of NR2A and NR2B, but the significance of this observation remains to be shown. None of the NR2 subunits form functional homomeric receptors; however, when co-expressed with the NR1 subunit they form heteromeric receptors with potentiated currents (Ikeda te al, 1992). Co-expression of NR1 along with one of the NR2 subunits produces a current with similar properties to neuronal N M D A receptors. Co-localization 19 of NR1 and NR2 subunits has also been observed suggesting that the N M D A receptors are likely to be heteromeric (Kutsuwada et al, 1992; Patralia et al, 1994). The receptors are highly calcium permeable and blocked by the inhibitors of N M D A receptors. Different subunit combinations produce different electrophysiological properties. For example, the M g 2 + block for NR1 / NR2C and NR1 / NR2D receptors is much weaker than for NR1 / NR2A and NR1 / NR2B (Kutsuwada et al, 1992). Subunit-specific differences are observed with respect to agonist-induced desensitization and current offset time after removal of glutamate. NR1 / NR2A offset time is considerably faster than the NR1 / NR2B or NR1 / NR2C receptor combination (Monyer et al, 1992). The desensitization rate is also much faster for the NR1 / NR2A subunit combination. Single channel studies reveal that both NR1 / NR2A and NR1 / NR2B combinations have two conductance levels, a main level of 50 pS and a sublevel of 38 pS, similar to the neuronal N M D A receptors. The NR1 / NR2C receptor combination gives rise to two different conductance levels of 36 pS and 19 pS (Stern et al, 1992). 1.3.8 Localization of NR1 and NR2 subunits In situ hybridization has revealed the presence of NR1 mRNA in almost all neuronal cells tested (Moriyoshi et al, 1991). Using an NI-specific probe in order to determine the regional expression of NR1 with the NI exon, Young and colleagues showed that this receptor splice variant is most prominent in the hippocampal CA3 region, the granular cell layer of the cerebellum, and some cells in the thalamic nuclei (Standaert et al, 1994). A probe directed to the C1-C2 exon showed that they are present in the 20 hippocampus and caudate. The expression of splice variants is also developmentally regulated, with high levels of expression of NR1 subunit without the NI exon at early stages and increasing levels at later stages (Laurie and Seeburg, 1994). NR1A and NR1B are the last to be expressed at P8, and reach peak levels at P14 in the hippocampus and cerebellum and at P31 in the thalamus. Using polyclonal antisera generated against the C-terminal of NR1, a 117 kDa and a 97 kDa band was observed on western blots (Chazot et al, 1992). This suggests that the NR1 subunit is usually highly glycosylated. In situ hybridization reveals that there is a differential distribution of the NR2 subunits. NR2A mRNA has been shown to express predominantly in cerebral and cerebellar cortex and in hippocampus (Moriyoshi et al, 1991). NR2B is more restricted to thalamic and telencephalic areas, while NR2C is highly expressed in cerebellar granule cells (Tolle et al, 1993; Hollmann and Heinemann, 1994). NR2D mRNA is largely restricted to diencephalic and lower brain stem regions and appears to be the only NR2 subunit in motor neurons of spinal cord (Tolle et al, 1993). The expression of NR2 subunits is also developmentally regulated, with the NR2B and NR2D being highly expressed before birth (Zhong et al, 1995). Postnatally, the NR2A and NR2C start to express and increase until adulthood, while NR2B and NR2D decrease in levels (Zhong et al, 1995). 1.4 Zinc in the central nervous system The importance of zinc during normal brain function and development was realized over two decades ago from observations reported in several areas of zinc research. 21 The association of this element with over a hundred enzymes (Chelebowski and Coleman, 1976), and the link between zinc deficiency in rat and abnormal fetal development (Swenerton, 1969) attracted the attention of many neurobiologists. The association of zinc with myelin marker enzyme 2'-3'-cyclic nucleotide phosphohydrolase and alkaline phosphatase led researchers to conclude that zinc might be important in the process of myelination and brain maturation (Dreosti, 1984). Zinc deficiency was also shown to reduce the activity of superoxide dismutase, an enzyme which protects against damage by superoxide radicals, linking zinc with a variety of neurological disorders (Ebadi and Pfieffer, 1984). The typical concentration of zinc in the gray matter was estimated to be -150-200 pM, making it more abundant than some classical neurotransmitters, such as ACh (Fredrickson, 1989). Various neurochemical studies suggest that there may be three separate pools of zinc in the central nervous system: protein-bound zinc, vesicular zinc, and free zinc. Protein-bound zinc constitutes 85-95% of total CNS zinc and is thought to be part of the structure of many zinc containing enzymes. This form of zinc is mainly thought to be involved in modulating the activity of many neuronal enzymes (Ebadi and Pfieffer, 1984). Vesicular zinc, which for unknown reasons is the only form of zinc that can be stained using various histochemical techniques, has been shown to constitute about 5-15% of total CNS zinc. Electron microscopy of Timm's stained neurons, has revealed that zinc is selectively localized to axonal boutons of certain neurons (Crawford and Conner, 1972). Interestingly, only a small fraction of the vesicles in the zinc- containing neurons were shown to stain for zinc (Crawford and Conner, 1972); these vesicles were 22 shown to be typically clear and round, raising the possibility that zinc containing vesicles are a small chemically distinct subclass of vesicles in these neurons (Crawford and Conner, 1972). Free zinc consists of a pool of ionic zinc in the cytosol or interstitial fluid, and its concentration is believed to be quite low (Feredrickson, 1989). It is believed that this form of zinc, possibly released from zinc-containing vesicles, regulates neuronal excitability in some regions of the CNS (see below). The mechanism by which zinc enters neuronal cells is unknown. Based on the concentration difference between the extracellular and intracellular zinc, the transport of zinc into the cell would require cellular energy. High affinity uptake of 6 5 Z n has been observed in hippocampal slices and in synaptosomes in vitro, and the rate of uptake has been shown to increase with electrical stimulation (Howell et al, 1984). The highest rate of uptake was shown to be in the CA3 pyramidal cells and the dentate gyrus (Howell et al, 1984). Histochemical staining has revealed that zinc-containing neurons are mainly localized to the limbic regions of the CNS, with sparsely scattered neurons also located in the striatum, lamina I-UI and V of cerebral cortex, and the thalamus (Haug, 1984). The trisynaptic circuit that leads from the lateral entorhinal cortex to CA1 pyramidal neurons are all zinc-containing (Slomianka, 1992). However, mossy fibers of the hippocampal formation have been shown to be especially rich in vesicular zinc. These fibers originate in the dentate granule cells and synapse with the apical dendrites of CA3 pyramidal neurons. Outside of the limbic system most other zinc-containing neurons are cerebrocervical and cerebrocortical neurons, which are the primary source of the zinc-23 containing neurons in striatum (Vera-Gil and Perez-Castejan, 1994). It appears that vesicular zinc is closely associated with excitatory amino acid neurotransmitters. The majority of the zincergic pathways have also been shown to be glutamatergic pathways (Slomianka, 1992), although the reverse is not true. Zinc release from hippocampal slice was first shown by Assaf and Chung (1984). In response to high K + (23.8 mM) or kainic acid (100 plM), an estimated 18 % of total zinc in hippocampal slices was shown to be extruded into the extracellular space, in a calcium-dependent manner, giving an evenly distributed concentration of -300 u M Z n 2 + at peak convulsive activity. This large release suggests a possible role for zinc either in signaling or modulation of neuronal excitability. Further experiments confirmed these results and showed that zinc release is voltage-dependent and is accelerated by electrical stimulation (Howell et al, 1984; Charton et al, 1985). Interestingly, zinc release was only observed in the CA3 region of hippocampus and not in the medial part of CA1, the fimbria, or the thalamus, even though these regions have also been shown to contain zinc (Anikszkejn et al, 1987). These results suggest a possible regulatory mechanism for release of vesicular zinc. Zinc release was further localized to regions innervated by mossy fibers i.e. striatum lucidum and hilar zone. Since zinc staining has been found in non-neuronal secretory cells such as pancreatic (3-cells, it has been proposed that zinc may have a role in stabilization of the secretory substance or as a storage cofactor (Perez-clausell and Danscher, 1985). Based on these results, two roles for vesicular zinc have been proposed: 1) it may act as a storage complex, and 2) it may act as a co-transmitter or modulator at glutamate synapses. 24 1.4.1 The physiological role of zinc in glutamatergic transmission A number of studies have indicated the possibility of Z n 2 + in modulation of neuronal excitability (Mayer et al, 1989; Westbrook and Mayer, 1987). The first set of experiments gave contradictory results in zinc activity. Danscher et al (1975) showed that in the presence of DDC, a chelator of vesicular zinc, a depression in synaptic activity was observed. On the other hand, an increase in paired-pulse facilitation in the CA3 region of hippocampus was observed in the presence of zinc (Khulusi et al, 1986). The mechanism of action of zinc in producing both the inhibitory and the excitatory effects was initially thought to be interference with neurotransmitter metabolism (Chung and Assaf, 1984). However, increased knowledge about the physiology and the pharmacology of neurotransmitter receptors and the use of more advanced experimental techniques led to the proposal of a different mechanism of action for zinc. Using patch-clamp recording it was shown that zinc has different effects on various ligand gated ion channels (Rassendren et al, 1990). A certain subtype of G A B A receptors as well as the N M D A receptor subtypes were shown to be potently inhibited by extracellular zinc (Peters et al, 1987; Westbrook and Mayer, 1987). In contrast, non-NMDA receptors were shown to be potentiated in the presence of low concentrations of zinc. Due to the involvement of N M D A receptors in a wide variety of processes in the central nervous system, the effect of Z n 2 + on N M D A receptors has received more attention than any other receptor. Zinc inhibition was shown to occur in a non-competitive manner with IC50 of -13 | i M in cortical cultures, which is within the proposed range of zinc concentration in the extracellular space (Mayer et al, 1989; Legendre and Westbrook, 25 1990). Zinc inhibition is voltage-independent, unlike Mg , suggesting that the binding site for Z n 2 + and M g 2 + are not the same. It has been suggested that the binding site for zinc is near the external surface of the membrane rather than deep in the channel, like M g 2 + (Legendre and Westbrook, 1990; Christine and Choi, 1990). Zinc had a small effect on the glycine affinity for N M D A receptors, suggesting that it causes allosteric modulation of glycine binding but probably does not bind directly to the binding site (Mayer et al, 1989). It is still not clear where the binding site for zinc is on the receptor. Since zinc appears to act as an allosteric inhibitor, it is possible that it binds to a site similar to zinc binding protein such as metaloenzymes. Zinc binding sites on other proteins such as "zinc finger" structure of D N A binding proteins, include two cysteine and two histidine residues separated by a loop of twelve amino acids. This structure has been shown to be present in ligand-gated ion channels as a cys-cys loop which forms a (3-structural loop in the extracellular domain (Hollman et al, 1993). To determine the specific mechanism of block of N M D A receptors, single channel analysis was performed on cultured hippocampal neurons. Using this technique, it was shown that zinc could have two different binding sites on N M D A receptors (Christine and Choi, 1990; Legendre and Westbrook, 1990). Low concentrations of zinc were shown to reduce the open probability of the larger conductance levels of N M D A receptors i.e. 50 pS, while the smaller conductance levels were largely unaffected (Legendre and Westbrook, 1990). The mean open time of the larger conductance level was markedly shortened in a dose-dependent manner. Increasing [Na+] and [Ca 2 +] did not result in any significant changes in the blocking potency of zinc, suggesting that the binding site for 26 zinc is not in the channel pore, and it does not interfere with the permeation of other ions (Christine and Choi, 1990). A second binding site for zinc has also been proposed on the N M D A receptor. This site has a lower affinity for zinc, since high concentrations were required to observe its effects. Flickering was observed in single-channel analysis of the N M D A receptor at higher concentrations of zinc, suggesting that this second binding site is probably in pore of the channel (Christine and Choi, 1990). The physiological significance of this second binding site is unclear. A physical basis for the behavior of divalent cations as blockers or permeators has been proposed to arise from differences in the rate constant for exchange of water molecules in the hydration shell surrounding the ions in the aqueous solution vs. the rate at which the ions dissociate from amino acid residues in the pore of the ion channel protein (Hille, 1992). Permeant ions have fast rates of exchange, while voltage-dependent blockers have slower rates (Hille, 1992). The rate of exchange of zinc lies between calcium and magnesium, which led to the suggestion that at high concentrations zinc can potentially enter the cell via N M D A or calcium permeable non-NMDA receptors (Koh and Choi, 1994; Yin and Weiss, 1995). Incubation of neurons in the presence of high concentrations of zinc has been shown to cause neuronal injury. Zinc toxicity has been shown to be attenuated by N M D A receptor antagonists, suggesting that one mode of entry of zinc is through the N M D A receptors (Koh and Choi, 1994). Increasing calcium concentrations were also shown to decrease zinc toxicity, suggesting that there is a competition between the two divalent ions. Since zinc affects the activities of many 27 neuronal proteins, intracellular zinc overload is thought to induce lethal disturbances in neuronal cell activity. Recent studies on neuronal cultures have shown that zinc may enter the cell via three routes: N M D A receptors, voltage gated calcium channels, and calcium permeable A M P A / Kainate receptors (Koh and Choi, 1994; Yin and Weiss, 1995; Shiraishi et al, 1993). However, since very high concentrations of zinc are required for permeation, the physiological role of zinc toxicity is still under question. 1.5 Excitotoxicity The possibility that excitatory amino acids could be involved in neuronal degeneration was first proposed by Lucas and Newhouse in 1957. They noticed that systemic injection of glutamate resulted in complete destruction of the inner neuronal layers of immature mouse retina. Further experiments confirmed the neurotoxic potential of systemic glutamate in the regions not protected by the blood brain, barrier, in rats , mice, and monkeys (Olney, 1978). In the four decades that have elapsed since these original observations, a tremendous amount of progress has been made in this field. It has become widely accepted that over-exposure to glutamate causes neuronal cell death by over excitation of cells, which led to the introduction of the term "excitotoxicity" for this type of cell death (Rothman and Olney, 1987). In spite of enormous efforts to elucidate the intracellular cascades leading to excitotoxicity, the molecular mechanisms underlying this type of cell death are still very much unclear. The basic characteristic of excitotoxic cell death as revealed by electron microscopy include an initial swelling of cell bodies that may or may not result in cell 28 death over the next several hours (Westbrook, 1993). Glial cells have been shown to undergo reversible cell swelling but were generally more resistant to glutamate toxicity (Rothman and Olney, 1987). A variety of evidence has linked the pathogenesis of acute CNS insults such as stroke and epilepsy, as well as certain adult onset neurodegenerative diseases such as Huntington's and Alzheimer's disease, to excitotoxic cell death (Albin and Greenamyre, 1992; Rothman and Olney, 1987; Kowall et al, 1987). As a result, there has been a large interest in elucidating the mechanisms underlying excitotoxic cell death in the hope of developing therapeutic agents to block neuronal cell death. 1.5.1 The mechanism of glutamate toxicity In general, studying the mechanism of glutamate-induced toxicity has been very difficult both in vivo and in vitro. The presence of glutamate uptake systems in glial cells and in synaptic terminals have largely interfered with the in vivo experiments. In order to circumvent these problems, experiments have been carried out in in vitro preparations such as cell culture, tissue culture, and brain slices. However, it has become quite clear that in different preparations, glutamate over-exposure activates different cascades which may or may not lead to cell death. A number of different pathways have been shown to be involved in inducing excitotoxic cell death (Choi, 1994). So far, no single intracellular pathway has been associated uniquely with cell death in each preparation. However, some factors have been shown to play a more important role in excitotoxicity in certain neurons than others. A number of extrinsic and intrinsic factors such as the type of glutamate 29 receptors, the mechanisms available to buffer Ca + or to metabolize free radicals have been speculated to play a role in determining the vulnerability of neurons to excitotoxic injury (Choi and Hartley, 1993). In general, in vitro studies indicate that over-exposure to glutamate can cause neuronal injury via two different mechanisms (Meldrum and Garthwaite, 1990; Rothman and Olney, 1987; Michaels and Rothman 1990). These mechanisms can be separated based on differences in ionic dependence and time course of injury. The first mechanism proposed is the acute type of cell death which is marked by immediate neuronal swelling in response to glutamate exposure (Rothman and Olney, 1987). Various reports indicate the importance of extracellular sodium and chloride in mediating this type of cell death (Garthwaite et al, 1986; Rothman et al, 1987). It is believed that over-activation of glutamate receptors leads to a large influx of extracellular sodium, accompanied passively by extracellular chloride and water, resulting in swelling of the cell and eventually leading to osmotic lysis (Rothman et al, 1987). However, this swelling is not necessarily lethal, and some neurons can potentially decrease the intracellular volume and survive. Depolarizing agents, such as high K + , also mimic this type of cell.death (Choi, 1987; Randall and Thayer, 1992). The second mechanism, which is thought to be the more important mechanism, is the delayed type of cell death. In response to low concentrations of glutamate for a short period of time, a gradual death over the next several hours is observed (Choi, 1987). Numerous reports have shown that calcium entry plays an important role in activating the machinery that leads to cell death. Calcium is thought to activate various enzymes such as 30 proteases, endonucleases or nitric oxide synthase, which in turn activate a cascade of second messenger systems leading to cell death (Orrenius et al, 1989; Choi and Hartley, 1993). Studies by different groups have shown that delayed cell death is mimicked by C a 2 + ionophores, such as A23187, and inhibited by intracellular infusion of the C a 2 + chelator, B A P T A (Choi, 1987; Tymianski et al, 1993). Whether this type of cell death is the same as apoptotic cell death or is necrosis is still controversial. Application of aurintricarboxylic acid (ATA), an inhibitor of endonuclease activity, has been shown to inhibit excitotoxic cell death in some preparations but not in others (Csernansky et al, 1994; Sagot et al, 1995; Dreyer et al, 1995). In a more recent study it was shown that the state of mitochondrial function might determine the mode of neuronal cell death. It was proposed that the immediate neuronal cell death is associated with extreme energy failure in mitochondria leading to necrosis, while the delayed cell death requires intact mitochondrial function and undergoes apoptotic cell death. Hence, the cells that survive the initial insult undergo apoptotic cell death (Anakarcrona et al, 1995). More experiments are required to confirm these findings. Pharmacological experiments indicate that glutamate receptor subtypes have different roles in inducing excitotoxicity. The calcium-dependent cell death is largely dependent on N M D A receptor activation (Meldrum and Garthwaite, 1990; Choi et al, 1987). N M D A receptors as mentioned previously are highly permeable to calcium, as well as sodium and potassium. Antagonists of N M D A receptors, such as A P V , have been shown to inhibit the delayed type of toxicity and reduce the amount of swelling (Simon et al, 1984; Rothman and Olney, 1987; Park et al, 1988). In contrast, inhibition of A M P A / 31 Kainate receptors with C N Q X mostly inhibits cell swelling (Rothman et al, 1987). Complete blockage of glutamate-induced cell death can only be induced in the presence of both N M D A and non-NMDA receptor antagonists. Although certain neurons, such as NADPH-diaphorase positive neurons and some hippocampal pyramidal neurons have been shown to be only susceptible to non-NMDA toxicity (Koh and Choi, 1988; Weiss et al, 1994b), the majority of neurons are more resistant to over-exposure to non-NMDA than N M D A receptor agonists. In order to observe cell death mediated by non-NMDA receptors, high concentrations of the agonists (AMPA or Kainate) and protracted exposure times are required (Westbrook, 1993). The underlying reason for this increased resistance is still unclear. Several subtypes of non-NMDA receptors have been shown to have a moderate calcium permeability; hence, calcium permeability alone could not explain the decreased susceptibility. Two possible explanations have been proposed for these observations (Westbrook, 1993). One is that the subcellular localization of the receptors does not allow calcium to activate the appropriate protein, and the second is that fast desensitization of these receptors does not allow sufficient calcium and sodium entry to cause cell death. In chapter 4, experiments to test these hypotheses have been described. 1.5.2 The role of calcium in excitotoxicity A number of studies have correlated the amount of calcium entry into the cell and the extent of neuronal degeneration (Hartley et al, 1993; reviewed by Frandsen and Schousboe, 1993). In response to glutamate, the intracellular calcium level has been 32 shown to increase and remain elevated for an hour, after which it declines to normal levels (Dubinsky, 1993). Several reports have shown that dantrolene, a blocker of intracellular C a 2 + release attenuates glutamate neurotoxicity, further confirming the calcium hypothesis (reviewed by Mody and McDonald, 1995). However, other studies have indicated that there may not be a simple linear relationship between elevated somatic [Ca 2 +]i and subsequent neuronal cell death (Witt et al, 1994). Furthermore, no direct relationship between the vulnerability of neurons to excitotoxic cell death and neuronal calcium-binding proteins, such as calbindin and parvalburriin, was found (Freund et al, 1991). It has been shown that the dentate hilus and the CA1 region of the hippocampus are the most susceptible to neuronal death due to ischemia or following siezure activity, while the CA3 pyramidal cells, which do not contain either of the calcium binding proteins, are largely resistant, suggesting that the level of intracellular calcium buffering is not the only determinant of excitotoxic vulnerability (Freund et al, 1991). Another theory proposed is that the calcium required to initiate cell death must be localized to specific compartments in the cell (Bindokas and Miller, 1995). In order to understand this correlation, a complete characterization of temporal and spatial parameters in intracellular calcium regulation is required. Based on the information available, a three stage model for the mechanism of excitotoxicity has been proposed which includes: induction, amplification, and expression (Choi, 1994). The induction stage includes all the immediate changes observed in response to glutamate such as increases in cytoplasmic Ca 2 + , Na + , K + , CI", and water, as well as IP3 and diacylglycerol generated after activation of metabotropic glutamate 33 receptors. Although the precise role of metabotropic glutamate receptors in excitotocity is not quite clear, they could augment the intracellular calcium levels by releasing intracellular calcium. The second stage, amplification, includes events that can cause further increase in the intensity of these initial changes, such as C a 2 + release from intracellular stores, activation of protein kinase C or calmodulin regulated enzymes, calpain or phospholipases (see Durkin et al, 1996). The final stage, expression, includes the activation of cascades ultimately leading to cell death. This stage is probably the most variable from cell to cell, and a number of different pathways have been shown to exist. It is unclear if activation of one of these cascades is sufficient to cause cell death or if combinations of them are involved. Discussion of all possible cascades proposed for induction of cell death is beyond the scope of this chapter. Current research has mainly focused on two pathways thought to be the more common cascades. The first is activation of various calcium activated proteases such as calpain I or phospholipase. These enzymes function to degrade major structural proteins or disrupt membrane integrity, subsequently leading to cell death (Siman et al, 1989). The second important pathway is the formation of free radicals, which could initiate many destructive processes such as peroxidation leading to cell death (Baughler and Hall, 1989; Coyle and Puttfarcken, 1993). Free radicals can be formed via three different routes in response to glutamate, including: 1) activation of nitric oxide (NO) synthase and release of NO, which in turn can react with superoxide to form peroxynitrite and ultimately the production of hydroxyl radicals (Dawson et al, 1991); 2) conversion of 34 Xanthine dehydrogenase to Xanthine oxidase (Dykens et al, 1987); and 3) activation of phospholipase A 2 leading to liberation of arachidonic acid (Chan et al, 1985). The role of NO in excitotoxicity is still unclear. Inhibition of NO synthase has been shown to reduce rapidly triggered excitotoxicity in rat cortical cultures (Dawson et al, 1991), while NADPH-diaphorase positive cells, which have NO synthase, have been shown to be generally more resistant to N M D A toxicity (Koh and Choi, 1988). More work has to be done to fully understand the role of each factor in excitotoxic cell death. Based on these results several attempts have been made to intervene with the cascade leading to cell death. One possible method is simple inhibition of N M D A receptors. Several non-competitive antagonists of NMDA-receptors have been developed and they have been able to inhibit neuronal cell death following various CNS insults (Kochhar et al, 1988; Sheardown et al, 1990). As mentioned previously zinc is present in several regions of the CNS and is co-released with glutamate. Since it blocks N M D A receptors in a non-competitive manner, it could play an important role in protection of neurons from excitotoxic cell death. 1.6 Research hypothesis A number of in vivo and in vitro studies have indicated that over-activation of N M D A receptors causes an increase in calcium entry into the cell, ultimately leading to cell death. However, the precise role of N M D A receptors has been difficult to study in neuronal systems due to the presence of a variety of other ligand-gated and voltage-gated ion channels. Several groups have reported that incubation of HEK-293 cells transiently 35 transfected with subunits of the N M D A receptor in medium containing serum, causes selective death of the transfected cells (Cik et al, 1993; Anegawa et al, 1995). Since HEK-293 cells do not express significant levels of other ligand-gated or voltage-gated ion channels, expression of recombinant glutamate receptors in these cells provides a system where the role of different subunits of glutamate receptors in mediating glutamate-induced cell death can be studied. The objective of this study is to characterize this system more carefully and to test the effect of extracellular zinc on different subunits of the N M D A receptor. Using patch-clamp recordings and cytotoxicity assays, an attempt to correlate the biophysical properties of glutamate receptors with their cytotoxic potentials will be made. Finally, the role of the metal chelator, serum albumin, in glutamate-induced toxicity will be examined. 36 CHAPTER 2 Materials and Methods 2.1 Cell Culture Human Embryonic Kidney-293 ( HEK-293) cells, obtained from American Type Culture Collection (CRL 1573), were maintained at 37°C and 5% C 0 2 in minimum essential medium (MEM) containing Earle's salts and supplemented with L-glutamine (2 mM), sodium pyruvate (1 mM), penicillin / streptomycin (100 U / ml) and 10% fetal bovine serum (FBS). The cells were passaged every three to four days and were plated at a density of ~2 x 106 / ml. In order to split the cells, the medium was aspirated and the cells were washed once with warmed phosphate-buffered saline (PBS) solution and incubated with 1 ml trypsin (0.25% in PBS) at 37°C and 5% C 0 2 for 1-2 minutes. Then, 5 ml of fresh medium was directly pipetted onto the culture and the cells were agitated until they detached from the plate. The cells were then divided into different dishes containing appropriate amounts of media. 2.2 PlasmidcDNA NR1A, NR2A, and NR2B were cloned into pRK5 ( a mammalian expression vector containing the C M V promoter). The NR1A and NR2B cDNAs were gifts from S. Nakanishi. NR2A (also called el) cDNA was a gift from M . Mishina. cDNA for GluRlj and GluR6 (the fully edited version) were both subcloned into the pRK5 vector and were gifts from P. Seeburg and S. Heinemann, respectively. The mutant NR1 subunits NR1 (N598Q) and NR1(N598R) were obtained from W. Tingley and R. Huganir. The 37 plasmid D N A containing the Escherichia coli (3-galactosidase ((3-gal) gene (pCMV(3) was obtained from Clontech Laboratories, Inc. (Palo Alto, CA). 2.3 Preparation of plasmid D N A 2.3.1 Transformation DH5a-competent cells ( Gibco BRL) were used for the preparation of plasmid DNA. 100 ul of the competent cells was added to the required number of pre-cooled eppendorf tubes. 30 ng of the plasmid D N A was added to each tube and the content was mixed by gentle tapping. The cells were incubated on ice for 30 minutes. They were then heat shocked for 45 seconds at 37°C and were immediately placed back on ice for 2 minutes. 0.95 ml of room temperature Luria-Bertani (LB) medium was added to each tube and they were shaken at 225 R P M for 1 hour at 37°C for expression. 100 ixl of the mixture was spread on L B plates containing 100 u,g / ml ampicillin and 50 ug / ml X-gal. The plates were incubated overnight at 37°C. 2.3.2 Growth of Transformant for Plasmid Preparation Single colonies of bacteria, transformed with plasmids were picked from plates and grown for ~6 hours in 5 ml of L B broth containing 100 u,g / ml ampicillin by shaking at 37°C at 250 R P M . The 5 ml cultures were added to 500 ml of terrific broth (TB) containing 100 mg of ampicillin. TB consists of 12 g trypton, 24 g yeast extract, 4 ml glycerol in 900 ml of H2O mixed with 100 ml of phosphate buffer containing 2.3 g KH2PO4 and 12.5g K 2 H P 0 4 . The mixture was shaken overnight at 37"C at 250 R P M . 38 2.3.3 Maxiprep of plasmid D N A For separation and purification of plasmid DNA, the Nucleobond ® Ax kit (Macherey-Nagel GmbH & Co., Germany) was used. The procedure outlined in the instruction book of the kit was followed. The D N A obtained was dissolved in 1 x TE (Tris-EDTA) buffer, pH 7.5 and the final concentration and purity determined by measuring absorbance at 260 and 280 nm wavelengths, using the Ultraspec 3000 (Pharmacia Biotech) spectrophotometer. 2.4 Transfections Cells were transfected by the method of calcium phosphate precipitation (Chen and Okayama, 1987). Cells were plated 12-24 hours prior to transfections. A total amount of 10-20 pg of plasmid D N A was used to transfect 10 cm plates. In experiments requiring expression of two different glutamate subunits, equal amounts of cDNAs for each subunit were added. In co-transfection of glutamate receptor subunits and lacZ (encoding bacterial (3-gal) cDNAs, the lacZ cDNA comprised one fourth of the total amount of transfected cDNA. The required combination of the plasmid cDNA was mixed in an eppendorf tube. 3 M sodium acetate and 100% ethanol were added to the mixture at volumes equaling one tenth and 3 times the total plasmid cDNA volume, respectively. The tube was then spun in a desktop centrifuge for a brief period of time. It was then filled with 100% ethanol and spun at 15,000 R P M for 15 minutes at 4°C. The ethanol was then aspirated, leaving a white pellet of cDNA. The cDNA pellet was 39 resuspended in 450 ul of sterile 0.1 x TE ( 0.045 M Tris, 0.001 M EDTA, pH 7.4). 50 ul of 2.5M CaCb was then added to the tube and the content was thoroughly mixed. 500 ul of the solution was transferred to a tube containing 500 JLX.1 of 2 x BES ( 50 mM BES, 280 mM NaCl, 1.5 mM Na2HP04, pH 6.96). The contents were mixed and kept in a sterile flow hood for 20 minutes. Thirty minutes prior to the transfection of cells, culture plates were transferred to a 37°C and 3% CO2 incubator. The contents of the tube were added to the plate and the cells were kept at 37°C and 3% CO2 for 10-14 hours. After the incubation period, the media was aspirated and the cells were washed twice with 5 ml of media. Cells were then incubated with 10 ml of fresh media. To the cells transfected with N M D A receptor subunits, 250 ul of (±) 2-amino-5-phosphopentanoic acid solution (APV; 40 m M A P V , 140 mM NaCl, 25 mM Hepes, pH 7.4) was also added. To prepare the cells for toxicity assays the cells in each 10 cm plate were divided into 6 poly-D-lysine pre-coated 35-mm plates (or, in some cases, two or three 10 cm plates were combined and divided into 12 or 18 smaller plates). In order to prepare the cells for electrophysiological recordings, the cells were cultured on glass coverslips. 2.5 Cytotoxicity Experiments Assessment of agonist induced cell death was performed by removal of the media from the transfected cells -40 hours following the start of the transfection. Cells were washed once with warm PBS and then incubated for six hours (5% CO2, 37"C) in solution containing, in mM: 140 NaCl, 5.4 KC1, 1.4 CaCl 2 , 1.2 NaH 2 P0 4 , 21 glucose, and 26 NaHC03, titrated to pH 7.4 with HC1 (osmolarity 325 mOsm). For experiments 40 examining the importance of extracellular sodium, we used a low sodium solution with the same components, except that 140 mM NaCl was replaced by 130 m M N-methyl glucamine chloride and it was titrated with K O H (osmolarity 320 mOsm). In chloride substitution experiments, the cells were incubated in solution containing, in mM: 70 Na 2 S0 4 , 21 glucose, 5.4 K 2 S 0 4 , 1.2 NaH 2 P0 4 , 26 NaHC0 3 , 1.4 CaCl 2 , 70 sucrose, titrated to pH 7.4 with NaOH. These solutions were pre-incubated for 30-60 minutes in C 0 2 5% and 37"C incubator and then N M D A (or A M P A or Kainate) as well as 50 uM glycine were added just prior to addition to the cells. 2.6 Immunocytochemistry At the end of the six hour incubation period with agonist containing buffered-salt solution, the cells were washed once with warm PBS and fixed in a 3% paraformaldehyde solution in 0.1 M K 2 P 0 4 , pH 7.3 for 30 minutes. Immunocytochemical staining for NR1 expression was performed using an affinity-purified rabbit polyclonal antibody generated against a 21 amino acid synthetic peptide corresponding to the C-terminus of NR1A (designated as anti-NRl). The cells were incubated with 0.5% Triton X-100 and 5% normal goat serum in PBS for 1 hour, followed by an overnight incubation at 4"C with 5 (ig / ml anti-NRl antibody in 1% goat serum / P B S , and a 1 hour incubation with a biocytin-linked goat anti-rabbit antibody (Amersham, Buckinghamshire, England). Finally, the cells were stained with the Vectastain A B C kit (Vector Laboratories, Inc., Burlingame, CA). The cells were then maintained in glycerol and the extent of cell death was determined. 41 2.7 P-gal staining Similarly, for P-gal staining the cells were washed with warm PBS after the six hour incubation period. They were then fixed by incubating in 0.4% glutaraldehyde / 2% formaldehyde in PBS for 5 minutes. The cells were then washed three times with PBS. P-gal expression was assessed by incubating the fixed cells for -12 hours at 37°C in freshly made solution containing 1.2 mM 5-bromo-4-chloro-3-indolyl P-D-galactopyranoside (X-gal), 5 mM potassium ferrocyanide, 5 mM potassium ferricyanide and 2 mM M g C l 2 . 2.8 Assessment of cell death The extent of transfected cell loss was assessed by counting NR1- or P-gal positive and negative cells and comparing the ratio of transfected cells to the total number of cells still remaining attached to the culture dish under different conditions. A total of at least 1000 cells (6-10 fields at 200x magnification) were counted for each plate and the results were normalized to the percentage of transfected cells alive for the control conditions. The percentage of cells alive in each plate was calculated as: Number of transfected cells x 100 Number of transfected cells + number of untransfected cells Each plate was counted twice, often by two different observers. A l l experimental conditions were replicated in at least three different transfections. Results of different 42 experimental conditions were compared using A N O V A , and significant differences were determined by the Fisher PLSD at the 95% confidence level. 2.9 Solution assay for p-gal activity After the six hour incubation period, the incubation buffer was transferred to an eppendorf tube and kept on ice. 1.5 ml of 0.1 M sodium phosphate, pH 7.5 ( prepared by mixing 41 ml of 0.2 M Na 2 HP0 4 , 9 ml of 0.2 M N a H 2 P 0 4 and 50 ml of H 2 0 ) was added to the culture dish and the cells were scraped off the dish using a cell scraper. This solution was then transferred to another eppendorf tube and was kept on ice. Both tubes containing the buffer solution and the cells were sonicated for 20 seconds using a cell sonicator (Ultrasonic W-10, Plainview, N.Y.) in two 10-second pulses. The tubes were immediately placed back on ice in order to avoid proteolysis. For each sample to be assayed, the following solutions were mixed: 100 x Mg solution 6 pi 1 x ONPG ( o-nitrophenyl-P-D-galactopyranosidase ) 132 u,l 0.1 M sodium phosphate (pH 7.5) 402 ill sample solution ( cell or buffer) 100 pJ 100 x Mg solution is 0.1 M MgCl 2 , 4.5 M (5-mecaptoethanol. 1 x ONPG is a 4 mg / ml solution of ONPG dissolved in 0.1 M sodium phosphate. The reaction mixture was incubated at 37°C for 10-30 minutes or until a faint yellow color was developed. The reaction was stopped by the addition of 1 ml of 1 M Na 2C03 to each tube. In each set of reactions, the incubation period for both the buffer 43 solution and the cell solution was equal. The optical density ( O.D.) of the reaction was read at a wavelength of 420 nm using the Ultraspec 3000 (Pharmacia Biotech) spectrophotometer. In each set of reactions a negative control was also included with the same mixture of solutions, with the exception of the sample solution which was replaced with 0.1 M sodium phosphate. The ratio of cell death was determined by: O . D .420 (buffer) x 100 O . D .420 (buffer) + O .D. 4 2 0 (cell) Cell death for each experimental condition was normalized to death obtained for the control, in which the cells were incubated with physiological salt solution. In order to determine the linear range of the reaction, instead of the sample solution, a solution with different concentrations of (3-galactosidase ranging from 0-12 units / ml was added to each reaction mixture and the O . D .420 measured just after a yellow color was developed. One unit of E. Coli [3-galactosidase is defined as the amount of the enzyme that will hydrolyze 1 umoles of ONPG in 1 minute at 37°C. 2.10 Electrophysiology 2.10.1 Recordings 40-48 hours following the start of transfections, the transfected HEK-293 cells plated on glass cover slips were transferred to the stage of an inverted microscope (Axiovert 100, Carl Zeiss, Inc.). Patch clamp recordings in the whole cell configuration (Hamill et al, 1981) were made under voltage-clamp ( V H = -60 mV) at room temperature. Electrodes were pulled from borosilicate glass (Warner Instruments Corp., Hamden, CT) 44 using the Narishige PP-83 electrode puller (Narishige scientific instruments, Japan). Electrodes with resistance of 1-5 M Q were used. The tip of the electrodes were fire-polished just prior to recording. Currents were generally sampled at 2-4 kHz, and filtered •a at 1-2 kHz. 100 p M glutamate was rapidly applied by a piezo-driven theta tube (Hilgenburg, Malsfeld, Germany). Control and agonist solutions were gravity-fed continuously through the two different sides of the theta tube which was positioned within -100 pm of the cell. Switching between control and agonist solutions was accomplished by computer-triggered piezo-mediated movement of the theta tube, and the 10-90% rise time for exchange of the two solutions was 1-2 ms. Control solution was also continuously gravity-fed directly into the chamber, while at the other end, solutions were drawn off at a rate which kept the fluid depth more-or-less constant. When changing solutions, both the bulk flow and the solutions from the theta tube were replaced. 2.10.2 Recording Solutions In the recording chamber, cells were kept in standard external solution containing, in mM: 145 NaCl, 5.4 KC1, 1.8 CaCl 2 , 11 glucose, and 10 Hepes, titrated to pH 7.3 with NaOH. In some experiments, the extracellular solution was replaced with a solution containing high C a 2 + and no Na + in order to determine the C a 2 + vs. K + reversal potential. In these situations the bathing solution contained, in mM: 110 CaCh, 5.4 KC1, 25 Hepes, and 11 glucose, titrated to pH 7.3 with Ca(OH)2. In all recordings, 50 p M glycine was added to the extracellular solution. 45 The solution in the patch pipette contained, in mM: 145 KC1, 5.5 B A P T A , 0.5 CaCl 2 , 2 M g C l 2 , 2 TEA, 4 MgATP, and 10 Hepes, titrated to pH 7.2 with K O H . Glutamate and glycine were prepared from 10 mM stock solution and were added to the appropriate solution to give the final concentration of 100 u M and 50 uM, respectively. ZnCl 2 was prepared from 1 m M stock solution to give the appropriate final concentration. In all cases concentration rather than activity was used. 2.10.3 Data Analysis Current signals were acquired and analyzed using p C L A M P software and the Axopatch 200 amplifier (Axon Instruments, Foster City, CA). Sets of different results were compared using the Student's t-test and significant differences were determined at 95% confidence intervals. Curve fitting was done using a least square regression routine using commercial software (Axum, Trimetrix, WA). The dose-response values were fitted to the equation I = I m a x x ( 1 / (1 + EC50 / [ agonists ] )n), where I m a x is the maximum current amplitude, EC50 is effective concentration at 50% block of the peak current, n is the Hil l coefficient. 46 CHAPTER 3 NMDA over-exposure causes selective death of transfected HEK-293 cells 3.1 Expression of functional NMDA receptors in HEK-293 cells HEK-293 cells were transiently transfected with cDNAs for NR1 and NR2A (the mouse homologue, el) using the calcium precipitation method (Chen and Okayama, 1987) described in chapter 2. At the end of the transfection period the DNA-containing medium was replaced with fresh medium containing 10% FBS. In order to protect transfected cells from over exposure to excitatory amino acids (glutamate, aspartate) present in the serum, 1 m M (±) A P V , a competitive blocker of N M D A receptors, was added to each 10 cm culture dish. -40 hours following the start of the transfection, the medium containing A P V was removed, and the cells were incubated for 6 hours (37°C, CO2 5%) with physiological salt solution containing 50 p M glycine. After the six hour period, cells were fixed and stained with NR1-specific antibodies. Figure 1A shows a typical experiment in which approximately 50% of the cells were transfected and were expressing the NR1 subunit of the N M D A receptor. On average, the transfection efficiencies of NR1 and NR2A plasmids varied between 20-80%, depending on the health of the cells and the pH of the BES solution used for transfections. A sister culture of transfected cells was incubated for a 6 hour period in physiological salt solution containing 3 m M N M D A and 50 p M glycine. Similarly, after the six hour period the 47 cells were fixed and stained for NR1 expression (figure IB). Interestingly, the cells incubated with the NMDA-containing solution appear to have a much lower transfection rate than the control cells; however, the number of untransfected cells was largely unaffected. Loss of NR1 expression was associated with a marked decrease in cell density, consistent with NMDA-induced death of NR1 / NR2A transfected cells. These results suggest that transfected HEK-293 cells express functional N M D A receptors and that over-exposure of these cells to N M D A decreases the survival rate of the transfected cells. 3.2 N M D A causes dose-dependent cell death in transfected HEK-293 cells Following transfection of HEK-293 cells, each 10 cm culture dish was divided into 6 poly-D-lysine pre-coated 35 mm plates. -40 hours after the start of the transfection, the medium containing 1 mM (±) A P V was removed and the cells were incubated for 6 hours (37°C and CO2 5%) in physiological salt solution containing different concentrations of N M D A in the presence of 50 u M glycine. After the incubation period, the cells were fixed and stained for NR1 expression. The percentage of transfected cells remaining on the dish was determined on each plate and compared to the percentage of transfected cells incubated in physiological solution lacking N M D A . Figure 2 shows the dose-response curve for N M D A toxicity obtained in this manner with the data fitted to the Hil l equation. The EC50 for N M D A toxicity was approximately 300 uM with slope factor of 1.1 and maximal cell death at 65%. Importantly, there was no transfected cell death detected at the end of the six hour incubation with physiological 48 Figure 1. Immunocytochemical staining of HEK-293 cells expressing NR1 and NR2A subunits of the N M D A receptors. At -40 hours following the start of the transfection, cells were incubated for 6 hours in physiological salt solution containing 50 u M glycine and 0 (A) or 3 m M N M D A (B). After which, the cells were fixed and stained for NR1 expression using the anti-NRl antibody. Both photographs were from a typical experiment and were taken at 200x magnification. A decrease in the percentage of transfected cells was observed when cells were incubated with N M D A solution, while the number of untransfected cells was largely unaffected. 49 50 Figure 2. Dose-response for NMDA-induced toxicity in transfected HEK-293 cells. At -40 hours following the start of the transfection, cells were incubated for 6 hours in physiological salt solution containing 50 p M glycine and varying concentrations of N M D A . The cells were then fixed and stained for NR1 expression using the anti-NRl antibody. The percentage of surviving transfected cells remaining attached to the dish was determined and normalized to the control condition (i.e. no N M D A ) . Each condition was repeated with at least three different transfections, and the mean values were fitted to the Hil l equation. The E C 5 0 for N M D A toxicity was -300 p M with slope factor of 1.1 and maximal cell death at 65%. 51 52 salt solution lacking N M D A as compared with cells which were kept in standard medium with 1 m M (±) A P V . Deprivation of medium (hence, growth factors) did not seem to cause selective death of transfected cells. However, a general decrease in the overall cell density was observed in cells incubated with physiological salt solution when compared to cells kept in standard medium containing the N M D A receptor inhibitor, A P V . 3.3 Using (3-galactosidase as a marker of transfected cells HEK-293 cells were transfected with cDNA encoding NR1 and NR2A subunits along with lacZ cDNA, encoding the bacterial protein (3-galactosidase ((3-gal). The relative amount of lacZ cDNA was one quarter of the total cDNA with equal amounts of the other two subunits. Following the transfection period the cells were treated as before, and incubated in fresh medium containing A P V . After -40 hours the cells were incubated in physiological salt solution containing 50 u M glycine in the presence or absence of 3mM N M D A . The cells were fixed and stained either with NR1-specific antibody (Figure 1) or for expression of (3-gal (Figure 3). Figure 3 shows the cells stained for (3-gal protein. The transfection rate of cells incubated in control solution (figure 3A) was about 50% and this rate was much reduced in cells incubated in 3mM N M D A solution. Importantly, similar rates of transfections were obtained using both staining methods (compare figure 1A with 3A and IB with 3B). In order to compare the N M D A dose-response curve obtained from the (3-gal staining with that for NR1-expressipn, each 10 cm plate of transfected cell was transferred to 6 poly-D-lysine pre-coated plates and incubated for 6 hours in physiological salt solution containing 50 uM 53 Figure 3. (3-galactosidase staining of HEK-293 cells expressing NR1, NR2A, and (3-gal protein. At -40 hours following the start of the transfection, cells were incubated for 6 hours in physiological salt solution containing 50 uM glycine and 0 (A) or 3 mM N M D A (B). After which, the cells were fixed and stained for (3-gal expression. Both photographs were from a typical experiment and were taken at 200x magnification. In the control condition, -50 % of the cells were transfected (A), while this rate was much reduced when the cells were incubated with 3 mM N M D A solution (B). 54 55 Figure 4. Dose-response for NMDA-induced toxicity in NR1 / NR2A / (3-gal-transfected HEK-293 cells generated by staining surviving cells for (3-gal expression. At -40 hours following the start of the transfection, cells were incubated for 6 hours in physiological salt solution containing 50 u M glycine and varying concentrations of N M D A . The cells were then fixed and stained for (3-gal expression. The percentage of surviving transfected cells remaining attached to the dish was determined and normalized to the control condition (i.e. no N M D A ) . Each condition was repeated with three to seven different transfections and the mean value was fitted to the Hil l equation. The EC50 for N M D A toxicity was -290 u M with slope factor of 1 and maximal cell death at 63%. 56 o o o o o o O 00 CO CM (%) s||93 6u|ssajdxe \ed-q jo |BAiAjns 57 glycine and different concentrations of N M D A . After which they were fixed and stained for (3-gal expression. The percentage of cell survival was determined as before and compared to the control solution. Figure 4 shows the N M D A dose-response obtained using (3-gal staining, with the data fitted to the Hil l equation. The EC50 for N M D A toxicity was -290 u M with a slope factor of 1 and maximal cell death at 67%. These results indicate that staining for (3-gal in NR1 / NR2A / (3-gal transfected HEK-293 cells accurately reflects the population of surviving N M D A receptor-expressing cells. Since the (3-gal staining was technically simpler and more rapid than the anti-NRl immunocytochemical staining method, the former staining method was used to assess the survival of cells expressing (3-gal and glutamate receptors in experiments described in the next two chapters. 3.4 Glutamate did not cause cell death in transfected HEK-293 cells Incubation of transfected cells with glutamate at concentrations up to 3 mM instead of N M D A did not seem to cause a significant decrease in transfected cell survival as compared to the control condition. However, when cells were incubated in salt solution containing 50 uM L-trans-pyrrollidine-2,4-dicarboxylic acid (trans-PDC), a glutamate uptake blocker, cell death was observed in the presence of glutamate and glycine. These results indicate that HEK-293 have an efficient glutamate transporter and cell death can only be obtained when the transporter is inhibited. Glutamate-induced cell death in the presence of 50 uM trans-PDC and 50 u M glycine had an EC50 of -300 uM with maximal cell death at 65% 58 Figure 5. The role of extracellular calcium in NMDA-induced cell death. HEK-293 cells were transfected with NR1 / NR2A and NR1 (N598Q) / NR2A, or NR1 (N598R) / NR2A, along with (3-gal. ~ 40 hours after transfections, the cells were incubated for six hours in physiological salt solution containing 50 uM glycine and different concentrations of N M D A . After which, the cells attached to the dish were fixed and stained for (3-gal expression. The percentage of transfected cell survival was determined as described in chapter 2. From the same batch of cells, some of the plates were kept in the medium containing A P V , and after the six hour incubation period, they were also fixed and stained for (3-gal expression. Incubation of cells with physiological salt solution did not cause selective death of transfected cells as compared to the cells kept in medium. Transfection of cells with the mutant subunits increased the survival rate when compared to the wild-type. Values are mean ± S E M from at least three different transfections. * indicates that the values differed significantly (p < 0.05) from that obtained for NR1 / NR2A transfected cells and + indicates that values for NR1 (N598Q) / NR2A-transfected cells are significantly different (p < 0.05) from NR1 (N598R) / NR2A-transfected cells. 59 09 Survival of B-gal expressing cells (%) ro o> oo o o o o o o o as determined by staining for (3-gal expression ( n=3 different transfections; data not shown). 3.5 Calcium and sodium influx contribute to NMDA-induced cytotoxicity In order to test the hypothesis that calcium influx through N M D A receptors is crucial for mediating toxicity in our system, mutant N M D A subunits that have previously been shown to have decreased calcium permeability, were transfected and the amount of cell death obtained was compared to that of the wild-type receptor. Two NR1 mutants were tested, one in which asparagine was replaced with glutamine (Q; NR1(N598Q)), that has been shown to decrease the calcium permeability, and another in which asparagine was replaced with arganine ( NR1(N598R)) which almost completely eliminates the calcium permeability. The mutant NR1 cDNAs were transfected along with NR2A and p-gal cDNAs, and the transfected cells were incubated in salt solution for 6 hours containing different concentrations of N M D A . Then, they were stained for P-gal expression and the percentage of cell survival under each condition was determined. Cells transfected with either of the mutant NR1 subunits exhibited a significantly reduced maximal cell death (figure 5). The maximal cell death for the glutamine mutant, NR1 (N598Q), was 35 ± 0.1 % (mean ± standard deviation for 10 mM N M D A ), and for the arginine mutant, NR1(N598R), was 23 ± 4% (10 mM N M D A ) , reduced from -65% observed for the wild-type receptor (figure 5). Interestingly, the EC50 for N M D A toxicity 61 Figure 6. The role of extracellular Na + in NMDA-induced cell death. HEK-293 cells were transfected with either NR1 / NR2A or NR1 (N598R) / NR2A along with (3-gal. About 40 hours after the transfection, the cells were incubated in low N a + solution ( N -methyl glucamine chloride was substituted for NaCl) containing 50 u M glycine and different concentrations of N M D A . The cells were then fixed and stained for (3-gal expression and the relative survival of transfected cells was determined. In cells transfected with the wild type N M D A receptor, the maximal cell death was reduced to 34%. Even at the highest concentration of N M D A , 3 mM, cell death was not observed in cells transfected with the calcium impermeable N M D A receptor, NR1 (N598R) / NR2A. Values are represented as mean ± S E M from three different transfections. * represents values significantly different (p < 0.05) from wild-type incubated in standard solution and + represents values significantly different between wild-type and mutant receptors incubated in low sodium solution. 62 £9 Survival of B-gal expressing cells (%) ro ^ o> oo o o o o o o o 3 I 1 1 1 1 1 1 1 1 1 r seems to have been decreased for the almost calcium-impermeable NR1 mutant (NR1 (N598R)). In order to test the importance of extracelluar sodium in N M D A cytotoxicity, two different incubation solutions were used. In one of the solutions N-methylglucamine chloride was used to substitute for NaCl, and H E K cells transfected with wild-type and calcium impermeable (NR1(N598R)) recombinant receptors were incubated for six hours in this solution. Figure 6 shows the N M D A dose-response obtained under this condition. No significant cell death was observed in cells transfected with calcium impermeable N M D A receptors, indicating that the residual cell death seen for this mutant receptor when incubated in the standard salt solution was dependent on extracellular sodium. Interestingly, in the low Na + solution, the maximal cell death for the wild type NR1 / NR2A transfected cells was reduced to 34 ± 3% at 10 mM N M D A , compared with -65% in the standard salt solution. Removal of extracellular chloride, instead of sodium, yielded similar results to those obtained in the sodium substitution experiments, except that the extent of maximal cell death for wild type transfected H E K cells was slightly larger ( 44 ± 2 %, n=3; data not shown). These results indicate that in this system both calcium and sodium influx contribute to the mechanisms causing cell death. 64 CHAPTER 4 Rapid desensitization of non-NMDA receptors is protective against agonist-induced cytotoxicity Numerous reports have indicated that neurons in culture are far more susceptible to N M D A than non-NMDA receptor mediated cell death (Choi, 1994). In general, in order to observe non-NMDA receptor mediated cell death, neurons require higher concentrations of the agonist ( A M P A / Kainate ) and protracted exposure times. The underlying reasons for this protection are still unclear. However, several neuron types, including cerebellar Purkinje neurons as well as NADPH-diaphorase positive neurons, have been shown to be vulnerable to kainate induced toxicity (Koh and Choi, 1988). Several subtypes of non-NMDA receptors, such as the Kainate receptor subunit GluR6 (the unedited version), as well as A M P A receptors lacking the GluR2 subunit exhibit significant calcium permeability in a similar range as the glutamine mutant NR1 (N598Q) / NR2A receptor combination, suggesting that decreased calcium permeability alone can not be the only reason for the protection. Experiments were designed to test the hypothesis that biophysical properties of the non-NMDA receptors, specifically their rapid desensitization, in large part underlie their reduced toxicity. HEK-293 cells were co-transfected with cDNAs encoding either GluRl ; or GluR6 (the edited version) along with p-gal. -40 hours following the transfection, the cells were incubated in physiological salt solution for 6 hours ( 37°C and CO2 5%) in the presence or absence of saturating concentrations of agonists. Cells 65 transfected with GluRl i were incubated with 100 uM A M P A , while cells transfected with GluR6 were incubated with 1 mM Kainate solution. The percentage of cell survival was determined by staining cells remained attached to the plate for (3-gal expression. No apparent decrease in the percentage of transfected cells were observed in either of those conditions when compared to the control condition. In parallel, some of the cells transfected with GluRl ; were incubated with physiological salt solution containing saturating concentrations of A M P A in the presence or absence of 100 u M cyclothiazide, an inhibitor of desensitization of AMPA-type receptors (Partin et al, 1993). Some of the GluRl ; transfected cells were also incubated in 1 mM Kainate solution for 6 hours. After the incubation period the percentage of transfected cell survival was determined by staining for (3-gal expression. AMPA-induced cell death was only evident in the presence of cyclothiazide (figure 7). 1 mM Kainate, which also produces non-desensitizing currents in A M P A receptors, caused 21 ± 1 % cell death. In parallel, some of the cells transfected with the Kainate receptor GluR6 subunit were treated for 10 minutes to 5 pJVI concanavalin A solution. Concanavalin A (con A) has been shown to inhibit desensitization of Kainate-type glutamate receptors (Partin et al, 1993; Patneau et al, 1993). These cells were then exposed to saturating concentrations of agonist for 6 hours. Following agonist exposure, the relative survival of the transfected cells was determined by staining for [3-gal expression. Kainate over-exposure was only toxic to transfected cells when they were pre-exposed to con A (figure 8). Six hour incubation of GluR6-transfected cells with saturating concentration of Kainate, following pretreatment with con A, resulted in a 18 ± 3 % decrease in the percentage of 66 transfected cells, while incubation of GluRlj transfected cells with saturating A M P A concentrations in the presence of cyclothiazide caused a 28 ± 2 % decrease. The increased toxicity observed for GluRl;-transfected cells compared with GluR6-transfected cells correlates well with the relative C a 2 + permeability of the two receptor subtypes (see discussion). These results indicate that the rapid desensitization of the non-N M D A receptors is protective against glutamate toxicity. 67 Figure 7. AMPA-induced toxicity is only evident in the presence of cyclothiazide. HEK-293 cells were transfected with the GluRlj subunit of the A M P A receptors along with p-gal. -40 hours after the transfection, the transfected cells were incubated with physiological salt solution containing either 100 uM A M P A , 100 p M cyclothiazide, 100 p M A M P A and cyclothiazide, or 1 m M kainate. After 6 hours, the cells were fixed and stained for P-gal expression. In parallel, some cells were left in standard medium and they were also stained for P-gal expression. The percentage of cell death was determined and normalized to the control condition. AMPA-induced cell death was only observed in the presence of cyclothiazide. 1 m M kainate, which produces non-desensitizing currents, also caused some cell death. Values are represented as mean ± S E M as in all graphs in this thesis. * represents values significantly different from cells incubated in Na + ringer solution and + represents values significantly different between cells incubated in A M P A + CTZ and Kainate containing solutions. 68 Figure 8. Kainate-induced toxicity is only evident in cells pre-treated with concanavalin A (con A). HEK-293 cells were transfected with the GluR6 (the fully edited version) subunit of the kainate receptors and (3-gal. -40 hours after the transfection, some of the transfected cells were pretreated with 5 u M con A for 10 minutes. Cells were then incubated with physiological salt solution containing 1 mM kainate. After 6 hours, the cells were fixed and stained for p-gal expression. "Medium" indicates control transfected cells maintained in standard medium. The percentage of cell death was determined and normalized to the control condition. Kainate induced cell death was only observed in cells pretreated with con A. Values represents mean ± S E M from at least three different transfections. * represents values significantly different from cells incubated in N a + ringer solution. 70 CHAPTER 5 Zinc protects against N M D A cytotoxicity in transfected HEK-293 cells 5.1 (3-galactosidase assay In the previous set of experiments, where cells remaining on the dish were counted, the assumption was that upon death of HEK-293 cells they detached from the culture plates and floated in the salt solution. This assumption was never tested. Also, in general, counting cells can be quite subjective even though proper measures were taken to ensure the most accurate count was taken. In order to more accurately determine the amount of cell death under each experimental condition, an assay system was designed that would determine the (3-gal activity of the cells remaining attached to the dish in relation to detached cells in the salt solution. Therefore, after the incubation period, the cells and the salt solution were transferred to separate tubes and sonicated to rupture the cell membranes. A sample of each solution was added to a mixture containing orthonitrophenyl galactoside (ONPG) as substrate. (3-galactosidase enzyme cleaves this substrate to produce galactose and orthonitrophenyl, which at basic pH has a distinct yellow color with maximum absorbance at 420 nm. Since the only rate determining step in the reaction is the presence of (3-galactosidase enzyme ( the other products are at saturating concentrations), the amount of product and, hence, the O.D. 4 2o readings reflect the approximate amount of (3-galactosidase. In order to show the relationship between the optical density at 420 nm and the (3-galactosidase activity, the reaction mixture containing different concentrations of (3-gal was incubated for 10 minutes in 37°C water 72 bath. Figure 9 shows the linear relationship between p-gal concentrations and the O.D. readings of 0.1 to 1.2 (for calculating the amount of cell death, O.D .420 values between 0.2 - 0.8 were used). Hence, determining the O.D. 4 2o within the linear range gives a good approximation of the total P-gal enzyme present in the sample. However, since in each experiment the rate of transfection and, hence, the amount of P-galactosidase enzyme, was different, it was important to make direct comparison only between plates from the same transfection. Therefore, in each experiment the amount of cell death was compared to controls from the same batch of transfected cells. HEK-293 cells were transfected with NR1, NR2A, and p-gal plasmids. The transfected cells were incubated for six hours in physiological salt solution containing 50 uM glycine and different concentrations of N M D A . Then, the solution and the cells remaining on the dish (collected in lysis buffer) were transferred to separate tubes and the amount of cell death was estimated by assaying for p-gal activity in the two sample solutions. The dose-response for N M D A cytotoxicity obtained in this manner is shown in figure 10. The maximal cell death was estimated to be 60%. However, it must be mentioned that 10-15% cell death was always observed in the absence of N M D A . This death was, in part, N M D A receptor-mediated, since addition of A P V to the physiological salt solution during washing reduced cell death in control ( 0 N M D A ) to'5-10%. It was suspected that this death came about due to a combination of two processes: 1) non-specific cell lysis and / or detachment during washes; and 2) a small amount of glutamate-induced cell death due to exposure to "burst" of glutamate release during lysis of other cells (APV sensitive). Although HEK-293 cells have an efficient glutamate transporter, the rate of glutamate uptake by the transporter is not known, and 73 Figure 9. P-gal activity is linearly related to O.D .420. Different concentrations of P-galactosidase enzyme, were added to reaction mixtures containing ONPG as substrate. The mixtures were buffered at pH 7.5 and incubated at 37 °C for 10 minutes. The reactions was stopped by addition of 1 M Na2C03. The O.D. at 420 nM wavelength were recorded using a spectrophotometer. The reaction catalyzed by P-gal had a linear range of O.D. 0.1-1.2, after which it began to level off. Values are represented as mean ± S E M (n = 3). 74 Figure 10. Dose-response for N M D A toxicity generated by assaying for (3-gal activity. At -40 hours following the start of the transfection, cells were incubated for 6 hours in physiological salt solution containing 50 p M glycine and varying concentrations of N M D A . The salt solution and the cells attached to the dish were then removed and assayed for (3-gal activity. The percentage of cell death at every experimental condition was determined and normalized to the control condition. The mean values ( ± S E M ; n = 3 to 7) were fitted to a Hi l l equation, with IC50 of 150 pM, slope factor of 1.1 and maximal cell death of 60%. 76 therefore, a sudden increase in glutamate concentration might trigger events leading to a small amount of delayed-type cell death. In all experiments, this amount of cell death for the control was subtracted from the observed amount of death for each experimental condition in order to determine the NMDA-dependent cell death in the specified incubation period. The E C 5 0 for N M D A toxicity was -150 u M N M D A with a slope factor of 1.1. These results are quite similar to the results obtained previously using the staining method, except for a lower EC50. The lower EC50 is possibly due to the more direct comparison between the amount of (3-gal loss, used in this method. The advantage of this method over the staining method is two fold. First, it proves our initial assumption that dead cells detach from the dish, and second, it is a far more accurate system, since human error in counting cells is virtually eliminated. The disadvantage of this method was that care must be taken not to saturate the p-gal activity especially at high transfection rates. In such conditions the incubation period for the reaction was reduced accordingly. This method was used to determine the amount of cell death in subsequent experiments. 5.2 Zinc protects against NMDA-induced toxicity Zinc has been shown to be a modulator of glutamate receptors, having inhibitory effects on N M D A receptors, while potentiating non-NMDA receptors (Mayer et al, 1989; Rassendren et al, 1990). A number of reports have shown that zinc is released in a voltage-dependent manner in various regions of the CNS (Anikszkejn et al, 1987; Charton et al, 1985). However, its role in glutamate-induced toxicity is still unclear. 78 Also, the relative sensitivity of the various heteromeric N M D A receptors to zinc is not known. In order to determine the effect of zinc on N M D A induced toxicity in our system, the cells transfected with NR1 / NR2A / P-gal cDNAs were incubated in physiological salt solution containing 50 uM glycine, 300 nM zinc, and different concentrations of N M D A . Figure 11 shows the dose-response obtained under these conditions. The E C 5 0 for N M D A toxicity was shifted from -150 p M to -800 p M , and maximal cell death was reduced from 60% to -40%. Hence, in our system zinc acted as a non-competitive blocker. The NR1 / NR2A / P-gal transfected cells were then incubated in 1 m M N M D A solution containing 50 p M glycine and varying concentrations of zinc. The amount of cell death at each different zinc concentration was then compared to the amount of death observed for cells incubated with 1 mM N M D A and 50 p M glycine alone. Results show that zinc protected against NMDA-induced toxicity in a dose-dependent manner with an IC50 of 500 nM (figure 12). Receptors composed of NR1 and NR2A subunits expressed in HEK-293 cells seemed to have a much higher affinity for zinc than neuronal N M D A receptors such that 5 u M Z n 2 + was sufficient to completely inhibit N M D A toxicity in our system. Incubation of the transfected cells in physiological salt solution containing different concentrations of zinc (without glycine or N M D A ) showed that in the concentration range used for our experiments, zinc did not have any toxic effects on the cells (data not shown). Cells were then transfected with plasmids containing NR1, NR2B, and lacZ cDNA, in order to compare the zinc sensitivity of this receptor subunit combination with 79 Figure 11. Zinc protects against N M D A toxicity in a noncompetitive manner. -40 hours following the start of the transfection, cells were incubated for 6 hours in physiological salt solution containing 50 u M glycine, 300 nM Zn 2 + , and varying concentrations of N M D A . The salt solution and the cells were then removed and assayed for P-gal activity. The percentage of cell death at every experimental condition was determined and normalized to the control condition. Mean values ± S E M (n = 5-8) were fitted to a Hil l equation. Presence of Z n 2 + shifted the EC50 for N M D A to -800 u M and reduced maximal cell death to 40%. 80 18 Survival of B-gal expressing cells (%) - L I O C O - ^ U l O - N l C O C O O o o o o o o o o o o o 1—i—1—i—1—r Figure 12. Zinc protects against N M D A toxicity in a dose-dependent manner. -40 hours following the start of the transfection, cells were incubated for 6 hours in physiological salt solution containing 1 mM N M D A , 50 u M glycine, and varying concentrations of zinc. The salt solution and the cells remaining on the dish were assayed for (3-gal activity. The percentage of cell death was determined at every concentration and normalized to the control condition ( i.e. no zinc). The mean values (± S E M ; n = 3-4) were fitted to a Hil l equation with IC50 of 500 nM and slope, factor of 1.1. 82 £8 that of NR1 / NR2A. Cells were incubated in solutions containing different concentrations of N M D A for 6 hours, after which the cells and the salt solution were assayed for p-gal activity. Surprisingly, no cell death was observed even at the highest N M D A concentrations during the 6 hour incubation period. The underlying reason for this result is speculated to be the small amplitudes of glutamate-evoked current obtained with expression of NR1 / NR2B in our system. Patch clamp recordings in the whole-cell configuration revealed that the average 100 p M glutamate-evoked current amplitude observed with NR1 / NR2B was about 10 fold smaller than that obtained for NR1 / NR2A (table 2). Therefore, it is possible that sufficient C a 2 + and N a + could not enter the cells in order to activate the mechanism responsible for cell death in our system. Hence, I was unable to perform the cytotoxicity experiments for cells expressing NR1 / NR2B subunit combination. 5.3 Electrophysiological properties of recombinant NMDA receptors Transfected cells were maintained in extracellular solution containing 50 p M glycine (and 0 Mg 2 + ) and were rapidly perfused with 100 u M glutamate solution. Current responses to a three second pulse of the agonist were recorded in the whole-cell mode under voltage clamp at -60 mV. The current responses from different transfected cells displayed a wide range of amplitudes, likely reflecting cell-to-cell variability in N M D A receptor expression levels. The current amplitudes recorded from cells expressing receptors composed of NR1 / NR2A ranged from 250 - 4150 pA with an average current of 1661 ± 1031 pA (n=33). However, the cells expressing NR1 / NR2B 84 apparent transfection rate, and the current amplitudes observed were much smaller, ranging from 16 - 662 pA with an average of 183 ± 151 pA (n=17). Based on the observation of others in the field, it is likely that the decreased expression of the latter receptor combination is due to the presence of an excessive portion of the untranslated region in the plasmid encoding the cDNA for NR2B. Decreased expression of one of the subunits has been shown to be sufficient to decrease overall expression of the receptor complex in heterologous expression systems (Mcllhinney et al, 1996). The two receptor combinations had different biophysical characteristics under our recording conditions (i.e. 100 u M glutamate, 50 u M glycine). Both receptors activated very rapidly in response to agonist application, however, their desensitization rates were markedly different. NR1 / NR2A desensitized in a biexponential manner, with the fast component of the desensitization having a time constant X D F of 142 ± 31 ms (n=21) and the slower component having a XDS of 1253 ± 433 ms (n=21). Current responses mediated by NR1 / NR2B desensitized with an apparent single exponential T D of 945 ± 309ms (n=17), but the extent of desensitization was quite, small (<10%) during the three second pulse. The time constant for dissociation of glutamate (t0ff) also differed between the two different receptor subunit combination. NR1 / NR2A exhibited a T 0ff of 270 ± 28 ms (n=12), while NR1 / NR2B showed a much slower glutamate dissociation, with T0ff of 950 ± 238 ms (n=17). In order to determine the peak current-voltage relationship for the two different subunit combinations, whole-cell voltage clamp recordings were made in which the 85 — + 1^ c is CJ " CJ I s -5 cn +1 cn +1 r t oo CN +1 CN O w r-CN w co +1 C N r t CN co cn +1 cn m CN cn cn cn o +l SO w CM +1 oo m +1 oo cn CN +1 £ O w ON ON o cn +1 m r t ON >/-> +1 00 < CN CN Oi Oi 2; I "3) = 8 8 o H V o a o " .a CD ca <u P , I-I 2 G o CD G G cd > -a CN « O G bO 0) CD c o -2 G CD bO -P .S 6 T3 O I.S c ,G I/) T3 O o CD r/5 CD +1 i? CD i/i G 1 0 G T3 w 3 a O S3* cd G CD tl 3 CD 0 (-! cd * g c§ U CD -G OH H ra * cd TS 3 o 0 o ^ CD —H • •*—> cd cd cd -G bO ^ cd -t_> T) 5 ^ O • M H CD G CD •i-H ^ CD O ^ C m " o 1 ' 3 s o o s— >. c« 4> 2 <U td . « 3 'Hb > t u O CD 0 si 2 S 1 ^ G CD G, ^ CD CD t—1 CD , v 1/3 <+H CD B CD CD CD ° •£ G c bO bO _C ^—» o "OH >^  T3 CD G •'cd -S o O cN a .s CD G CD s CD '— CD •S c o G CD 1-CD ^ CD & >^ •4—» >^ -G Q O G - a G cd cd u > bO .s 8 .s a cd -G O CD O 1/) t/5 O C i — i •1+ ^ o CD * i CD § bO ^ «2 ' "n ^ ^ 1/) § 3 3 CD > 3 .2 %H ^ G O 2 ^ ' 3 ^ 9 c ^ CD -t—> J 3 O . G CL, & Id CD CD cd CD a ^ ^ H cd % v; G G O O \S  3 r G -  G o o 86 holding potential was stepped in 20 mV increments, from -100 to +100 mV, and the current responses to a 500 ms pulse of the agonist were recorded at each holding potential. Both receptor types had a linear current-voltage relation with reversal potentials of 4 ± 1 mV (n=5) and 5 ± 1.5 mV (n=5) for NR1 / NR2A and NR1 / NR2B receptors, respectively. The relative calcium permeability of the two receptors was determined by measuring the glutamate-evoked current-voltage relation in high calcium external solution. The reversal potential under these conditions was 37 ± 3 mV (n=3) and 38 ± 2 mV (n=3) for NR1 / NR2A and NR1 / NR2B, respectively. 5.4 Currents mediated by NR1 / NR2A are more sensitive to zinc than those of NR1 / NR2B The inhibitory effects of zinc on neuronal N M D A receptors and its potentiating effects on homomeric NR1 receptors have been shown previously (see introduction). However, little is known about the sensitivity of the different heteromeric N M D A receptors to zinc. In order to directly observe the effects of zinc on recombinant N M D A receptors and to correlate its biophysical effects on N M D A receptors of different subunit composition with its protective effect on NMDA-induced cytotoxicity, patch-clamp recordings were done in the whole-cell configuration in HEK-293 cells expressing the NR1 / NR2A or NR1 / NR2B subunit combination along with (3-gal protein. Extracellular zinc inhibited the glutamate-evoked whole-cell currents mediated by both types of N M D A receptors in a dose-dependent manner. However, NR1 / NR2A showed a much higher sensitivity to zinc (IC 5 0 of -500 nM- figure 15) than did NR1 / NR2B(IC 5 0 of -10 87 j i M - Figure 17, 19). Moreover, the IC50 for zinc inhibition of glutamate-evoked NR1 / NR2A mediated currents matched that obtained for zinc inhibition of NMDA-induced cytotoxicity (compare figures 12 and 15). Extracellular zinc also affected the time constant of desensitization exhibited by both receptors. The fast component of NR1 / NR2A desensitization (t D F) was decreased to just 0.19 ± 0.04 (n=3) of control in the presence of 10 p M zinc (figure 16). The T D S and x0ff were largely unaffected by zinc concentrations up to 10 p M . However, the percentage of current desensitizing by the fast vs. slow time constant was also affected. In the absence of zinc, -61 % of the current desensitized by the faster time constant, whereas that percentage was reduced to -38 % in 300 nM zinc and to -12 % in 10 p M zinc (figure 13). The time constant of desensitization of NR1 / NR2B was also decreased in the presence of extracellular zinc. To was 286 ± 77 ms in the presence of 100 p M zinc compared to -945 ms in the absence of zinc. The effect of zinc on desensitization was dose-dependent as shown in figure 18. A comparison of the glutamate-evoked current-voltage relationships in whole-cell patch clamp recordings from cells transfected with NR1 / NR2A vs. NR1 / NR2B confirmed that N a + vs. K + and C a 2 + vs. K + reversal potentials were not changed in the presence of extracellular zinc (see figure 14, 18). Hence, zinc did not interfere with the ion permeability of these recombinant receptors. 88 Figure 13. The effect of extracellular zinc on glutamate-evoked whole-cell currents in cells expressing NR1 / NR2A and p-gal. -40 hours following the start of the transfection, cells were voltage-clamped at VH= -60 mV and the current responses to three second pulses of 100 p M glutamate in the presence of 50 p M glycine were recorded. A and B show the effect of different concentrations of zinc on glutamate-evoked currents from two different cells. The leak currents were subtracted from these traces. Extracellular zinc inhibited the peak current and decreased the contribution of the fast-desensitizing current. Note the change in scale between the right and the left panel in B . 89 < OH O o O 4 < OH O o < co OH O O o r t C3 90 Figure 14. Current-voltage relationship for NR1 / NR2A in the control condition as well as in the presence of Z n 2 + . Current responses to 500 ms pulses of 100 p M glutamate ( + 50 p M glycine) at holding potentials ranging from -100 to +100 mV were recorded. The leak currents were subtracted and the peak current at each holding potential was determined. NR1 / NR2A showed a linear current-voltage relationship which was not affected in the presence of 1 p M Z n 2 + . Zinc inhibition of the peak current was not voltage-dependent. 91 Figure 15. Dose-response for zinc inhibition of peak current in cells transfected with NR1 / NR2A and (3-gal. Transfected HEK-293 cells were voltage-clamped at VH=-60 mV and the current responses to 100 p M glutamate (+ 50 p M glycine) were recorded in the presence of different concentrations of zinc. The peak current measured in the presence of zinc was normalized to that of control ( no zinc) for the same cell and the mean values were fitted to a Hil l equation. Each point represents the mean ± S E M from n = 3 to 6 different cells. Because of the relative irreversibility of zinc inhibition, only a single zinc concentration was tested on each different cell. The EC50 for zinc inhibition was 500 nM with slope factor of 1. 93 Figure 16. Zinc decreases the time constant of fast desensitization (TDF) of current responses from cells transfected with NR1 / NR2A. Transfected HEK-293 cells were voltage-clamped at VH=-60 mV and the current responses to 100 p M glutamate (+ 50 p M glycine) were recorded in the presence of different concentrations of zinc. The time constant of fast desensitization of currents in the presence of zinc was determined and normalized to that rate measured in the absence of zinc. Values are represented as mean ± S E M (n =3-4). Zinc decreased the time constant of fast desensitization of NR1 / NR2A current responses in a dose-dependent manner. * represents a significant difference from control by the paired t-test (p<0.05). 95 96 o Figure 17. The effect of extracellular zinc on glutamate-evoked whole cell currents in cells expressing NR1 / NR2B and (3-gal. -40 hours following the start of the transfection, cells were voltage-clamped at VH= -60 mV, and the current responses to three second pulses of 100 p M glutamate in the presence of 50 p M glycine were recorded. A and B show the effect of different concentrations of zinc on glutamate-evoked currents from two different cells. The leak current was subtracted from these traces. Extracellular zinc inhibited the peak current and increased the rate of desensitization of the current. 97 86 Figure 18. Current-voltage relationship for NR1 / NR2B in the control condition as well as in the presence of Z n 2 + . Current responses to 500 ms pulses of 100 p M glutamate ( + 50 p M glycine) at holding potentials ranging from -100 to +100 mV in 20 mV increments were recorded. The leak currents were subtracted, and the peak current at each holding potential was determined. NR1 / NR2B had a linear current-voltage relationship which was not affected in the presence of 10 p M Z n 2 + . Zinc inhibition of the peak current was not voltage-dependent. 99 001 Figure 19. Dose-response for zinc inhibition of peak current in cells transfected with NR1 / NR2B and (3-gal. Transfected HEK-293 cells were voltage-clamped at VH=-60 mV, and the current responses to 100 p M glutamate (+ 50 p M glycine) were recorded in the presence of different concentrations of zinc. The peak currents were normalized to the control currents and the mean values were fitted to the Hil l equation. Each point represents the mean ± S E M (n = 3 different cells for each point). The EC50 for zinc inhibition was 10 p M with slope factor of 1.1. 101 Figure 20. Zinc decreases the time constant of desensitization (xD) of currents responses from cells transfected with NR1 / NR2B. Transfected HEK-293 cells were voltage-clamped at VH=-60 mV, and the current responses to 100 p M glutamate (+ 50 p M glycine) were recorded in the presence of different concentrations of zinc. The time constant of desensitization of current in the presence of zinc was determined and normalized to that for the control response. Values are represented as mean ± S E M (n = 3 different cells for each point). Zinc increased the rate of desensitization of NR1 / NR2B current responses in a dose-dependent manner. * represents a significant difference from control by the paired t-test (p<0.05). 103 CHAPTER 6 Serum Albumin inhibits the protective effect of zinc against NMDA-induced toxicity 6.1 Serum Albumin did not potentiate N M D A toxicity It has been previously noted that incubation of NR1 / NR2A-transfected HEK-293 cells in media containing 10 % FBS is toxic to the transfected cells (Cik et al, 1993; Anegawa et al, 1995; Raymond et al, 1996). The excitatory amino acids present in the serum were thought to be the underlying reason for death of the transfected cells. However, the concentration of glutamate and aspartate in the medium (-60 u M and -48 uM, respectively), measured by HPLC (Schramm et al, 1990), seemed too low to cause such complete death of the transfected cells when compared with the dose-response for N M D A induced toxicity. Hence, it was suspected that some additional factors present in the medium were perhaps potentiating agonist-induced cytotoxicity. Serum albumin has previously been shown to cause a dose-dependent increase in glutamate toxicity in cerebellar granule cells (Eimerl and Schramm, 1991). Since albumin is also a major component of the serum, we decided to examine its effect on NMDA-induced toxicity in NR1 / NR2A-transfected HEK-293 cells. HEK-293 cells were tranfected with cDNAs encoding NR1 and NR2A subunits of N M D A receptors along with (3-gal as a marker of the transfected cells. - 40 hours following the start of the transfection the cells were incubated in physiological salt solution containing 1 mM N M D A , 50 u M glycine, and different concentrations of bovine 105 serum albumin. Three different concentrations were chosen to represent the amount in cerebrospinal fluid ( 0.15 mg / ml ), medium ( 4 mg / ml ), and blood ( 40 mg /ml ). Serum albumin, even at the highest concentration of 40 mg / ml, did not cause significant potentiation of the NMDA-induced toxicity (figure 22). Similar results were obtained using lower concentrations of N M D A (not shown). Furthermore, incubation of transfected cells with physiological salt solution containing only serum albumin ( without N M D A ) did not cause any cell death compared to control (not shown). 6.2 Serum albumin reduced the protective effect of zinc on NMDA-induced toxicity Transfected HEK-293 cells were incubated in physiological salt solution containing saturating concentrations of the N M D A receptor agonists ( 1 m M N M D A , 50 p M glycine ) to cause maximal death of transfected cells. In parallel (i.e. from the same batch of cells ) some cells were incubated with the above solution containing different concentrations of zinc. As shown previously, zinc inhibited the N M D A toxicity in a dose-dependent manner. Interestingly, when the salt solution also included serum albumin, the inhibitory effect of zinc was much reduced. For example, incubation of transfected cells with 3 p M zinc normally reduced the NMDA-induced toxicity to just 6 ± 1 %, however, cell death was increased to 23 ± 6 % (n=3) in the presence of 4 mg / ml serum albumin (figure 21). In order to determine if the effect of serum albumin was dose-dependent, transfected cells were incubated with solution containing saturating concentrations of N M D A and glycine and 3 p M zinc, along with different concentrations of serum 106 albumin. Figure 22 shows that serum albumin reduced the effect of zinc in a dose-dependent manner. Under control conditions (1 mM N M D A and 50 u M glycine only), -60 % of the transfected cells died, while the presence of Z n 2 + reduced the amount of cell death. Serum albumin caused a dose-dependent increase in the amount of cell death, such that incubation of cells with 40 mg / ml serum albumin increased N M D A toxicity to -38 % in the presence of 3 u M zinc. These results indicate that serum albumin could act to decrease the protective effect of zinc against N M D A toxicity. 107 Figure 21. Serum albumin inhibits the protective effect of zinc against N M D A toxicity. -40 hours following the start of the transfection, cells were incubated for 6 hours in physiological salt solution containing 1 mM N M D A , 50 p M glycine and varying concentrations of zinc. In parallel, some of the plates were incubated with the above solutions containing 4 mg / ml bovine serum albumin. The salt solution and the cells remaining on the dish were assayed for (3-gal activity. The percentage of cell death was determined at every concentration. Values are represented as mean ± S E M (n = 3-6 different transfections). The percentage of cell death was increased in the presence of serum albumin. * represents a significant difference from control by the paired t-test (p<0.05). 108 601 Percentage of P-gal expressing cells killed ro co ^ oi o> o o o o o o o Figure 2 2 . Albumin decreased the inhibitory effects of zinc in a dose-dependent manner. At -40 hours following the start of the transfection, cells were incubated for 6 hours in physiological salt solution containing 1 mM N M D A , 50 p M glycine, 3 p M zinc and varying concentrations of serum albumin. The salt solution and the cells remaining on the dish were then assayed for (3-gal activity. The amount of cell death was determined and the mean values ( ± SEM) from at least three different transfections are shown. Serum albumin only potentiated N M D A toxicity in the presence of zinc. * represents a significant difference from control by the paired t-test (p<0.05). 110 CHAPTER 7 Discussion 7.1 Characterization of N M D A toxicity in transfected HEK-293 cells The ability of HEK-293 cells to take up and express foreign plasmids has been well established. Previous reports have suggested that incubation of HEK-293 cells expressing functional N M D A receptors in medium containing 10% FBS causes selective death of transfected cells (Anegawa et al, 1995; Cik et al, 1994). This cell death was proposed to be due to the presence of excitatory amino acids in the serum; however, direct experimental evidence for this idea has not been shown. Also, these two groups assayed only total cell death (by trypan blue uptake) and therefore did not measure the extent of transfected cell death. Hence, in order to fully characterize this system, selective death of NR1 / NR2A-transfected HEK-293 cells was assayed under conditions of defined agonist concentration and exposure time. A six hour incubation period was selected as shorter periods did not produce sufficient cell death. By cell staining and counting methods, the EC50 for N M D A toxicity was shown to be -300 p M and maximal cell death was -65%. In these initial experiments, the percentage of transfected cells remaining attached to the dish in all experimental conditions was compared to the control plate from a sister culture of cells; hence, directly determining the amount of transfected cell death. By using the marker gene, lacZ, it was shown that the amount of cell death determined by staining for (3-gal expression was similar to the value obtained using anti-N R l specific antibody. Using calcium imaging, it has been shown that >85% of (3-gal positive HEK-293 cells co-transfected with LacZ and N M D A receptor subunits also 112 respond to N M D A (S. Duffy and T.H. Murphy, personal communication). Hence, staining for (3-gal expression gives a good estimate for the percentage of cells expressing N M D A receptors. Using this technique, it was noticed that even at the highest concentrations of the agonist (10 m M N M D A and 50 p M glycine), a significant percentage of the transfected cells still survived. The underlying reason for this resistance is unclear, but it is speculated that these cells may express the N M D A receptor subunits at relatively low levels, and hence, produce currents with relatively small amplitudes. Thus, the C a 2 + and Na + influx would be below the threshold for induction of cell death. Further experiments are required to test this hypothesis. In the staining assays, it was assumed that dead cells detach from the culture plate and only the cells that survive the N M D A exposure remain attached to the dish. In order to more accurately determine the amount of cell death, we developed an assay system that would eliminate human error inherent in counting cells and support our initial assumption regarding cell death. By comparing P-gal activity in the salt solution and the cells attached to the dish, the amount of cell death was estimated at every condition and normalized to the control condition from the same batch of cells. The EC50 for N M D A toxicity was -150 p M with maximum cell death at 60%. The sensitivity of N M D A -induced toxicity in our system is comparable to that observed in neurons following agonist exposure (Meldrum and Garthwaite, 1990). Incubation of transfected cells in medium containing 10% FBS, has been reported to cause -89% cell death (Raymond et al, 1996). However, the expected concentration of 113 excitatory amino acids in the serum (see Schramm et al, 1990) seem too low to cause such complete death of transfected cells, when compared to the N M D A dose-response. Two possible explanations are proposed for this discrepancy: 1) the medium may contain factors that potentiate toxicity, such as polyamines and serum albumin (Durand et al, 1993; Traynelis et al, 1993; Eimerl and Schramm, 1993); and 2) acute exposure of transfected cells to low concentrations of agonist could trigger events leading to a delayed type of cell death, similar to neuronal systems (Choi, 1988). Surprisingly incubation of transfected cells with glutamate at concentrations as high as 10 mM, did not result in significant cell death. Glutamate toxicity was only observed in the presence of 50 u M trans-PDC, a blocker of glutamate transport. Kidney cells have been shown to highly express the neuronal glutamate transporter (Kanai and Hediger, 1992). Also, rapid uptake of radiolabeled glutamate from the medium has been observed in HEK-293 cells by J. Rothstein (unpublished data). It is speculated that other amino acid analogs of glutamate which could activate NR1 / NR2A subunits but are not taken up by the transporter, underlie the cell death observed in incubation with medium (see Kanai and Hediger, 1992). Also, since the actual kinetics of glutamate uptake by the transporter are not known, the possibility that a sudden increase in glutamate concentration triggering events leading to delayed type cell death can not be ruled out. Further experiments are required in our system to determine whether a large percentage of NMDA-expressing cells are destined to die 24 hours following agonist exposure. Several reports have shown that the affinity of NR1 / NR2A is significantly higher for glutamate than for N M D A (Moriyoshi et al, 1991; Ishii et al, 1993; Laurie and 114 Seeburg, 1994). Hence it was expected that the EC50 for glutamate be lower than NMDA-induced cell death. In our study, no difference in the sensitivity of the receptor for glutamate and N M D A was observed. It is possible that higher concentration of trans-PDC are required in order to completely block glutamate uptake by the transporter. 7.2 Both sodium and calcium play a role in inducing cell death In order to assess the role of extracellular calcium and sodium in mediating cell death in this system, we used two different mutants of NR1 subunits with decreased calcium permeability. In chapter 3, it was shown that the survival of HEK-293 cells transfected with the mutant receptors was increased relative to the wild-type receptor, suggesting that calcium entry plays an important role in mediating cell death in our system. Although the conditions we used to assess cell death tend to favor Na + -dependent cell death in neurons, in our system -65% of cell death was calcium-dependent. The role of extracellular Na + in mediating cell death was further confirmed by substitution of N a + with N-methylglucamine. Incubation of cells transfected with NR1 / NR2A subunits of the N M D A receptors in low Na + solution containing different concentrations of the agonists, resulted in increased survival of the transfected cells relative to control. Incubation of cells transfected with the calcium impermeable N M D A receptor in this solution, did not show any significant cell death. Interestingly, CI" substitution experiments resulted in more cell death than Na + substitution. It is possible 115 that N-methylglucamine which was substituted for Na + , might have some blocking action on the receptor. Further experiments are required to confirm this idea. These results indicate that in this system both sodium and calcium play a role in inducing cell death. However, calcium influx seems to be the more important mediator of cell death. Further experiments are needed to determine whether short agonist exposure combined with delayed assessment will result in a more extensive cell death that is purely calcium-dependent. 7 . 3 Rapid desensitization of non-NMDA receptors is protective against agonist toxicity Results described in chapter 4 indicate that one possible explanation for the decreased susceptibility of neurons to over-exposure to non-NMDA receptor agonist is the fast desensitization of these receptors. Incubation of HEK-293 cells expressing either GluRl i or GluR6 with saturating concentrations of agonists did not result in significant cell death compared to control. However, in the presence of inhibitors of desensitization, cell death was observed. Also, incubation of GluRl j transfected cells with 1 mM kainate, which produces non-desensitizing currents, caused significant cell death compared to control. Apparently, the rapid desensitization of non-NMDA receptors does not allow sufficient sodium and calcium entry into the cell to activate the mechanism leading to cell death. These results are in agreement with previously published reports in neurons. AMPA-receptor mediated toxicity in cultured cerebellar Purkinje cells has been shown to be potentiated by cyclothiazide (Zorumski et al, 1990; May and Robinson, 1993). 116 Similarly, conA and cyclothiazide treatment of cultured hippocampal neurons has been shown to enhance non-NMDA toxicity (Moudy et al, 1994; Brorson et al, 1995). However, it must also be mentioned that several reports have shown non-NMDA toxicity in the absence of desensitization inhibitors (Choi, 1994; Weiss et al, 1994; Yin et al, 1994). It is speculated that the resistance observed in the transfected HEK-293 cells may depend on the glutamate receptor subunit composition, differences between HEK-293 cells and neurons in vulnerability to cell death, or on our experimental conditions. It has been shown that the receptors we tested, GluRl ; and GluR6, desensitize completely in response to saturating concentrations of agonists (Raymond et al, 1996). On the other hand, several groups have shown that some subtypes of A M P A receptors do not desensitize completely (Brorson et al, 1995). It is also possible that exposure to A M P A or kainate only activates the delayed type of cell death and little cell death is observed after 6 hours. Although this hypothesis is quite unlikely due to the decreased calcium permeability of non-NMDA receptors, the possibility cannot be ruled out. The relative calcium permeability of non-NMDA and N M D A subunit composition has been determined (Burnashev et al, 1992). It is of special interest to note that the calcium permeability of GluRl is comparable to the mutant N M D A receptor NR1 (N598Q) / NR2A and the calcium permeability of the edited GluR6 subunit is similar to the nearly calcium impermeable N M D A mutant, NR1 (N598R) / NR2A (Egebjerg and Heinemann, 1993). Not surprisingly the extent of cell death in the mutant N M D A receptors matches the corresponding non-NMDA receptors in the presence of inhibitors of desensitization (18 ± 3 % vs. 23 ± 4% for GluR6 and NR1 (N598R) / NR2A 117 and 28 ± 2 % vs. 35 + 0.1 % for GluRl and NR1 (N598Q) / NR2A). The slight difference observed might be due to the different current amplitudes of the receptor subtypes. Hence, the extent of agonist induced cell death in this system seemed to correlate well with the relative calcium permeability of these recombinant receptors. 7.4 Electrophysiological properties of N R 1 / N R 2 A and N R 1 / N R 2 B Patch-clamp recording in the whole-cell configuration was performed on HEK-293 cells expressing either NR1 / NR2A / (3-gal or NR1 / NR2B / (3-gal. Results in chapter 5 show that the current responses to a three second pulse of 100 p M glutamate in the presence of 50 p M glycine differ significantly for the two subunit combinations. Both receptors showed fast onset in response to the agonist but had different kinetics of desensitization, with NR1 / NR2A having a fast and a slow component of desensitization, while the NR1 / NR2B showed only a slow desensitization. These results are in agreement with previously published results (Monyer et al, 1992). Another difference between the two receptors was the shorter off-set decay time course observed following the three second pulse of glutamate NR1 / NR2A vs. NR1 / NR2B. This observation is in agreement with previously published results, except for the time constant for current off-set decay. Previous reports indicate TOFF'S of -120 and -400 ms for NR1 / NR2A and NR1 / NR2B, respectively (Monyer et al, 1992; Monyer et al 1994), while we observed TOFF'S of -270 ms and -930 ms for the two subunit combinations, respectively. In our laboratory, a wide range of values for T0ff have been observed. It is speculated that the larger values observed in my experiments are due to suboptimal positioning of the theta 118 tube, such that complete removal of glutamate is relatively slow. The offset decay time constant is thought to be important for the coincidence detection of pre- and postsynaptic activities (Bourne and Nicoll, 1993). Hence, the longer decay time in NR1 / NR2B could be important in detection of low synchronicity of presynaptic activities and postsynaptic depolarization. The current-voltage relationship for both receptor subtypes was linear with reversal potentials of -4 mV and the relative calcium permeability as estimated by C a 2 + vs. K + reversal potential was similar for both receptors. Interestingly, in our system the relative expression of NR1 / NR2B was much less than NR1 / NR2A. The current amplitude for NR1 / NR2B was -10 fold smaller than NR1 / NR2A. As discussed in chapter 5, it is speculated that the presence of sequence of base pairs in the 5' untranslated region of the NR2B gene inhibits translation of NR2B receptors. It has been shown that in order to obtain functional glutamate receptors in HEK-293 cells, expression of both NR1 and NR2 subunits are required (Mcllhinney et al, 1996). Hence, more work is needed to determine the sequence of the untranslated region of the gene causing the decreased expression. 7.5 Differential sensitivity of N M D A receptor subunits to zinc The effect of zinc on N M D A induced cell death was analyzed. Results in chapter 5 showed that zinc inhibits N M D A toxicity in NR1 / NR2A transfected cells in a non-competitive manner with an IC50 of -500 nM. Previous reports have shown similar results in cultured cortical and cerebellar granule cells. However, the IC50 reported for 119 the inhibition of glutamate toxicity, ranged from 500 nM in cerebellar granule cells to 80 p M in cortical cultures (Eimerl and Schramm, 1991; Koh and Choi, 1988). Using patch-clamp recordings, we showed that the IC50 for inhibition of peak current in cells transfected with NR1 / NR2A was similar to the value obtained in the toxicity assays, suggesting that the mechanism of action of zinc is simple blockage of the receptor. Since inhibition of the peak current mediated by NR1 / NR2A by extracellular zinc was not voltage-dependent, and did not interfere with the ionic permeability of the receptor, it is speculated that the binding site for zinc is not in the pore of the receptor-ion channel. A similar mechanism of inhibition by zinc was observed for NR1 / NR2B mediated currents, however, the IC50 was -20 fold larger for this subunit combination. The I C 5 0 for inhibition of the peak current was -10 pM, similar to the value reported for inhibition of peak currents and single channel currents in neuronal N M D A receptors (Christine and Choi, 1990; Legendre et al, 1990). In support of our results, previous reports have shown that NR1 / NR2A shows a higher affinity for certain N M D A receptor antagonists such as D-APV, than does NR1 / NR2B, while the reverse is true for N M D A receptor agonists (Laurie and Seeburg, 1994). Hence, NR1 / NR2A and NR1 / NR2B are termed antagonist-preferring and agonist-preferring, respectively. Another interesting effect of zinc on the whole cell current was the change in the rate of desensitization. The rate of the fast component of the desensitization of NR1 / NR2A current was increased, while the slower component was relatively unchanged. The rate of desensitization of currents in NR1 / NR2B transfected cells was also 120 increased. This increase in the desensitization rate was dose-dependent in both receptor subtypes, and it is speculated that it might play a role in modulating the receptor and hence, calcium entry into the cell. The net effect of zinc on neuronal N M D A toxicity is unclear. With increasing concentrations of zinc, a decrease in the ratio of the fast-desensitizing current was observed for NR1 / NR2A-transfected cells, such that at 10 u M zinc, only 12% of the current desensitized with the fast rate. Previous work had shown that zinc preferentially inhibits the larger conductance channels (50 pS) while, the smaller amplitude conductance channels (<25 pS) were relatively unaffected (Christine and Choi, 1990). It is possible that the proportion of the current desensitizing with the faster rate mainly constitutes the larger conductance channels, and hence, they are mainly inhibited. Single channel analysis is required to confirm this hypothesis. Hence, different subunit compositions of N M D A receptors have different sensitivities to zinc. The decreased affinity of NR1 / NR2B for zinc could result in decreased inhibition of the current during development, since this receptor combination is predominantly expressed before birth (Monyer et al, 1994). On the other hand, the increased sensitivity of NR1 / NR2A could play an important role in the inhibition of excitotoxic cell death in adulthood. A recent study by Lou-Vallano and colleagues (1994) suggest that cerebellar granule cells express different subunits of the N M D A receptors when cultured under different conditions. It is believed that alteration in C a 2 + influx through N M D A receptors may provide a feed-back system that regulates subunit expression. Incubation of cerebellar granule cells in solutions containing N M D A or 25 mM KC1 was shown to cause a rapid induction of NR2A and down-regulation of NR2B, 121 followed by a gradual induction of NR2C. Expression of NR1 / NR2A subunits could explain the low IC50 for zinc observed in these cells. Furthermore, our results indicate that even at very low concentrations, zinc could modulate the N M D A response in neurons expressing NR1 / NR2A subunit composition. 7.6 Serum albumin inhibits the protective effect of zinc against NMDA toxicity N M D A receptors have been shown to be modulated by a number of different factors. Various reports have shown that some of these factors are present in the culture medium and might potentiate glutamate toxicity in neurons and transfected cells. Eimerl and Schramm (1991, 1993) have shown that albumin, which is a major component of serum, potentiates glutamate toxicity in cerebellar granule cells in a dose-dependent manner. In chapter 6, experiments that contradict these results were described. Serum albumin at the highest concentration of 40 mg / ml did not significantly potentiate N M D A toxicity in transfected HEK-293 cells. However, the presence of serum albumin shifted the dose-response for zinc in a dose-dependent manner. At the highest concentration of serum albumin used, which corresponds to the concentration of albumin in blood, it was shown that there was a ~4 fold decrease in the inhibition of N M D A -2+ 2+ toxicity by zinc. It has been shown that serum albumin chelates Cu and Zn and is believed to be the major transporter of these ions in the body (Masuoka et al, 1993; Goumakos et al, 1991). It is speculated that the presence of serum albumin in the brain could potentiate excitotoxic cell death by chelating free zinc. These results suggest that 122 disruption of the blood-brain barrier following ischemia or trauma, could exacerbate the extent of excitotoxic neuronal death in part by albumin-mediated chelation of zinc. 7.7 Conclusion In this project we have developed a system where the biophysical properties of glutamate receptors can be directly correlated with their cytotoxic potential. I showed that both Na + and C a 2 + were responsible for mediating N M D A toxicity and that rapid desensitization of non-NMDA receptors is protective against agonist-induced toxicity. Inhibition of desensitization in non-NMDA receptors resulted in significant cell death which correlated well with the relative calcium permeability of the receptor. The effect of zinc on N M D A toxicity was also analyzed. Zinc inhibited N M D A toxicity in cells transfected with NR1 / NR2A with an IC50 -20 fold lower than previously reported for inhibition of N M D A currents in neurons. Using patch-clamp recordings, it was shown that NR1 / NR2A and NR1 / NR2B have differentia] sensitivities to zinc. These results suggest that even at very low concentrations, zinc may modulate N M D A receptors composed of NR1 / NR2A subunits. Also, the different sensitivity of the N M D A receptor subtypes to zinc could play a role in determining the susceptibility of cells to glutamate toxicity, such that neurons expressing the NR1 / NR2A subunit composition would be more resistant to glutamate toxicity. Several reports have indicated that the CA3 region of the hippocampus is relatively resistant to ischemic cell death. This region of the hippocampus has been shown to contain large amounts of vesicular zinc and is the only region in which release 123 of zinc in a calcium-dependent and voltage-dependent manner has been shown. It is possible that zinc release in this region underlies, at least in part, the relative resistance of CA3 to glutamate toxicity. More work is required to confirm this possibility. Finally, the effect of serum albumin was tested on N M D A toxicity, and it was shown that the protective effect of zinc against N M D A toxicity is inhibited in the presence of albumin. Changes in the blood-brain barrier observed in old age (Kleine at al, 1993) or under conditions of ischemia and trauma, could cause a large increase in the serum albumin concentration in the brain, leading to disruption in the modulatory effect of zinc. Expression of recombinant receptors in HEK-293 cells is a promising model system for further analysis of the molecular mechanisms underlying N M D A receptor-mediated cytotoxicity. This system is useful for further testing the role of other intracellular proteins, such as NOS or calcium binding proteins, in modulating glutamate induced excitotoxicity. Furthermore, the protective effects of various pharmacological agents can be readily tested. Our results suggests that N M D A receptor subtypes have different sensitivities to zinc which might be important in limiting glutamate-induced cell death. .124 Literature cited Albin, R.L., and Greenamyre, J.T. (1992). Alternative excitotoxic hypotheses. Neurology 42:733-738. Anegawa, N.J., Lynch, D.R., Verdoorn, T.A., and Pritchett, D.B. (1995). Transfection of N-methyl-D-aspartate receptors in a non-neuronal cell line leads to cell death. J. Neurochem. 64:2004-2012. Aniksztejn, L. , Charton, G., Ben-Ari, Y . (1987). Selective release of endogenous zinc from the hippocampal mossy fibers in situ. Brain Research. 404: 58-64. Ankarcrona, M . , Dypbukt, J .M., Bunfoco, E., Zhivotovsky, B. , Orrenius, S., Lipton, S.A., Nicotera, P. (1995) Glutamate-induced neuronal death: a succession of necrosis or apoptosis depending on mitochondrial function. Neuron. 15: 961-973. Ascher, P. and Nowak, L.(1986). C a 2 + permeability of the channels activated by N -methyl-D-aspartate in isolated mouse central neurons. J.Physiol. 377 : 35-45. Ascher, P. and Nowak, L . (1988). Quisqualate- and kainate-activated channels in mouse central neurones in culture. J. Physiol. 399:227-245. Assaf, S.Y. and Chung, S.H. (1984). Release of endogenous Z n 2 + from brain tissue during activity. Nature. 308 : 734-736. Baughler, J .M. and Hall, E.D.(1989). Central nervous system trauma and stroke. I. Biochemical consideration for oxygen radical formation and lipid peroxidation. J.Free.Radic.Bio.Med. 6:289-301. 125 Bernard, E.A. and Henley, J .M. (1990). The non-NMDA receptors: types, protein structure and molecular biology. TiPS. 11: 500-507. Bettler, B. , Boulter, J., Hermans-Borgmayer, I., O'Shea-Greenfield, A. , Deneris, E.S., Mol l , C , Borgmayer, U . , Hollmann, M . , Heineman, S. (1990). Cloning of the novel glutamate receptor subunit, GluR5: Expression in the nervous system during development. Neuron. 5:583-595. Bettler, B. , Egebjerg, J., Sharma, G., Pecht, G., Hermans-Borgmayer, I., Mol l , C , Stevens, C , Heineman, S. (1992). Cloning of the putative glutamate receptor: a low affinity kainate-binding subunit. Neuron. 8:257-265. Bindokas, V . P., and Miller, R.L. (1995). Excitotoxic degeneration is initiated at non-random sites in cultured rat cerebellar neurons. J. Neurosci. 15:6999-7011. Bourne, H.R., and Nicoll, R. (1993). Molecular machines integrate coincidence synaptic signals. Cell/Neuron. 72 / 10: 65-75. Brorson, J.R., Manzolillo, A. , and Miller, R J . (1994). C a 2 + entry via A M P A / K A receptors and excitotoxicity in cultured cerebellar Purkinje cells. J. Neurosci. 14:187-197. Brorson, J.R., Manzolillo, P.A., Gibbons, S.J., and Miller, R.J. (1995). A M P A receptor desensitization predicts the selective vulnerability of cerebellar Purkinje cells to excitotoxicity. J. Neurosci. 15:4515-4524. Burke, R.E., Baimbridge, K . G . (1993). Relative loss of striatal triosome compartment, defined by calbindin-D28K immunostaining, following developmental hypoxic-ischemia injury. Neuroscience. 56: 305-315. 126 Burnashev, N . , Schoepfer, R., Monyer, H. , Ruppersberg, J.P., Gunther, W., Seeburg, P.H., Sakmann, B . (1992a). Control by asparagine residues of calcium permeability and magnesium blockade in the N M D A receptor. Science 257:1415-1419. Burnashev, N . , Monyer, H. , Seeburg, P.H., and Sakmann, B . (1992b). Divalent ion permeability of A M P A receptor channels is dominated by the edited form of a single subunit. Neuron 8:189-198. Burnashev, N . , Khodorava, A. , Jonas, P., Helm, P.J., Wisden, W., Monyer, H. , Seeburg, P.H., Sakmann, B. (1992c). Calcium-permeable AMPA-kainate receptors in fusiform cerebellar glial cells. Science. 256: 1566-1570. Burnashev, N . (1993). Recombinant ionotropic glutamate receptors: functional distinctions imparted by different subunits. Cell Physiol. Biochem. 3: 318-331. Burnashev, N . , Zhou, Z., Neher, E., Sakmann, B. (1995). Fractional calcium currents through recombinant GluR channels of the N M D A , A M P A , and kainate receptor subtypes. J. Physl. 485: 403-418. Chan, P.H., Fishman, R.A., Longar, S., Chen, S., Yu, A . (1985). Cellular and molecular effects of polyunsaturated fatty acids in brain ischemia and injury. Prog.Brain Res.63: 227-235. Chazot, P.L., Cik, M . , Stephenson, F.A. (1992). Immunological detection of the NMDAR1 glutamate receptor subunit expressed in human embryonic kidney 293 cells and in rat brain. J. Neurochem. 59: 1176-1178. Chen, C. and Okayama, H . (1987). High-efficiency transformation of mammalian cells by plasmid D N A . Mol. Cell. Biol. 7:2745-2752. 127 Charton, G., Rovira, C., Ben-Ari, Y . , Leviel, V . (1985). Spontaneous and evoked release of endogenous Z n 2 + in the hippocampal mossy fiber zone of the rat in situ. Exp. Brain. Res. 58: 202-205. Chelebowski, J.F. and Coleman, J.E. (1976). Zinc and its role in enzymes. In Siegel H (ed.): "metal ions in biological systems" New York, Mercel Dekker, PI . Christine, C.W. and Choi, D. (1990). Effect of Zinc on N M D A receptor-mediated channel currents in cortical neurons. / . Neurosci. 10: 108-116. Choi, D.W. (1987). Ionic dependence of glutamate neurotoxicity. / . Neurosci. 7:369-379. Choi, D.W. (1988). Glutamate neurotoxicity and diseases of the nervous system. Neuron 1:623-634. Choi, D.W. and Hartley, D . M . (1993). Calcium and glutamate-induced cortical neuronal death. In molecular and cellular approaches to the treatment of neurological diseases, edited by S.G. Waxman. Raven Press. New York. Choi, D.W. (1994). Glutamate receptors and the induction of excitotoxic neuronal death. Prog. Brain Res. 100:47-51. Choi, D.W., Koh, J.-Y., and Peters, S. (1988). Pharmacology of glutamate neurotoxicity in cortical cell culture: attenuation by N M D A antagonists. / . Neurosci. 8:185-196. Chung, S.H. and Assaf, S.Y. (1984). Emerging role of zinc ions in neuronal transmission. In The neurobiology of zinc. Ala R. Liss, Inc., N .Y . 128 Cik, M . , Chazot, P.L., and Stephenson, F.A. (1993). Optimal expression of cloned N M D A R 1 / N M D A R 2 A heteromeric glutamate receptors: a biochemical characterization. Biochem. J. 296:877-883. Cik, M . , Chazot, P.L., and Stephenson, F.A. (1994). Expression of N M D A R l - l a (N598Q)/NMDAR2A receptors results in decreased cell mortality. Eur. J. Pharmacol. 266:R1-R3. Coyle, J.T., and Puttfarcken, P. (1993). Oxidative stress, glutamate, and neurodegenerative disorders. Science 262:689-695. Crawford, L L . and Conner, J.D.(1972). Zinc in maturing rat brain: hippocampal concentration and localization. J. Neurochem. 19: 1451-1458. Csernansky, C.A., Canzoniero, L.M.T. , Sensi, S.L., Yu, S.P., Choi, D.W. (1994). Delayed application of auritricarboxylic acid reduces glutamate-induced cortical neuronal injury. J. Neurosci. Resch. 38: 101-108. Cull-Candy, S.G. and Usowicz, M.M.(1987). Multiple conductance channels activated by excitatory amino acids in cerebellar neurons. Nature. 325:525-528. Cull-Candy, S.G., Howe, J., Ogden, D.C. (1987). Noise and single channel activated by extracellular amino acids in rat cerebellar granule neurons. J. Physiol. 400: 189-222. Curtis, D.R., Phillis, J.W., Watkins, J.C. (1959). Chemical excitation of spinal neurons. Nature. 183: 611-612. Curtis, D.R., Phillis, J.W., Watkins, J.C. (1960). Chemical excitation of spinal neurons by certain acidic amino acids. / . Physiol. (London). 150: 656-682. 129 Curtis, D.R. and Watkins, J.C. (1960). The excitation and depression of spinal neurons by structurally related amino acids. J. Neurochem. 6: 117-41 Dani, J.A. and Mayer, M . L . (1995). Structure and function of glutamate and nicotinic acetylcholine receptors. Curr. Opin. Neurobio. 5:310-317. Danscher, G., Shipley, M.T., Anderson, P.(1975). Persistent function of mossy fiber synapses after metal chelation with DEDTC. Brain Res. 85: 522-526. Dawson, V . L . , Dawson, T.M. , Bredt, D.S., Synader, S.H. (1991). NO mediates glutamate neurotoxicity in primary cortical cultures. Proc.Natl.Acad.Sci.USA 88:6368-6371. Dreosti, I.E. (1984). Zinc in the central nervous system: the emerging interaction. In the neurobiology of zinc. Alan R.Liss Inc. New York.pp 1-26. Dreyer, E.B., Zhang, D., Lipton, S.A. (1995). Transcriptional or translational inhibition blocks low dose N M D A mediated cell death. NeuroReport. 6: 942-944. Dubinsky, J .M. (1993). Intracellular calcium levels during the period of delayed excitotoxicity. J. Neurosci. 13: 623-631. Durand, G .M. , Gregor, P., Zheng, X . , Bennett, M . V . L . , Uhl, G.R., Zukin, R.S. (1992). Cloning of an apparent splice variant of the N-methyl-D-aspartate receptor NMDAR1 with altered sensitivity to poly amines and activators of protein kinase C. Proc. Natl. Acad. Sci. USA. 89: 9359-9363. Durand, G .M. , Bennett, M . V . L . , and Zukin, R.S. (1993). Splice variants of the N-methyl-D-aspartate receptor NR1 identify domains involved in regulation by poly amines and protein kinase C. Proc. Natl. Acad. Sci. USA 90:6731-6735. 130 Durkin, J.P., Tremblay, R., Buchan, A. , Blosser, J., Chakravarthy, B. , Mealing, G., Morley, P., Song, D. (1996). An early loss in membrane protein kinase C activity precedes the excitatory amino acid acid-induced death of primary cortical neurons. J. Neurochem. 66: 951-962. Dykens, J.A., Stern, A. , Trenkner, E. (1987). Mechanism of kainate toxicity to cerebellar neurons in vitro is analogous to perfusion tissue injury. J.Neurochem. 49: 1222-1228. Ebadi, M . and Pfieffer, R.F. (1984). Zinc in neurological disorders and in experimentally induced epileptiform seizures. In the neurobiology of zinc. Alan R.Liss Inc. New York.pp 1-26. Egebjerg, J., and Heinemann, S.F. (1993). Ca^+ permeability of unedited and edited versions of the kainate selective glutamate receptor GluR6. Proc. Natl. Acad. Sci. USA 90:755-759. Egebjerg, J., Bettler, B. , Hermans-Borgmeyer, I., and Heinemann, S. (1991). Cloning of cDNA for a glutamate receptor subunit activated by kainate but not A M P A . Nature 351:745-748. Ehlers, M.D. , Tingley, W.G., Huganir, R.L. (1995). Regulated subcellular distribution of the NR1 subunit of the N M D A receptor. Science. 269: 1734- 1737. Eimerl, S., and Schramm, M . (1991a). Acute glutamate toxicity in cultured cerebellar granule cells: agonist potency, effects of pH, Zn2+ and the potentiation by serum albumin. Brain Res. 560:282-290. Eimerl, S., and Schramm, M . (1991b). Acute glutamate toxicity and its potentiation by serum albumin are determined by the C a 2 + concentrations. Neuro. Lett. 130: 125-127. 131 Eimerl, S., and Schramm, M . (1992). An endogenous metal appear to regulate the N M D A receptor mediated 4 5 C a 2 + influx and toxicity in cultured cerebellar granule cells. Neuro. Lett. 137: 198-202. Eimerl, S., and Schramm, M . (1993). Potentiation of 4 5 C a 2 + uptake and acute toxicity mediated by the N-methyl-D-aspartate receptor: the effect of metal binding agents and transition metal ions. J. Neurochem. 61: 518-525. Evans, R.H., Evan, S.J., Pook, P.C., Sunter, D.C. (1987). A comparison of excitatory amino acid antagonists acting at different C-fibers and motor neurons of isolated spinal cord of the rat. Br. J.Pharmac. 91: 531-537. Frandsen, A. , and Schousboe, A. (1993). Excitatory amino acid-mediated cytotoxicity and calcium homeostasis in cultured neurons. 7. Neurochem. 60:1202-1211. Frederickson, C. (1989). Neurobiology of zinc and zinc-containing neurons. Int. Rev. Neurobio. 380: 145-238. Freund, T.F., Ylinen, A. , Miettinen, R., Pitkanen, A. , Lahtinen, H. , Baimbridge, K .G. , Riekkinen, P.J. (1991). Pattern of neuronal death in the rat hippocampus after status epilepticus. Relationship to calcium binding protein content and ischemic vulnerability. Brain Res. Bull. 28:27-38. Garthwaite, G., Hajos, F., and Garthwaite, J. (1986). Ionic requirements for neurotoxic effects of excitatory amino acid analogues in rat cerebellar slices. Neuroscience 18:437-447. Gasic, G.P., and Hollmann, M . (1992). Molecular neurobiology of glutamate receptors. Anna. Rev. Physiol. 54:507-536. 132 Goumakos, W., Loussac, J., Sarkar, B. (1991). Binding of cadmium(II) and zinc(II) to human and dog serum albumins. An equilibrium dialysis and 1 1 3 C d - N M R study. Biochem. Cell Biol. 69: 809-820. Hack, N . , Balazs, R. (1995). Properties of A M P A receptors expressed in rat cerebellar granule cell cultures: C a 2 + influx studies. J.Neurochem. 65: 1077-1084. Hamill, O.P., Marty, A. , Neher, E., Sakmann, B., and Sigworth, F.J. (1981). Improved patch-clamp techniques for high resolution current recording from cells and cell free membrane patches. Pflugers Arch. 391:85-100. Hartley, D . M . , Kurth, M.C. , Bjerkness, L. , Weiss, J.H., and Choi, D.W. (1993). Glutamate receptor-induced 45^ a2+ accumulation in cortical cell culture correlates with subsequent neuronal degeneration. J. Neurosci. 13:1993-2000. Haug, (1984). In " The Neurobiology of Zinc". Vol A. Edited by Fredrickson C.J., Howell, G.A. Allan R.Liss . New York. pp213-228. Hayashi, T. (1954). Effect of sodium glutamate on nervous system. Keio. J.Med. 3: 183-192 Herb, A. , Burnashev, N . , Werner, P., Sakmann, B., Wisden, W., Seeburg, P.H. (1992). The KA-2 subunit of the excitatory amino acid receptors shows wide spread expression in brain and forms ion channels with distantly related subunits. Neuron. 8:775-785. Hille, B. (1992). Ionic channels of excitable membranes. Sinauer Associates Inc., Sunderland, Massachusette. 133 Hollmann, M . , Maron, C , Heinemann, S. (1994). N-glycosylation site tagging suggests a three transmembrane domain topology for the glutamate receptor G l u R l . Neuron. 13: 1331-1343. Hollmann, M . , and Heinemann, S. (1994). Cloned glutamate receptors. Annu. Rev. Neurosci. 17:31-108. Hollmann, M . , Hartley, M . , and Heinemann, S. (1991). Ca^+ permeability of K A -AMPA-gated glutamate receptor channels depends on subunit composition. Science 252:851-853. Hollmann, M . , Boulter, J., Maron, C , Beasley, L. , Sulivan, J., Pecht, G., Heineman, S. (1993). Zinc potentiates agonist induced currents at certain splice variants of the N M D A receptors. Neuron. 10: 943-954. Howell, A .G. , Welch, M.G. , Frederickson, C.J. (1984). Stimulation-induced uptake and release of zinc in hippocampal slices. Nature. 308: 736-738. lino, M . , Ozawa, S., Tsuzuki, K. (1990). Permeation of calcium through excitatory amino acid receptor channels in cultured rat hippocampal neurons. J.Physiol. 424: 151-165. Ikeda, K., Nagasawa, H. , Mori, H. , Araki, K., Sakimura, K., Watatiabe, M . , Inoue, Y. , Mishina, M . (1992). Cloning and expression of the e4 subunit of the N M D A receptor channels. FEBS Lett. 313: 34-8. Ishii, T., Moriyoshi, K. , Sugihara, H., Sakurada, K., Kadotani, H. , Yokoi, M . , Akazawa, C , Shigemoto, R., Mizuno, N . , Masu, M . , and Nakanishi, S. (1993). Molecular characterization of the family of the N-methyl-D-aspartate receptor subunits. 7. Biol. Chem. 268:2836-2843. 134 Jahr, C.E. and Stevens, C.F. (1987). Glutamate activates multiple single channel conductance in hippocampal neurons. Nature. 325: 529-531. Johnson, J.W. and Ascher, P. (1987). Glycine potentiates the N M D A response in cultured mouse brain neurons. Nature. 325: 529-531. Kanai, Y. , and Hediger, M . A . (1992). Primary structure and functional characterization of a high-affinity glutamate transporter. Nature 360:467-471. Kashiwabushi, N . , Ikeda, K., Araki, K., Hirano, T., Shibuki, K., Takayama, C., Inoue, Y. , Kutsuwada, T., Yagi, T., Kang, Y. , Aizawa, S., Mishina, M . (1995). Impairment of motor coordination, purkinje cell synapse formation, and cerebellar long-term depression in GluR82 mutant mice. Cell. 81: 245-252. Keinanen, K. , Wisden, W., Sommer, B. , Werner, P., Herb, A. , Verdoorn, T.A., Sakmann, B., Seeburg, P.H. (1990). A family of AMPA-selective glutamate receptors. Science. 249: 556-560. Khulusi, S.S., Brown, M.W., Wright, D . M . (1986). Zinc and paired-pulse potentiation in the hippocampus. Brain. Res. 363: 152-155. Kochhar, A. , Zivin, J.A., Lyden, P.D., and Mazzarella, V. (1988). Glutamate antagonist therapy reduces neurologic deficits produced by focal central nervous system ischemia. Arch. Neurol. 45:148-153. Koh, J.Y, Choi, D.W. (1988).Vulnerability of cultured cortical neurons to damage by excitotoxinc: differential susceptibility of neurons containing NADPH-diaphorase. J.Neurosci. 8:2153-2163. 135 Koh, J.Y. and Choi, D.W. (1988).Zinc alters Excitatory amino acid neurotoxicity on cortical neurons. J.Neurosci. 8:2164-2171. Koh, J.Y., Choi, D.W. (1994). Zinc toxicity on cultured cortical neurons: involvement of N-methyl-D-aspartate receptors.Neurosci. 60: 1049-1057. Kohler, M . , Burnashev, N . , Sakmann, B., and Seeburg, P.H. (1993). Determinants of Ca2+ permeability in both TM1 and TM2 of high affinity kainate receptor channels: diversity by R N A editing. Neuron 10:491-500. Kowall, N.W., Ferrante, R.J., and Martin, J.B. (1987). Patterns of cell loss in Huntington's disease. Trends Neurosci. 10:24-29. Kutsuwada, T., Kashiwabuchi, N . , Mori, H. , Sakimura, K., Kushiya, E., Araki, K., Meguro, H. , Masaki, H. , Kumanishi, T., Arakawa, M . , and Mishina, M . (1992). Molecular diversity of the N M D A receptor channel. Nature 358:36-41. Lambolez, B. , Audinat, E., Bochert, P., Crepel, F., Rossier, J. (1992). A M P A receptor subunits expressed by purkinje cells. Neuron. 9: 247-258. Laurie, D.J. and Seeburg, P.H. (1994). Ligand affinities at recombinant N M D A receptors depend on subunit composition. Euro. J. Pharm. 268: 335-345. Le Bourdelles, B., Wafford, K . A . , Kemp, J.A., Marshall, G., Bain, C , Wilcox, S.A., Sikela, J .M., Whiting, P.J. (1994). Cloning, functional coexpression, and pharmacological characterization of human cDNAs encoding N M D A receptors NR1 and NR2A subunits./. Neurochem. 62: 2091-2098. Legendre, P., Westbrook, G.L. (1990). The inhibition of single N-methyl-D-aspartate-activated channels by zinc ions on cultured rat neurones. J. Phsiol. 429: 429-449. 136 Lomeli, H. , Spreglel, L. , Laurie, D.J., Kohr, G., Herb, A. , Seeburg, P., Wilsen, W., (1993). The rat delta-1 and delta-2 subunits extend the excitatory amino acid receptor family. FEBS lett. 315: 318-322. Lou Vallano, M . , Lambolez, B., Audinat, E., Rossier, J. (1996). Neuronal activity differentially regulates N M D A receptor subunit expression in cerebellar granule cells. / . Neurosci. 16: 631-639. Lucas, D.R. and Newhouse, J.P. (1957). The toxic effects of sodium L-glutamate on the inner layer of the retina. Arch Opthalmol. 58: 193-201. Martin, L.J. , Blackstone, C.D., Levey, A.I., Huganir, R.L., Price, D.L.(1993). A M P A glutamate receptor subunits are differentially distributed in rat brain. Neuroscience. 53: 327-358. May, P.C., and Robison, P .M. (1993). Cyclothiazide treatment unmasks A M P A excitotoxicity in rat primary hippocampal cultures. J. Neurochem. 60:1171-1174. Mayer, M.L . , and Westbrook G.L. (1987). The physiology of excitatory amino acids in the vertebrate nervous system. Prog. Neurobiol. 28:197-276. Mayer, M . , Vyklicky, L. , Westbrook, G.L. (1989). Modulation of excitatory amino acid receptors by group IIB metal cations in cultured mouse hippocampal neurons. J. Physiol. 415: 329-350. Masuoka, J., Saltman, P. (1994). Zinc (II) and copper (II) binding to serum albumin. A comparative study of dog, bovine, and human albumin. J. Biol. Chem. 269: 25557-25561. 137 Mcllhinney, R.A.J. , Molnar, E., Atack, J.R., Whiting, P.J. (1996). Cell surface expression of the human N-methyl-D-aspartate receptor subunit la requires the co-expression of the NR2A subunit in transfected cells . Neuroscience. 70: 989-997. Meldrum, B., and Garthwaite, J. (1990). Excitatory amino acid neurotoxicity and neurodegenerative disease. TiPS 11:379-387. Michaels, R.L., and Rothman, S.M. (1990). Glutamate neurotoxicity in vitro: antagonist pharmacology and intracellular calcium concentrations. J. Neurosci. 10:283-292. Mody, L, MacDonald, J.F. (1995). N M D A receptor-dependent excitotoxicity: the role of intracellular C a 2 + release. TiPS. 16: 356-359. Molnar, E., Baude, A. , Richmond, S.A., Patel, P.B., Somogyi, P., Mcllhinney, R.A. (1993). Biochemical and immunocytochemical characterization of antipeptide antibodies to a cloned GluRl glutamate receptor subunit: cellular and subcellular distribution in the rat forebrain. Neuroscience. 53: 307-326. Monaghan, D.T., Bridges, R.J., and Cotman, C.W. (1989). The excitatory amino acid receptors: their classes, pharmacology, and distinct properties in the function of the central nervous system. Annu. Rev. Pharmacol. Toxicol. 29:365-402. Monyer, H. , Seeburg, P.H., Wisden, W. (1991). Glutamate-operated channels: developmentally early and mature form arise by alternative splicing. Neuron. 6: 799-810. Monyer, PL, Sprengel, R., Schoepfer, R., Herb, A. , Higuchi, M . , Lomeli, H. , Burnashev, N . , Sakmann, B., Seeburg, P.H. (1992). Heteromeic N M D A receptors: molecular and functional distinction of subtypes. Science. 256: 1217-1221. 138 Monyer, H. , Burnashev, N . , Laurie, D.J., 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. Moriyoshi, K., Masu, M . , Ishii, T., Shigemoto, R., Mizuno, N . , Nakanishi, N . (1991). Molecular cloning and characterization of the rat N M D A receptor. Nature. 354: 31-37. Moudy, A . M . , Yamada, K .A . , and Rothman, S.M. (1994). Rapid desensitization determines the pharmacology of glutamate neurotoxicity. Neuropharmacology 33:953-962. Muller, T., Moller, T., Berger, T., Schnittzer, J., Kettenmann, H. (1992). Calcium entry through kainate receptors and resulting potassium channel blockade in Bergmann glial cells. Science. 256: 1563-1566. Nakanishi, S. (1992). Molecular diversity of glutamate receptors and implications for brain function. Science 258:597-603. Nowak, L. , Bregestovski, P.,Ascher, P., Herbert, A. , Pochiantz, M . (1984). Magnesium gates glutamate activated channels in mouse central neurons. Nature. 37: 462-465. Olney, J.W. (1978). In "kainic acid as a tool in neurobiology". Edited by McGeer, E.G., Onley, J.W., McGeer, P.J.. Raven press. 15-171. Orrenius, S., McConkey, D.J., Bellomo, G., and Nicotera, P. (1989). Role of C a 2 + in toxic cell killing. TiPS 10:281-285. 139 Park, C.K., Nehls, D.G., Graham, D.I., Teasdale, G.M. , and McCulloch, J. (1988). The glutamate antagonist MK-801 reduces focal ischemic brain damage in the rat. Ann. Neurol. 24:543-551. Partin, K . M . , Patneau, D.K., Winters, C.A., Mayer, M.L . , and Buonanno, A. (1993). Selective modulation of desensitization at A M P A versus kainate receptors by cyclothiazide and concanavalin A. Neuron 11:1069-1082. Patneu, D.K., Vyklicky, L. , Mayer, M . (1993). Hippocampal neurons exhibit cyclothiazide-sensitive rapidly desensitizing responses to kainate. J. Neurosci. 13: 3496-3509. Perez-clausell, J., Danscher, G. (1985). Intravesicular localization of zinc in rat telencephalic boutons. A histochemical study. Brain. Res. 337: 91-98. Peters, S., Koh, J., Choi, D.W. (1987). Zinc selectively blocks the action of N-methyl-D-aspartate on cortical neurons. Science. 236: 589-593. Portera-Cailliau, C , Price, D.L., Martin, L.J. (1996). N-methyl-D-aspartate receptor proteins NR2A and NR2B are differentially distributed in the developing rat central nervous system as revealed by subunit-specific antibodies. / . Neurochem. 66: 692-700. Randall, R.D., and Thayer, S.A. (1992). Glutamate-induced calcium transient triggers delayed calcium overload and neurotoxicity in rat hippocampal neurons. J. Neurosci. 12:1882-1895. Ransom, R., Stec, N.L . (1988). Cooperate modulation of [3H] MK-801 binding to the N -methyl-D-aspartate receptor-ion channel complex by L-glutamate, glycine, and polyamines. J.Neurochem. 51: 830-837. 140 Rassenderen, F., Lory, P., Pin, J., Nargeot, J. (1990) Zinc has opposite effects on N M D A and non-NMDA receptors expressed in Xenopus Oocytes. Neuron. 4: 733-740. Raymond, L .A . , Blackstone, C D . , and Huganir, R.L. (1993). Phosphorylation and modulation of recombinant GluR6 glutamate receptors by cAMP-dependent protein kinase. Afatare 361:637-641. Raymond, L .A . , Moshaver, A. , Tingley, W.G., Shalaby, I., Huganir, R.L. (1996). Glutamate receptor ion channel properties predict vulnerability to cytotoxicity in a transfected non-neuronal cell line. Mol. Cell. Neurosci. 7 : 102-115. Roche, K.W., Raymond, L .A. , Blackstone, C., and Huganir, R.L. (1994). Transmembrane topology of the glutamate receptor subunit GluR6. J. Biol. Chem. 269:11679-11682. Rothman, S.M., Olney, J.W. (1987) Excitotoxicity and the N M D A receptor. TINS, 10: 299-302. Rothman, S.M., Thurston, J.H., and Hauhart, R.E. (1987). Delayed neurotoxicity of excitatory amino acids in vitro. Neuroscience 22:471-480. Sagot, Y. , Dubois-Dauphin, M.,tan, S.A., de Bilbao, F., Aebischer, P., Martinou, J.C., Kato, A .C . (1995). Bcl2 overexpression prevents ,motorneuron cell body loss but not axonal degeneration in a mouse model of a neurodegenerative disease. J. Neurosci. 15: 7727-7733. Sakimura, K. , Morita, T., Kushiya, E., Mishina, M . (1992). Primary structure and expression of the fl subunit of the glutamate receptor channel selective for kainate. Neuron. 8: 261-21 A. 141 Sakurada, K., Masu, M . , and Nakanishi, S. (1993). Alteration of Ca^+ permeability and sensitivity to Mg^+ and channel blockers by a single amino acid substitution in the N-methyl-D-aspartate receptor. / . Biol. Chem. 268:410-415. Sansom, M.S.P. and Usherwood, P.N.R. (1990). Single channel studies of glutamate receptors. Int. Rev. Neurobio. 32: 51-106. Schanne, F .A.X. , Kane, A .B . , Young, E.E., and Farber, J.L. (1979). Calcium dependence of toxic cell death: a final common pathway. Science 206:700-702. Schramm, M . , Eimerl, S., and Costa, E. (1990). Serum and depolarizing agents cause acute neurotoxicity in cultured cerebellar granule cells: role of the glutamate receptor responsive to N-methyl-D-aspartate. Proc. Natl. Acad. Sci. USA 87:1193-1197. Schoepp, D.D. and Conn, P J . (1993). Metabotropic glutamate receptors in brain function and pathology. TiPS 14: 13-20. Seeburg, P.H. (1993). The molecular biology of mammalian glutamate receptor channels. Trends Neurosci. 16:359-365. Seeburg, P.H. (1996). The role of R N A editing in controlling glutamate receptor channel properties. J.Neurochem. 66:1-5. Sheardown, M.J . , Nielsen, E.O., Hansen, A.J. , Jacobsen, P., and Honore, T. (1990). 2,3-Dihydroxy-6-nitro-7-sulfamoyl-benzo(F)quinoxaline: a neuroprotectant for cerebral ischemia. Science 247:571-574. Shiriashi, K. , Nakazawa, S., Ito, H. (1993). Zinc enhances kainate neurotoxicity in the rat brain. Neur. Res. 15: 113-116. 142 Siegel, S.J., Janssen, W.G., Tullai, J.W., Rogers, S.W., Moran, T., Heinemann, S., Morrison, J.H. (1995). Distribution of the excitatory amino acid receptor subunits GluR2(4) in monkey hippocampus and colocalization with subunits GluR5-7 and N M D A R 1 . J.Neurosci. 15:2707-2719. Siman, R., Neszek, J.C., Kegerise, C. (1989). Calpain I activation is specifically related to excitatory amino acid induction of hippocampal damage. J. Neurosci. 9: 1579-1590. Simon, R.P., SwanJ.H., Griffiths, T., and Meldrum, B.S. (1984). Blockade of N-methyl-D-aspartate receptors may protect against ischemic damage in the brain. Science 226:850-852. Slomianka, L . (1992). Neurons of origin of zinc-containing pathways and the distribution of zinc-containing boutons in the hippocampal region of the rat. Neuroscience. 48: 325-352. Sommer, B. , Keinanen, K., Verdoorn, T.A., Wisden, W., Burnashev, N . , Herb, A. , Kohler, M . , Takagi, T., Sakmann, B., and Seeburg, P.H. (1990). Flip and flop: a cell-specific functional switch in glutamate-operated channels of the CNS. Science 249:1580-1585. Stanaert, D.G., Testa, C M . , Young, A.B. , Penney, J.B. (1994). Organization of N -methyl-D-aspartate glutamate receptor gene expression in the basal ganglia of the rat. J. Comp. Neurol. 343: 1-16. Stern, P., Behe, P., Schoepfer, R., and Colquhoun, D. (1992). Single-channel conductances of N M D A receptors expressed from cloned cDNAs: comparison with native receptors. Proc. R. Soc. Lond. B 250:271-277. 143 Sugihara, H. , Moriyoshi, K., Ishii, T., Masu, M . , and Nakanishi, S. (1992). Structures and properties of seven isoforms of the N M D A receptor generated by alternative splicing. Biochem. Biophys. Res. Commun. 185:826-832. Swenerton, H. , Shrader, E.R., Hurley, L.S. (1969) Zinc deficient embryos reduced thymidine incorporation. Science. 166: 1014-17. Tingley, W.G., Roche, K.W., Thompson, A .K . , and Huganir, R.L. (1993). Regulation of N M D A receptor phosphorylation by alternative splicing of the C-terminal domain. Nature 364:70-73. Traynelis, S.F., Hartley, M . , and Heinemann, S.F. (1995). Control of proton sensitivity of the N M D A receptor by R N A splicing and polyamines. Science 268:873-876. Tolle, T.R., Berthele, A. , Zieglgansberger, W., Seeburg, P.H., Wisden, W. (1993). The differential expression of the 16 N M D A and non-NMDA receptor subunits in the rat spinal cord and in the periaquadactal grey. J. Neurosci. 13:5009-5028. Tsuji, K. , Nakamura, Y. , Ogata, T., Shibata, T., Kataoka, K. (1995). Transient increase of cyclic A M P induced by glutamate in cultured neurons from rat spinal cord. J. Neurochem. 65: 1816-1822. Tymianski, M . , Wallace, M.C. , Spigelman, I., Uno, M . , Carlen, P.L., Tator, C.H., and Charlton, M.P. (1993). Cell-permeant Ca^+ chelators reduce early excitotoxic and ischemic neuronal injury in vitro and in vivo. Neuron 11:221-235. Vera-Gil, A., Perez-Castejon, M.C . (1994). 6 5 Z n in studies of the neurobiology of zinc. Histl Histopath. 9: 413-420. 144 Verdoorn, T.A., Burnashev, N . , Monyer, H. , Seeburg, P.H., Sakmann, B.(1991). Structural determinants of ion flow through recombinant glutamate receptor . channels. Science. 252: 1715-1718. Wang, L . Y . , Taverna, F.A., Huang, X.P. , MacDonald, J.F., Hampson, D.R. (1993). Phosphorylation and modulation of a kainate receptor (GluR6) by cAMP-dependent protein kinase. Science. 259: 1173-1175. Watkins, J.C. (1962). The synthesis acidic amino acids processing neuropharmocological activity. J.med.pharm.chem.16: 113-119. Weiss, J.H., Yin, H.-Z., and Choi, D.W. (1994a). Basal forebrain cholinergic neurons are selectively vulnerable to AMPA/kainate receptor-mediated neurotoxicity. Neuroscience 60:659-664. Weiss, J.H., Turetsky, D., Wilke, G., and Choi, D.W. (1994b). AMPA/kainate receptor-mediated damage to NADPH-diaphorase-containing neurons is Ca 2 + dependent. Neurosci. Lett. 167:93-96. Westbrook, G.L., Mayer, M . L . (1987). Micromolar concentrations of Zn + antagonize N M D A and G A B A response of hippocampal neurons. Nature. 328: 640-643. Westbrook, G.L.(1993). Glutamate receptors and Excitotoxicity. In molecular and cellular approaches to treatment of neurological diseases, edited by S.G. Waxman. P 35-50. Raven Press. New York. Westergaard, N . , Banke, T., Wahl, P., Sonnewald, U . , Schousboe, A . (1995). Citrate modulates the regulation by zinc of N-methyl-D-aspartate channel currents and neurotransmitter release. Proc. Natl. Acad. Sci. USA. 92: 3367-3370. 145 Wisden, W., Seeburg, P.H. (1993). Mammalian ionotropic glutamate receptors. Curr. Opin. Neurobio. 3: 291-8. Witt, M.R., Dekermendjian, K. , Frandsen, A. , Schousboe, A. , Nielsen, M . (1994). Complex correlation between excitatory amino acid induced increase in the intracellular C a 2 + concentration and subsequent loss of neuronal function in individual neocortical neurons in culture. Proc. Natl. Acad.Sci.USA. 91: 12303-12307. Wood, M.W., Vandongen, H.M.A. , Vandongen, A.M.J . (1995). Structural conservation of ion conduction pathways in K channels and glutamate receptors. Proc. Natl. Acad. Sci. USA. 92: 4882-4886. Yamazak, M . , Arai, K., Shibata, A. , Mishina, M . (1992). Molecular cloning of a cDNA encoding a novel member of the mouse glutamate receptor family. Biochem. Biophys. Res. Commun. 183: 886-892. Yin, H.-Z., Lindsay, A.D. , and Weiss, J.H. (1994). Kainate injury to cultured basal 2+ forebrain cholinergic neurons is Ca dependent. NeuroReport 5:1477-1480. Yin, H.Z. and Weiss, J.H. (1995). Z n 2 + permeates C a 2 + permeable A M P A / Kainate channels and triggers selective neuronal injury. NeuroReport. 6: 2553-2556. Zarei, M . M . and Dani, J.A. (1994). Ionic permeability characteristics of the N-methyl-D-aspartate receptor channel. J. Gen. Phyl.103: 231-248. Zarei, M . M . and Dani, J.A. (1995). Structural basis for explaining open channel blockade of the N M D A receptor. J. Neurosci. 15: 1446-1454. 146 Zhong, J., Carrozza, D.P., Williams, K., Prichett, D.B., Molinoff, P.B. (1995). Expression of mRNA encodes subunits of the N M D A receptor in developing rat brain. / . Neurochem. 64: 531-539. Zorumski, C.F., Thio, L .L . , Clark, G.D., and Clifford ,D.B. (1990). Blockade of desensitization augments quisqualate excitotoxicity in hippocampal neurons. Neuron 5:61-66. Zukin, R.S. and Bennett, M . V . L . (1995). Alternative spliced isoforms of the N M D A R 1 receptor subunit. TINS. 18: 306-313. 147 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0099031/manifest

Comment

Related Items