Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Astrocytes release small molecules to protect neurons from oxidative stress Wang, Xue-Feng 2001

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

Item Metadata


831-ubc_2001-71554X.pdf [ 6.93MB ]
JSON: 831-1.0090451.json
JSON-LD: 831-1.0090451-ld.json
RDF/XML (Pretty): 831-1.0090451-rdf.xml
RDF/JSON: 831-1.0090451-rdf.json
Turtle: 831-1.0090451-turtle.txt
N-Triples: 831-1.0090451-rdf-ntriples.txt
Original Record: 831-1.0090451-source.json
Full Text

Full Text

A S T R O C Y T E S R E L E A S E S M A L L M O L E C U L E S T O P R O T E C T NEURONS F R O M O X I D A T I V E STRESS By Xue-Feng Wang M.D., Beijing Medical University, 1987 M . S c , Chinese Academy o f Medical Sciences, 1990 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY In The Faculty of Graduate Studies Graduate Program In Neuroscience We accept this thesis as conforming to the required standard The University o f British Columbia September 2001 © Xue-Feng Wang, 2001 U B C Special Collections - Thesis Authorisation Form Page 1 of 1 I n p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f t h e r e q u i r e m e n t s f o r an advanced degree a t t h e U n i v e r s i t y o f B r i t i s h C olumbia, I agree t h a t t h e L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and s t u d y . I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y p u rposes may be g r a n t e d by t h e head o f my department o r by h i s o r her r e p r e s e n t a t i v e s . I t i s u n d e r s t o o d t h a t c o p y i n g o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n . ) The U n i v e r s i t y o f B r i t i s h C olumbia Vancouver, Canada 16/09/2001 ABSTRACT The aim of the present experiments is to explore the mechanisms by which astrocytes may protect neurons from oxidative stress with a focus on small molecules. A non-contact astrocyte-neuron co-culturing method was first developed. It was observed that neuronal survival was promoted substantially when neurons were co-cultured with a confluent astrocyte feeder layer. The mechanisms by which a stable cysteine level is maintained by astrocytes have been explored in recent years. However, it is still unclear how cysteine is released. In my experiments, cysteine, glutathione, and related compounds in astrocyte conditioned medium were analyzed with HPLC methods. My data suggest that glutathione is released by astrocytes directly and that cysteine is generated from the extracellular thiol/disulfide exchange reaction of cystine and glutathione. Neuron conditioned medium, CSF, and plasma of the carotid artery and internal jugular vein were also analyzed. The results indicate that cystine, rather than cysteine, is transported from blood to the CNS and that the thiol/disulfide exchange reaction occurs in the brain in vivo. Though cysteine plays a critical role in regulating intracellular levels of glutathione, its cytotoxicity has also been noted, particularly in neurons. In the experiments of this thesis, I found that copper substantially accelerates the autoxidation rate of cysteine even at submicromolar levels, whereas iron and other transition metal ions, including manganese, chromium, and zinc, are less efficient. In tissue culture tests, it was found that cysteine toxicity depends highly on its autoxidation rate and on the total amount of cysteine being oxidized, suggesting that the toxicity can be attributed to the free radicals produced from cysteine autoxidation, but not to cysteine itself. The in vivo mechanisms that protect against cysteine toxicity were further explored. Catalase and pyruvate were each found to inhibit the production of hydroxyl radicals generated by cysteine autoxidation. In tissue culture, they both protected primary neurons against cysteine toxicity catalyzed by copper. Pyruvate, but not catalase or glutathione peroxidase, was detected in astrocyte conditioned medium and CSF. Herein, a concerted mechanism of neuroprotection by astrocytes is outlined. TABLE OF CONTENTS ABSTRACT » TABLE OF CONTENTS »» LIST OF TABLES vi LIST OF FIGURES vii ABBREVIATIONS ix ACKNOWLEDGEMENTS xii Chapter 1 GENERAL INTRODUCTION 1 1. Oxygen and Our Life 1 2. Oxidative Stress in the CNS and the Vulnerability of Neurons 2 3. Free Radical Chemistry 3 4. Multiple Roles of Astrocytes 10 5. Antioxidative Systems of the Organisms 14 6. Theories of Thiol Provision by Astrocytes 17 7. Theories of Cysteine Toxicity 22 8. Neurological Diseases and Oxidative Stress 25 9. Experimental Objectives 29 Chapter 2 MATERIALS AND METHODS 32 Materials 32 1. Animals 32 2. Chemicals and Reagents 32 3. Equipment 33 Methods 33 1. Preparation of Solutions and Media 33 2. Tissue Culture 35 3. Detection of Neuronal Death and Apoptosis 39 4. HPLC Assay of Thiols, Disulfides, and Related Compounds 41 5. Spectrophotometric Assay of Cysteine with Ellman's Reagent (DTNB) 42 6. Immunohistochemical Staining 43 7. Enzymatic Assays 43 8. Procedures of Obtaining CSF and Plasma 45 9. Fluorimetric Assay of Hydroxyl Radicals 45 10. Pyruvate Assay 46 11. Stati stical Analysis 46 Chapter 3 EFFECTS OF ASTROCYTES ON NEURONAL SURVIVAL AND ATTACHMENT SHOWN IN A CO-CULTURE SYSTEM 47 Introduction 47 Results 49 1. Effects of Astrocytes on Neuronal Survival 49 2. Neuronal Apoptosis Induced by Deprivation of Astrocyte Feeder Layer 51 3. Effects of Astrocytes on Neuronal Attachment 52 Discussion 54 Chapter 4 PROVISION OF CYSTEINE BY ASTROCYTES TO NEURONS BY WAY OF RELEASING GLUTATHIONE 59 Introduction 59 Results 60 1. Cysteine-glutathione Disulfide (CSSG) Found in Cystine-containing Astrocyte Conditioned Medium 60 2. GSH Released by Astrocytes in Cystine-free Medium 63 3. Analysis of Cysteine and Related Compounds in CSF and in Plasma of Artery and Vein 64 4. Thiols not Found in Neuron Conditioned Medium 65 5. Contents of GSH and Related Compounds in Astrocytes and Neurons 66 6. Autoxidation of Cysteine and GSH 67 Discussion 67 Chapter 5 NEUROTOXICITY INDUCED BY CYSTEINE AUTOXIDATION AND THE CENTRAL ROLE OF COPPER 73 Introduction 73 Results... 73 1. Cysteine Autoxidation in the Presence of Transition Metal Ions 74 2. Cysteine Autoxidation in CSF 77 3. Neurotoxic Effects of Cysteine 78 4. Generation of Hydroxyl Radicals During Cysteine Autoxidation 80 Discussion..... 80 Chapter 6 NEUROPROTECTIVE EFFECTS OF PYRUVATE ON CYSTEINE TOXICITY AND THE RELEASE OF PYRUVATE BY ASTROCYTES 85 Introduction 85 Results 86 1. Effects of Catalase and Pyruvate on Cysteine Toxicity 86 2. Effects of Catalase and Pyruvate on Generation of Hydroxyl Radicals 86 3. Pyruvate Released by Astrocytes 88 4. Effects of Cysteine Provision on Neuronal Survival 89 Discussion 90 Chapter 7 GENERAL DISCUSSION AND CONCLUSIONS 94 General discussion 94 1. Thiol Sources in the CNS 94 2. Mechanisms of Cysteine Toxicity 95 3. Free Radical Sources in the CNS 96 iv 4. Antioxidative Mechanisms of Astrocytes 97 5. Significance of the Present Findings 98 6. Perspectives 100 Conclusions 101 R E F E R E N C E S 103 v LIST OF TABLES Table 1-1. Half-lives of ROS and Related Species 4 Table 1-2. Standard Redox Potentials 9 Table 1-3. Observations of Neuroprotective Effects of Astrocytes 14 Table 2-1. HPLC Retention Times of the Standard Compounds 41 Table 3-1. Effects of Coating Methods on Neurite Outgrowth and Cell Viability 54 Table 4-1. Content of Cysteine and Related Compounds in A C M Made from Cystine-Free Medium 63 Table 4-2. Content of Cysteine and Related Compounds in CSF, artery and vein 65 Table 4-3. Content of Thiols, Disulfides and Acidic Amino Acids in Astrocytes and Neurons 66 Table 6-1. Pyruvate Content in A C M , N C M and CSF 89 vi LIST OF FIGURES Figure 1-1. Atmospheric Oxygen and the Course of Evolution 2 Figure,l-2. Free Energy Change of Stepwise Reduction of Oxygen 10 Figure 1-3. Molecular Structures of Glutathione and Related Compounds 18 Figure 1-4. GSH Synthesis and Some of Reactions of GSH 20 Figure 2-1. Micrographs of Subcultured Astrocytes 35 Figure 2-2. A Side View of the Glial-Neuron Non-Contact Co-Culture Situation 38 Figure 2-3. Schematic Mechanism of Assay of Glutathione Peroxidase 44 Figure 3-1. Effects of Astrocyte Feeder Layers on Neuronal Viability 49 Figure 3-2. Micrographs of the Co-Cultured Neurons : 50 Figure 3-3. Neuronal Death Induced by Glial Cell Deprivation 51 Figure 3-4. Morphological Characterization and TUNEL Staining of Apoptotic Neurons 53 Figure 3-5. Nuclear Morphology of Apoptotic Neurons 53 Figure 3-6. Micrographs of Cultured Neurons under Different Coating Conditions 55 Figure 4-1. HPLC Graphs of Cysteine and Related Compounds 61 Figure 4-2. Time Course of Cysteine and Related Compounds' Content in A C M 62 Figure 4-3. Time Course of Cysteine and CSSG Content in N C M 66 Figure 4-4. Autoxidation of Cysteine and GSH 68 Figure 4-5. A Model Showing How Astrocytes may Provide Cysteine to Neurons ....... 69 Figure 5-1. The Factors Influencing Cysteine Autoxidation 75 Figure 5-2. Comparisons of Cysteine Autoxidation in the Presence of Several Transition Metal Ions 76 Figure 5-3. Cysteine Autoxidation Rates in CSF 77 Figure 5-4. Neurotoxic Effects of Cysteine in the Presence of Copper 78 Figure 5-5. Effect of Total Amount of Cysteine Being Oxidized on Neuronal Survival 79 Figure 5-6. Generation of Hydroxyl Radical by the Autoxidation of Cysteine and Other Thiols 81 Figure 6-1. Protective Effects of Catalase and Pyruvate on Cysteine Neurotoxicity 87 vii Figure 6-2. Effects of Catalase and Pyruvate on the Generation of -OH from Cysteine Autoxidation 88 Figure 6-3. Time Course of Pyruvate Release by Astrocytes 90 Figure 6-4. Effect of Cysteine Supplement on Neuronal Survival 91 Figure 6-5. Diagram of the Proposed Mechanism of Protection by Astrocytes 93 viii ABBREVIATIONS •o2- superoxide radical •OH hydroxyl radical 2-ME 2-mercaptoethanol 7-OHCCA 7-hydroxylcoumarin-3 -carboxylic acid AP P-amyloid peptide A C M astrocyte conditioned medium A D Alzheimer's disease A L S amyotrophic lateral sclerosis A N O V A analysis of variance ApoE Apolipoprotein E APP amyloid precursor protein ASC system Alanine-Serine-Cysteine transport system B B B blood-brain barrier B M A A P-methylamino-L-alanine B O A A P-oxaloamino-L-alanine C A cysteic acid C C A coumarin-3-carboxylic acid C D - M E M chemically defined M E M CNS central nervous system CSA cysteine sulfinic acid CSF cerebrospinal fluid CSH cysteine esse cystine CSSG cysteine-glutathione disulfide CysGly cysteine-glycine dipeptide D A B 3,3'-diaminobenzidine tetrahydrochloride DIV days in vitro DTNB 5,5'-dithiobis(2-nitrobenzoic acid) DTT dithiothreitol EBSS Earle's balanced salt solution FBS fetal bovine serum FDNB l-fluoro-2,4-dinitrobenzene GFAP glial fibrillary acidic protein GPx glutathione peroxidase GSH glutathione GSSG glutathione-disulfide H2O2 hydrogen peroxide HBSS Hank's balanced salt solution HPLC high-performance liquid chromatography HSH homocysteine L D H lactic dehydrogenase M E M minimum essential medium MTT 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide N A C N-acetyl-cysteine N C M neuron conditioned medium NSE neuronal specific enolase PBS phosphate buffered solution PD Parkinson's disease P L L poly-L-lysine ROS reactive oxygen species SOD superoxide dismutase TBARS thiobarbituric acid reactive species TBI traumatic brain injury TBS tris buffered solution t-BuOOH ter/-butyl hydroperoxide TUNEL terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end labeling P-NADH P-nicotinamide adenine dinucleotide reduced form p-NADPH P-nicotinamide adenine dinucleotide phosphate reduced form y-GT y-glutamyl-transpeptidase ACKNOWLEDGEMENTS I wish to express my deep gratitude to my supervisor, Dr. Max Cynader, for his support, guidance, understanding and tolerance throughout. His encouragement always gives me confidence to step further ahead. I also enjoy the inspiring and pleasant discussion with him. I owe a great debt of gratitude to Joan Martin, not only for her generous provision of the HPLC equipment at the crossroads of my experiments, but also for her willingness to help me in every way that she could. My thanks extend to Dr. John Phillips for his generosity and kindness. I will never forget my first tissue culture experiment, guided by Dr. Yihong Wang. She helped me to kick off my research project with great patience and time. I wish her all the best in future. I sincerely thank my supervisory committee members, Dr. Vanessa Auld, Dr. Tim Murphy and Dr. Wolfram Tetzlaff, for their constructive criticism, helpful discussion and valuable suggestions. It is impossible for me to list all of the help which I have received from the members of my department. To merely mention their names, I am very grateful to Drs. Joanne Matsubara, Qiang Gu, William Jia, Shiv Prasad and Jing Cui. I am indebted to Virginia Booth, Howard Meadows, Zheng Chen, Bing Zhu, Lilian Luo, Tara Stewart, Christian Wong and Eugene Au. Thanks also go to Faye Pileberg, Swarni Sunner, Chris Crossfield and Barry Gibbs. I must thank Mr. Spencer Kong of the BC Cancer Agency for the use of the luminescence spectrometer. I thank Dr. Steven Vincent for being on my final doctoral exam committee and Dr. Kenneth Baimbridge for being on my comprehensive exam committee. I appreciate their knowledge and help. I thank Liz Wong for her warm-hearted secretary assistance throughout. Finally, I thank my wife Ying and my daughter Kelly for their boundless support, and for sharing my anxieties, pains and joys. Chapter 1 Introduction C H A P T E R 1 G E N E R A L INTRODUCTION 1. Oxygen and Our Life We cannot live without oxygen (O2), whose atmospheric levels are rather stable (21%). Life on the earth, however, was not born under aerobic conditions. Isotope labeling studies suggest the first living cells were born about 3.8 x 109 years ago (Mojzsis et al., 1996). O2 was absent from the atmosphere at that time. It is not until about 3 x 109 years ago that O2 was released to the air by the appearance of the photosynthetic cells, cyanobacteria (or blue-green algae) (Nunn, 1997; Xiong et al., 2000). The increase in atmospheric O2 was very slow at first because 0 2 produced by early cyanobacteria was almost completely utilized by converting Fe(II) to Fe(III), causing the precipitation of huge amounts of ferric oxides. After exhaustion of ferrous ion, 0 2 levels began to rise, and reached their current levels about 1 x 109 years ago (Fig. 1-1). With the availability of O2, some bacteria evolved the ability to respire, i.e. making use of the high redox potential O2 to oxidize completely the food they ingested for the synthesis of ATP. Mitochondria appeared when primitive eucaryotic cells endocytosed such respiration-dependent bacteria and formed a symbiotic relationship about 1.5 x 109 years ago. They provided the cells highly efficient utilization of chemical fuels. Oxidation of each molecule of glucose by O2 can produce 30 molecules of ATP in mitochondria, compared with only 2 molecules of ATP by anaerobic glycolysis. Today, mitochondria have become the major cell energy-converting organelles, and 0 2 has become indispensable for life in virtually all eucaryotes (Alberts et al., 1994; Rosen et al., 1999). However, O2 also has been shown to be toxic to living cells. The toxicity arises from oxidative damage to biologically important macromolecules (Gilbert, 1981). Over the course of evolution, organisms have developed various protective mechanisms against oxidation toxicity. Life is a dynamic balance between challenge and adaptation. Organisms have evolved such a complicated and efficient antioxidative system in order to survive. However, our antioxidative systems are not perfect. Mammals, along with other organisms, are still bothered everyday by numerous diseases closely related with reactive oxidative species (ROS) damage. Oxidative stress has been found to be involved in 1 Chapter 1 Introduction PUFAs are particularly easy targets of free radical attack, and facilitate the initiation and propagation of the free radical chain reaction (Aust, 1987; Gutteridge, 1987). This process leads to lipid peroxidation, which is a common and significant contributor to oxidative stress and tissue damage in vivo (Mattson, 1998). c). Neurons are excitable cells. Glutamate is the major excitatory neurotransmitter. Excitotoxicity of glutamate has been found to be mediated by massive C a 2 + influx through the N M D A receptors (Choi, 1987; Michaels and Rothman, 1990; Tymianski et al., 1993). Further evidence supports the notion that elevated [Ca2+]j induces sustained mitochondrial depolarization and collapses mitochondrial membrane potential (Ai|/) (Schinder et al., 1996; White and Reynolds, 1996; Brustovetsky and Dubinsky, 2000). Impairment of mitochondrial function increases the generation of free radicals, and is a primary event in glutamate neurotoxicity (Coyle and Puttfarcken, 1993; Castilho et al., 1998; Luetjens et al., 2000). d). The antioxidative capacity of neurons is weak. The activities of antioxidative enzymes, such as glutathione transferase and y-glutamylcysteine synthetase, were lower in cultured neurons than in cultured astrocytes (Makar et al., 1994). Peroxide detoxification ability is less efficient in neurons than in astrocytes (Dringen et al., 1999b). Intracellular glutathione (GSH) levels were lower in neurons than in astrocytes (Raps et al., 1989; Sagara et al., 1993). Taken together, these lines of evidence suggest that neurons of the CNS face a great oxidative stress and are vulnerable to free radical injury. 3. Free Radical Chemistry A free radical is defined as a molecule or a chemical species that has one or more unpaired electrons. Each orbital within atoms or molecules can hold two electrons, which spin in opposite directions. If an orbital contains only one electron, that electron is said to be unpaired. This broad definition of free radical encompasses hydrogen atom (H), most transition metals, oxygen molecule (O2), superoxide anion (O2"), hydroxyl radical (OH), nitric oxide ONO), nitric dioxide ONO2), alkoxyl radical (RO), peroxyl radical (ROO), semiquinone radical (HQ-), cysteinyl radical (CS-), carbon radical (L-), tocopherol radical (TocO), ascorbyl radical (AscO), etc. Though this vast variety of chemical species can all be called free radicals by definition, their reactivity and stability differ greatly. Among them, the oxygen-centered partial reduced oxygen products, such 3 Chapter 1 Introduction as O2", OH, RO-, and ROO- etc., are the most reactive species, and are implicated in various physiological and pathological processes (Halliwell and Gutteridge, 1984; Symons, 1991; Roberfroid and Calderon, 1995). Other species are either relatively stable or less involved in tissue injury. On the other hand, some important species in oxidative stress, such as hydrogen peroxide (H 2 0 2 ) , lipid peroxide (ROOH) and peroxynitrite (ONOO), do not belong to free radicals by definition. They do not have unpaired electrons, and generally have longer half-lives (Pryor, 1986; Borg, 1993). However, they are chemically reactive, and participate in many important reduction-oxidation reactions. The term of reactive oxidative species (ROS) encompasses these non-radical oxidants and those partial reduced oxygen radicals. Even this term is still not good enough to include all the biologically important species, such as carbon-centered radical (L ) and thiol-centered radical (RS-). The term "ROS and related species", therefore, are often used to indicate all the species, though mainly oxygen radicals, which have the potential to cause radical reaction to occur in biological system. Here, I alternately use free radicals, ROS, or oxygen radicals, to implicate the same species without distinction in particular. The chemical characteristic of free radicals and related species is their high reactivity with other molecules. As free radicals have unpaired electrons, they tend to lose or take electrons from other molecules to become paired. Non-radical oxidants, such as H2O2, are electrophilic as well, and tend to abstract electrons. Reactivity of free radicals and related species is different from species to species. The half-life of these species is one of the indexes reflecting the reactivity of free radicals (Table 1-1). Table 1-1. Estimate of the half-lives of ROS and related species (modified from Pryor, 1986) Species Half-life Substrate3 •OH 10"9 sec 1 M Lenoleate RO- 10"6 sec 100 mM Lenoleate ROO- 10 sec 1 m M Lenoleate L- 10"8 sec 0.02 m M 0 2 4 Chapter 1 Introduction H 2 0 2 Stable •02" b "Substrate chosen as representative of typical reactive target molecules for ROS and related species. bHalf-life varies according to different substrates and enzymes. O2 and partial reduced oxygen intermediates play the central role in oxidative stress. The ground-state O2, designated as 32 , g", has two unpaired electrons. Its electronic structure is as follows: (a Is)2 (o* Is)2 (a 2s)2 (a* 2s)2 (a 2p z) 2 (Tt 2p x) 2 (n 2p y) 2 (71* 2px)' (71* 2py)' The two unpaired electrons spin in parallel. This characteristic of O2 makes it hard to accept a pair of electrons in an orbital, which spin in opposite, from another molecule at a time (Kanfer and Turro, 1981). Instead, O2 prefers to accept one electron at a time. The stepwise reduction of O2 results in the formation of -02, H2O2, OH, and H2O in sequence, with addition of total four electrons (Halliwell and Gutteridge, 1984; Ingraham and Meyer, 1985): 0 2 -> 02" -> H 2 0 2 -> -OH -> H 2 0 . Peroxynitrite (ONOO") is another important ROS causing tissue injury. It is formed from the reaction of nitric oxide NO and O2": -NO + 0 2 " —» ONOO". NO is synthesized from L-arginine under the catalysis of NO synthetase (NOS) (Bredt et al., 1991; Vincent and Hope, 1992). Unlike the weak reactivity of NO and -02", ONOO" is a potent oxidant and nitrating agent capable of modifying proteins, lipids and D N A (Bartosz, 1996; Demiryurek et al., 1998). Evidence has shown that ONOO" is involved in the initiation of neurodegenerative diseases (Smith et al., 1997; Torreilles et al., 1999). Lipid peroxidation is one of the major oxidative injuries. PUFAs (e.g., linoleic, linolenic, and arachidonic acids) are readily attacked by free radicals, eventually forming lipid peroxides. The whole process can be described as the following steps (Gutteridge, 1987; Borg, 1993): (1) Initiation: a free radical, such as OH, can initiate the process of chain reaction by abstracting a hydrogen atom (H) from a methylene carbon of a PUFA. L H + OH -> L- + H 2 0 5 Chapter 1 Introduction (2) Propagation: the resulting carbon radical (L-) reacts rapidly with O2 to form a peroxy radical (LOO). L- + 0 2 -» LOO-LOO- can further abstract an H- from another PUFA, leaving an L- and lipid peroxide (LOOH). L H + LOO- -> LOOH + L-L- can either abstract an H- from a L H or react with O2 to give a LOO-, and reinitiate the process, producing more lipid peroxides. (3) Termination: the chain reaction will propagate until two free radicals react with each other to terminate the chain. L- + L- -> L L LOO- + LOO- -> LOOL + 0 2 L- + LOO- -> LOOL Another form of termination is the reaction of free radicals with antioxidants, such as Vitamin E (tocopherol). This process is also called interception. L- + TocOH -> L H + TocO-LOO- + TocOH -> LOOH + TocO-TocO- is a stable and low reactive radical. It can be recycled to TocOH by other antioxidants (Acworth et al., 1997). The produced lipid peroxides can break down into a great diversity of aldehydes in biological system. The most intensively studied aldehydes are malonaldehyde (MDA), and 4-hydroxyalkenal, such as 4-hydroxynonenal (HNE) and 4-hydroxyhexenal. These aldehydes, unlike free radicals, have long half-lives (minutes to hours) and can diffuse to relatively distant sites. They are the chemically active end products of lipid peroxidation, and can form adducts with proteins by binding covalently to cysteine, lysine and histidine residues via Michael addition and Schiff-base crosslink (Nair et al., 1986; Esterbauer et al., 1991; Uchida and Stadtman, 1992), altering protein functions. These aldehydes are cytotoxic (Esterbauer, 1993; Kruman et al., 1997; Mark et al., 1997). DNA damage by free radicals has long been considered as an important contributor to aging and aging-related diseases, such as cancer and neurodegeneration (Ames and 6 Chapter 1 Introduction Shigenaga, 1992; Mecocci et al., 1993). The major damaging free radical was found to be OH, as neither O2" nor H2O2 causes any modifications in D N A strand i f transition metal ions are completely removed (Rowley and Halliwell, 1983; Mello Filho and Meneghini, 1984; Aruoma et al., 1989a; Aruoma et al., 1989b). Though most experiments used iron, copper was also found to be efficient in inducing D N A damage (Dizdaroglu et al., 1990). The role of transition metal ions in the reaction system is to catalyze the formation of -OH, which will be discussed below. -OH can attack all four D N A bases by adding hydroxyl groups onto the base residues, producing hydroxyl bases. 8-hydroxy-2'-deoxyguanosine (8-OH-2-dG) represents one of the major products of free radical attack (Fraga et al., 1990; Dizdaroglu, 1991). Besides -OH, aldehydes (Douki and Ames, 1994) and peroxynitrite (Douki et al., 1996) can attack D N A as well. Free radical attack can also cause other kinds of modifications, such as methylation of D N A bases (Augusto, 1993) and DNA-protein cross-linking (Gajewski and Dizdaroglu, 1990). The resulting D N A adducts can lead to D N A mutation (Echols and Goodman, 1991; Shibutani et al., 1991) or inhibit D N A replication (Moore and Strauss, 1979). Protein damage by free radicals mostly involves modification of amino acid residues. Like the attack on lipid and DNA, the major damaging free radical is OH, which can oxidize proteins to produce protein carbonyls (Amici et al., 1989). ONOO" can also attack proteins, forming carbonyls and oxidizing thiols to disulfides or nonreducible products (Radi et al., 1991). These modifications will generally induce increased cellular proteolysis and removal of the damaged proteins. The roles of transition metal ions in oxidative stress have been extensively studied. Transition elements belong to the d-block in the periodic table. The important transition metals are the elements that have incompletely filled 3d-subshell in oxidation states. Among these elements, iron (Fe), copper (Cu), manganese (Mn), zinc (Zn), nickel (Ni), and chromium (Cr), are biologically important. Transition metal ions have the ability to combine with O2, O2" or H2O2, to form complexes. The biological significance of this characteristic is to facilitate and control the oxygen utilization in vivo. Transition metal ions are usually chelated by proteins, and exert the well-controlled role in various physiological processes, such as reduction-oxidation reaction (in oxidases, superoxide dismutase, peroxidases) and transport of O2 (in hemoglobin, myoglobin). The 7 Chapter 1 Introduction importance of transition metal ions in oxidative stress is largely due to their involvement in the formation of free radicals, particularly -OH. Iron-dependent formation of -OH has been widely considered the major source of -OH in vivo. Iron(II)-catalyzed generation of •OH from H 2 0 2 is called the Fenton reaction (1898) (Fenton, 1898): H 2 0 2 + Fe 2 + -> -OH + OH" + Fe 3 + Copper has a similar effect (Fenton-like reaction): H 2 0 2 + C u + -> OH + OH" + C u 2 + The Haber-Weiss reaction was first considered as another important source of -OH without transition metals: •02" + H 2 0 2 -> OH + OH" + 0 2 Later, it was found that this reaction is too slow to induce any toxic effect i f the transition metal ions were carefully removed, and that this reaction can be greatly accelerated by iron (Halliwell, 1978; McCord and Day, 1978). Redox potential is an important term in reduction-oxidation reaction. Electrons are transferred from one compound to another in reduction-oxidation reaction. Reducing agent (reductant) is the electron donor, and oxidizing agent (oxidant) is the electron acceptor: Oxidant + e ±5 Reductant Each paired electron donor/acceptor is rated according to its propensity to give up or take up electrons. Such ratings are termed redox (reduction-oxidation) potential. The standard reduction potential (EQ) is a measure of the electromotive force generated at 25°C and pH 7.0 by a sample half-cell (containing paired electron donor/acceptor) with respect to a reference half-cell. The reference half-cell contains 1 M H + in equilibrium with H 2 at a pressure of 1 atm. The redox potential of H + / H 2 couple is defined to be 0 volt (V). A strong reductant has a negative redox potential, whereas a strong oxidant has a positive redox potential. The standard redox potential of some important biological compounds are listed below: 8 Chapter 1 Introduction Table 1-2. Standard redox potentials of some oxidant/reductant couples Electron acceptor Electron donor n E6 (V) Reference •OH H 2 0 1 +2.33 (Fee and Valentine, 1977) •o2- H 2 0 2 1 +0.87 (Fee and Valentine, 1977) V202 H 2 0 2 +0.82 (Fee and Valentine, 1977) Fe 3 + Fe 2 + 1 +0.77 (Prohaska, 1997) H 2 0 2 •OH 1 +0.38 (Fee and Valentine, 1977) Cytochrome c-Fe 3 + Cytochrome c-Fe 2 + 1 +0.22 (Stryer, 1988) C u 2 + C u + 1 +0.16 (Prohaska, 1997) Ubiquinone (ox) Ubiquinone (red) 2 +0.10 (Stryer, 1988) Dehydroascorbate Ascorbate 2 +0.08 (Stryer, 1988) Cytochrome b-Fe Cytochrome b-Fe 1 +0.07 (Stryer, 1988) 2 H + H 2 2 0 Pyruvate Lactate 2 -0.19 (Stryer, 1988) esse 2 C S H 2 -0.22 (Jocelyn, 1967) GSSG 2 GSH 2 -0.24 (Jocelyn, 1967) N A D + N A D H 2 -0.32 (Stryer, 1988) N A D P + N A D P H 2 -0.32 (Stryer, 1988) o2 •02" 1 -0.33 (Fee and Valentine, 1977) DTT(ox) DTT(red) 2 -0.331 (Jocelyn, 1987) Succinate a-Ketoglutarate 2 -0.67 (Stryer, 1988) The free energy change ( A G 0 ' ) for a reduction-oxidation reaction can be calculated from the standard redox potential: A G ° ' = -ftFAEo = -nF(Eo, acceptor - EQ, donor) Where n is the number of electrons transferred and F is Faraday's constant (23 kcal/V). The utility of determining A G ° ' lies in that its magnitude can predict whether or not a given chemical reaction can take place spontaneously. If A G ° ' = 0, the system is at equilibrium; if A G ° < 0, the reaction will proceed spontaneously as written; if A G ° ' > 0, 9 Chapter 1 Introduction the reaction will not proceed as written, but will, instead, proceed in the opposite direction. The free energy change of stepwise reduction of O2 can be calculated from the redox potential shown in Table 1-2 (Fig 1-2). The addition of the first electron to O2 is an uphill and unfavorable process, which is crucial for the stabilization of O2. It is reversible between O2 and 0 2". A l l subsequent steps to H2O are downhill processes, and irreversible. Energy is released during reduction of O2. AG°' can also be calculated from the equilibrium constant, K, of the reaction: AG°' = - RT ln K R is the gas constant (1.987 cal/deg-mole), and T is the absolute temperature (0°C = 273°K). If the reactants and products are present in concentrations other than 1 M each, the actual redox potential (£") can be calculated from the Nernst equation: £ , _ £ , + R T j [oxidized form] 0 nF [reduced form] 4. Multiple Roles of Astrocytes For decades, glial cells, originally viewed as just the glue that holds the brain together, have been considered as passive structural and metabolic support elements. Accumulating evidence now indicates that these cells, and especially astrocytes, can nourish, protect, and interact with surrounding neurons. Neuroglia outnumber neurons 10- to 50-fold, and comprise about half the total volume of the brain and spinal cord (Travis, 1994). There are four major types of neuroglia in the CNS: (1) astrocytes, (2) oligodendrocytes, (3) microglia, and (4) ependymal cells. Oligodendrocytes are mainly responsible for the formation of the myelin sheath of nerve fibers in the CNS (Szuchet, 1995). Schwann cells in the peripheral nervous system (PNS) are the functional equivalents of oligodendrocytes. Microglial cells serve as the immune cells of the CNS. 10 Chapter 1 Introduction They are inactive in the normal brain. In inflammation, they proliferate and become actively phagocytic (Streit, 1995; Kreutzberg, 1996). Ependymal cells line the cavities of the brain and the central canal of the spinal cord, assisting in the circulation of the CSF (Reichenbach and Robinson, 1995). Unlike other types of neuroglia, the functions of astrocytes are quite diverse. They can serve as scaffolding for the migration of immature neurons (Rakic, 1995), participate in the formation of B B B (Wolburg and Risau, 1995), regulate extracellular K + ions (Newman, 1995), absorb glutamate and y-aminobutyric acid (GABA) released by the nerve terminals (Martin, 1995), release neuroactive amino acids and neurotrophic factors (Banker, 1980; Levi and Gallo, 1995), and supply substrates of energy (Coles, 1995). The microenvironment of the CNS is quite different from the external environment of the body. For example, not only the concentrations of amino acids of CSF are lower than that of plasma, but also the CSF/plasma ratios of amino acids are different from each other (Perry et al., 1975; McGale et al., 1977), indicating the selective absorption or regulation by the brain. Except for small lipophilic molecules and gaseous agents, most nutrients enter the brain via transporters, channels, or pinocytosis (Pardridge and Oldendorf, 1977; Rowland et al., 1991). The various components of CSF are maintained at stable levels while plasma components may fluctuate widely (Cutler and Spertell, 1982). Given the diverse functions of neuroglia, particularly astrocytes, and the special microenvironment of the brain, we could consider neuroglia functioning as neuronal care-givers, maintaining the optimal internal milieu of the brain. The glia-neuron co-culturing experiments provide a direct proof of neurosupportive effects of astrocytes. The data from different researchers have repeatedly confirmed the fact that astrocytes can improve neuronal survival. Banker (Banker, 1980) reported that the cultures of hippocampal neurons from rat fetuses have higher number of process-bearing cells when they are co-cultured with astrocytes or cultured in astrocyte conditioned medium (ACM), than that cultured alone. Similarly, astrocytes can also support the survival of cortical neurons (Walsh et al., 1992; Sass et al., 1993) and retinal ganglion cells (McCaffery et al., 1984) in co-culture systems. Different co-culture methods have been used. The primary neurons and confluent astrocytes can be co-cultured either by directly plating neurons on the surface of an already confluent glial 11 Chapter 1 Introduction monolayer (McCaffery et al., 1984; Baughman et al., 1991), or by using the non-contact methods (Goslin and Banker, 1991; Walsh et al., 1992; Sass et al., 1993). The evidence from astrocyte conditioned medium (ACM) also supports a neurosupportive role of astrocytes. A C M can improve neuronal survival and neurite outgrowth (Vaca and Wendt, 1992; Yoshida et al., 1995). Some researchers have found that the trophic effects of astrocytes have regional specificity (Yuzaki et al., 1993; Le Roux and Reh, 1994; Yoshida et al., 1995). For example, cerebellar astrocytes specifically support the survival of Purkinje cells in culture (Yuzaki et al., 1993). As neurosupportive effects of astrocytes are well supported, it is agreed upon that astrocytes release or secrete some trophic factors to exert those functions. A large number of compounds have been found to be released by astrocytes. They include amino acid transmitters (Bowery et al., 1976; Holopainen and Kontro, 1990; Kimelberg et al., 1990), eicosanoids (Hartung and Toyka, 1987; Jeremy et al., 1987), neuropeptides (Deschepper et al., 1986; Melner et al., 1990), and neurotrophic factors. The neurotrophic factors that can be secreted by astrocytes includes nerve growth factor (NGF) (Furukawa et al., 1986), glial cell line-derived neurotrophic factor (GDNF) (Lin et al., 1993), basic fibroblast growth factor (bFGF) (Oh and Yong, 1996), epidermal growth factor (EGF) (Gomes et a l , 1999), interleukin-6 (IL-6) (Lee et al., 1993; Maeda et al., 1994), granulocyte-macrophage colony-stimulating factor (GM-CSF) (Ohno et al., 1990), and erythropoietin (Masuda et al., 1994; Sakanaka et al., 1998), all of which can improve neuronal survival with varying degrees of effects. The role of neurotrophic factors in glial-neuron communication has been studied intensively over the past 10 years (Barres and Barde, 2000). It has become increasingly clear that neurotrophic factors not only have important influence on neuronal survival (Goldberg and Barres, 2000), but also participate in a variety of nervous activities, such as regulating synaptic plasticity (Thoenen, 1995) and activating sodium channels (Kafitz et al., 1999). Though so many factors have been found, there is still no agreement as to which one or which category may play a more important role in supporting neuronal survival. Perhaps a synergic action of the many factors is essential for astrocytes to exert their neurotrophic effects. Varon et al. (1984) found that a trophic activity of glial cells resides with low molecular weight agents rather than with protein macromolecules. While some 12 Chapter 1 Introduction researchers indicated that each neurotrophic factor has a specific supportive effect on distinct classes of neurons (for review see Lindsay, 1995; Schatzl, 1995; Davies, 1997). The effort of searching for more potent neurosupportive factors has never stopped, and the new members continue to be discovered. Astrocytes have shown the capacity to protect neurons from oxidative stress, as shown in Table 1-3. Various compounds were used in these experiments to induce oxidative stress. However, the mechanisms by which astrocytes protect neurons from oxidative stress are still uncertain. Desagher et al (1996) reported that the neurotoxic effect of H2O2 on neurons co-cultured with astrocytes was strongly attenuated compared with that observed with a pure population of neurons. These researchers found that catalase activity was enriched in astrocytes compared with neurons, and attributed neuroprotective effect of astrocytes to their strong H202-scavenging capacities. Data from many other experiments also identified the stronger antioxidative capacities of astrocytes than neurons (Makar et al., 1994; Lucius and Sievers, 1996; Iwata-Ichikawa et al., 1999). However, the evidence that astrocytes are more resistant to oxidative stress than neurons cannot automatically prove the neuroprotective functions of astrocytes. If astrocytes can only scavenge the oxidants that entered their cytoplasm, their role will be limited to attenuate, but not efficiently prevent, oxidative stress that neurons could face. For example, extracellular H2O2 can cross cell membranes without distinction. No matter how much H2O2 astrocytes may take and neutralize, there is still a certain proportion of H20 2 that will enter neurons. H20 2 permeating cell membrane itself is risky for the neurons. Therefore, in order to protect neurons efficiently, astrocytes are assumed not only to defend themselves, but also to set up a defense for neurons and/or to enhance neuronal antioxidative capacities. So far, only a few reports have addressed this question. Tanaka et al. (1999) reported that astrocytes secrete laminin and fibronectin, which attenuate reactive oxygen/nitrogen species-induced neuronal damage. Neurotrophic factors, which can be secreted by astrocytes, can protect neurons from oxidative stress to a certain degree (Kirschner et al., 1996; Mena et al., 1997). Iwata-Ichikawa et al. (1999) found that glial conditioned medium enhances GSH synthesis of neurons by up-regulating neuronal mRNA expression of y-glutamylcysteine synthetase. Astrocytes were also found to release GSH (Yudkoff et al., 1990) and cysteine (Sagara et al., 1993), 13 Chapter 1 Introduction whose (cysteine) levels regulate neuronal GSH synthesis. The latter will be discussed in detail in section 6 of this chapter. Given the above evidence, it is reasonable to hypothesize that astrocytes release or secrete some factors that could either exert their antioxidative functions extracellularly or enhance antioxidative capacities of neurons. Table 1-3. Observations of neuroprotective effects of astrocytes against various stimuli-induced oxidative stress Neurons Toxins References Cortical neurons •OH and ONOO" (Tanaka et al., 1999) y-radiation (Noel and Tofilon, 1998) Retinal ganglion cells •OH and ONOO" (Lucius and Sievers, 1996) Cerebellar neurons Dopamine (Hochman et al., 1998) Striatal neurons H 2 0 2 (Desagher et al., 1996) Mesencephalic neurons H 2 0 2 (Langeveld et al., 1995) L-Dopa (Han et al., 1996; Mena et al., 1997) 6-hydroxydopamine (Bronstein et al., 1995; Hou et al., 1997) 5. Antioxidative Systems of the Organisms As mentioned at the beginning, ever since 0 2 was generated in the atmosphere, organisms started developing defenses against oxidative stress. In the long odyssey of evolution, many efficient antioxidative systems appeared and have been adopted by most organisms. These systems can be largely classified as large molecule antioxidative defenses and small molecule antioxidative defenses. Antioxidative enzymes play a very important role in cellular antioxidative defenses. They include superoxide dismutase (SOD), catalase, and glutathione peroxidase (GPx). These enzymes are present in almost all cell types, indicating a universal requirement of these antioxidative enzymes. SOD catalyzes the dismutation of 02": 2O2" + 2 H + —> H 2 0 2 + 0 2 . Cu/Zn SOD is highly expressed in cytoplasm and some organelles, such as lysosomes and peroxisomes, in virtually all cell types, including neurons and non-14 Chapter 1 Introduction neuronal cells (Slot et al., 1986; Pardo et al., 1995). C u 2 + in Cu/Zn SOD is at the active site of the protein, while Z n 2 + helps stabilizing the enzyme (Fridovich, 1995). Mn SOD, catalyzing the same reaction as Cu/Zn SOD, exists largely in the mitochondria. M n 3 + is at the active site of the enzyme (Fridovich, 1995). Gene knockout animals have shown the importance of SOD for survival. For example, Mn SOD deficient mice suffered severe neurodegeneration and perinatal death (Lebovitz et al., 1996). H2O2 is usually removed in human tissues by two enzymes: catalase and GPx. Catalase decomposes H2O2 directly: 2H2O2 2 H 2 O + O2. Catalase is largely present in subcellular organelles known as peroxisomes of all major tissues, being especially concentrated in liver (Chance et al., 1979). GPx removes H 2 0 2 by using tripeptide glutathione (GSH): H 2 0 2 + 2 G S H —» GSSG + H2O. GPx requires selenium as its cofactor. Glutathione disulfide (GSSG) is converted back to GSH by glutathione reductase with N A D P H : GSSG + N A D P H + H + —» 2 G S H + N A D P + . The GPx family is comprised of several distinct forms of Se-dependent GPx. They are (1) classical GPx existing in the cytoplasm of almost all cell types (Levander, 1987), (2) phospholipid hydroperoxide GPx in the cell membrane (Ursini et al., 1985), and (3) plasma or extracellular GPx in plasma and lung fluid (Avissar et al., 1989; Avissar et al., 1996). Besides these enzymes that remove ROS directly, there are some other enzymes, which do not remove free radicals directly but are related to antioxidative defenses, such as glutathione reductase (GR), glutathione synthetase, glutathione S-transferase, and glucose-6-phosphate dehydrogenase ( G 6 P D ) . Proteins that bind transition metal ions are considered as very important antioxidative defenses. Transition metal ions, such as iron and copper, can mediate electron transfer, and their physiological functions are usually well controlled. Free or low-molecular-weight complexes of transition metal ions can catalyze autoxidation of reducing agents and the Fenton reaction, generating ROS (Halliwell and Gutteridge, 1984). The organism has evolved many binding proteins for these transition metal ions to minimize their availability as free forms (Murray et al., 1993). Transferrin is a glycoprotein that transports iron to sites where it is needed. One molecule of transferrin can bind two molecules of Fe 3 + (Bates and Schlabach, 1973), preventing them from participating in catalyzing reactions. Intracellular iron is stored as Fe 2 + in the binding form with ferritin. Heme, released from hemoglobin, can stimulate the generation of 15 Chapter 1 Introduction ROS. Hemopexin, a plasma protein, can bind heme tightly and inhibit heme-induced lipid peroxidation (Gutteridge and Smith, 1988). Plasma copper is tightly bound to ceruloplasmin, each molecule of which binds 6 molecules of copper. Albumin can also bind, but with weak ability, to heme and copper. Metallothioneins (MTs) are a family of single chain molecules, which are rich in cysteine content and mainly bind zinc and copper (Dunn et al., 1987). There are four isoforms of MTs. MT-I and MT-II are extensively expressed in all mammalian tissues, including astrocytes in the brain and spinal cord (Blaauwgeers et al., 1993). MT-III and MT-IV are predominantly expressed in the CNS (Palmiter et al., 1992; Kobayashi et al., 1993). MTs are implicated in the storage of metal ions in a non-toxic form and regulation of cellular zinc and copper metabolism. Their expression in the CNS is up-regulated during oxidative stress and inflammation (Aschner, 1998; Hidalgo and Carrasco, 1998). Taken together, organisms have evolved many mechanisms to bind transition metal ions, preventing them from catalyzing the generation of ROS. Our body contains a complex mixture of small molecule antioxidants. These compounds can be classified into those synthesized in vivo and those obtained from diet. Glutathione is the major synthesized cellular antioxidant. It can remove H2O2 with GPx, or neutralize free radicals by itself. The metabolism of glutathione will be discussed in the next section. ct-Lipoic acid, previously classified as a vitamin, can be synthesized by humans (Carreau, 1979). Lipoic acid and its reduced form, dihydrolipoic acid, have antioxidative functions, such as scavenging free radicals, chelating transition metals, and regenerating other antioxidants (Packer et al., 1995). Uric acid is the oxidized end product of purine bases, such as adenine, guanine, hypoxanthine, and xanthine. It can react with free radicals, and its plasma levels are considerably high (-300 uM) (Ames et al., 1981). cc-Ketoacids, including pyruvate, a-ketoglutarate, a-ketobutyrate and oxaloacetate, are H 2 0 2 scavengers. Though the reactivity of a-ketoacids with H 2 0 2 was first noticed many years ago (Holleman, 1904; Hatcher and Hil l , 1928; Bunton, 1949), the protective effects of pyruvate against oxidative stress in biological system have not been reported until relatively recently (O'Donnell-Tormey et al., 1987; Desagher et al., 1997; Giandomenico et al., 1997). CSF has high levels of pyruvate (-200 uM) (Vannucci and Duffy, 1976). There are also some other small molecule antioxidants 16 Chapter 1 Introduction synthesized in vivo, such as bilirubin, ubiquinone (coenzyme Q), melatonin, and melanin, all of which have capacities of scavenging free radicals. There are many small molecule antioxidants, which cannot be synthesized by humans but can be obtained from the diet. This class of antioxidants includes vitamins. Ascorbic acid (vitamin C) is a water-soluble vitamin. As a reducing agent, ascorbic acid can donate electrons and be oxidized to dehydroascorbic acid (DHA). It participates in many reduction-oxidation reactions, such as synthesis of epinephrine from tyrosine and collagen synthesis (Murray et al., 1993). The antioxidative functions reside in its capacities of scavenging various free radicals, including O2", OH, ROO-, ONOO" etc., and regenerating ct-tocopherol (Halliwell, 1996). It is postulated that ascorbic acid is especially important in the brain, as its concentration in CSF is about 10 times higher than that in plasma (Spector and Eells, 1984). However, it often acts as a pro-oxidant, usually by interaction with transition metal ions (Halliwell, 1996). In fact, many in vitro experiments make use of the mixtures of ascorbic acid and iron to generate OH (Barrier et al., 1998; Tanaka et al., 1999). a-Tocopherol (vitamin E) and carotenoids are lipid-soluble vitamins. Vitamin E is probably the most important inhibitor of lipid peroxidation. It scavenges lipid peroxyl radicals and breaks chain-reaction of lipid peroxidation (Burton et al., 1982). Vitamin E supplements have been reported to be beneficial for preventing cardiovascular disease, cataracts, various cancers (Landvik, 1997), as well as for the treatment of Alzheimer's disease (Grundman, 2000; Pratico and Delanty, 2000). Carotenoids, which are a group of compounds that are rich in plant tissues, can also scavenge free radicals (Everett et al., 1996). Besides these vitamins, there are many other plant antioxidants, such as flavanols, flavones, anthocyanidins, and phenylpropanoids. Although these non-vitamin antioxidants continue to be identified, the importance of their antioxidative roles in biological systems has yet to be elucidated. 6. Theories of Thiol Provision by Astrocytes Glutathione (GSH) is a tripeptide made up of glutamate, cysteine, and glycine. The structures of GSH and related compounds are shown in Fig. 1-3. It is the major cellular antioxidant. GSH is ubiquitously present in mammalian cells at intracellular concentrations ranging from 0.5 to 10 mM (Kosower and Kosower, 1983; Meister and 17 Chapter 1 Introduction Glutatftione Glutathione disulfide f H 3 N \ ^CH — CH2 - CH^s. rOOC X C I O I N . Cysteine* Cystine •—*• 1 H O i | ! - -Y coo H y Cysteine-glutathione 1 disulfide O C H - C H 2 C H 2 OOC f <fH2 \ y k C H 2 Figure 1-3. Molecular structures of glutathione and related compounds. Anderson, 1983). As GSH is the major cellular thiol, it is considered as a storage form of thiols. GSH plays diverse roles in cellular antioxidative defenses and other functions (Meister and Anderson, 1983; Beutler, 1989; Cooper and Kristal, 1997). 1). GSH acts as a cofactor of GPx and decomposes H2O2, with the concomitant conversion of GSH to GSSG. 2). As a reducing agent, GSH can directly react with free radicals, stopping the free radical chain-reaction. 3). GSH acts as a cofactor of GSH S-transferase to convert electrophilic xenobiotics (RX) to glutathione conjugates, which are excreted into bile. The reaction is as follows: R X + GSH -> RSG + H X . 4). GSH can regenerate ascorbic acid (Meister, 1992; Rose, 1993) and tocopherol (Heslop et al., 1996), and exert antioxidative roles synergistically. 5). GSH has been found to chelate copper ions and diminish their ability to generate free radicals in the reaction mixture of Cu (I) and H2O2 (Hanna and Mason, 1992). 6). By regulating disulfide bonds, GSH can maintain protein - S H groups, or play roles in protein folding and the degradation of proteins (Carelli et al., 1997). 7). GSH plays the critical roles in cellular redox signaling (Sen, 1998; Janaky et al., 1999). 8). GSH is involved in uptake of amino acids into cells by the y-glutamyl cycle (Orlowski and Meister, 1970; Meister and Anderson, 1983). It is now realized that 18 Chapter 1 Introduction though y-glutamyl cycle does operate in vivo, it may not play a major role in amino acid transport. Though GSH is solely synthesized in the cytoplasm, it also functions in some other organelles. For example, GSH is transported into the mitochondria, which are the major intracellular sources of ROS, scavenging free radicals and maintaining the normal mitochondrial membrane potential (Avy) (Beatrice et al., 1984; Garcia-Ruiz et al., 1995; Reed and Savage, 1995; Marchetti et al., 1997). GSH is synthesized from glutamate, cysteine and glycine (Fig. 1-4). The synthesis of GSH has two steps. First, y-glutamylcysteine synthetase catalyzes the formation of y -glutamylcysteine. Next, y-glutamylcysteine is joined by glycine catalyzed by glutathione synthetase, forming GSH. Both reactions require the presence of ATP. y -Glutamylcysteine synthetase is feedback inhibited by GSH. The availability of sulfur amino acids influences GSH content, and is considered the rate-limiting substrate for GSH synthesis (Beutler, 1989; Phelps et al., 1992). y-Glutamyltranspeptidase can break down GSH and transfer the glutamate residue onto other amino acids such as cysteine, methionine, and glutamine. The enzyme is located on the plasma membrane with its active site facing outwards (Meister, 1995). The resulting cysteinyl-glycine can further be hydrolyzed by a peptidase. GSH is oxidized to GSSG when it exerts antioxidative functions. GSSG is converted back to GSH by glutathione reductase (GR) with N A D P H as cofactor. GSH can be released from cells. Liver is the major organ releasing GSH to plasma (Fariss and Reed, 1983; Aw et al., 1986). In summary, GSH metabolism is very active, mirroring its critical roles in antioxidative defenses and other functions. The roles of GSH in the CNS have been extensively studied. Much evidence indicates that GSH is critical for neuronal survival. GSH depletion results in increased generation of ROS, loss of mitochondrial function, and rapid cell death (Jain et al., 1991; Wullner et al., 1999). Supplementation of N-acetylcysteine to neuronal cultures can increase intracellular GSH, and prevent PC 12 cell death caused by withdrawal of serum and NGF (Ferrari et al., 1995). Ratan et al. (1994) reported that inhibition of cystine uptake by homocysteate decreases intracellular GSH levels and induces neuronal death concurrently. They also found that macromolecular synthesis inhibitors prevent neuronal death by a mechanism of shunting cysteine from protein synthesis to glutathione synthesis. Murphy et al. (1990) found that glutamate toxicity of immature cortical 19 Chapter 1 Introduction GSSG H 2 0 11 Glu Cys Gly y-Glu-Cys H 2 0 2 II N A D P H N A D P + X R J a s * y-Glu-Cys-Gly (GSH) GSR + X H Figure 1-4. GSH synthesis and some of reactions of GSH. 1, y-glutamylcysteine synthetase. 2, glutathione synthetase. 3, glutathione peroxidase. 4, glutathione reductase. 5, glutathione-S-transferase. neurons involves the inhibition of cystine uptake. L-Buthionine sulfoximine (BSO), which irreversibly blocks y-glutamylcysteine synthetase, can abolish the astrocyte-mediated enhancement of neuronal survival (Drukarch et al., 1997). A l l these data demonstrate the importance of GSH in the CNS. It has been found that neurons and astrocytes have different preferences in the utilization of amino acids for GSH synthesis (Kranich et al., 1996). These researchers found that refeeding cultured neurons temporarily deprived of amino acids with cysteine, but not with cystine, can restore GSH content to a maximal degree. For astrocytes, it is cystine, but not cysteine, that restores a maximal GSH content. The utilization of cysteine by neurons requires a stable extracellular thiol source. As thiols are easily oxidized under aerobic conditions, the maintenance of stable thiol levels in the extracellular fluid is very important. Cysteine and GSH can be transported across the blood-brain barrier (BBB) (Wade and Brady, 1981; Kannan et a l , 1990). Plasma cysteine concentration of human venous blood was reported to be 11.2 + 0.9 uM (Chawla et al., 1984) or 11.9 uM (ranging from 4.96 to 21.1 uM) (Kleinman and Richie, 2000). {Cystine concentration of human venous plasma is 46 ± 8 jaM (Chawla et al., 1984) or 90.8 uM (Felig et al., 1973)}. GSH concentration of human venous plasma is 4.5 ± 0.4 uM (Chawla et al., 1984) or undetectable (lower than 8 nM) (Kleinman and Richie, 20 Chapter 1 Introduction 2000). Whether plasma thiols contribute to the thiol pool of CNS is uncertain, as no data have shown arterio-venous differences across the brain yet. Blood thiol levels fluctuate a lot under different nutritional conditions (Boebel and Baker, 1983; Suberville et al., 1987). Depending on blood as thiol supply would be a poor way to maintain stable levels of thiols in the CNS. The mechanisms of thiol maintenance by astrocytes have been explored in recent years. Yudkoff et al. (1990) followed the incorporation of [ 1 5N] glutamate into GSH, and found that GSH is released by cultured astrocytes, and the GSH concentration in the medium can reach to ~ 3 uM. Bannai and his colleagues have contributed a lot to unraveling the mechanisms of thiol maintenance in the CNS. They reported that the cysteine concentration in glial cultured medium is 2.6 ± 0.4 uM, and neuronal cultured medium cysteine is 0.6 ± 0.2 uM (Sagara et al., 1993). They also found that the uptake rate of cysteine by neurons is much faster than cystine. They suggest that neurons maintain their GSH levels by taking up cysteine provided by glial cells (Sagara et al., 1993). Later, they observed a GSH efflux from cultured astrocytes, and studied the kinetics of the GSH efflux (Sagara et al., 1996). But they admitted that the physiological significance of the GSH efflux is not clear, as they had confirmed that neurons do not take up GSH directly to replenish their GSH pool. The studies from analyzing perfusates of brain slices are not quite consistent. Keller et al. (1989) found that brain slices release cysteine on Ca2+-dependent depolarization. The same research group later reported that more compounds, including cysteine, GSH, glutamate and aspartate, are released under the same conditions (Zangerle et al., 1992). L i et al. (1999) have shown more compounds are released by hippocampal slices during oxygen/glucose deprivation, which include cysteine, GSH, cysteine sulfinate, glutamate, y-GluGlu, y-GluGln, and y-GluCys. Thus, there are two unsolved questions: first, what compound(s) is (are) released by astrocytes, cysteine, or GSH, or both? Second, i f astrocytes release GSH, how and what compound will it be converted into for neuronal uptake? Dringen et al. (1999a) suggest that GSH released by astrocytes is decomposed by ectoenzyme y-glutamyl transpeptidase to generate dipeptide CysGly that is subsequently used by neurons as a precursor for GSH synthesis. Their experiments show that CysGly, like cysteine, can serve as a GSH precursor when it is supplemented in the culture medium. Acivicin, an inhibitor of the astrocyte ectoenzyme y-glutamyl transpeptidase, decreases 21 Chapter 1 Introduction the GSH content of neurons in neuron-astrocyte co-cultures. However, there is no evidence for the existence of CysGly in CSF or astrocyte conditioned medium (ACM). The finding that y-glutamyl transpeptidase is mainly located in the endfeet of astrocytes (Zhang et al., 1997) also raises a question that this enzyme may be not accessible to extracellular GSH. In summary, GSH plays a critical role in cellular antioxidative defenses. Cysteine has been found to be the rate-limiting precursor of GSH synthesis. Thiol sources in the CNS must be maintained at a stable levels, which is a distinguishing feature of GSH metabolism in the CNS. What antioxidative compounds are released by astrocytes and provided to neurons are important questions and deserve to be investigated in depth. 7. Theories of Cysteine Toxicity Cysteine is naturally present in the human body and in the environment. It is critical for GSH synthesis. However, it has long been noted that cysteine can cause neurotoxicity. Though this finding was first reported more than three decades ago (Olney and Ho, 1970), the mechanism is still uncertain. There are several theories about cysteine neurotoxicity. 1. Excitotoxic metabolites of cysteine. Neuronal death was first reported in the infant mouse retina and hypothalamus following oral administration of high doses of cysteine (3 mg/g weight) (Olney and Ho, 1970), or subcutaneous injection (1.5 mg/g weight) (Olney et al., 1971). Widespread degenerative changes, including cerebral cortex, hippocampus, amygdala, caudate, thalamus, pons, medulla oblongata, spinal cord and cerebellum, were reported later in infant and fetal mice (Olney et al., 1972), and infant rats (Karlsen et al., 1981). In their studies, Olney et al. (1971; 1972) compared the pathological patterns of cysteine with other excitatory amino acids, such as glutamic acid, aspartic acid, cysteic acid (CA), cysteine sulfinic acid (CSA), N-methyl DL-aspartic acid and N-methyl DL-glutamic acid. Some differences were observed. For example, cysteine toxicity destroys neurons in several major brain areas never affected by glutamic acid and other acidic amino acids, such as amygdala, some areas of hippocampus and cerebral cortex and several neuronal groups within the thalamus. On the other hand, some structures, like the midline periventricular area, are damaged by acidic amino acids, but not affected by 22 Chapter 1 Introduction cysteine. Even though there exist discrepancies between cysteine and other excitatory amino acids, the first explanation of the mechanism of cysteine neurotoxicity was that the conversion of - S H terminal of cysteine to acidic group (SO2H or SO3H) could account for its neurotoxicity (Olney et al., 1971). Cysteine has one rather than two acidic groups and does not obey the structure-activity requirements for excitotoxins, while CSA (-S0 2 H) and C A (-SO3H) are NMDAmimetic. Lehmann et al. (1993) studied whether cysteine toxicity depends on its conversion to CSA or CA. They found that the distribution of brain injury after subcutaneous administration of cysteine (1 mg/g) to infant rat is quite different from that caused by CSA (3 mg/g). In addition, the concentrations of CSA and C A converted from injected cysteine are much lower than the toxic dose levels of CSA and C A (Lehmann et al., 1993). Similar findings from other researchers also showed that cysteine is about 10-fold more toxic to neuronal cell lines than CSA and CA, suggesting that it is not cysteine metabolites that mediate cysteine neurotoxicity (Pean et al., 1995). 2. Bicarbonate-dependent excitotoxicity. Olney et al. (1990) reported that excitotoxic potency of cysteine is substantially increased in the presence of physiological concentrations of bicarbonate ion, and that N M D A antagonists can block cysteine toxicity. In parallel, Weiss and Choi (1988) reported that beta-N-methylamino-L-alanine (BMAA) , an environmental toxin that causes Guam ALS-parkinsonism-dementia, requires bicarbonate as a cofactor to produce neurotoxic effects. It was suggested that cysteine, similar to B M A A , forms noncovalent hydrogen bond with bicarbonate, simulating the acidic Q terminal of excitatory amino acids, to act as a glutamate agonist (Olney et al., 1990). However, whether cysteine can form a hydrogen bond with bicarbonate has not been proven using analytical chemistry. 3. Redox modulation of the N M D A receptor. It has been found that the responses of the N M D A receptor are modulated by reduction and oxidation (Aizenman et al., 1989; Janaky et al., 1993; Kohr et al., 1994; Gozlan and Ben-Ari, 1995; Choi and Lipton, 2000). The extracellular part of N M D A receptor channel contains several redox sites. While the reducing agent dithiothreitol (DTT) potentiates the electrophysiological response of cultured neurons to N M D A , the thiol oxidant 5,5-dithio-bis-2-nitrobenzoic acid (DTNB) decreases the magnitude of the response (Aizenman et al., 1989). Similar 23 Chapter 1 Introduction results were confirmed with the recombinant N M D A receptors with GSH and DTT (Kohr et al., 1994). GSH was also found to increase the glutamate- and NMDA-induced C a 2 + influx (Janaky et al., 1993), and to increase neuronal vulnerability to azide-induced chemical hypoxia and glucose deprivation (Regan and Guo, 1999). Cysteine, like DTT and GSH, is a reducing agent and can modulate redox state of proteins as well. Puka-Sundvall et al. (1995) reported that cysteine and its N-acetylated analog, N-acetyl-cysteine (NAC), can enhance glutamate toxicity in both neuronal cultures and in vivo experiments. These results show that reducing agents, such as cysteine and GSH, can sensitize the N M D A receptor and enhance glutamate excitotoxicity. However, high doses or elevated levels of glutamate are the prerequisite of cysteine toxicity in all these experiments. Evidence showed that cysteine injection (s.c.) cannot itself induce a significant glutamate release (Puka-Sundvall et al., 1995). As cysteine toxicity can be induced just by cysteine administration itself and no glutamate is required, the significance of sensitized N M D A receptor in cysteine toxicity is still in question. 4. Free radical theory. Cysteine is a very unstable amino acid. It will be oxidized under aerobic conditions, during which process free radicals are produced (Hanaki and Kamide, 1975; Gardner and Jursinic, 1981; Munday, 1988, 1989). Saez et al. (1982) detected spin adducts of hydroxyl radicals by electron paramagnetic resonance (EPR) during the spontaneous autoxidation of cysteine. Ions of iron and copper increase the yield of spin adducts. They also found that incubation of cysteine (4 mM) with cultured hepatocytes causes release of lactate dehydrogenase (LDH) and decrease of intracellular ATP and GSH concentrations. Nath and Salahudeen (1993) found that cysteine autoxidation generates H2O2, which is catalyzed by plasma, kidney extract, and copper ions. Autoxidation of cysteine (4 mM) in the presence of copper is toxic to cultured kidney cell lines. However, the cysteine concentrations used in these experiments are very high. It is still uncertain whether it can reflect the in vivo cysteine toxicity. In addition, there are no reports showing the effect of cysteine autoxidation on neuronal survival thus far. Cysteine is a double-edged sword. It is an essential and rate-limiting precursor of GSH synthesis. On the other hand, different research groups have proven its cytotoxicity repeatedly in different cell types. The existence of different theories reflects the 24 Chapter 1 Introduction complexity of cysteine toxicity. Only if the mechanisms are fully understood, can we make use of the beneficial role of cysteine while harnessing its side effects. 8. Neurological Diseases and Oxidative Stress Though the pathogenesis of neurodegeneration is still elusive, accumulative evidence has indicated the important role of oxidative stress and free radical damage in neurodegenerative diseases, such as Alzheimer's disease (AD), Parkinson's disease (PD), amyotrophic lateral sclerosis (ALS), Huntington's disease, Friedrich's disease, multiple system atrophy, and olivopontocerebellar atrophy (OPCA) (Gotz et al., 1994; Markesbery, 1997; Behl, 1999; Smith et al., 2000). Alzheimer's disease (AD) is the leading cause of senile dementia, and is characterized by progressive decline in memory, language, and other cognitive functions. The pathological findings include the accumulation of extracellular senile plaques containing (5-amyloid peptides (AP), intracellular neurofibrillary tangles containing polymerized and hyperphosphorylated tau proteins, and severe neuronal loss (Martin, 1999). The pathogenesis of A D remains unknown, though there are several hypotheses for A D , such as genetic defects, latent virus infection, energy metabolism deficits, excitotoxicity, mitochondrial defects, and oxidative stress (for review see Markesbery, 1992). There has been heightened interest in the oxidative stress hypothesis in recent years. Different lines of evidence have shown the widespread free radical damages in A D . For instance, protein carbonyl, the protein oxidation products, is increased in some regions, such as the frontal lobe and the hippocampus, in A D subjects (Smith et al., 1991; Hensley et al., 1995). The oxidized nucleoside, 8-hydroxy-2'-deoxyguanosine (8-OHdG), the D N A oxidation product, is 3-fold higher in mitochondria and significantly increased in the nuclei of A D subjects compared with age-control subjects (Mecocci et al., 1994). Lipid peroxidation, as measured by thiobarbituric acid reactive substances (TBARS), is also increased in A D brain compared with controls (Hajimohammadreza and Brammer, 1990; Subbarao et al., 1990; Balazs and Leon, 1994; Lovell et al., 1995). Nitration (Smith et al., 1997; Torreilles et al., 1999) and advanced glycosylation end products (AGE) (Vitek et al., 1994; Yan et al., 1994; Smith et al., 1995) are also oxidative modifications found in A D . 25 Chapter 1 Introduction The sources of oxidative stress in A D are still uncertain. It is generally believed that aging is associated with an increased production of free radicals and a decreased ability of antioxidative defenses. The modified proteins can further form intra- and intermolecular crosslinking, and cause subsequent accumulation of damaged proteins. Ap is a heterogeneous peptide with 42 (or 43) amino acids, derived from amyloid precursor protein (APP). Though Ap neurotoxicity has been extensively investigated, there still lack of consensus regarding its toxic mechanism. Recent studies have shown that AP and APP is involved in the production of free radicals. Ap can produce H2O2 in the presence of Fe 3 + and C u 2 + (Huang et al., 1999a; Huang et al., 1999b). APP was found to have copper binding sites and to produce H2O2 and -OH in the presence of Cu (Multhaup et al., 1998; White et al., 1999). C u 2 + can also bind Ap (Atwood et al., 2000). Pathological findings show that senile plagues are rich in C u 2 + (Lovell et al., 1998), which contain Ap. Sayre et al. (2000) reported that neurofibrillary tangles and senile plaques are major sites for catalytic redox reactivity. These lesions catalyze the ^ i n -dependent oxidation of 3,3'-diaminobenzidine (DAB), and pretreatment with chelators of iron and copper abolishes the ability of these lesions. The apolipoprotein E-4 (Apo E-4) genotype is an important risk factor in late-onset familial and sporadic A D (Corder et al., 1993; Saunders et al., 1993). ApoE has been reported to be a strong chelator of iron and copper (Miyata and Smith, 1996), and to be susceptible to crosslinking by products of lipid peroxidation (Montine et al., 1996). Oxidative theory has lead to the application of antioxidants in the treatment of A D . In fact, clinical trials have demonstrated potential benefits of the antioxidants, such as Vitamin E, co-enzyme Ojo, flavonoids, selegiline, and melatonin (Pratico and Delanty, 2000). Amyotrophic lateral sclerosis ( A L S ) is a progressive degenerative disorder affecting upper and lower motor neurons. Up to 10% of all A L S cases are familial forms, and the others are sporadic. Though the pathogenesis is still to be ascertained, there has been important progress in recent years. Rosen et al. (1993) identified the genetic linkage of mutations in the Cu/Zn superoxide dismutase (SOD-1) gene with the development of some forms of familial A L S . Since the initial report, approximately 50 mutations of the SOD-1 gene have been found in 20% of familial A L S cases (Siddique and Deng, 1996). Mechanistic studies have consistently pointed to a gain of function of mutant SOD-1, 26 Chapter 1 Introduction which involves the generation of free radicals, as the damaging properties of these mutations (Louvel et al., 1997; Morrison and Morrison, 1999). Transgenic mice expressing the mutant SOD-1 gene have normal or elevated enzyme activities, and yet develop the symptoms and pathology resembling A L S (Borchelt et al., 1994; Gurney et al., 1994; Ripps et al., 1995; Bruijn et al., 1997). On the other hand, SOD-1 knockout mice that are completely deficient of this enzyme do not develop motor neuron degeneration and A L S symptoms (Reaume et al., 1996). Further investigation revealed that the loosely bound copper to the mutant SOD-1 plays a critical role in the acquisition of the toxicity of the mutant gene (Carri et al., 1994; Estevez et al., 1999; Gabbianelli et al., 1999). Excitotoxicity is another important theory of ALS. Lathyrism, a form of motor neuron disease, was found to be causally associated with eating chickling pea, which contains P-oxaloamino-L-alanine (BOAA), an excitatory amino acid (Spencer et al., 1986). These researchers also found that ALS-Parkinsonism-dementia complex of Guam can be linked to the digestion of cycad component p-methylamino-L-alanine (BMAA) (Spencer et al., 1987). B M A A does not possess the Q acid, as shared by glutamate, B O A A , and other excitotoxins. Weiss and Choi (1988) explained that B M A A requires bicarbonate as a cofactor. The two molecules form a hydrogen bond, mimicking the structure of excitotoxin. However, Duncan et al. (1990) indicated that the levels of B M A A in cycad flour are too low to cause the delayed and widespread neurodegeneration. An important finding reported by Rothstein et al. (1995) showing that there is a selective loss of glial glutamate transporter GLT-1 in motor cortex and spinal cord of postmortem A L S cases. GLT-1 and GLAST are selectively localized to astrocytes (Rothstein et al., 1994; Lehre et al., 1995), and are critical for the uptake of extracellular glutamate. Inhibition of these glutamate transporters using chronic antisense oligonucleotide administration produced elevated extracellular glutamate levels, neurodegeneration characteristic of excitotoxicity, and a progressive paralysis of rat (Rothstein et al., 1996). However, it is not explained why and how the selective loss of glutamate transporters occurs. There is a lack of genetic and etiological evidence to support this hypothesis. 27 Chapter 1 Introduction In stroke and traumatic brain injury (TBI), brain tissue develops a series of pathological events that evolve in time and space. Oxidative stress has been found to play a critical role in secondary neuronal damage (Juurlink and Paterson, 1998; Dirnagl et a l , 1999; Lewen et al., 2000). In animal model studies, free radicals (Peters et al., 1998) and the D N A damage product, 8-OHdG (Nagayama et al., 2000), were detected in the peri-infarct region (or penumbra) after focal cerebral ischemia. Generation of free radicals after TBI was also detected (Awasthi et al., 1997). Excitotoxicity is an important pathogenic factor in the secondary neuronal injury of stroke and TBI. Impairment of cerebral blood flow restricts the delivery of oxygen and glucose. In the core region of the affected brain tissue, neurons undergo energy depletion and anoxic cell death. In the penumbral region, energy supply is partially impaired, which rapidly leads to modest decrease of ionic gradients and depolarization of membrane potential. Depolarized membrane potential will cause glutamate release into the extracellular space (Katsura et al., 1994; Martin et al., 1994). Over-stimulation of N M D A receptors, as discussed earlier, causes C a 2 + influx and overload, which subsequently leads to mitochondrial C a 2 + accumulation and failure of mitochondrial membrane potential (A^F) (Schinder et al., 1996; White and Reynolds, 1996). Mitochondria are the major source of free radicals, and the dysfunction of mitochondria increases the generation of free radicals (Luetjens et al., 2000). Free radicals, therefore, are the executioners in the excitotoxicity process. Inflammation is another pathogenic factor in stroke and TBI, and is closely interconnected with oxidative stress. Free radicals are the major mediators of inflammation. In stroke and TBI, microglial cells are rapidly activated. Neutrophils and macrophages enter the injury site via the compromised BBB. During phagocytosis, 0 2 consumption of these inflammatory cells substantially increases, which is known as respiratory burst. The N A D P H oxidase complex in the plasma membrane of inflammatory cells makes use of N A D P H to reduce 0 2 , generating 02~: N A D P H + 2 0 2 -» N A D P + + H + + 2-02". 0 2" will be further converted to other inflammatory oxidants, including H 2 0 2 , OO", and OH. These free radicals are responsible for the oxidative degradation of engulfed particles, and release of the produced free radicals helps phagocytes to attack extracellular targets (Conner and Grisham, 1996). As free radicals' 28 Chapter 1 Introduction attack is indiscriminate, they will cause tissue injury, as in the cases of stroke and TBI (Awasthi et al., 1997; Nagayama et al., 2000). On the other side, free radicals produced in inflammation may play a role in chemotaxis. H2O2 produced by activated phagocytes can up-regulate adhesion molecules, such as ICAM-1, to facilitate phagocyte adherence to endothelium (Lo et al., 1993). Free radicals can also act as second messengers to propagate inflammatory signal (Suzuki et al., 1997; Hensley et al., 2000). For example, the important oxidative stress-related transcription factor N F - K B can be activated by free radicals (Schreck et al., 1991; Pahl and Baeuerle, 1994). These results demonstrate an important role of free radicals in inflammation-induced secondary tissue injury in stroke and TBI. Taken together, different lines of evidence suggest that oxidative stress is involved in many neurological diseases, and is a major pathogenic factor. Neuroprotective effects of antioxidants in the treatment of neurological diseases have been widely reported (Gotz et al., 1994; Louvel et al., 1997; Cuajungco et al., 2000; Pratico and Delanty, 2000). Further clarification of some basic oxidative stress-related mechanisms will be of benefit to the antioxidative intervention, and may lead to the final success of disease control. 9. Experimental Objectives The overall goal of my experiments is to explore the mechanisms by which astrocytes enhance antioxidative capacities of neurons via small molecule pathways. The overall hypothesis is that astrocytes can release small molecules to protect neurons from oxidative stress. Through experiments, I will test the following hypotheses: 1) . Astrocytes improve neuronal survival. 2) . Astrocytes provide cysteine to neurons. 3) . Free radicals are generated during cysteine autoxidation. 4) . Cysteine toxicity can be prevented by antioxidants. 5) . Astrocytes release antioxidant to protect neurons from cysteine toxicity. The experiments will be done in the following sequence: First, a tissue culture system will be set up for the assessment of neurosupportive effects of astrocytes. This will be the basis of further experiments. Neuronal survival 29 Chapter 1 Introduction with or without co-cultured astrocyte feeder layer will be compared. Neurosupportive effects of astrocytes will also be observed from another perspective, i.e., the neuronal death induced by deprivation of astrocyte feeder layer. Apoptotic neuronal death will be assessed. Second, the provision of cysteine and its mechanisms by astrocytes will be investigated. These mechanistic studies are based on the evidence that astrocytes improve neuronal survival, and that cysteine is critical for GSH synthesis. As discussed earlier, it is still unclear how cysteine is released and what the mechanisms of cysteine maintenance by astrocytes entail. Cysteine, GSH, and related compounds in astrocyte conditioned medium (ACM), neuron conditioned medium (NCM), plasma, and cerebrospinal fluid (CSF) will be analyzed using HPLC. A hypothesis of indirect cysteine provision by astrocytes is proposed. Third, the mechanism of cysteine neurotoxicity will be addressed. Cysteine is critical for GSH synthesis, particularly for neurons. However, its toxicity has long been noted. To solve this paradox, I will study the relationship of cysteine autoxidation and the generation of hydroxyl radicals. The factors influencing cysteine autoxidation will be investigated. In these experiments, it was found that copper is the most efficient catalyst of cysteine autoxidation. In further experiments, the influence of copper on cysteine neurotoxicity will be studied in cell culture. The generation of hydroxyl radicals during cysteine autoxidation in the absence or presence of copper will be assayed with fluorimetric methods. A proposed mechanism of cysteine toxicity will be defined based on these results. Fourth, the in vivo mechanisms which defend against cysteine toxicity will be explored. It is known that cysteine is essential for neurons and naturally present in the CNS, whereas it has neurotoxic effects. Therefore, the organism must have developed some defensive mechanisms to prevent the side effects of cysteine. Pyruvate has antioxidative capacity by neutralizing H2O2, as does catalase. It was found in experiments to be reported here that both pyruvate and catalase prevent cysteine toxicity. In order to identify which mechanism may represent the in vivo situation, pyruvate concentrations and activities of antioxidative enzymes, including catalase and glutathione peroxidase, will be assayed in A C M and CSF. 30 Chapter 1 Introduction From these experiments, a concerted mechanism of neuroprotection by astrocytes will be outlined, with a focus on small molecule antioxidants. The significance of the findings in neurological diseases will be discussed. 31 Chapter 2 Materials & Methods CHAPTER 2 MATERIALS AND METHODS M A T E R I A L S 1. Animals Both pregnant and adult Long-Evans rats were obtained from Charles River (Laval, QC). Pregnant rats were bred until the embryos were 18 day old. 2. Chemicals and Reagents The following items were obtained from Boehringer Mannheim (Mannheim, Germany): DNase I and biotin-16-dUTP. The following items were purchased from GIBCO BRL (Grand Island, NY): fetal bovine serum (FBS), 0.25% trypsin solution, Hank's balanced salt solution (HBSS), 10 mM Dulbecco's phosphate buffered saline (PBS), terminal deoxynucleotidyl transferase, recombinant (rTdT) and 5x TdT buffer. The following items were made by Molecular Probes (Eugene, OR): Calcein AM, ethidium homodimer, Hoechst 33258, and 7-hydroxycoumarin-3-carboxylic acid (7-OHCCA). The following items were obtained from Sigma (St. Louis, MO): cysteine (CSH), glutathione (GSH), glutathione disulfide (GSSG), glutathionesulfonic acid (GA), glutamic acid (Glu), aspartic acid (Asp), CysGly, cysteic acid (CA), homocysteic acid (HCA), sodium pyruvate, sodium lactate, cupric sulfate, cuprous chloride, ferric chloride, ferrous sulfate, manganese sulfate, chromium chloride, zinc chloride, hemin, 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), 0.4% trypan blue solution, iodoacetic acid, l-fluoro-2,4-dinitrobenzene (FDNB), 5,5'-dithiobis(2-nitrobenzoic acid) (DTNB), 3,3'-diaminobenzidine tetrahydrochloride (DAB), isopropanol, dithiothreitol, 2-mercaptoethanol, tert-butyl hydroperoxide (t-BuOOH), hydrogen peroxide (H2O2), catalase, glutathione peroxidase, glutathione reductase, lactic dehydrogenase (LDH), p-nicotinamide adenine dinucleotide reduced form ((3-NADH), p-nicotinamide adenine dinucleotide phosphate reduced form (P-NADPH), anti-human 32 Chapter 2 Materials & Methods glial fibrillary acidic protein (GFAP) polyclonal antibody from rabbit, Earle's balanced salts (EBSS), Eagle's Minimum Essential Medium (MEM), cystine-free Eagle's M E M (M-2289), phenol red-free Eagle's M E M (M-4144), insulin, transferrin, selenium, fibronectin, and PLL. The following items were obtained from Vector Labs (Burlingame, CA): biotinylated anti-rabbit IgG from goat, A B C kit (avidin-biotinylated horseradish peroxidase complex. Cysteine-glutathione disulfide (CSSG) was purchased from Toronto Research Chemicals (North York, ON). Coumarin-3-carboxylie acid (CCA) was obtained from Aldrich (Milwaukee, WI). Anti-rat neuronal specific enolase (NSE) polyclonal antibody from rabbit was made by Polysciences (Warrington, PA). 40% formaldehyde solution was from B D H (Toronto, ON). 3. Equipment C O 2 water-jacketed incubator was manufactured by NuAire (Plymouth, MN). The inverted microscope with phase contrast and fluorescence was made by Nikon (Tokyo, Japan). Orbit shaker was produced by Lab-Line (Melrose, IL). HPLC was performed using the 712 Gilson gradient system from Gilson Medical Electronics (Middleton, WI). The 3-amino propyl ion-exchange column, with a particle size of 5 um and dimensions of 4.6 x 200 mm, was obtained from CEL Associates (Houston, TX). Fluorescent measurement was accomplished using the luminescence spectrometer Model LS 50B from Perkin Elmer (Buckinghamshire, England). The Angiocath catheter was obtained from Becton Dickinson (Sandy, UT). Centricon-3 was purchased from Amicon (Beverly, MA) . Tissue culture dishes were manufactured by Corning (Corning, NY) . Glass coverslips were produced by FGR-Steinmetz (Vancouver, BC). Microfilament glass was made by Frederick Haer (Brunswick, ME). METHODS 1. Preparation of Solutions and Media 33 Chapter 2 Materials & Methods Tissue Culture: • Brain dissecting buffer: HBSS was supplemented with NaHC03 (0.375 g/1), glucose (5 g/1), HEPES (3.57 g/1), and penicillin/streptomycin (100 units/ml of each). This buffer was Ca /Mg free. • Poly-L-lysine (PLL) solution: PLL was dissolved in dH 2 0 as stock solution (1 mg/ml) and stored at -20°C. The working solution was prepared by 1:10 dilution of the stock PLL in 0.15M, pH 8.4 sodium tetraborate solution. • DNase I solution: DNase I was dissolved in HBSS as stock solution (1 mg/ml) and stored at -20°C, and the final concentration was 0.1 mg/ml. • Basal M E M : M E M was supplemented with glucose (33 mM), glutamine (2 mM), NaHC0 3 (26 mM). • Chemically defined M E M (CD-MEM): basal M E M was supplemented with insulin (10 mg/1), transferrin (5.5 mg/1), sodium selenite (5 ug/1), and pyruvate (1 mM). • 10% fetal bovine serum-supplemented M E M (10% FBS-MEM): basal M E M was supplemented with 10% (v/v) FBS. HPLC and Others: • m-cresol purple solution (0.2 mM): m-cresol purple was dissolved in dH 2 0, and stored in a dark container. • Iodoacetic acid solution (100 mM): it was dissolved in 0.2 mM m-cresol purple solution, and stored at 4°C for at least 7 days. • 1 -Fluoro-2,4-dinitrobenzene (FDNB) solution (1% v/v): it was dissolved in ethanol, and stored at 4°C in a dark container. • DTNB stock solution (10 mM): DTNB was dissolved in methanol, and stored at 4°C for at least 2 weeks. • Hemin stock solution (10 mM): hemin was dissolved in 50 mM NaOH, and stored at 4°C. • The stock solutions of transition metal ions (100 uM to 10 mM): these transition metals, including FeS0 4, FeCl 3 , CuCl, CuS0 4 , MnS0 4 , CrCl 3 , and ZnCl 2 , were dissolved in dH 2 0. They were freshly prepared before use. 34 Chapter 2 Materials & Methods astrocytes. The astrocytes were identified by morphology and GFAP immunochemical staining (Fig. 2-1). The confluent astrocyte cultures in the flasks were ready for the preparations of A C M and subcultures, and were maintained in 10% F B S - M E M . 2.1.2. Preparation of A CM Two- to six-week astrocyte-enriched cultures in the flask were used to make A C M . The cultures were first rinsed with Hank's solution 3 times. Twenty milliliters of basal M E M was then added to each flask. The medium was collected 24 hr later and filtered-" through a 0.45-pm pore-size filter. The A C M was stored at 4°C for use. The attachment-improving effect of the A C M can be observed for at least 3 months. To prepare A C M for HPLC analysis, serum-free M E M was added to the confluent astrocyte cultures at a concentration of 1.33 ml per 1 x 106 cells (10 ml per flask). The density of confluent astrocytes was ~1 x 105 cells/cm2. The A C M was collected 24 hr later. 2.1.3. Subculturing Astrocytes onto Coverslips For the preparation of astrocyte feeder layer, the confluent astrocytes in the flasks were further subcultured onto coverslips. Two- to six-week-cultured astrocytes were first rinsed with Hank's solution three times, then digested with 5 ml of 0.25% trypsin for 5 min. The flask was gently shaken to help detach the cells. The dissociated cells were collected into a tube and centrifuged at 300 x g for 2 min. The supernatant was discarded. The pellet cells were rinsed with 10% F B S - M E M . Repeat this step once and suspend the cells in 10% F B S - M E M . The dissociated cells were plated onto the PLL coated coverslips at a density of 4 x 104 cells/cm2 in 10% F B S - M E M . After 1 week, the subcultured astrocytes become confluent and ready to use. 2.2. Tissue Culture of Primary Neurons 2.2.1. Coating the Cover si ips with PLL and A CM To make the surfaces of coverslips hydrophilic were required for neuronal cultures. The coverslips were immersed in concentrated nitric acid overnight. The next day, the 36 Chapter 2 Materials & Methods coverslips were fully washed in tap water for 1 hr and distilled water for 15 min in a glass container. The coverslips were autoclaved at 121 °C for 15 min and dried in an oven. The working solution of PLL was prepared in borate solution before use. The coverslips were incubated with 0.5 ml of PLL solution for 30 min, then rinsed with dri^O three times and dried in the flow culture hood for 0.5 hr. The dried PLL coated coverslips can be used subsequently or stored at 4°C for several weeks. For astrocyte cultures, the dissociated cells in the serum-supplemented medium can be plated directly onto the PLL-treated coverslips or flasks. The procedures for coating flasks with PLL were the same as those of coverslips. For neuronal cultures, further coating the coverslips with A C M was required. Before plating the cells, each PLL-treated coverslip was incubated with 0.5 ml of A C M for 5 min. Then, the A C M was withdrawn. The coverslips were rinsed with HBSS twice and aspirated. Alternatively, serum-supplemented medium and fibronectin can replace A C M to coat PLL-treated coverslips. The procedures were similar to that of A C M . The coverslips were coated with PLL and dried. Before plating the cells, the coverslips were covered with 0.5 ml of 10% F B S - M E M each for 5 min, and then rinsed with HBSS twice and aspirated. Fibronectin is a plasma protein and is often used to promote cell attachment. The coating procedures with fibronectin (40 pg/ml) were the same as for A C M and serum-supplemented medium. 2.2.2. Preparation of Cell Suspensions E-18 Long-Evans rat embryos were used. The cerebral cortices were taken and placed in Ca 2 +/Mg 2 +-free brain dissecting buffer. The meninges were removed. The tissue was cut into small pieces with a pair of scissors. The cortical tissue was transferred to a tube with 5 ml of 0.25% trypsin and 0.1 mg/ml DNase mixed solution. The tissue was digested for 15 min in a 37°C water bath. The tube was gently shaken every 2 min. After digestion, 1 ml of ice-cold FBS was added to inactivate the trypsin activity. The suspension was then centrifuged at 300 x g for 2 min. The supernatant was discarded. The pellet was resuspended in basal M E M and gently dispersed with the pipet five times. It then stood in ice for 5 min. This step allowed the undigested small tissue debris to deposit at the bottom. The cell suspension was carefully transferred to a new tube. The 37 Chapter 2 Materials & Methods suspension was again centrifuged and the supernatant was discarded. The pellet was resuspended in C D - M E M . The suspended cells were counted in a hematocytometer and the concentration adjusted as needed. 2.2.3. Culturing Primary Cortical Neurons The dissociated cells (0.5 ml) in C D - M E M were plated onto the coverslips, previously coated with PLL and A C M . The cell density was 4 x 104 cells/cm2 or 1 x 105 2 cells/cm , according to different situations. The cultures were maintained in a humidified incubator of 5% CO2 and 95% air. For different purposes, some neuronal cultures were used directly, and some others were co-cultured with astrocyte feeder layer. 2.3. Co-cultures of Neurons with Astrocyte Feeder Layer Two hours after plating the dissociated neurons, 2.5 ml of C D - M E M was added to each culture dish (35 mm diameter x 10 mm high). Then a glass holder was inserted and the astrocyte feeder layer was applied facing down. As shown in Fig. 2-2, the two coverslips bearing primary neurons and confluent astrocytes respectively were separated by a V-shaped glass holder of 5 mm height. The medium was changed twice a week. The glass holder was used to support the astrocyte feeder layer in the co-culture dish. It was made from microfilament glass. The thin glass tube was heated and bent into a glass holder (Fig. 2-2). First, the middle part of glass tube was heated over a gas flame and bent to a " V " . After cooling, the top of the V-shaped tube was again heated and allowed to fall down a little bit to form a foot of the holder. Lastly, the two ends of the Figure 2-2. A side view of the glial-neuron non-contact co-culture situation. The primary cortical neurons are plated on the bottom coverslip. The confluent astrocytes are attached to the top coverslip facing down. A V-shaped holder separates the two coverslips at a height of 0.5 cm. As shown on the right side, the holder is bent from a thin glass microfilament. It contains two arms and three feet. It is removable and can be reused. 38 Chapter 2 Materials & Methods tube were heated and bent at a 90° angle to form the two other feet. The holder has two arms and three feet, with a height of 5 mm. It is removable, and can be used repeatedly after sterilization. When preparing neuron-enriched cultures, the dissociated cells were first plated onto the bottom coverslip. After adding 2.5 ml of medium, the glass holder was seated on the bottom coverslip, and then the coverslip that carried the confluent astrocyte feeder layer was put on top of the glass holder facing down. 3. Detection of Neuronal Death and Apoptosis 3.1. Trypan Blue Staining Trypan blue staining was used to count the viable cell number. While live cells do not take up trypan blue, dead cells do. Before counting the cells, 0.2 ml of 0.4% trypan blue solution was added to 1 ml serum-free medium. If medium contains serum, it requires changing the medium to serum-free medium. Trypan blue was incubated with cell cultures for 15 min. Trypan blue negative cells represent the live cells and were counted. Three random fields in each dish were counted, and the average was calculated. Neuronal viability was expressed as a percentage of the control, usually the average number at DIV 0. This method was used in the experiments of Chapter 3. Later, the MTT method was applied to measure the cell viability. 3.2. MTT Colorimetric Assay of Neuronal Viability The MTT method was modified from Denizot and Lang (1986). The tetrazolium salt MTT is reduced into a blue formazan by the mitochondrial enzyme succinate-dehydrogenase (Slater et al., 1963), and the amount of formazan produced is proportional to the number of living cells. The cultured neurons were incubated with 0.5 mg/ml of MTT in phenol red-free M E M . After a 3 hr incubation at 37°C in a humidified atmosphere of 5% C O 2 , the medium was removed, and 1 ml of propanol was added to each dish to solubilize the formazan. The optical density was measured at 560 nm with 690 nm as a reference (OD56o-69o)- The cell viability was normalized as a percentage of control. 39 Chapter 2 Materials & Methods 3.3. TUNEL Staining The terminal deoxynucleotidyl transferase (TdT)-mediated d U T P - b i o t i n n ick end labeling ( T U N E L ) assay is an in situ detection o f D N A fragments in apoptotic cells (Gavriel i et al . , 1992). Th is method is based on the specific binding o f T d T to 3 ' - O H ends o f D N A . T d T catalyzes template-independent addition o f labeled deoxyuridine triphosphate ( d U T P ) , wh ich is then visual ized using immunohistochemical stains. The procedures are as fol lows: cultured neurons on coverslips were rinsed with P B S 3 times, and fixed with 4% paraformaldehyde overnight at 4 ° C . After that, 3 % H 2 0 2 was used to inactivate intrinsic peroxidase activity for 5 min . The cells were rinsed with Tr is Buffered saline ( T B S ) 3 times. T h e n the fo l lowing reaction mixture was added onto the coversl ips: 80 units o f T d T , 54 ul o f T d T buffer, 10 ul o f 1 m M d U T P , and 200 ul d H 2 0 . The mixture was incubated at 3 7 ° C for 1 hr. After r insing the cells with T B S , add avidin-biotinylated horseradish peroxidase complex ( A B complex) for 30 m i n at room temperature. Af ter r insing with T B S , the cells were stained with D A B . The mixture included 2.8 m g D A B , 12 ul o f 3 0 % H 2 0 2 , and 4 m l T B S . The cells were then counterstained with methylgreen for 10 min . F inal ly , the cells were dehydrated in a series o f gradient ethanol solutions, and mounted. 3.4. Fluorescent Probes Estimation o f cell death was also assisted with fluorescent dyes. Ca lce in A M is an esterase substrate that is hydrolyzed to a green fluorescent product, calcein. It serves as a viabil ity probe that measures both esterase activity and cel l -membrane integrity. E th id ium homodimer is a red fluorescent dye that has high affinity to nucleic acids. It can only pass through the compromised membranes, and serves as a dead indicator. Hoechst 33258 is cell-permeant and D N A binding stain that fluoresce bright blue upon binding to D N A . It stains nuclei o f both l ive cells and dead cells. Because o f this property, Hoechst 33258 can be used to compare the morphological differences o f normal cells f rom apoptotic cells, wh ich is featured by chromatin condensation. Ca lce in A M and ethidium homodimer cannot be used to determine the differences o f cell death, i.e., apoptotic or necrotic. These probes were dissolved in D M S O as 1 m M stock solutions, 40 Chapter 2 Materials & Methods and stored at -20°C. They were added to culture medium to a final concentration of 10 uM, and observed under fluorescence microscope. 4. HPLC Assay of Thiols, Disulfides, and Related Compounds Cysteine and related compounds were analyzed using the HPLC method developed by Reed and co-workers (Reed et al., 1980; Fariss and Reed, 1983) with minor modifications. In principle, thiols of the sample are first reacted with iodoacetic acid to block free thiol groups, and then with fluoro-dinitrobenzene (FDNB) to produce N -dinitrophenyl (DNP) derivatives, which are analyzable by spectrophotometric detection at 365 nm. This method is capable of evaluating thiols, disulfides, and acidic amino acids, etc. The components of the sample are determined by comparing the retention time of each peak with that of standard chemicals. In our procedure, 200 ul of the sample was first reacted with 25 ul of 100 mM iodoacetic acid in 0.2 mM m-cresol purple and 25 (al of N a H C 0 3 (0.24 M) / NaOH (0.12 M) buffer for 30 min. Then, 225 ul of 1% (v/v) FDNB in ethanol was added, and the mixture was stored at 4°C overnight. Thereafter, 25 ul of 1 M lysine was added to eliminate unreacted FDNB and the sample was then ready for analysis. In cases in which only a small amount of sample was available, such as CSF, the volumes of the above reactants were adjusted proportionately according to the sample amount. HPLC solvent A was 80% methanol in water. Solvent B was 0.8 M sodium acetate in 64% methanol (for preparation of solvent B, refer to Reed et al., 1980). 20 ul of the samples was injected into HPLC column. The mobile phase was maintained at 80% A, 20% B for 5 min followed by a gradient elution to 1% A, 99% B over 10 min, and held for 5 min. The flow rate was 1.5 ml/min. The retention values for several important compounds in our experiments are shown in Table 2-1. Table 2-1. HPLC retention times for thiols, disulfides and acidic amino acids. Chemicals Retention time (min) Glutamic acid (Glu) 6.29 Cystine (CSSC) 7.78 41 Chapter 2 Materials & Methods Homocysteic acid (HCA) 8.48 CysGly 9.32 Cysteic acid (CA) 9.47 Aspartic acid (Asp) 9.80 Cysteine (CSH) 10.75 Cysteine-glutathione disulfide (CSSG) 13.13 Glutathione (GSH) 14.23 Glutathione disulfide (GSSG) 16.02 The listed standard chemicals are reacted with iodoacetic acid and l-fluoro-2,4-dinitrobenzene to produce N-dinitrophenyl derivatives, which are spectrophotometrically detected at 365 nm. The HPLC mobile phase starts with 5 min of isocratic elution at 80% A, 20% B, followed by 10 min of gradient elution to 1% A, 99% B, and held at 1% A, 99% B for 5 min. The flow rate is 1.5 ml/min. The concentrations of cysteine and related compound in the samples were calculated from the peak area on the basis of external standards. Two injection loops were used. For 20 pi injection volume, the calculating equations are: thiol concentration (uM) = Peak Area (arbitrary unit)/1054, and disulfide concentration (uM) = Peak Area (arbitrary unit)/2108. The minimal detectable concentrations were 0.50 uM for thiols and acidic amino acids, and 0.25 uM for disulfides. For 100 pi injection volume, thiol concentration (uM) = Peak Area (arbitrary unit)/5017, and disulfide concentration (pM) = Peak Area (arbitrary unit)/10034. The minimal detectable concentrations were 0.10 uM for thiols and acidic amino acids, and 0.05 p M for disulfides. 5. Spectrophotometric Assay of Cysteine with Ellman's Reagent (DTNB) This is an alternative method for quantitating cysteine. It is based on the thiol/disulfide reaction of thiol and DTNB, a disulfide, liberating the chromophore 5-mercapto-2-nitrobenzoic acid (Ellman, 1959; Hu, 1994). The advantage of this reagent is that its reaction with thiols is faster (in seconds) than iodoacetic acid (in minutes), which is employed in our HPLC analysis. Though it cannot detect multiple components of thiols and disulfides, it is suitable for simple systems, such as the study of cysteine autoxidation in pure solutions. The reaction mixture included transition metal ions and 42 Chapter 2 Materials & Methods cysteine in PBS. After the reaction, 0.1 ml of DTNB was added to 0.9 ml of the reaction mixture. It was measured against a reference of 1 mM of DTNB in PBS. An additional blank control containing all components except cysteine was evaluated to correct the absorption of transition metal ions. Cysteine concentrations were measured from absorbance at 412 nm, and calculated on the basis of cysteine standards: cysteine concentration (uM) = 80.0 x Absorbance. 6. Immunohistochemical Staining Anti-NSE and anti-GFAP antibodies were used to stain primary neuron cultures and astrocyte cultures respectively. The procedures for NSE staining are as follows: 1). The coverslips bearing cultured neurons were washed in PBS and fixed with 4% paraformaldehyde in PBS for 10 min at room temperature. 2). Endogenous peroxidase activity was blocked by incubating the coverslips with 1% H2O2 solution for 5 min. 3). Non-specific binding was blocked by incubating with 5% normal goat serum for 30 min. 4). The cultures were incubated with primary anti-NSE antibody (1:2000) in PBS for 2 h at room temperature. 5). After removing the primary antibody and rinsing with PBS, the cultures were incubated with biotinylated goat anti-rabbit IgG (1:800) for 1 h at room temperature. 6). The cultures were incubated with avidin and biotinylated horseradish peroxidase complex (1:1000) for 1 h at room temperature and rinsed 3 times with PBS. 7). The cells were then stained with 1 ml of nickel DAB solution, which was composed of 49.5 ml Nickel solution, 0.5ml 2% DAB and 20 (al 1% H 20 2 , for 10 min. Nickel solution is prepared from 0.5 g nickel ammonium sulphate, 2.5 ml of 1 M imidazole, and 47 ml of 0.05 M Tris (pH 7.6). 8). The cultures were dehydrated in graded ethanol solutions, and transferred to xylene. Mounted the coverslips onto glass slides with a polyvinyl alcohol-based mountant. For the anti-GFAP immunostaining of astrocyte cultures, the primary anti-GFAP antibody was diluted 1:200. The procedures were the same as above. 7. Enzymatic Assays 43 Chapter 2 Materials & Methods 7.1. Glutathione Peroxidase Glutathione peroxidase (GPx) activity was determined using a method modified from Takahashi et al. (1987). GPx activity was assessed from the oxidation of N A D P H in the presence of glutathione reductase and oxidized glutathione formed by GPx. The reaction mixture included Tris-HCl (0.1 M , pH 8.0), N A D P H (0.2 mM), EDTA (0.5 mM), GSH (2 mM), GSH reductase (1 U/ml), sample (100 ul), and t-BuOOH (100 uM). The total volume was 1 ml, and t-BuOOH was added last. The oxidation of N A D P H was monitored against a reference mixture without sample and t-BuOOH. A negative control containing all components except the sample was used to correct the oxidation of GSH and N A D P H by t-BuOOH. The samples were adult rat CSF and A C M . The latter was 1 Ox concentrated A C M prepared by filtering A C M in a centricon with 3 K D molecular-weight cutoff. Calculation of GPx activity was based on the consumption of N A D P H at A 3 4 o (ENADPH = 6220 M " 1 cm"1 at 340 nm) (Fig. 2-3). Reaction mixture: Tris-HCl: 0.1 M NADPH: 0.2 mM GSH: 2mM EDTA: 0.5 mM GSH reductase: 1 units t-BuOOH: 100 uM Sample: CSF or ACM. The decrease of NADPH is measured at A 3 4 0 . Figure 2-3. Schematic mechanism of assay of glutathione peroxidase (GPx). 7.2. Catalase Catalase activity was measured by monitoring the absorbance change of H2O2 (Duffy et al., 1998). The reaction mixture included PBS, EDTA (0.5 mM), H 2 0 2 (10 mM), and 100 ul of sample. The total volume was 1 ml. The reference contained all components except the sample. The samples were adult rat CSF and the concentrated A C M as same as above. Absorbance was monitored at 240 nm from 300 to 600 s. The minimal detectable enzyme activity of this test was 5 mU/ml. One unit of enzyme activity was defined as the amount of the enzyme that decomposes 1 umole of H2O2 per minute. Mechanism: N A D P N A D P H GSH GSSG t-BuOOH H 2 0 44 Chapter 2 Materials & Methods 8. Procedures of Obtaining CSF and Plasma 8.1. CSF CSF was obtained from the rat cerebellomedullary cistern. Three-month-old male Long-Evans rats were anesthetized with a mixture of ketamine (80 mg/kg) and xylazine (5 mg/kg). The rat was placed in a stereotaxic frame. The skin was incised along the midline over the occipital crest, and the muscles were separated. A puncture through occipital foramen magnum was made with an Angiocath catheter. About 100 pi of CSF was slowly drawn over 2 min. The CSF was centrifuged at 300 g for 2 min to remove a tiny amount of contaminating blood cells. To examine the cysteine autoxidation rate in CSF, 10 pi of 1 m M cysteine solution was added to 90 pi of CSF, to reach a final concentration of 100 uM. The original cysteine concentration in the CSF (-1.12 uM) can be ignored. The CSF was incubated at 37°C in a humidified atmosphere with 5% CO2 and 95% air. 8.2. Plasma of Artery and Vein For obtaining the blood samples, the anesthetized 3-month-old rat was laid on its back. The carotid artery or internal jugular vein was carefully exposed by separating the surrounding muscles and connective tissues. A vessel puncture was made with the Angiocath catheter. About 1 ml of blood was drawn into the heparin rinsed syringe. The blood was then centrifuged at 300 x g for 2 min. Plasma was taken and processed for HPLC assay. 9 . Fluorimetric Assay of Hydroxyl Radicals Production of OH was estimated by using coumarin-3-carboxylic acid (CCA). Nonfluorescent C C A was converted by -OH to highly fluorescent 7-hydroxycoumarin-3-carboxylic acid (7-OHCCA) (Collins et al., 1994a). For measuring -OH produced during autoxidation of 100 uM cysteine in the presence of 0.2 uM C u 2 + , the reaction mixture was prepared as follows in sequence: 0.885 ml of PBS, 0.1 ml of 10 m M C C A (final concentration: 1 mM), 10 pi of 10 mM cysteine, and 5 pi of 40 u M C u 2 + . The solution 45 Chapter 2 Materials & Methods was left at 37°C and 100% air to finish the reaction completely. After 4 hr incubation, the samples were measured using a fluorescence spectrometer, with excitation wavelength of 400 nm and emission of 450 nm. C C A (1 mM) in PBS was used as the reference. A standard curve was calculated by measuring the fluorescence intensities of a series of concentrations of 7-OHCCA. The produced OH from cysteine autoxidation was represented by the corresponding 7-OHCCA concentrations. 10. Pyruvate Assay Pyruvate in A C M , N C M , and the CSF was measured using an enzymatic method described by Von Korff (1969). Lactic dehydrogenase (LDH) catalyzes the conversion of pyruvate to lactate with N A D H : pyruvate + N A D H U lactate + N A D + . The reaction mixture was prepared as follows: 198 pi of PBS, 100 pi of 1 mM N A D H , 100 pi of the sample, and 2 pi of 0.4 units/ul L D H . The decrease of N A D H was monitored at 340 nm with a spectrophotometer for 300 s until a constant value was obtained. Pyruvate concentration was calculated from the oxidized N A D H (ENADH = 6290 M " 1 cm"1 at 340 nm). 11. Statistical Analysis The data are analyzed by using analysis of variance (ANOVA). The post hoc tests with Fisher's PLSD to determine significant differences between means of individual groups are followed when necessary. 46 Chapter 3 Neuronal Survival and Attachment CHAPTER 3 EFFECTS OF ASTROCYTES ON NEURONAL SURVIVAL AND ATTACHMENT SHOWN IN A CO-CULTURE SYSTEM INTRODUCTION It is our aim to discover the laws of nature, which are hidden behind phenomena. A solid fact may constitute a basis of further exploratory work. My experiments started from tissue cultures of embryonic cortical neurons. The initial results were not satisfactory. Repeated experiments showed that the cultured neurons hardly survive more than several days. Increasing the plate density of the dissociated cells does improve neuronal survival. It was noticed that the cells died rapidly during the first several days even in the high-density cultures in 10% FBS supplemented M E M . After DIV 3 or 4, the flat, polygonal glial cells appear and start to proliferate, forming isolated islands. The cultured neurons aggregate around these glial islands, and the number of neurons tends to be stable. The glial cells proliferate until a confluent monolayer is formed within two weeks. The neurons grow on top of this layer, and survive for several weeks. The neurosupportive effects of astrocytes have long been recognized (Banker, 1980; Walsh et al., 1992; Sass et al., 1993). These pioneering researchers have designed different methods to make use of this property of astrocytes to culture neurons. Neurons can be directly plated on an already confluent astrocyte feeder layer (McCaffery et a l , 1984; Baughman et al., 1991). In this method, neurons are grown with astrocytes, and cannot be manipulated separately. Another disadvantage is that one cannot tell whether neurons benefit from a direct attachment to astrocytes or from the factors released by astrocytes. The non-contact co-culture method has been developed to solve these problems. The coverslip bearing the neurons is dotted with paraffin drops as supporters, and inverted against a confluent glial cell feeder layer (Goslin and Banker, 1991); Or astrocytes and neurons are co-cultured in transwells (Walsh et al., 1992); Or two coverslips bearing primary neurons and astrocytes respectively can be put together in a larger dish (Sass et al., 1993). These methods have been tried, and there existed various 47 Chapter 3 Neuronal Survival and Attachment problems. For example, the height of paraffin dots was so low that gas exchange in the space between the two coverslips was severely compromised. It was also inconvenient to sterilize and repetitively use the coverslips. For transwell co-culture method, it was hard to subculture astrocytes into the inside plastic membrane. For the method of leaving two coverslips bearing astrocytes and neurons respectively in one big dish, the neurosupportive effects of astrocytes were not obvious, speculating that released factors by astrocytes were diluted in the relatively large volume of medium. In these co-culture methods, the media used are either serum-supplemented medium (Goslin and Banker, 1991; Sass et al., 1993) or N 2 chemically defined medium (Walsh et al., 1992). The disadvantage in the use of serum-supplemented medium is that the complicated components of serum make it difficult to analyze released factors of astrocytes. Also, addition of antimitotic agents was required to inhibit the proliferation of non-neuronal cells. Likewise, N 2 chemically defined medium contains many additives (Bottenstein and Sato, 1979), which make it hard to differentiate the neurosupportive effects of astrocytes from that of supplemented components. In this part of experiments, I optimized the astrocyte-neuron non-contact co-culture method. An astrocyte feeder layer was used to promote neuronal survival. A specific glass holder was devised to support the coverslip bearing the astrocyte feeder layer in the co-cultures. A simple chemically defined medium was used in this co-culture method. In addition, the neurosupportive effects of astrocytes were studied from another perspective, i.e. neuronal death induced by deprivation of astrocyte feeder layer. Neuronal survival was estimated after removal of the co-cultured astrocyte feeder layer. Morphological changes of the whole neurons and their nuclei were observed during the process of cell death. TUNEL staining was used to estimate the apoptotic cells in situ. Healthy attachment of the dissociated cells onto the surfaces of glass coverslips or petri dishes is the prerequisite of neuronal cultures. Usually, the dissociated cells were suspended in serum-supplemented medium, and plated in the PLL coated coverslips. Serum is thought to contain factors to help neuronal attachment (Barnes and Sato, 1980; Romijn et al., 1984). However, serum proteins seldom penetrate the B B B . It is assumed that the true "attachment factors", which exist in the extracellular matrix of the CNS, are 48 Chapter 3 Neuronal Survival and Attachment Astrocyte feeder layer Control 0 1 2 3 4 5 7 14 days in vitro Figure 3-1. Effects of astrocyte feeder layers on neuronal viability. E-18 rat embryo cortical neurons were cultured in C D - M E M at the density of 4x l0 4 cells/cm2. Results show that co-culture with the astrocyte feeder layer substantially improved neuronal survival. Control cultures were grown with the blank coverslips. The medium was changed twice a week. Neuronal viability was quantified by counting the number of trypan blue negative stained cells and expressed as a percentage of the average number at DIVO. Each point represents the mean ± S E M of five dishes. * p < 0.001 compared with control; ** p < 0.02 compared with control. produced inside. Therefore, whether astrocytes release the factors that can promote neuronal attachment was investigated. RESULTS 1. Effects of Astrocytes on Neuronal Survival The effects of astrocytes on neuronal survival were observed by co-culturing primary neurons with astrocyte feeder layer in chemically defined M E M (CD-MEM). The cell density is 4 x 104 cells/cm2. The co-culture method has been described in Materials and Methods 2.3. A V-shaped glass holder was made (see Fig. 2-2) to support astrocyte feeder layer. The C D - M E M was M E M supplemented with glucose (33 mM), glutamine (2 mM), insulin (10 mg/1), transferrin (5.5 mg/1), sodium selenite (5 jag/1), and pyruvate (1 mM). Blank coverslips, instead of the astrocyte feeder layers, were put on top of neurons as controls. As shown in Fig. 3-1, neuronal survival by co-culturing with astrocyte feeder layer was compared with that of blank controls. While the neurons of 49 Chapter 3 Neuronal Survival and Attachment cell death was assessed by TUNEL staining and morphological observations. Cell body observation under amplified phase-contrast microscope showed the cell shrinking and blebbing of the plasma membrane (Fig. 3-4A, B). Apoptotic neurons were revealed in situ by TUNEL staining (Fig. 3-4C, D). Nuclear staining with Hoechst 33258 revealed chromatin condensation (Fig. 3-5). These results are consistent with the characterization of apoptotic cells. 3. Effects of Astrocytes on Neuronal Attachment Neuronal attachment was observed by examining neuron shape, neurite outgrowth, and 24 hr cell viability after plating. The cell suspensions in serum-free C D - M E M were plated onto the surfaces of coverslips treated with different methods. The neurite outgrowth after cell plating is assessed in terms of percentage of neurons, each of which has one or more extended neurites longer than 10 urn, an arbitrary length equal to about one neuronal diameter, (a) For the coverslips treated only with PLL, most of the cells at 4 hr after plating showed round bodies, the plasma became flat and transparent under the phase-contrast microscope (Fig. 3-6A). The neurons had few extended neurites, and died within 24 hr of plating (Table 3-1). (b) For PLL- and ACM-treated coverslips, the neurons showed a polygonal cell body shape with one or more neurites extended (Fig. 3-6B). There was a high proportion of cells with neurites of more than 10 urn long at 4 hr, and the survival was high 24 hr after plating (Table 3-1). (c, d) After P L L precoating, further alternative treatments were tried, including serum-supplemented medium (10% FBS-MEM) and fibronectin (40 ug/ml). These coating methods also produced healthy neuronal attachment as shown by polygonal cell bodies and extended neurites (Fig. 3-6C, D). Neurite length and 24-hr survival revealed no obvious differences compared with that of ACM-coated coverslips (Table 3-1). (e) If the coverslips were coated only with A C M without precoating with PLL, most of the neurons did not adhere to the coverslips (Fig. 3-6E). Only a few neurons extended neurites and the 24-hr survival was low (Table 3-1). (f) I also prepared cell suspensions in 10% F B S - M E M and plated them onto PLL-coated coverslips. The results showed healthy attachment using the assay of 4-hr neurite outgrowth and 24-hr survival rate (Fig. 3-6F and Table 3-1). 52 Chapter 3 Neuronal Survival and Attachment These experiments provide several methods to plate dissociated neurons. Coating the PLL-pretreated coverslips with A C M , or serum-supplemented medium or fibronectin, can all produce healthy attachment and neurite outgrowth. The same effects can also be obtained by plating cell suspensions in serum-supplemented medium, which is frequently employed in many serum-free culture protocols. Table 3-1. Effects of different coating methods on neurite outgrowth and cell viability Coating methods Culture mediums Neurons (%) with one Cell viability of cell suspensions or more neurites (> 10 nm) (4hr) (%) (24 hr) A. PLL only C D - M E M 4.4 ± 0.9 a 0C B. PLL+ACM C D - M E M 84.7 ± 2.3 84.5 ±5 .4 C. PLL+10%FBS-MEM C D - M E M 82.7 ± 1.8 86.6 ± 4.3 D. PLL+Fibronectin C D - M E M 79.5 + 3.1 83.0 ±6.4 E. A C M only C D - M E M . 17.0 ± 4.0 b 4.7 ± 2.2d F. PLL only 10%FBS-MEM 84.2 ±2.5 81.6 ± 5.8 The dissociated cells were suspended in serum-free chemically defined MEM (CD-MEM) (A-E) or 10% FBS-MEM (F). Different coating methods were used as shown above. 4 hr after plating, neurite outgrowth was measured by counting the number of neurons with one or more neurites longer than 10 urn, and expressed as a percentage of total cells. Cell viability was assayed by measuring trypan blue-negative stained live cells at 24 hr, and expressed as a percentage of D1V0 cell number. The results were means ± SEM. a'V<0.001 v s . g r 0 U p B (PLL+ACM); c'dp<0.001 vs. group B (PLL+ACM). No obvious differences were found among group B, C, D and F. D I S C U S S I O N This co-culture system shares with other astrocyte-neuron co-culture methods the advantage that primary neurons can be cultured at low density and with high viability. In addition, my protocol possesses the following characteristics: (a) A removable glass holder was employed, which can be made easily from a glass microfilament, to separate the two coverslips in a culture dish. Therefore, the coverslips can be handled 54 Chapter 3 Neuronal Survival and Attachment overflow of the culture dish, require extra medium unnecessarily, and lower neuronal viability, (c) A simplified formula of C D - M E M , which is supplemented with insulin, transferrin, selenium, and pyruvate, was developed. It is simpler than N 2 and many other chemically defined media, while it maintains a high neuronal survival. A simple formula will facilitate the analysis of astrocyte-released factors. In addition, pure neuronal populations can be obtained because glial cells do not proliferate in this serum-free medium. More than 95% of the cells in this culture are neurons, as identified by neuron specific enolase (NSE) immunostaining. No mitotic inhibitors are required with this method. This experiment has generated in vitro evidence that neuronal survival is highly astrocyte-dependent. The mechanisms by which astrocytes support neuronal survival remain unclear though extensive studies have been done in this field. It has been found that astrocytes can play diverse functions. They can optimize the composition of extracellular fluids, by regulating extracellular potassium (Newman, 1995), inactivating neurotransmitters (Martin, 1995), and secreting bioactive factors, such as amino acids, substrates of energy, neuropepetides, and neurotrophic factors (Vernadakis, 1988; Martin, 1992; Schmalenbach and Muller, 1993). In addition, astrocytes have been found to protect neurons from oxidative stress (Langeveld et al., 1995; Desagher et al., 1996; Hochman et al., 1998; Tanaka et al., 1999). The neurosupportive effects of astrocytes may reflect a synergistic action of multiple factors. Each factor may exert different influence according to particular circumstances. As noticed in my initial experiments, the cultured primary neurons die rapidly during the first several days in vitro in serum-supplemented medium. Then glial cells start to proliferate, and neurons tend to aggregate around the islands of glial cells. This phenomenon could be explained by the neurosupportive effect of astrocytes. The cultured neurons are very fragile due to the insufficient support from the small number of glial cells at the beginning of cultures. Once the glial cells, mainly astrocytes, proliferate and grow to a confluent layer, the cultured neurons obtain sufficient support for survival and live in a healthy state. Apoptosis has been known to be a form of cell death distinct from necrosis, and the term was first coined more than 20 years ago (Kerr et al., 1972). Apoptosis has been suggested to be involved in the abnormal neuronal death that occurs following axonal 56 Chapter 3 Neuronal Survival and Attachment injury (Edstrom et al., 1996; Kogawa et al., 2000) or in neurodegenerative diseases (Nishimoto et al., 1997; Desjardins and Ledoux, 1998; Tatton and Olanow, 1999). While necrotic cell death is typified by rapid cell swelling and lysis, apoptotic cell death is characterized by cell shrinkage, membrane blebbing, and nuclear condensation (Kerr et al., 1972; Wyllie et al., 1980). A variety of strategies have been applied to induce experimental apoptosis, including treatment with oxidants (Slater et al., 1995), hypoxia (Beilharz et al., 1995), heat shock (Fairbairn et al., 1995), glutamate (Ankarcrona et al., 1995), calcium (Trump and Berezesky, 1995), glucocorticoids (Schwartzman and Cidlowski, 1994), hormone withdrawal (Gould and McEwen, 1993), and growth factor withdrawal (Raff et al., 1993; Collins et al., 1994b). In the present experiment, neuronal apoptosis was induced by withdrawal of the astrocyte feeder layer. This can be used as a model of neuronal apoptosis. The mechanisms of cell death in this model can be attributed to the loss of diverse neurosupportive effects of astrocytes. This apoptosis model induced by withdrawal of astrocytes may reflect the physiological and pathological conditions in vivo. Cautionary notes have been reported about the TUNEL staining for the detection of apoptosis, showing that TUNEL assay is non-specific for apoptosis (Charriaut-Marlangue and Ben-Ari, 1995; Grasl-Kraupp et al., 1995). In addition, more investigations have focused on the mechanistic studies of cell death, other than the differentiation between apoptosis and necrosis in recent years. In my further experiments, apoptosis is no longer emphasized. It has significant physiological meaning that A C M helps neuronal attachment. The factors in A C M may represent the true extracellular matrix proteins surrounding neurons in vivo. In many serum-free culture protocols, the dissociated cell suspensions are first plated in serum-supplemented medium, which is then replaced by serum-free medium several hours or 1 day later (Bottenstein and Sato, 1979; Romijn et al., 1984; Pardo et al., 1997). Serum contains fibronectin or other attachment factors (Barnes and Sato, 1980), which may contribute to this function of serum. However, these serum proteins cannot penetrate the blood-brain barrier, and therefore may not reflect the in vivo condition of neuronal growth. As for the unknown "attachment factor" in A C M , our initial analysis with an HPLC size exclusion column suggests that it has a high molecular weight (> 670 57 Chapter 3 Neuronal Survival and Attachment kDa). My observations reveal that plating dissociated cells in serum-supplemented medium has a similar effect on neuronal attachment and 24 hr cell viability as simply coating the coverslip with A C M , serum-supplemented medium, or fibronectin. This result supports the hypothesis that the critical role of serum played in the first-step culture procedure is simply due to its functions in neuronal attachment and neurite outgrowth. Therefore, the procedure of plating cells in serum-supplemented medium can be replaced by the simple coating with A C M , which is closer to the in vivo situation. It is also shown that PLL is indispensable for A C M , serum, or fibronectin, etc., to exert their attachment-improving effects, though PLL itself cannot produce a healthy attachment of neurons. This positively charged polymer is presumed to assist the firm coating of those attachment factors to the surface of the coverslip. 58 Chapter 4 Astrocytes Provide Cysteine to Neurons CHAPTER 4 PROVISION OF CYSTEINE BY ASTROCYTES TO NEURONS BY WAY OF RELEASING GLUTATHIONE I N T R O D U C T I O N In order to study the neurosupportive mechanisms of astrocytes, I have tested many chemicals and growth factors in neuronal cultures. The neurosupportive effects were observed with the supplements of cc-tocopherol or cysteine. Addition of ct-tocopherol to the culture medium improved neuronal survival, and the results were stable. The results obtained from cysteine supplements, however, were unstable. Addition of cysteine to the culture medium could be beneficial or cytotoxic, varying with cysteine concentrations. Also, cysteine was toxic in basal M E M , while non-toxic in D M E M . These contradictory data implicate that the role of cysteine is complicated. Cysteine is a very special amino acid. Unlike other amino acids, cysteine is very unstable owing to autoxidation under aerobic conditions. It is also the rate-limiting precursor for the synthesis of GSH (Beutler, 1989), which is the major cellular antioxidant. The extracellular abundance of thiol-containing compounds substantially influences intracellular GSH levels (Meister, 1989). Neurons, in particular, require cysteine, instead of cystine, to synthesize GSH (Kranich et al., 1996). In the following experiments, the mechanisms of cysteine maintenance in the CNS were investigated. The mechanisms by which a stable cysteine level is maintained by astrocytes have been explored in recent years. Release of thiols by astrocytes has been reported and is considered to play an important role to increase neuronal GSH synthesis (Yudkoff et al., 1990; Sagara et al., 1993; Sagara et al., 1996; Dringen et al., 1999a). However, it is still uncertain as to how thiols are provided. Yudkoff et al. (1990) found GSH in conditioned medium of astrocytes. Sagara et al. (1993) reported that glial cells can release cysteine to the cultured medium to supply neurons. In a later report (Sagara et al., 1996), they found GSH efflux from cultured astrocytes. Recently, Dringen et al. (1999a) suggested that the ectoenzyme y-glutamyl transpeptidase uses GSH released by astrocytes as a substrate to generate the dipeptide CysGly that is subsequently used by neurons as a precursor for 59 Chapter 4 Astrocytes Provide Cysteine to Neurons GSH synthesis. In the present experiments, I tried to determine whether astrocytes release cysteine, or GSH, or both. What is the relationship of cysteine and GSH? What mechanism represents the in vivo situation? I have analyzed cysteine, GSH and other related compounds in A C M , neuron conditioned medium (NCM), CSF, plasma of the carotid artery and internal jugular vein of rat using HPLC methods developed by Reed et al. (1980). These methods are capable of evaluating thiols (cysteine, GSH, homocysteine, CysGly), disulfides (cystine, glutathione disulfide, cysteine-glutathione disulfide), and acidic amino acids (glutamic acid, aspartic acid, cysteic acid, homocysteic acid), etc. This advantage makes it possible to analyze the relationships among these compounds. These experimental results suggest that astrocytes release GSH directly, and that cysteine is generated by an extracellular thiol exchange reaction: Glutathione (GSH) + Cystine (CSSC) U Cysteine (CSH) + Cysteine-glutathione disulfide (CSSG). This reaction can occur non-enzymatically (Jocelyn, 1967). The results show that astrocytes can provide small molecules to enhance GSH synthesis, and therefore the antioxidative capacities of neighboring neurons. RESULTS 1. Cysteine-glutathione Disulfide (CSSG) Found in Cystine-containing Astrocyte Conditioned Medium I first analyzed the components and concentrations of cysteine and related compounds in A C M . M E M was added to the confluent astrocyte cultures at a concentration of 1.33 ml / lx lO 6 cells (10 ml per flask). As shown in Fig. 4-1, the control medium contains mainly cystine (0 hr in culture dishes). Analysis of the control medium after it has been in the incubator for 48 hr shows no difference with that at 0 hr. Several new peaks appeared in A C M compared with control medium. Three of them were clearly identified as cysteine, CSSG and GSH. The finding of CSSG in A C M was meaningful. The direct efflux of CSSG is unlikely, because no CSSG was detected in astrocytes (see Table 4-3). On the other hand, extracellular CSSG can be produced when GSH reacts with cystine non-enzymatically: GSH + CSSC ^ CSH + CSSG. 60 Chapter 4 Astrocytes Provide Cysteine to Neurons o'.OO 8'.OO lo'.OO 16 '.00 2o'.00 Time (min) Figure 4-1. HPLC graphs of cysteine and related compounds in control medium, A C M and the standards. Control medium was serum-free cystine-containing M E M . The A C M was made by adding the culture medium to the confluent astrocytes for 24 hr. The peaks in the control medium and A C M were identified by referring to the standards. The graphs show that: A . Control M E M contains mainly cystine. B. Several new peaks besides cystine were found in 24 hr A C M . Three of them can be identified as CSH, CSSG and GSH. C. The standards were CSSC, CSH, CSSG, GSH and GSSG. The concentration of each chemical is 10 uM. The peak area of disulfide is about double that of thiol. AUFS, absorbance units of full scale. 61 Chapter 4 Astrocytes Provide Cysteine to Neurons At 37°C and pH 7.4, the equilibrium constant K of the above reaction is 3.2 and velocity constant k is 610 liter/moles x min (Jocelyn, 1967). That means the reaction can proceed forward rapidly. The appearance of CSSG in A C M leads us to suggest that cysteine may not be released, but instead is produced extracellularly by the reaction of GSH, which is released by astrocytes, and the stable amino acid cystine. This suggestion is consistent with the fact that GSH, not cysteine, is maintained at a high intracellular level. CysGly was not detected in the A C M . Time of incubation (hr) Figure 4-2. Time course of cysteine and related compounds' content in A C M . Confluent astrocytes in flasks were rinsed with Hank's solution 3 times. The serum-free cystine-containing M E M was added at a concentration of 1.33 ml/1 x 106 cells. Samples of the A C M were taken at 0, 6, 12, 24 and 48 hr, and processed immediately for HPLC analysis. The calculated values of the equilibrium constant K for the equation, GSH + CSSC ±5 CSH + CSSG, are as follows: K, 2 h=4.66, K24h=4.19, K48h=4.57. Data are shown as mean ± S.E.M. of five determinations. The changes in concentrations of cysteine and related compounds in A C M were measured as shown in Fig. 4-2. Cystine decreased with time of incubation, with a beginning concentration of 76.9 ±3.4 uM. Cysteine and CSSG increased rapidly from zero, then decreased when cystine content diminished at 24 hr and 48 hr. GSH was barely detectable at 6 hr. This can be explained by the high concentration of cystine in the medium. The theoretical value of GSH calculated from the K value (3.2) and concentrations of cystine, cysteine and CSSG is 0.07 uM at 6 hr, well below the minimal 62 Chapter 4 Astrocytes Provide Cysteine to Neurons detectable level, 0.5 uM. Therefore, GSH may not be detected in this situation. The values of the equilibrium constant K calculated from Fig. 4-2 are 4.66 at 12 hr, 4.19 at 24 hr, and 4.57 at 48 hr. A small amount of GSSG was found at 48 hr. GSSG may be generated by two possible mechanisms: autoxidation of GSH, and the thiol/disulfide exchange reaction. The latter can also occur non-enzymatically (Jocelyn, 1967): CSSG + GSH ±5 CSH + GSSG. 2. GSH Released by Astrocytes in Cystine-free Medium In order to obtain direct evidence for GSH release from astrocytes, I analyzed the conditioned medium from cystine-free M E M . Cystine-free M E M (M-2289) was added to the confluent astrocyte cultures at a concentration of 0.66 ml / lx lO 6 cells (5 ml per flask). In this case, the extracellular thiol/disulfide exchange reaction between cystine and GSH can be excluded. As shown in Table 4-1, cysteine and CSSG were no longer found, while GSH was the only component detected. This result indicates that there is no direct release of cysteine and CSSG by astrocytes. Astrocytes release GSH to the medium from an intracellular reservoir of GSH. Because there is no cystine as a source for GSH synthesis, the detected GSH concentration may not reflect the potential levels the astrocytes can release. The small content of GSSG is assumed to be the autoxidation product from its reduced form. The dipeptide CysGly was not detected in these experiments. Table 4-1. Content (uM) of cysteine and related compounds in A C M made from cystine-free medium. Time esse CSH CSSG GSH GSSG CysGly 2 h N D N D ND 0.66 ±0.11 N D N D 4 h ND N D N D 1.06 ±0.03 N D N D 6 h N D N D N D 1.43 ±0.09 N D N D 12h N D N D N D 3.18 ±0.09 0.32 ±0.19 N D The confluent astrocyte cultures were rinsed with Hank's solution 3 times, and the cystine-free MEM was added at a concentration of 0.66 ml/lxlO6 cells. The samples were collected at 2, 4, 6 and 12 hr, and 63 Chapter 4 Astrocytes Provide Cysteine to Neurons analyzed using HPLC methods. Data are shown as mean ± S.E.M (uM) of five determinations. ND, not detectable. 3 . Analysis of Cysteine and Related Compounds in CSF and in Plasma of Artery and Vein It is of course critical to determine whether the in vitro mechanisms studied above are applicable to the in vivo situation. Plasma of the carotid artery and of the internal jugular vein, which represent blood supply and return of the brain, and CSF were studied. Our results (Table 4-2) showed that the components of cysteine and related compounds in CSF and blood are surprisingly different. No cystine was detected in CSF, while it has a high concentration in plasma. Vein plasma clearly has a lower level of cystine than artery plasma. GSH showed a relatively high concentration in CSF, while it was not detected in plasma. Cysteine and CSSG were detected in both CSF and plasma, with a low level in CSF. No clear differences in the concentrations of cysteine and CSSG were detected in the artery and vein. GSSG was detected at a low concentration in CSF. No CysGly was detected in CSF, artery or vein. A supplementary experiment was done later on to analyze cysteine and related compounds in CSF and in plasma of artery and vein with enhanced sensitivity. The injection volume of the sample in HPLC analysis was changed from 20 pi to 100 pi. The detection sensitivity was correspondingly increased about 5 times. Three-month-old rats and the analytical procedures used the same as in the previous experiment. Some components, which were not detected with 20 pi injector, were found with 100 pi injector. In CSF, CSSC is 0.09 ± 0.02 (mean ± SEM) uM, ranging from 0.05 to 0.13 uM (n = 5). In plasma of carotid artery, GSH is 1.26 ± 0.24 uM, ranging from 0.16 to 1.68 uM (n = 5); GSSG is 0.19 ± 0.08 uM, ranging from 0.05 to 0.34 uM (n = 5). In plasma of internal jugular vein, GSH is 1.17 ± 0.31 uM, ranging from 0.65 to 1.71 p M (n = 5); GSSG is 0.20 ± 0.01 uM, ranging from 0.19 to 0.21 uM (n = 5). There are no significant differences of GSH and GSSG concentrations between artery and vein (p > 0.05). CysGly was not detected in CSF, artery and vein. The results of other components have no obvious differences from that listed in Table 4-2. 64 Chapter 4 Astrocytes Provide Cysteine to Neurons Table 4-2. Content (uM) of cysteine and related compounds in CSF, and in plasma of the carotid artery and internal jugular vein CSSC CSH CSSG GSH GSSG CysGly CSF Mean ± SEM N D b 1.12 ± 0.14 0.50 ± 0 . 0 8 5.87 ± 0 . 2 9 0.22 ± 0 . 1 2 N Db n=7 Range 0.68-1.75 0.28-0.83 4.92-7.11 0-0.80 Artery Mean ± SEM 35.2 ± 1.2 4.94 ± 0.69 3.17 ± 0.21 N D b N D b N Db n=7 Range 30.8-40.1 3.32-7.49 2.37-3.72 Vein Mean ± SEM 26.3 ± 0.9" 5.42 ± 0.79 3.06 ± 0 . 3 5 N D b N D b N D b n=5 Range 23.9-29.2 2.48-7.00 2.34-4.35 Three-month-old rats were used. CSF was taken from the cerebellomedullary cistern. Plasma of artery and vein was taken from the carotid artery and internal jugular vein respectively. All the samples were centrifuged at 300 x g for 2 min, and immediately processed for HPLC analysis. Values represent mean ± S.E.M. (uM) and concentration range with 5 to 7 samples as indicated. The contents of cysteine and CSSG have no difference between artery and vein. ND, not detectable. a p < 0.001, different from artery value. b, see results of supplementary experiments in text. 4 . Thiols not Found in Neuron Conditioned Medium Whether neurons can provide thiols by themselves was studied in order to determine whether astrocytes are the major releasers of thiols. Neuron-enriched cultures were prepared by using the glial-neuron co-culture method in serum-free medium as described in the methods. The final density is 1 x 105 cells/cm2. At DIV 7, the astrocyte feeder layers were removed. Neuron-enriched cultures were rinsed with Hank's solution 3 times. Cystine-supplemented or cystine-free M E M (M-2289) was added at a concentration such that the medium volume/cell number ratio was the same as that of A C M . As shown in Fig. 4-3, no cysteine or GSH was detected in cystine-supplemented N C M . Cystine levels showed no obvious change. Low levels of CSSG were found in the cystine-supplemented N C M , suggesting that there is still a tiny amount of GSH released by neurons and the extracellular thiol/disulfide exchange reaction, but the amount of thiols is too small to be detected. It cannot be excluded that residual glial contamination (<5%) of these cultures account for the GSH release observed. 65 Chapter 4 Astrocytes Provide Cysteine to Neurons 100 5 a. O </) (0 o o w (0 o 12 24 36 Time of incubation (hr) Figure 4-3. Time course of cysteine and CSSG content in N C M . Neuron-enriched cultures were prepared with the glial-neuronal co-culture method in serum-free medium. At DIV 7, the cultures were rinsed with Hank's solution 3 times. The serum-free cystine-containing M E M was added at a concentration of 1.33 ml / lx lO 6 cells, the same as medium volume/cell number ratio of A C M . CSH, GSH, GSSG and CysGly were not detected. Data are shown as mean ± S.E.M. of five determinations. N C M prepared from cystine-free M E M was also analyzed. The medium was added at a concentration of 0.66 ml / lx lO 6 cells. The samples were collected at 6 hr and 12 hr. The thiols and disulfides, including cysteine, cystine, CSSG, GSH, GSSG, and CysGly, were not detected in five determinations. 5. Contents of GSH and Related Compounds in Astrocytes and Neurons As the cross-membrane concentration gradient influences the kinetics of membrane transport, the intracellular contents of GSH and related compounds were investigated. Confluent astrocytes (2 to 6 weeks in culture) and neuron-enriched cultures (DIV 7) were analyzed. As shown in Table 4-3, there are high levels of GSH in astrocytes and neurons, with higher concentrations in astrocytes. Low levels of cysteine and GSSG were found in astrocytes, but not in neurons. GSH/CSH ratio is 11.8:1, and GSH/GSSG ratio is 81.5:1 in astrocytes. Cystine, CSSG and the dipeptide CysGly were not detectable in either astrocytes or neurons. In contrast with the relatively lower levels of GSH, neurons contain much higher levels of glutamate and aspartate than astrocytes (Table 4-3). Table 4-3. Content (nmol/mg of protein) of thiols, disulfides and acidic amino acids [glutamic acid (Glu) and aspartic acid (Asp)] in astrocytes and neurons 66 Chapter 4 Astrocytes Provide Cysteine to Neurons GSH CSH GSSG Glu Asp Astrocytes 59.1 ±6.0 5.02 ±0.38 0.73 ±0.12 33.7 ±6 .6 4.50 ±0.94 Neurons 50.8 ± 1.5 N D N D 194 ± 6 63.0 ±3 .6 The confluent astrocyte cultures and neuron-enriched cultures at DIV 7 were ruptured and lysed in dH 20 by freezing and thawing. Protein content was determined by the Bradford method. The samples were immediately analyzed by HPLC. The results represent mean ± S.E.M. (nmol/mg of protein) of five determinations. Cystine, CSSG and CysGly were not detected. ND, not detectable. 6. Autoxidation of Cysteine and G S H Cysteine and GSH are unstable under aerobic conditions. We assayed the autoxidation rates of cysteine and GSH, 100 uM each, in phosphate buffered solution. As shown in Fig. 4-4, cysteine was oxidized to cystine, and GSH oxidized to GSSG under conditions of 37°C, pH 7.4 and 100% air. A very small amount of cysteic acid (~ 1-2 uM) was also found as a byproduct of cysteine autoxidation. DISCUSSION In this study, I demonstrate the mechanism by which astrocytes provide cysteine to neurons indirectly by releasing GSH. 1). In the absence of cystine, as the results from cystine-free A C M show (Table 4-1), it is GSH, not cysteine, that is found in the A C M . This result indicates that GSH is the direct product released by astrocytes. Astrocytes release GSH from their intracellular reservoirs without requiring further supplies of thiol in the cystine-free medium. Hence, the GSH concentrations measured here under these conditions may not reflect the potential levels of secretion that astrocytes may reach. 2). In the presence of cystine, as the results from cystine-containing A C M show (Fig. 4-1 & 4-2), cysteine, CSSG and GSH were found in the A C M . Intracellular CSSG was not detected in astrocytes, and cysteine was present at much lower intracellular levels than GSH (Table 4-3). The high levels of CSSG and cysteine in the A C M cannot be explained by direct cellular release. However, extracellular CSSG and cysteine can be produced when GSH reacts with cystine non-enzymatically: GSH + CSSC ±5 CSH + CSSG. The reaction can rapidly proceed forward with an equilibrium constant K 3.2 67 Chapter 4 Astrocytes Provide Cysteine to Neurons S •3- 7 5 o W 50 CO 5 0 o I 25 o o M — • 0 • GSH • GSSG 0 6 12 24 48 Time(hr) B Figure 4-4. Autoxidation of cysteine and GSH. 100 uM of cysteine, or GSH, in phosphate buffered solution was incubated under conditions of pH 7.4, 37°C and 100% air. The samples were collected at 0, 6, 12, 24, and 48 hr, and analyzed by HPLC. Values represent mean ± S.E.M. of four determinations. A . Cysteine is oxidized to cystine. B. GSH is oxidized to GSSG. (Jocelyn, 1967). The data suggest that astrocytes release GSH directly, which then reacts with the stable amino acid cystine extracellularly to generate cysteine and CSSG (Fig. 4-5). It should be noted that CSSG does not accumulate in A C M (Fig. 4-2). I found that CSSG can replace cystine to support astrocyte survival, while astrocytes cannot survive in cystine-free medium without the addition of other thiols or disulfides. It is possible that CSSG might either be taken up by astrocytes directly or further react with GSH to generate cysteine and GSSG (Jocelyn, 1967). This hypothesis is supported by the results of transporter studies, and provides a new understanding of the physiological significance of GSH efflux. GSH efflux has been found in fibroblasts (Bannai and Tsukeda, 1979), hepatocytes (Fariss and Reed, 1983; Aw et al., 1986), and astrocytes (Sagara et al., 1996). Kinetic studies from astrocytes suggest that GSH efflux is carrier-mediated, and the Km value is 127 nmol/mg of protein (Sagara et al., 1996). GSH transporters of astrocytes are ion-independent, and net transport depends on the concentration gradient of GSH. Since the intracellular • CSH • esse 5 12 24 Time (hr) 48 3 - 75 CO 5 0 X CO o 25 68 Chapter 4 Astrocytes Provide Cysteine to Neurons Blood Vessel Astrocyte Neuron Figure 4-5. A model showing how astrocytes may provide cysteine to neurons. My data suggest that blood mainly supplies cystine to the brain. Cystine is taken up by astrocytes, and utilized to synthesize GSH. Astrocytes release GSH to the extracellular fluid, where GSH reacts with cystine to generate cysteine and CSSG non-enzymatically. Cysteine is then taken up by neurons for GSH synthesis. concentration of GSH is several orders of magnitude greater than the extracellular concentration, the transport of GSH operates as an efflux process. Cysteine transporters, in contrast, belong to the Na+-dependent system ASC (Kilberg et al., 1981; Franchi-Gazzola et al., 1982). Recently, two members of the ASC transporter family, ASCT1 and ASCT2, have been cloned and functionally studied (Arriza et al., 1993; Shafqat et al., 1993; Utsunomiya-Tate et al., 1996). Cysteine, as well as some other neutral amino acids, is taken up by cells with high affinities, coupling to the cotransport of Na + . These results suggest a cross-membrane movement of GSH efflux and cysteine influx. Consistent with the transporter data, I show that it is GSH, not cysteine, that is released by astrocytes. Cysteine is generated extracellularly, and then taken up by neurons for GSH synthesis. Cooper and Kristal (1997) have previously indicated that transfer of sulfur amino acids and peptides between astrocytes and neurons occurs, and that astrocytes play an important role in sulfur homeostasis in the CNS. It was found that astrocytes and neurons have completely different capacities in the utilization of cystine and cysteine. The uptake rate of radio-labeled cystine by astrocytes is much higher than that of neurons (Sagara et al., 1993). Kranich et al. (1996) found that cultured astrocytes could make use of cystine 69 Chapter 4 Astrocytes Provide Cysteine to Neurons to synthesize GSH to their maximal degree. Cultured neurons, instead, make use of cysteine, rather than cystine. Similar to these results, my data show that cystine was used rapidly within 48 hr in astrocyte-enriched cultures (Fig. 4-2), while cystine levels showed almost no change in neuron-enriched cultures (Fig. 4-3). (Note: the ratios of medium volume/cell number of A C M and N C M are the same). These results indicate that astrocytes have a strong reducing ability, converting cystine to cysteine in their cell bodies. This ability may account for the enhanced synthesis and release of GSH by astrocytes. These data show that astrocytes release substantial GSH in A C M , while neurons release little. Whether y-glutamyl transpeptidase (yGT) is involved in thiol generation is still unclear. Dringen et al. (1999a) found that adding the dipeptide CysGly, the product of yGT catalyzed reaction, to neuron-enriched primary cultures increases the GSH content of neurons. Acivicin, a yGT inhibitor, suppresses the astrocyte-mediated increase in neuronal GSH content. They suggested that the ectoenzyme yGT uses GSH released by astrocytes to generate CysGly that is subsequently used by neurons. However, no evidence thus far shows the existence of CysGly in astrocyte conditioned medium or CSF. The localization of yGT in the CNS is crucial for the study of its function. yGT is used extensively as a marker of brain microvessels. It is abundant in the choroid plexus (Tate et al., 1973). yGT positive cells include endothelial cells (Orte et al., 1999), pericytes (Risau et al., 1992), and the endfeet of astrocytes (Zhang et al., 1997). This type of localization is consistent with the suggested role of yGT in the transport of amino acids across the blood-brain barrier (BBB) (Meister and Anderson, 1983; Cooper and Kristal, 1997). However, it does not prove that y-GT can generate thiol in the extracellular space of the CNS. GSH released by astrocytes into the extracellular space may not reach the ectoenzyme yGT due to the restrictive distribution of this enzyme (only around brain microvessels). A recent experiment has shown the detection of several y-glutamyl derivatives of amino acids in the perfusates of brain slices (Li et al., 1999). This result could be explained by the existence of a large amount of exposed brain microvessels in brain slices. GSH released by astrocytes could react with the ectoenzyme y-GT on the surface of these microvessel cells, which may not represent the situation of 70 Chapter 4 Astrocytes Provide Cysteine to Neurons integral tissue. Further screening of y-glutamyl derivatives in conditioned medium and CSF is required to clarify the involvement of yGT in thiol generation. I am most interested in the applicability of the in vitro results to in vivo situations. By analyzing cysteine, GSH and related compounds in CSF, and in plasma of the carotid artery and internal jugular vein, the following conclusions are obtained. First, cystine is substantially transported from blood to the CNS. Both artery and vein contain high concentrations of cystine, and there is also an obvious positive arterio-venous (A-V) concentration difference, indicating an uptake by the brain. Second, blood GSH cannot be the thiol source of the brain. GSH concentrations were very low in plasma of the carotid artery (1.26 ± 0.24 uM) and internal jugular vein (1.17 ± 0.31 uM), while GSH has a relatively high level (5.87 ± 0.29 uM) in CSF. Third, the other substances considered, such as cysteine, CSSG and CysGly, are less likely to be the thiols or disulfides transported from blood to the brain. Cysteine and CSSG have no A - V differences, suggesting no net uptake across the BBB within the CNS. The dipeptide CysGly was not detectable in blood and CSF, nor was it found in A C M in our in vitro studies. Fourth, the thiol/disulfide exchange reaction occurs in the CNS. Cysteine, cystine, GSH, and CSSG were all found in CSF. Cystine in CSF was low (0.09 ± 0.02 uM), and was detected only with 100 pi injector. This result is consistent with the data reported by Perry (1975) (0.1 ± 0.2 uM of human CSF). Several discrepancies exist in the study of thiol and disulfide transport across the B B B . Wade and Brady (1981) found that little [35S]-cystine is taken up by the brain using a carotid injection technique. By contrast, Hwang et al. (1980) showed [ 3 5S]-cystine uptake by isolated brain capillaries. My finding of cystine levels in plasma is not unique, as Felig et al. (1973) observed a similar positive arterial-jugular venous concentration difference of cystine (9.0 + 4.2 pM) in humans. For cysteine and GSH, my data from plasma levels suggest no substantial uptake of cysteine and GSH by the brain, in contrast with the observation of [35S]-cysteine uptake (Wade and Brady, 1981) and [ 3 5S]-GSH uptake (Kannan et al., 1990). It is postulated that cysteine and GSH are permeable across the B B B , and that the movement is bidirectional. GSH is also important in the y-glutamyl cycle. However, their net transport across the B B B contributes little to the thiol/disulfide pool of the CNS. 71 CHAPTER 4 ASTROCYTES PROVIDE CYSTEINE TO NEURONS It should be mentioned that there exists a sharp gradient of GSH in different blood vessels. It has long been known that the liver is the major organ releasing GSH into blood (Anderson et al., 1980; Fariss and Reed, 1983). The hepatic vein plasma has the highest levels of GSH observed (26 uM) (Anderson et al., 1980). I found that GSH levels are very low in the carotid artery and internal jugular vein. The explanation for this could be that GSH released from the liver rapidly proceeds to the thiol/disulfide exchange reaction in blood. Due to its critical role in intracellular GSH synthesis, cysteine must be maintained at a stable level. However, cysteine is unstable and will be oxidized to cystine under aerobic conditions. Likewise, GSH and other thiols will also be oxidized to the corresponding disulfides. My experiments show that the stable levels of thiols in the brain are maintained by astrocytes. The choroid plexus might also play a role in the formation and recycling of GSH in the CSF (Anderson et al., 1989). Extensive studies have shown that astrocytes play diverse functions in maintaining the stable internal milieu of the brain. In this study, I show that astrocytes indirectly provide cysteine to neurons for GSH synthesis. These data indicate that astrocytes play a critical role in protecting neurons from oxidative stress. 72 Chapter 5 Cysteine Neurotoxicity and Copper C H A P T E R 5 NEUROTOXICITY INDUCED BY CYSTEINE AUTOXIDATION AND T H E C E N T R A L R O L E OF COPPER I N T R O D U C T I O N Though cysteine plays a pivotal role in regulating intracellular levels of GSH, its toxicity has long been noted, particularly to neurons. Extensive degenerative changes in the CNS are induced following subcutaneous injection of cysteine to newborn mice (Olney et al., 1972) and rats (Karlsen et al., 1981). Cysteine is also toxic to cultured neurons (Puka-Sundvall et al., 1995), hepatocytes (Saez et al., 1982), and kidney cell lines (Nath and Salahudeen, 1993). The underlying mechanisms of cysteine toxicity have been studied, and there are several theories. 1). Olney et al. (1990) found that N M D A antagonists can prevent cysteine toxicity, and that cysteine is a bicarbonate-sensitive excitotoxin. They suggest a direct toxicity of cysteine. 2). Some researchers reported that glutathione potentiates glutamate toxicity by modulating the redox site of the N M D A receptor-channel complex (Sucher and Lipton, 1991; Janaky et al., 1993; Regan and Guo, 1999). Like GSH, cysteine can also regulate redox status by participating in the thiol/disulfide exchange reaction, thus exerting an indirect toxic effect. 3). Cysteine autoxidation can generate free radicals, which are cytotoxic (Saez et al., 1982; Nath and Salahudeen, 1993). This last mechanism has not been studied in neurons. Because neurons are very vulnerable to oxidative stress, due to the high oxygen consumption of the brain, their high proportion of membrane polyunsaturated fatty acids, and the weak activities of their antioxidative enzymes (Makar et al., 1994), free radicals may play an important role in cysteine neurotoxicity in brain. In the present experiments, I explore the free radical mechanisms of cysteine toxicity. I investigate the influence of cysteine autoxidation, the generation of free radicals from cysteine autoxidation, and its neurotoxicity to cultured neurons. R E S U L T S 73 Chapter 5 Cysteine Neurotoxicity and Copper 1. Cysteine Autoxidation in the Presence of Transition Metal Ions To study the factors determining cysteine autoxidation, I first investigated the autoxidation rate of cysteine in the following three solutions: Earle's balanced salt solution (EBSS), cystine-free M E M (MEM-CSSC"), and MEM-CSSC" with cultured neurons. EBSS contains the basal salts of M E M . The medium of the primary cortical neurons was replaced by MEM-CSSC" before testing. Only a small amount of cell debris was present in the cell cultures. A l l these solutions were incubated at 37°C and pH 7.4 in a humidified atmosphere of 5% C O 2 . Cysteine was added to a final concentration of 100 pM. Cysteine and cystine were measured by HPLC at 0, 5, 15, 30, 60, and 120 min after cysteine was added to the three solutions. The results showed that cysteine was gradually oxidized to cystine. Cysteine concentrations in MEM-CSSC" and neuronal culture medium had no significant differences from the corresponding values in EBSS (p > 0.05) (Fig. 5-1 A). This result demonstrates that the components of culture medium and cultured cells have no obvious influence on cysteine autoxidation. Transition metal ions, such as Fe 3 + or C u 2 + , are considered to catalyze the oxidation of reducing agents, including thiol-containing compounds (Halliwell and Gutteridge, 1985). The effects of Fe 3 + and C u 2 + on cysteine autoxidation were measured under the same conditions as described above. Fe 3 + , at concentrations of 0.1, 1.0, and 10 uM, had no obvious effects on the autoxidation rate of cysteine, as assessed by comparing cysteine concentrations with the corresponding value in EBSS (p > 0.05) (Fig. 5-IB). There were also no obvious differences among the three different concentrations (p > 0.05). Our results showed that C u 2 + substantially accelerated the autoxidation rate of cysteine at concentrations between 0.1 to 10 uM (Fig. 5-1C). The efficacy of submicromolar levels of C u 2 + in catalyzing cysteine autoxidation has significant physiological meaning as these are well within the physiological range. The concentrations of loosely bound copper in human CSF have been reported to be in the range of 0.13 to 0.75 p M (Gutteridge, 1984). I compared the effects of variable oxidation valences of iron (Fe 2 + and Fe 3 +) and copper (Cu + and Cu 2 + ) on cysteine autoxidation. The effects of hemin, the low-molecular-weight complex of Fe 3 + , and some other biologically important transition metal ions, including M n 2 + , Cr 3 + , and Zn 2 + , on cysteine autoxidation were also studied. As shown in Fig. 5-2, cysteine autoxidation with Fe 2 + was faster than that with Fe 3 + at 74 Chapter 5 Cysteine Neurotoxicity and Copper concentrations of 100 and 200 uM, but was not significantly different at concentrations of 1 and 10 uM. The effects of Cu + and C u 2 + on cysteine autoxidation were not significantly different at concentrations of 0.01 to 1 uM. Hemin was more efficient in catalyzing cysteine autoxidation than Fe 2 + and Fe 3 + , but still much less efficient than C u + and C u 2 + . The concentration for hemin to catalyze the half oxidation of 100 uM cysteine in 60 min was -10 uM, while the concentration for C u + and C u 2 + was between 0.1 and 0.2 uM, about a 50- to 100-fold difference. M n 2 + was efficient at concentrations of 100 1 10 100 Fe(ll), Fe(lll), Hemin ( ( l M) 1 10 100 Mn(ll), Cr(lll), Zn(ll) (nM) 120 1000 1000 0.01 0.1 Cu(l), Cu(ll) ^ M ) Figure 5-2. Comparisons of cysteine autoxidation in the presence of several transition metal ions. A series of concentrations of FeSC»4, FeCl 3 , hemin, MnS0 4 , C rC l 3 , ZnCl 2 , CuCl, and CuS0 4 , were reacted with 100 uM cysteine respectively. The control was 100 uM cysteine without transition metal ions. The reaction mixtures were incubated in PBS at pH 7.4 and 37°C in a humidified 100% air. Cysteine concentrations were determined with Ellman's reagent after 60 min. A . Cysteine autoxidation with Fe 2 + , Fe 3 + (1 to 200 uM), and hemin (1 to 100 uM). B. Cysteine autoxidation with M n 2 + , C r 3 + , and Z n 2 + (1 to 200 uM). C. Cysteine autoxidation with C u + and C u 2 + (0.01 to 1 uM). Data are the mean ± S E M of three independent experiments in duplicate. *p<0.0\, significantly different from the control. abp<0.0\ vs. corresponding values of Fe J groups. 3+ 76 Chapter 5 Cysteine Neurotoxicity and Copper Figure 5-3. Cysteine autoxidation rates in CSF and comparisons with those in Cu -supplemented solutions. CSF was taken from 3-month-old rats as described in Methods. Cysteine was added to the CSF with a final concentration of 100 pM. The CSF was incubated under conditions of pH 7.4 and 37°C in a humidified atmosphere of 5% CO2 and 95% air. Samples were taken at several time points for HPLC assays. A . Time course of the concentrations of cysteine and related compounds in the CSF during cysteine autoxidation. B. Comparison of cysteine autoxidation in CSF and in C u 2 + -supplemented solutions. C u 2 + was prepared in EBSS at concentrations of 0.1 uM, 0.2 uM and 0.3 p.M, and incubated under the same conditions as the CSF. Data are the mean ± S E M of three independent experiments in duplicate. CSSC, cystine; CSSG, cysteine-glutathione disulfide. and 200 pM. No catalyzing effects of C r J + and Z n 2 + were observed at concentrations between 1 to 200 uM. 2. Cysteine Autoxidation in CSF Cysteine autoxidation in CSF was examined. As described in the methods, cysteine was added to rat CSF to a concentration of 100 uM. The samples were taken for test at 15, 30, 60, and 120 min after this addition. The results showed that cysteine was oxidized at a moderate rate in the CSF (Fig. 5-3 A). Most of cysteine was converted to cystine (-80%). The other small portion of cysteine participated in the thiol/disulfide 77 Chapter 5 Cysteine Neurotoxicity and Copper CSH (pM) Figure 5-4. Neurotoxic effects of cysteine in the presence of copper. Primary cortical neurons were cultured in serum-free M E M . Cysteine was added at concentrations as indicated in the presence of 0, 0.2, and 1.0 uM C u 2 + . Neuronal viability was estimated 24 hr later using the MTT assay. Results are expressed as the percentage of surviving neurons compared with control cultures (without addition of cysteine and Cu 2 + ) . Data represent the mean ± S E M of three independent experiments in triplicate. exchange reaction and formed mixed disulfides, such as cysteine-glutathione disulfide and cysteine-protein disulfides. I compared cysteine autoxidation in CSF and in Cu -supplemented solutions (Fig. 5-3B). The autoxidation rate of cysteine in CSF was roughly equal to that observed in solutions of 0.2 to 0.3 uM C u 2 + . The kinetics of cysteine autoxidation in Cu solution and in CSF, however, was different. Cysteine autoxidation in Cu2+-supplemented solution was a zero-order reaction, and that in CSF was a first-order reaction (Fig. 5-3B). In the discussion I propose that copper is likely the major catalyst in CSF for cysteine autoxidation. I chose 0.2 uM C u 2 + to mimic the cysteine autoxidation in CSF in the following toxicity experiments. 3. Neurotoxic Effects of Cysteine Cysteine neurotoxicity was investigated in the cell culture system. The effects of the autoxidation rate of cysteine, the amount of cysteine being oxidized, as well as cysteine concentrations on cysteine neurotoxicity were observed. In Fig. 5-4, primary neurons were cultured in the presence of a series of concentrations of cysteine (0, 10, 20, 50, 100, 200, 500, 1000, and 2000 pM) and in the absence or presence of copper (0, 0.2, and 1.0 78 Chapter 5 Cysteine Neurotoxicity and Copper uM) in 2 ml culture medium (MEM). Neurotoxicity was estimated 24 hr later by the MTT assay. The role of copper in this experiment is to modulate the autoxidation rate of cysteine. Without addition of cysteine, copper itself had no visible effect on neuronal survival. The neuronal survivals in the presence of copper and absence of cysteine were 99.6 ± 11.3% (mean ± S E M % , C u 2 + - 0.2 uM) and 105 ± 5% (Cu 2 + = 1.0 uM), no significant differences from the control 100 ± 1.5% (Cu 2 + = 0) (p > 0.05). Without C u 2 + , cysteine was toxic only at relatively high concentrations (EC5o -600 uM). With addition of C u 2 + , the toxic concentrations of cysteine decreased (EC50 -30 uM at 0.2 uM C u 2 + and EC50 -12 uM at 1.0 uM Cu 2 + ) . This result demonstrates that cysteine toxicity is closely related to its autoxidation rate. The effect of the total amounts of cysteine being oxidized on its toxicity was further investigated. C u 2 + (0.2 pM) was added for the purpose of catalyzing cysteine autoxidation. Cysteine (10 uM) was added to the culture medium either once, 5, or 10 times at a rate of once every 5 min, respectively. 10 uM cysteine was completely 120 & 100 ^ 2 80 O 60 O 40 "~ 20 0 1 2 3 4 5 Tim e (m in) 1 Control 1 1x 5x 1 10x Cu 2 * CSH 0.2uM 0.2nM 1x10nM 0.2nM 5x10nM 0.2nM 10x10nM Figure 5-5. Effect of total amount of cysteine being oxidized on neuronal survival. Primary cortical neurons were cultured in serum-free M E M . The concentration of C u 2 + was 0.2 uM. 10 u M of cysteine was added each time, and was added once or repetitively every 5 min for a total of 5 and 10 times. 10 p M of cysteine was completely oxidized within 5 min in the presence of 0.2 uM C u 2 + (upper right figure). Neuronal viability was estimated 24 hr later using the MTT assay. Results are expressed as the percentage of surviving neurons compared with control cultures (without addition of cysteine). Data represent the mean ± S E M of three independent experiments in triplicate. * p < 0.01, significantly different from the control. 79 Chapter 5 Cysteine Neurotoxicity and Copper oxidized in 5 min (Fig. 5-5). While a single dose of cysteine showed no obvious toxicity (91 ± 5.1% of viability), multiple additions of cysteine substantially increased cysteine toxicity. Neuronal survival was 91.5 ± 5.1% (mean ± SEM%,/? > 0.05 vs. control) with addition of 10 uM cysteine once, and decreased to 28.9 ± 4.4% (p < 0.01 vs. control) with 5 additions of 10 uM cysteine, and 12.6 ± 1.5% (p < 0.01 vs. control) with 10 applications. Note the repeated addition increased the total amounts of cysteine being oxidized, while the maximal concentration of cysteine never exceeded 10 uM at any one time. These results demonstrate that cysteine toxicity is closely related with the total amount of cysteine being oxidized. Taken together, these data suggest that cysteine autoxidation, rather than cysteine itself, is responsible for cysteine toxicity. 4. Generation of Hydroxyl Radicals During Cysteine Autoxidation Using coumarin-3-carboxylic acid (CCA) as a probe, I measured the production of •OH generated from cysteine autoxidation. The hydroxyl radical reacts with C C A to generate 7-hydroxycoumarin-3-carboxylic acid (7-OHCCA). Cysteine in PBS was incubated at 37°C and pH 7.4 for 4 hr. As shown in Fig. 5-6A, little -OH was produced from cysteine autoxidation without the presence o fCu 2 + . In the presence of 0.2 uM C u 2 + , hydroxyl radicals were generated in substantial amounts during cysteine autoxidation, and the amount of OH increased with cysteine concentration. I also compared the generation of OH with some other thiols. Glutathione, N-acetyl-cysteine, homocysteine, dithiothreitol, and 2-mercaptoethanol, can all generate -OH with C u 2 + as a catalyst to variable degrees (Fig. 5-6B). Disulfides (cystine and glutathione disulfide) and sulphur-containing amino acid (cysteic acid) did not generate -OH. DISCUSSION The biological role of cysteine is double-edged. It is a very important amino acid for the synthesis of glutathione, which is the major cellular antioxidant. Neurons, in particular, prefer to take up cysteine, rather than cystine, to synthesize glutathione 80 Chapter 5 Cysteine Neurotoxicity and Copper (Kranich et al., 1996). On the other hand, cysteine has been found to be cytotoxic. Subcutaneous injection of cysteine in newborn mice (1 mg/g weight) (Olney et al., 1972) and rats (1.2 mg/g weight) (Karlsen et al., 1981) caused extensive neuronal death. Cysteine was toxic to cultured hepatocytes (4 mM) (Saez et al., 1982), kidney cell lines (4 mM) (Nath and Salahudeen, 1993), and primary neurons (1 mM) (Puka-Sundvall et al., 1995). In my neuronal cultures, 1 mM cysteine decreased neuronal survival to 30.5% (EC50 -600 uM) in the absence of copper (Fig. 5-4), similar to other findings (Puka-Sundvall et al., 1995). Importantly, my results showed that cysteine toxicity was greatly increased in the presence of even submicromolar levels of copper, while copper itself was 0.04 50 100 150 C S H ( | iM) 200 Cu 0 Cu 0.2 11M 0.03 3 0.02 < o o X o I r-B 0.01 J 0 Figure 5-6. Generation of hydroxyl radical (OH) by the autoxidation of cysteine and other thiols. Cysteine and related compounds were incubated in PBS under the conditions of pH 7.4 and 37°C in a humidified atmosphere of 100% air. C C A (1 mM) was added to react with the generated -OH, producing 7-OHCCA. Fluorescence was measured 4 hr after the reaction began. A . Cysteine, at concentrations of 0 to 200 uM, was incubated in the presence or absence of 0.2 uM C u 2 + . B. Generation of -OH from the autoxidation of thiols. Thiols, disulfides and sulphur-containing amino acid (100 uM of each) were incubated with 0.2 uM C u 2 + . The thiols include cysteine (CSH), glutathione (GSH), N-acetyl-cysteine (NAC), homocysteine (HSH), dithiothreitol (DTT) and 2-mercaptoethanol (2-ME). The disulfides include cystine (CSSC) and glutathione disulfide (GSSG). The sulphur-containing amino acid is cysteic acid (CA). Data represent the mean ± S E M of three independent experiments in duplicate. 81 Chapter 5 Cysteine Neurotoxicity and Copper non-cytotoxic without cysteine (Fig. 5-4). Copper appears to function as a catalyst to accelerate cysteine autoxidation. Cysteine can be oxidized to form cystine, and donate an electron to an oxidant. O2 generally acts as an oxidant, accepting electrons one by one to form reactive oxygen species: superoxide anion (-02"), hydrogen peroxide (H2O2), and hydroxyl radical (OH). However, the electron transfer from cysteine to oxygen requires transition metal ions (Halliwell and Gutteridge, 1985), which can act as the electron carriers to catalyze cysteine oxidation. Our experiment showed that cysteine was oxidized to cystine very slowly in three kinds of solutions, including basal salt solution, culture medium and neuronal culture medium (Fig. 5-2A). Iron is generally considered the major transition metal ion in mediating the production of free radicals. My experiments show that copper is much more efficient in catalyzing cysteine autoxidation than iron and some other transition metal ions, such as manganese, chromium, and zinc. Variable valences of copper do not influence its catalyzing efficacy. For iron, the soluble low-molecular-weight Fe 3 + complex, hemin, is more efficient in catalyzing cysteine autoxidation than Fe 2 + and Fe 3 + . Whether the solubility causes the differences of catalyzing capacities of these iron forms is unknown. While FeCl 3 is soluble in d H 2 0 (74.4 g/lOOml), it is insoluble in buffered salt solution as dictated by the low solubility (K s p = 3.16 x 10"38) of Fe(OH) 3 formed at neutral pH. My FeCl 3 stock solution was prepared in d H 2 0 . When FeCl 3 is added into the final solution, the availability of FeCl 3 in the reaction mixture is accurate, though FeCl 3 will form deposits in the reaction solution. F e S 0 4 is soluble at neutral pH in the testing concentration range. However, the iron oxidation status changes between Fe 2 + and Fe 3 + in iron-catalyzed cysteine autoxidation. In addition, Fe 2 + will be oxidized to Fe 3 + in aerobic conditions. Therefore, though F e S 0 4 is dissolved in the reaction solution completely at first, it gradually forms deposits. This could explain why the catalyzing efficacy of F e S 0 4 is just moderately higher than that of FeCl 3 . The insolubility could be the factor hampering the catalyzing effect of both F e S 0 4 and FeCl 3 . However, even the catalyzing efficacy of hemin is still much lower than copper. My finding that copper is a more potent catalyst than iron is not unique. It has been reported that copper is more 82 Chapter 5 Cysteine Neurotoxicity and Copper efficient than iron in mediating paraquat toxicity (Chevion, 1988), and in catalyzing dialuric acid autoxidation (Munday, 1988). My data show that cysteine was oxidized at a moderate rate in CSF. The autoxidation rate of 100 uM cysteine in CSF is equivalent to that in the presence of-0.2 uM Cu (Fig. 5-3). According to different authors, total copper concentrations in human CSF are in the range of 0.22 to 1.7 uM (14.2 to 109 ug/1) (Kapaki et al., 1997; Jimenez-Jimenez et a l , 1998; Joergstuerenburg et al., 1999; Stuerenburg, 2000). CSF also contains some other physiologically important transition metal ions, such as iron, manganese, chromium, and zinc. Their concentrations in human CSF are 1.1 to 3.8 uM for Fe (Jimenez-Jimenez et al., 1998; LeVine et al., 1998), 12 to 60 nM for M n (D'Amico and Klawans, 1976; Kapaki et al., 1997; Jimenez-Jimenez et al., 1998), 0.28 p M for Cr (Aguilar et al., 1998), and 0.16 to 2.6 uM for Zn (Palm and Hallmans, 1982; Kapaki et al., 1997; Jimenez-Jimenez et al., 1998). The binding status of copper in CSF is of particular importance with regard to its catalyzing effect, because only protein-unbound or loosely-bound copper, which can form a low-molecular-weight complex with cysteine, can catalyze cysteine autoxidation (Halliwell and Gutteridge, 1985). The concentration of loosely bound copper detected in human CSF, ranges from 0.13 to 0.75 p M (Gutteridge, 1984). By comparing the efficient catalyzing concentrations of these transition metal ions and their physiological ranges, I hypothesize that copper is the major determinant influencing cysteine autoxidation rate in CSF. The kinetics of cysteine autoxidation in CSF was a first-order curve, different from the zero-order reaction in copper-supplemented solutions. The first-order kinetics may reflect the regulation of copper chelating status. It is quite possible that protein-copper binding is modulated by the redox status of the CSF. The mechanism of copper-catalyzed cysteine autoxidation has been extensively investigated (Cavallini et al., 1969; Munday, 1989; Kachur et al., 1999). It has been suggested that the intermediate cysteine-Cu complex is initially formed in a 2:1 ratio. Cysteine, as well as other low-molecular-weight thiol compounds, can donate electrons via catalysts. 0 2 generally acts as an oxidant, accepting electrons one by one to generate reactive oxygen species: 0 2", H 2 0 2 , and OH. 83 Chapter 5 Cysteine Neurotoxicity and Copper Like cysteine, other low-molecular-weight thiol compounds, including glutathione, N-acetyl-cysteine, homocysteine, etc., can generate -OH during copper-catalyzed autoxidation (Fig. 5-6B). In CSF, cysteine (1.12 ± 0.14 pM) and glutathione (5.87 ± 0.29 uM) are the major low-molecular-weight thiols. The constant autoxidation of these thiols will place neurons in a situation of oxidative stress if no mechanisms exist to remove the generated oxygen radicals. The hydroxyl radical, which has an extremely short half-life 10"9s (Pryor, 1986), is the major damaging radical. Once OH is produced, it rapidly attacks poly-unsaturated fatty acids to initiate the chain reaction of lipid peroxidation, as well as DNA, proteins and carbohydrates. H2O2 itself is stable and non-toxic, and is the one-step precursor of OH. Eliminating H2O2 can therefore block the formation of -OH. In the following experiments, I will explore the in vivo mechanisms preventing the cysteine neurotoxicity. 84 Chapter 6 Neuroprotective Effects of Pyruvate CHAPTER 6 NEUROPROTECTIVE EFFECTS OF PYRUVATE ON CYSTEINE TOXICITY AND T H E RELEASE OF PYRUVATE BY ASTROCYTES I N T R O D U C T I O N The above data show that the extracellular fluids of the CNS contain a certain level of cysteine, and cysteine will constantly generate reactive oxygen intermediates via an autoxidation process. Therefore, the organism must have developed certain mechanisms to remove the free radicals produced. H2O2 itself is stable and non-toxic, and is the one-step precursor of OH. Eliminating H2O2 can therefore block the formation of -OH. Though cytoplasmic enzymes, such as GPx and catalase, can eliminate membrane-permeable H2O2 intracellularly, H202-permeating the cell membrane itself is risky for cells. In addition, extracellularly generated H2O2 will be reduced to OH if it is not scavenged immediately. Therefore, it will be more efficient and beneficial i f extracellularly derived H2O2 can be eliminated in situ in the extracellular space. In this chapter, I will explore two possible extracellular protective mechanisms: an enzymatic mechanism and a small molecule antioxidant mechanism. Of the two enzymes, which can catalyze the degradation of H2O2, catalase is known to be located in subcellular particles known as peroxisomes (Houdou et al., 1991). High activities of catalase are present in liver, kidney and erythrocytes but there is very little in brain (Halliwell and Gutteridge, 1985). Glutathione peroxidase (GPx), however, exists ubiquitously in cytoplasm of all cell types. The secreted forms of GPx have also been found in plasma (Avissar et al., 1989) and lung extracellular fluid (Avissar et al., 1996). Here, I will investigate whether astrocytes can secrete catalase and GPx. Pyruvate, as well as other a-ketoacids, can react with H2O2 nonenzymatically, being converted to carbon dioxide and the carboxylic acid with one less carbon: R-COCOOH + H 2 0 2 -> R-COOH + H 2 0 + C 0 2 (Holleman, 1904; Bunton, 1949). Removing H 2 0 2 will prevent the formation of OH, which is the major damaging radical. The protective effects of pyruvate against oxidative stress in biological systems have been reported 85 Chapter 6 Neuroprotective Effects of Pyruvate (O'Donnell-Tormey et al., 1987; Desagher et al., 1997; Giandomenico et al., 1997). Pyruvate, and the enzymatic activities of glutathione peroxidase and catalase were examined in A C M and in the CSF. RESULTS 1. Effects of Catalase and Pyruvate on Cysteine Toxicity When cysteine is oxidized, 0 2 accepts electrons one by one, and reactive oxygen intermediates are produced. The hydroxyl radical is the major reactive oxygen species to cause tissue damage, and is generated from H2O2. It is therefore reasonable to expect that catalase, which decomposes H2O2, could prevent -OH generation from cysteine autoxidation, and thus reduce cysteine toxicity. Likewise, pyruvate would also be expected to have the same effect owing to its reactivity with H2O2. The effects of catalase and pyruvate on cysteine neurotoxicity were tested in our primary neuronal cultures. In the presence of C u 2 + (0.2 uM), cysteine (100 uM) was toxic to neurons (16.8 ± 2.8% of viability) (Fig. 6-1 A). Addition of catalase (10 units/ml) or pyruvate (1 mM) completely prevented cysteine toxicity, with 101 ± 4 and 104 ± 5% of viability, respectively. Pyruvate and lactate are both glucose metabolites, and important energy suppliers to neurons (Selak et al., 1985; Pellerin and Magistretti, 1994; Tsacopoulos and Magistretti, 1996). To demonstrate that the preventive effect of pyruvate against cysteine toxicity is due to its specific reactivity with H202, other than its energy supplying effect, sodium lactate was used as control. The results showed that lactate could not prevent cysteine neurotoxicity (Fig. 6-1 A). The protective capacity of pyruvate was dose-dependant. In the presence of 0.2 uM C u 2 + and 50 or 100 uM cysteine, increasing concentrations of pyruvate (up to 1 mM) produced a progressive enhancement of neuronal protection (Fig. 6-IB). 2. Effects of Catalase and Pyruvate on Generation of Hydroxyl Radicals Owing to their ability to remove H2O2, catalase and pyruvate are expected to prevent the formation of -OH generated from cysteine autoxidation. In the presence of 0.2 uM C u 2 + and 100 u M cysteine, the generation of -OH was measured in PBS at 37°C and pH 86 Chapter 6 Neuroprotective Effects of Pyruvate 120 -, 100 -o .2 +* 80 -> c ra o u 60 -c o o 40 -3 Q> 20 -Z 0 -u i Contr i C S H I Cat Pyr L a c C u 2 + 0 . 2 u M + + + + + C S H 1 0 0 u M + + + + Other addition - - Cat Pyr Lac B o 1 § ra « o ° «- s= a> Z CSH 0 •CSH 50nM • CSH 100nM 100 Pyruvate (\xM) 1000 Figure 6-1. The protective effects of catalase and pyruvate on cysteine neurotoxicity. Primary cortical neurons were cultured in serum-free M E M . Cysteine toxicity was induced by addition of 100 uM cysteine and 0.2 uM C u 2 + . Neuronal viability was estimated 24 hr later using the MTT assay. Results are expressed as the percentage of surviving neurons compared with control cultures (without addition of cysteine). A . Catalase (10 U/ml), pyruvate (1 mM) and lactate (1 mM) were added immediately before addition of cysteine and C u 2 + . Data represent the mean ± SEM of three independent experiments in triplicate. B. Dose-response curve illustrating the neuroprotective effect of pyruvate. Cysteine concentrations were 0, 50, and 100 uM, respectively. Data represent the mean ± SEM of three independent experiments in duplicate. * p < 0.01, significantly different from the control. 87 Chapter 6 Neuroprotective Effects of Pyruvate 7.4. Catalase and pyruvate substantially decreased -OH production, while lactate did not (Fig. 6-2A). Lactate even increased the OH production, and the reason for this is unknown. The inhibitory effect of pyruvate on OH production from cysteine autoxidation was dose-dependent (Fig. 6-2B). 3. Pyruvate Released by Astrocytes I explored two potential extracellular antioxidative mechanisms, which may prevent the toxic effects of cysteine autoxidation: (1) enzymatic mechanisms and (2) small molecule antioxidant mechanisms. For the former, I tested the activities of glutathione peroxidase (GPx) and catalase in A C M and CSF. The concentrated A C M prepared from 0.04 C S H Cat Pyr Lac s < o o X o I B 0.03 0.02 0.01 0.1 1 Pyruvate (mM) 10 Figure 6-2. Effects of catalase and pyruvate on the generation of -OH from cysteine autoxidation. Cysteine (100 uM) and C u 2 + (0.2 uM) were incubated in PBS under the conditions of pH 7.4 and 37°C in a humidified atmosphere of 100% air. C C A (1 mM) was added to react with the generated OH, producing 7-OHCCA. The fluorescence was measured 4 hr after the reaction. Data represent the mean ± SEM of three independent experiments in duplicate. A . Catalase (10 U/ml), pyruvate (1 mM), and lactate (1 mM) were added immediately before addition of cysteine and C u 2 + . B. Pyruvate, at concentrations of 0, 0.1, 1, and 10 mM, was added immediately before addition of cysteine and C u 2 + . *,**/?< 0.01, significantly different from the control (cysteine only). Chapter 6 Neuroprotective Effects of Pyruvate primary confluent astrocyte cultures was used as described in the Methods. CSF was taken from 3-month-old rats. The results showed that the activities of the two enzymes were not detected in A C M and CSF. GPx activity of A C M was 1.6 ± 0.2 mU/ml (mean ± S E M , n = 6), not significantly different from the negative control (1.9 ± 0.4 mU/ml, n = 6, p > 0.05). GPx activity of CSF was 2.2 ± 0.2 mU/ml (n = 6), not significantly different from the negative control (1.7 ± 0.2 mU/ml, n = 6, p > 0.05). Catalase activity was not detected in A C M and CSF (< 5 mU/ml, n = 6, respectively). As pyruvate can scavenge H2O2, it may act as an extracellular antioxidant in vivo. Pyruvate concentration was therefore assayed in A C M , N C M , and CSF. Relatively high levels of pyruvate were found in A C M (254 ± 15 uM) and in CSF (131 ± 9 pM). Pyruvate was also detected in N C M at relatively low concentrations (45.6 ± 4.4 uM) (Table 6-1), suggesting that pyruvate pool of the extracellular space and CSF was mainly contributed by astrocytes. The time course of pyruvate release by astrocytes was also measured (Fig. 6-3). In the A C M made from pyruvate-free medium, pyruvate concentration increased rapidly within 12 hr, reached a peak at 24 hr. Table 6-1. Pyruvate content in A C M , N C M , and CSF Pyruvate concentrations (juM) Mean ± S E M Range A C M (n = 6) 254 ± 15 184-297 N C M (n = 6) 45.6 ±4.4 36.9-62.3 CSF (n = 6) 131 ± 9 98.9-145 ACM was prepared from confluent astrocyte cultures, and NCM was from 7 D1V neuronal cultures in serum-free MEM. The conditioned media were collected 24 hr later for pyruvate assay. CSF was obtained from 3-month-old male rats. 4. Effects of Cysteine Provision on Neuronal Survival As cysteine is critical for GSH synthesis, it is essential to inhibit the neurotoxic effect of cysteine in order to exert its beneficial effects. I observed the effect of cysteine provision on neuronal survival in pyruvate-containing C D - M E M . Cysteine concentration was 20 u.M, and was added every 24 hr from the beginning of neuronal cultures. The 89 Chapter 6 Neuroprotective Effects of Pyruvate 300 Time (hr) Figure 6-3. Time course of pyruvate release by astrocytes. Confluent astrocytes in flasks were rinsed with Hank's solution. The serum-free M E M was added at a concentration of 1.33 ml per 1 x 106 cells. Samples of the A C M were taken at 0, 6, 12, 24, and 48 hr and used for pyruvate assays. Data represent the mean ± S E M of three independent experiments in triplicate. control was C D - M E M without cysteine. As shown in Fig. 6-4, addition of cysteine clearly increased the survival of cultured neurons as compared with controls. This beneficial effect of cysteine supplement, however, is still less efficient than that of astrocyte feeder layer, suggesting that neurosupportive mechanisms of astrocytes are much more complicated. DISCUSSION I have explored two possible extracellular H202-eliminating mechanisms, including enzymatic mechanisms and small molecule antioxidant mechanisms. The activities of the two most plausible antioxidative enzymes, GPx and catalase, were not detected in A C M or CSF. Pyruvate is generally considered as an energy substrate for cultured neurons, though its trophic effect on neuronal survival has long been reported (Selak et al., 1985; Katoh-Semba et al., 1988; Izumi et al., 1994; Matsumoto et al., 1994). Recently, pyruvate was reported to protect neurons (Desagher et al., 1997) or cell lines (Giandomenico et al., 1997) from H 20 2-induced toxicity. I found that pyruvate, as well as catalase, can inhibit the production of -OH generated from cysteine autoxidation, by removing H 2 0 2 . The neuroprotective effect of pyruvate cannot be attributed to its role in 90 Chapter 6 Neuroprotective Effects of Pyruvate pyruvate from astrocytes to the extracellular pool, then to neurons. In agreement with this, there are reports showing that astrocytes release pyruvate (Pellerin and Magistretti, 1994), and neurons can utilize pyruvate for their function recovery or survival (Selak et al., 1985; Matsumoto et al., 1994; Yoshioka et al., 2000). Given the evidence above, it is suggested that the extracellular pool of pyruvate is supplied mainly by astrocytes rather than neurons. It is still unknown whether other brain cells such as choroidal epithelial cells or ependymocytes contribute to the pyruvate pool. Another possible source of pyruvate is from blood. Pyruvate can be transported through the blood-brain barrier by the monocarboxylate transporter (Pardridge and Oldendorf, 1977; Miller and Oldendorf, 1986). However, the arterio-venous difference of pyruvate across the brain of adult fed rats was found to be negligible (arterial blood, 141 ± 14 uM; sinus blood, 138 ± 14 uM), whereas the arterio-venous differences for glucose was remarkable (510 ± 50 uM) (Hawkins et al., 1971). It suggests that though pyruvate is permeable across the blood-brain barrier, the net transport from blood to brain contribute little to the pyruvate pool of the CNS under normal conditions. Instead, the brain utilizes glucose and produces pyruvate on its own. Taken together, these findings lead to the set of mechanisms outlined in Fig. 6-5 as those underlying the neurosupportive effect of astrocytes against cysteine toxicity. It should be noted that cysteine and some other reducing agents play a dual role. They generate reactive oxygen species when they are oxidized under the catalysis of transition metal ions, functioning as prooxidants. Meanwhile, they can donate electrons to electrophilic substances, such as 02~ and OH, functioning as antioxidants. Due to the extremely short half-life of OH, these reducing agents cannot completely prevent the damage induced by -OH. That may explain why cysteine itself cannot prevent the toxicity induced by its autoxidation. Pyruvate, however, is not a reducing agent. It functions only as an antioxidant, but not prooxidant, by reacting with H2O2. From this point of view, pyruvate is a special and indispensable antioxidant of the CNS. It needs to be noted that though addition of cysteine can improve neuronal survival when neurotoxicity is prevented, this neurosupportive effect of cysteine addition is still far less than that of astrocyte feeder layer. Two reasons are considered: 1). Due to the limit of our techniques, it is impossible to keep a stable low level of cysteine by adding 92 Chapter 6 Neuroprotective Effects of Pyruvate Figure 6-5. Diagram of the proposed mechanism of protection by astrocytes in preventing cysteine toxicity catalyzed by copper. Astrocytes release glutathione and indirectly produce cysteine in the extracellular fluid of the CNS. Cysteine, as well as glutathione or other thiols, will be oxidized to disulfide under the catalysis of protein-unbound or loosely bound copper. Molecular oxygen, as the oxidant, accepts electrons step by step to produce superoxide radicals, hydrogen peroxide and hydroxyl radicals. The latter is the major damaging free radical to the cells. In parallel, astrocytes also release pyruvate, which can react with hydrogen peroxide, preventing the formation of hydroxyl radicals. cysteine to the culture dishes. Addition of a high dose of cysteine and the rapid autoxidation will produce a fluctuation of cysteine concentration. Considering the complex effects of cysteine, this difference may affect neuronal survival. 2). More likely, the neurosupportive effect of astrocytes must reflect a synergistic action of multiple factors. The antioxidative effect of astrocytes described here via small bioactive molecules may represent only one of the several mechanisms. Other factors, such as neurotrophins, hormones, energy substrates, uptake of excitatory amino acids, and regulation of pH and ions, may all contribute to the neurosupportive effect of astrocytes. Each factor may exert different influences according to special circumstances. 93 Chapter 7 General Discussion & Conclusions C H A P T E R 7 G E N E R A L D I S C U S S I O N A N D C O N C L U S I O N S G E N E R A L DISCUSSION I have studied the neuroprotective effect of astrocytes from the following perspectives: a) the neurosupportive effect of astrocytes in a co-culture system, b) the mechanism of thiol maintenance in the CNS, c) the effect of cysteine neurotoxicity on cultured neurons and its mechanism, d) the in vivo mechanism that protect against cysteine toxicity. The experimental evidence has shown astrocytes can release small molecules that are able to protect neurons from oxidative stress. As this area has not been extensively investigated, some important questions have to be clarified. 1. Thiol Sources in the CNS It has been found that cysteine is the rate-limiting precursor of GSH synthesis, and that extracellular cysteine regulates neuronal GSH levels (Beutler, 1989; Meister, 1989; Kranich et al., 1996). Cysteine, unlike other amino acids, is unstable and will be oxidized under aerobic conditions. The mechanisms by which the cysteine's level is maintained in the CNS have been explored by early researchers, as discussed in section 6 of Introduction. My results, consistent with most other reports, suggest that astrocytes release GSH, but not cysteine. However, there is no agreement so far on the functions of the released GSH. Sagara et al. (1996) found that neurons do not take up GSH directly, whereas they did not explain the physiological meaning of GSH efflux. As regard to the y-GT hypothesis proposed by Dringen et al. (1999a), it is not supported by several critical experiments, as discussed in Discussion of Chapter 4. My present results suggest a mechanism of thiol/disulfide exchange. Evidence from the analysis of conditioned medium and CSF supports the hypothesis that GSH released by astrocytes constantly converts cystine to cysteine and keeps cysteine at a stable level. It is conceivable that the functions of the released GSH are not limited to the maintenance of cysteine. GSH is also able to regulate the redox state of many proteins, including, for example, the N M D A receptor, and to maintain a stable redox potential of the CNS. From a broad point of 94 Chapter 7 General Discussion & Conclusions view, GSH released by astrocytes can be considered as a source of reducing power in the CNS. My results suggest that the ultimate thiol/disulfide source of the brain is cystine, taken from blood. The present data show that plasma cystine concentration is much higher than the concentrations of cysteine and GSH (Table 4-2). There is also an A - V difference of cystine concentration, but not of cysteine and GSH. This demonstrates that the brain does not depend on blood as a thiol provider. Instead, the brain depends on astrocytes to provide thiols from disulfide sources and to regulate thiol/disulfide ratios. The brain, by this mechanism, can avoid the influence of fluctuations in the blood nutrient concentrations on its microenvironment, and maintain stable thiol levels and redox state in the CNS. 2. Mechanisms of Cysteine Toxicity Cysteine neurotoxicity has been reported many years ago (Olney and Ho, 1970). So far, there exist at least four theories about cysteine toxicity, as discussed in section 7 of Introduction: a) excitotoxic metabolites of cysteine, b) bicarbonate-dependent excitotoxicity, c) redox modulation of the N M D A receptors, d) free radical toxicity. My experiments have investigated the last theory in cultured neurons. A l l the other three theories are related to excitotoxicity. First, CSA and C A are metabolites of cysteine autoxidation and are NMDAmimetic. However, my experiments show that the major oxidized product of cysteine is cystine (> 98%), rather than CSA and CA. The results from other researchers show that the distribution of brain injury caused by cysteine administration is quite different from that caused by CSA (Lehmann et al., 1993), and that cysteine is about 10-fold more toxic to neurons than CSA and C A (Pean et al., 1995). A l l these data do not support the hypothesis that CSA and C A mediate cysteine toxicity. Second, so far there is no evidence that proves the existence of a hydrogen bond between cysteine and bicarbonate. The bicarbonate-dependent excitotoxicity theory therefore cannot be verified. Third, it has been found that the reducing agents can sensitize N M D A receptors, i.e. increasing glutamate-induced response (Aizenman et al., 1989; Janaky et al., 1993; Kohr et al., 1994; Regan and Guo, 1999; Choi and Lipton, 2000). Cysteine, like GSH and other 95 Chapter 7 General Discussion & Conclusions reducing agents, is assumed to have a similar effect on the N M D A receptors. An important question that has to be answered is whether the sensitized N M D A receptors will kill neurons even with normal glutamate levels and metabolism, as cysteine administration cannot trigger a significant glutamate release (Puka-Sundvall et al., 1995). My experiments did not test this theory. However, there is no conflict i f both the redox modulation mechanism and the free radical mechanism are verified. Fourth, the data presented here suggest that cysteine neurotoxicity is a result of hydroxyl radicals generated during cysteine autoxidation. This mechanism can also explain the early observation of cysteine neurotoxicity in animals. The high dose administration of cysteine in infant mice or rats will subsequently increase cysteine levels in CSF. Presumably it is easier for cysteine to penetrate to the brain areas due to the incomplete B B B in infant animals. Cysteine will undergo autoxidation under the catalysis of loosely bound copper, whose levels in CSF are 0.13 to 0.75 uM (Gutteridge, 1984). It is suggested that the brain damage is caused by free radicals generated from autoxidation of the large amount of cysteine under those conditions. Since cysteine is an endogenous naturally occurring constituent in the extracellular fluid of the CNS and in all cells, the complete understanding of cysteine toxicity is very important. Further experiments are required to elucidate the unsolved questions. 3. Free Radical Sources in the CNS Mitochondria are probably the major sources of free radicals in vivo. Considering over 90% of 0 2 taken up by organisms is used in mitochondria, even a small percentage of electron leakage from electron transport chain will generate rather large amounts of reactive oxygen intermediates (Fisher, 1987; Barja et al., 1994). Some 0 2-utilizing enzymes, such as xanthine oxidase, tryptophan dioxygenase, indoleamine dioxygenase and nitric oxide synthetase, can generate free radicals directly (Halliwell and Gutteridge, 1985; Culcasi et al., 1994; Gotz et al., 1994). Autoxidation of some biologically important molecules, such as levodopa, dopamine and cysteine, can also generate free radicals (Saez et al., 1982; Nath and Salahudeen, 1993; Lai and Yu, 1997; Shen and Dryhurst, 1998; Soto-Otero et al., 2000). Iron is generally considered as the major transition metal ion that catalyzes autoxidation reactions. 96 Chapter 7 General Discussion & Conclusions The studies presented here demonstrate that copper, but not iron or other transition metal ions, is the most important transition metal ion in catalyzing the generation of free radicals from autoxidation reactions. The toxic effects of copper in extracellular fluids or cytoplasm could be quite different. Extracellularly, copper is able to catalyze the autoxidation of cysteine, GSH, as well as other reducing agents, such as ascorbic acid in the CNS. This kind of reactions could occur under physiological conditions owing to the existence of loosely bound copper in CSF. The copper-catalyzed autoxidation is suggested to be an important extracellular source of free radicals in the CNS. Intracellular, copper is well bound to proteins and there is no free copper under normal conditions (Rae et al., 1999). Under pathological conditions, however, copper's binding status could be compromised and copper could unexpectedly catalyze the generation of free radicals from autoxidation reactions. This assumption may explain the toxic mechanism of SOD mutants in certain familial A L S cases. 4. Antioxidative Mechanisms of Astrocytes Ever since neurons were born at the perinatal stage when neuroblasts undergo their final round of division to give rise to postmitotic neurons, they started to set up cell-cell connections. We cannot afford the death of neurons, which will result in the loss of memory and many other neural functions, as in the cases of neurodegenerative diseases. As discussed earlier, neurons of the CNS face a great oxidative stress. To live or to die is always a critical question for neurons. Fortunately, astrocytes have evolved to serve as neuronal protectors. Evidence obtained from other reports and the present experiments shows that astrocytes play diverse functions to protect neurons from oxidative stress. 1). Glutamate uptake by astrocytes. Glutamate is released into the synaptic cleft to activate the next neuron. Overstimulation of glutamate receptors will lead to the generation of free radicals, as discussed earlier. Released glutamate is taken up by glutamate transporters, which are Na+-dependent systems, and five of which have been cloned (Kanai, 1997). Selective depletion of the astrocyte glutamate transporter results in elevated extracellular glutamate levels and neuronal death (Rothstein et al., 1996). 2). Provision of cysteine by astrocytes. Release of thiols by astrocytes has been noted for a decade (Yudkoff et al., 97 Chapter 7 General Discussion & Conclusions 1990; Sagara et al., 1993). My experiments demonstrate a mechanism by which astrocytes provide cysteine to neurons. The significance of astrocytes in maintenance of thiol levels in the CNS is explained. 3). Extracellular H2O2 scavenging mechanism. Removal of H2O2 is the efficient and last defense in preventing -OH attack. Pyruvate has been found to react with H2O2. Though catalase can remove H2O2 as well, my data show that astrocytes have selected pyruvate as the extracellular H2O2 scavenger. Pyruvate is maintained at relatively high levels by astrocytes in the extracellular fluid of the CNS. Taken together, these mechanisms have revealed the neuroprotective role of astrocytes against oxidative stress. It is speculated that this antioxidative role of astrocytes is ubiquitous in the CNS. The juxtaposition of astrocytes and neurons in the tissue allows astrocytes to exert their functions efficiently. 5. Significance of the Present Findings The present data clearly indicate the importance of copper in oxidative stress. This finding is useful for studying the pathogenic mechanisms of neurological diseases. Wilson's disease is an inherited disorder characterized by hepatic cirrhosis and neuronal degeneration due to an impairment of copper excretion. The Wilson's disease gene has been found to encode a copper transporting P-type ATPase (Bull et al., 1993; Tanzi et al., 1993; Yamaguchi et al., 1993). This ATPase is expressed predominantly in the liver and is responsible for the excretion of copper into the bile. Mutations of the gene result in copper accumulation in the hepatocytes and subsequently in extrahepatic tissues, such as the brain and the cornea (Loudianos and Gitlin, 2000). The pathological changes in the CNS include copper deposition (ten- to fifteen-fold over normal) in virtually all parts of brain with resulting extensive neuronal loss and gliosis in the gray matter (Scheinberg and Sternlieb, 1983). Plasma loosely bound copper and free radical production have been found to be markedly increased in Wilson's disease (Ogihara et al., 1995). However, the molecular mechanisms of copper-induced oxidative stress and cytotoxicity have not been explained. My studies suggest that elevated free copper in the CNS accelerates the autoxidation of cysteine and other reducing agents, resulting in increased production of free radicals and subsequent cytotoxicity in Wilson's disease. 98 Chapter 7 General Discussion & Conclusions Recently, several research groups have intensively studied copper's role in the pathogenesis of Alzheimer's disease. It was found that copper could form a complex with Ap (Huang et al., 1999b) or APP (Multhaup et al., 1998; White et al., 1999), catalyzing a site-specific reduction-oxidation reaction. Neurofibrillary tangles and senile plaques in tissue slices of Alzheimer's cases show the high redox reactivity, as compared with controls (Sayre et al., 2000). The physiological functions of A(3 and APP, however, are still unknown. One possibility is that they participate in copper transport. These results indicate that copper-induced toxicity is a critical pathogenic factor. In fact, copper chelators have been suggested for the treatment of Alzheimer's disease (Cuajungco et al., 2000) . Mutations of Cu/Zn SOD have been identified in patients with familial amyotrophic lateral sclerosis (FALS) (Rosen et al., 1993). Further studies have found that oxidative stress caused by altered copper coordination is the major pathogenic factor in this FALS model (Estevez et al., 1999; Gabbianelli et al., 1999). Copper chelation has been used in the treatment of A L S in these transgenic mice (Nagano et al., 1999; Andreassen et al., 2001) . In studies of apoptosis, the release of cytochrome c from mitochondria is generally considered as a critical trigger of the apoptosis process (Liu et al., 1996; Green and Reed, 1998). However, the role of copper, the major cofactor of cytochrome c, has not been extensively studied. It is shown that cysteine and other reducing agents will be oxidized under physiological conditions in vivo. This may explain why cysteine is maintained at a relatively low level in the CSF, compared with other amino acids (McGale et al., 1977). The organism requires minimizing the generation of free radicals from cysteine autoxidation. However, the existence of certain levels of extracellular cysteine and its autoxidation are inevitable, as cysteine is essential for GSH synthesis within neurons. The brain utilizes pyruvate to scavenge H2O2 as an efficient defense. At present, some thiols, including GSH, N-acetyl-cysteine (NAC), have been applied or are being tested as therapeutic drugs or components of nutritional products. A caution in their application should be raised based on above considerations. Large doses of thiols could increase the generation of free radicals from their autoxidation in vivo. GSH and N A C can undergo an autoxidation reaction just as can cysteine, simply with varying degrees. Pyruvate, on 99 Chapter 7 General Discussion & Conclusions the other hand, should be considered as a good candidate for antioxidant defense in clinical applications. It is a rare antioxidant without prooxidant properties. The beauty of great nature is beyond imagination. Over billions of years, the living world has become such a complicated, dynamic and well-arranged system. The organisms have faced countless challenges from the environment. To survive, they must evolve certain mechanisms to overcome these challenges. Adaptation is a great gift of life. The living species may adopt an existing system from another species, or develop a new system by themselves. It is unknown when copper was picked up by the living species from the environment as an enzyme cofactor, exerting its powerful role in redox reactions. Presumably, this event occurred after the appearance of 0 2 in the atmosphere. The organisms must have found that the uncontrolled catalyzing effect of copper ions is devastating and intolerable in vivo. Then the specialized proteins appeared in the body fluids to carefully chelate copper ions, making full use of their catalyzing property in a controlled manner. It is unknown how the living species obtained the capacities to synthesize GSH, and to absorb so many antioxidants they cannot synthesize by themselves. When pyruvate was reacted with H2O2 in an organism for the first time, the organism must have sensed the benefit of pyruvate, then employed this chemical as a defense later on, and passed on this mechanism to the next generation. Along the history of evolution, living species have constructed many efficient and delicate antioxidative systems. To elucidate these mechanisms will have significant value for the health of humanity. 6 . Perspectives In the present study, copper was found to play a pivotal role in neurotoxicity. However, the knowledge of copper metabolism in the brain is still patchy. As free copper can cause cell damage, it has to be carefully transported. Ceruloplasmin gene expression has been detected in glial cells of the brain and retina (Klomp et al., 1996; Levin and Geszvain, 1998). Ceruloplasmin levels are increased in subjects of Alzheimer's disease, Parkinson's disease, and Huntington's disease (Loeffler et al., 1996). However, it is unknown what the relationship of ceruloplasmin is with other copper-binding peptides or proteins, such as Ap, APP or apoE. The copper chaperone for 100 Chapter 7 General Discussion & Conclusions SOD (CCS), which binds copper and delivers it to target protein, has recently been investigated (Rosenzweig and O'Halloran, 2000). Its role in A L S and other neurodegenerative diseases is unclear. A critical question about copper's role in neurodegeneration is what are the initial factors that could trigger the abnormal copper coordination. The therapeutic value of copper chelators deserves to be explored. Copper chelators in the treatment of Alzheimer's disease and A L S have been mentioned earlier. In stroke and traumatic brain injury (TBI), it is expected that copper levels at the site of damage will substantially increase owing to the breakdown of blood-brain barrier. The copper concentration in blood is much higher than in CSF, with ratio of CSF/plasma equal to 1.2% (Joergstuerenburg et al., 1999). Chelation of copper could decrease copper-induced oxidative stress in these pathological situations. It has been shown that astrocytes use diverse mechanisms to protect neurons from oxidative stress. It is not hard to imagine that once the functions of astrocytes are weakened, the survival capacity of neighbouring neurons will be severely compromised. Senescence can lead to the gradual functional decrease of many cell types, particularly mitotic cells like astrocytes. Senescent astrocytes may have lower capacities of uptaking glutamate and of releasing GSH and pyruvate. Once these changes in astrocytes reach beyond threshold, neurons will be affected. It is also known that aging is the common etiological factor of neurodegenerative diseases. Even for cases with hereditary tendency, like early onset Alzheimer's disease, most of them strike only after maturity. Based on these considerations, the functions of senescent astrocytes and their role in the pathogenesis of neurodegeneration should be examined. C O N C L U S I O N S The following conclusions can be drawn from the experiments: 1) . Astrocytes improve neuronal survival. 2) . Astrocytes promote neuronal attachment. 3) . Astrocytes provide cysteine to neurons indirectly by releasing GSH. 4) . Autoxidation of cysteine, as well as other reducing agents, generates free radicals, which are responsible for their cytotoxic effects. 101 Chapter 7 General Discussion & Conclusions 5) . Copper is the major catalyst of cysteine autoxidation in the CNS. 6) . Catalase and pyruvate can prevent cysteine toxicity by removing H2O2. 7) . Astrocytes release pyruvate. 9 ) . Cysteine can improve neuronal survival when its neurotoxic effects are prevented. 102 References R E F E R E N C E S Acworth IN, McCabe DR, Maher TJ (1997) The analysis of free radicals, their reaction products, and antioxidants. In: Oxidants, antioxidants, and free radicals (Baskin SI, Salem H, eds), pp 23-78. Washington, DC: Taylor & Francis. Aguilar MV, Jimenez-Jimenez FJ, Molina JA, Meseguer I, Mateos-Vega CJ, Gonzalez-Munoz MJ, de Bustos F, Gomez-Escalonilla C, Ort-Pareja M, Zurdo M, Martinez-Para MC (1998) Cerebrospinal fluid selenium and chromium levels in patients with Parkinson's disease. J Neural Transm 105:1245-1251. Aizenman E, Lipton SA, Loring RH (1989) Selective modulation of NMDA responses by reduction and oxidation. Neuron 2:1257-1263. Alberts B, Bray D, Lewis J, Raff M, Roberts K, Watson JD (1994) Molecular biology of the cell, 3rd Edition. New York: Garland Publishing. Ames BN, Shigenaga MK (1992) DNA damage by endogenous oxidants and mitogenesis as causes of aging and cancer. In: Molecular biology of free radical scavenging systems (Scandalios JG, ed), pp 1-22. New York: Cold Spring Harbor Laboratory Press. Ames BN, Cathcart R, Schwiers E, Hochstein P (1981) Uric acid provides an antioxidant defense in humans against oxidant- and radical-caused aging and cancer: a hypothesis. Proc Natl Acad Sci U S A 78:6858-6862. Amici A, Levine RL, Tsai L, Stadtman ER (1989) Conversion of amino acid residues in proteins and amino acid homopolymers to carbonyl derivatives by metal-catalyzed oxidation reactions. J Biol Chem 264:3341-3346. Anderson ME, Bridges RJ, Meister A (1980) Direct evidence for inter-organ transport of glutathione and that the non-filtration renal mechanism for glutathione utilization involves gamma-glutamyl transpeptidase. Biochem Biophys Res Commun 96:848-853. Anderson ME, Underwood M, Bridges RJ, Meister A (1989) Glutathione metabolism at the blood-cerebrospinal fluid barrier. Faseb J 3:2527-2531. Andreassen OA, Dedeoglu A, Friedlich A, Ferrante KL, Hughes D, Szabo C, Beal MF (2001) Effects of an inhibitor of poly(ADP-ribose) polymerase, desmethylselegiline, trientine, and lipoic acid in transgenic ALS mice. Exp Neurol 168:419-424. Ankarcrona M, Dypbukt JM, Bonfoco E, Zhivotovsky B, Orrenius S, Lipton SA, Nicotera P (1995) Glutamate-induced neuronal death: a succession of necrosis or apoptosis depending on mitochondrial function. Neuron 15:961-973. Arriza JL, Kavanaugh MP, Fairman WA, Wu YN, Murdoch GH, North RA, Amara SG (1993) Cloning and expression of a human neutral amino acid transporter with structural similarity to the glutamate transporter gene family. J Biol Chem 268:15329-15332. 103 References Aruoma OI, Halliwell B, Dizdaroglu M (1989a) Iron ion-dependent modification of bases in DNA by the superoxide radical-generating system hypoxanthine/xanthine oxidase. J Biol Chem 264:13024-13028. Aruoma OI, Halliwell B, Gajewski E, Dizdaroglu M (1989b) Damage to the bases in DNA induced by hydrogen peroxide and ferric ion chelates. J Biol Chem 264:20509-20512. Aschner M (1998) Metallothionein (MT) isoforms in the central nervous system (CNS): regional and cell-specific distribution and potential functions as an antioxidant. Neurotoxicology 19:653-660. Atwood CS, Scarpa RC, Huang X, Moir RD, Jones WD, Fairlie DP, Tanzi RE, Bush AI (2000) Characterization of copper interactions with alzheimer amyloid beta peptides: identification of an attomolar-affinity copper binding site on amyloid betal-42. J Neurochem 75:1219-1233. Augusto O (1993) Alkylation and cleavage of DNA by carbon-centered radical metabolites. Free Radic Biol Med 15:329-336. Aust SD (1987) Sources of iron for lipid peroxidation in biological systems. In: Oxygen radicals and tissue injury (Halliwell B, ed), pp 27-33. Bethesda: Upjohn. Avissar N, Whitin JC, Allen PZ, Wagner DD, Liegey P, Cohen HJ (1989) Plasma selenium-dependent glutathione peroxidase. Cell of origin and secretion. J Biol Chem 264:15850-15855. Avissar N, Finkelstein JN, Horowitz S, Willey JC, Coy E, Frampton MW, Watkins RH, Khullar P, Xu YL, Cohen HJ (1996) Extracellular glutathione peroxidase in human lung epithelial lining fluid and in lung cells. Am J Physiol 270:L173-182. Aw TY, Ookhtens M, Ren C, Kaplowitz N (1986) Kinetics of glutathione efflux from isolated rat hepatocytes. Am J Physiol 250:G236-243. Awasthi D, Church DF, Torbati D, Carey ME, Pryor WA (1997) Oxidative stress following traumatic brain injury in rats. Surg Neurol 47:575-581; discussion 581-572. Balazs L, Leon M (1994) Evidence of an oxidative challenge in the Alzheimer's brain. Neurochem Res 19:1131-1137. Banker GA (1980) Trophic interactions between astroglial cells and hippocampal neurons in culture. Science 209:809-810. Bannai S, Tsukeda H (1979) The export of glutathione from human diploid cells in culture. J Biol Chem 254:3444-3450. Barja G, Cadenas S, Rojas C, Perez-Campo R, Lopez-Torres M (1994) Low mitochondrial free radical production per unit 02 consumption can explain the simultaneous presence of high longevity and high aerobic metabolic rate in birds. Free Radic Res 21:317-327. Barnes D, Sato G (1980) Serum-free cell culture: a unifying approach. Cell 22:649-655. Barres BA, Barde Y (2000) Neuronal and glial cell biology. Curr Opin Neurobiol 10:642-648. Barrier L, Page G, Fauconneau B, Juin F, Tallineau C (1998) Autoxidation of rat brain homogenate: evidence for spontaneous lipid peroxidation. Comparison with the characteristics of Fe2+- and ascorbic acid-stimulated lipid peroxidation. Free Radic Res 28:411-422. 104 References Bartosz G (1996) Peroxynitrite: mediator of the toxic action of nitric oxide. Acta Biochim Pol 43:645-659. Bates GW, Schlabach MR (1973) The reaction of ferric salts with transferrin. J Biol Chem 248:3228-3232. Baughman RW, Huettner JE, Jones KA, AleemKhan A (1991) Cell culture of neocortex and basal forebrain from postnatal rats. In: Culturing nerve cells (Banker G, Goslin K, eds), pp 227-249. Cambridge: MIT Press. Beatrice MC, Stiers DL, Pfeiffer DR (1984) The role of glutathione in the retention of Ca2+ by liver mitochondria. J Biol Chem 259:1279-1287. Behl C (1999) Alzheimer's disease and oxidative stress: implications for novel therapeutic approaches. Prog Neurobiol 57:301-323. Beilharz EJ, Williams CE, Dragunow M, Sirimanne ES, Gluckman PD (1995) Mechanisms of delayed cell death following hypoxic-ischemic injury in the immature rat: evidence for apoptosis during selective neuronal loss. Brain Res Mol Brain Res 29:1-14. Beutler E (1989) Nutritional and metabolic aspects of glutathione. Annu Rev Nutr 9:287-302. Bittar PG, Charnay Y, Pellerin L, Bouras C, Magistretti PJ (1996) Selective distribution of lactate dehydrogenase isoenzymes in neurons and astrocytes of human brain. J Cereb Blood Flow Metab 16:1079-1089. Blaauwgeers HG, Sillevis Smitt PA, De Jong JM, Troost D (1993) Distribution of metallothionein in the human central nervous system. Glia 8:62-70. Boebel KP, Baker DH (1983) Blood and liver concentrations of glutathione, and plasma concentrations of sulfur-containing amino acids in chicks fed deficient, adequate, or excess levels of dietary cysteine. Proc Soc Exp Biol Med 172:498-501. Borchelt DR, Lee MK, Slunt HS, Guarnieri M, Xu ZS, Wong PC, Brown RH, Price DL, Sisodia SS, Cleveland DW (1994) Superoxide dismutase 1 with mutations linked to familial amyotrophic lateral sclerosis possesses significant activity. Proc Natl Acad Sci U S A 91:8292-8296. Borg DC (1993) Oxygen free radicals and tissue injury, a reference outline. In: Oxygen free radicals in tissue damage (Tarr M, Samson F, eds), pp 12-53. Cambridge, MA: Birkhauser Boston. Bottenstein JE, Sato GH (1979) Growth of a rat neuroblastoma cell line in serum-free supplemented medium. Proc Natl Acad Sci U S A 76:514-517. Boveris A, Chance B (1973) The mitochondrial generation of hydrogen peroxide. General properties and effect of hyperbaric oxygen. Biochem J 134:707-716. , Bowery NG, Brown DA, Collins GG, Galvan M, Marsh S, Yamini G (1976) Indirect effects of amino-acids on sympathetic ganglion cells mediated through the release of gamma-aminobutyric acid from glial cells. Br J Pharmacol 57:73-91. Bredt DS, Hwang PM, Glatt CE, Lowenstein C, Reed RR, Snyder SH (1991) Cloned and expressed nitric oxide synthase structurally resembles cytochrome P-450 reductase. Nature 351:714-718. Broer S, Rahman B, Pellegri G, Pellerin L, Martin JL, Verleysdonk S, Hamprecht B, Magistretti PJ (1997) Comparison of lactate transport in astroglial cells and monocarboxylate transporter 1 (MCT 1) 105 References expressing Xenopus laevis oocytes. Expression of two different monocarboxylate transporters in astroglial cells and neurons. J Biol Chem 272:30096-30102. Bronstein DM, Perez-Otano 1, Sun V, Mullis Sawin SB, Chan J, Wu GC, Hudson PM, Kong LY, Hong JS, McMillian MK (1995) Glia-dependent neurotoxicity and neuroprotection in mesencephalic cultures. Brain Res 704:112-116. Bruijn LI, Becher MW, Lee MK, Anderson KL, Jenkins NA, Copeland NG, Sisodia SS, Rothstein JD, Borchelt DR, Price DL, Cleveland DW (1997) ALS-linked SOD1 mutant G85R mediates damage to astrocytes and promotes rapidly progressive disease with SOD 1-containing inclusions. Neuron 18:327-338. Brustovetsky N, Dubinsky JM (2000) Dual responses of CNS mitochondria to elevated calcium. J Neurosci 20:103-113. Bull PC, Thomas GR, Rommens JM, Forbes JR, Cox DW (1993) The Wilson disease gene is a putative copper transporting P-type ATPase similar to the Menkes gene. Nat Genet 5:327-337. Bunton CA (1949) Oxidation of a-diketones and ot-keto-acids by hydrogen peroxide. Nature 163:444. Burton GW, Joyce A, Ingold KU (1982) First proof that vitamin E is major lipid-soluble, chain-breaking antioxidant in human blood plasma. Lancet 2:327. Carelli S, Ceriotti A, Cabibbo A, Fassina G, Ruvo M, Sitia R (1997) Cysteine and glutathione secretion in response to protein disulfide bond formation in the ER. Science 277:1681-1684. Carreau JP (1979) Biosynthesis of lipoic acid via unsaturated fatty acids. Methods Enzymol 62:152-158. Carri MT, Battistoni A, Polizio F, Desideri A, Rotilio G (1994) Impaired copper binding by the H46R mutant of human Cu,Zn superoxide dismutase, involved in amyotrophic lateral sclerosis. FEBS Lett 356:314-316. Castilho RF, Hansson O, Ward MW, Budd SL, Nicholls DG (1998) Mitochondrial control of acute glutamate excitotoxicity in cultured cerebellar granule cells. J Neurosci 18:10277-10286. Cavallini D, De Marco C, Dupre S, Rotilio G (1969) The copper catalyzed oxidation of cysteine to cystine. Arch Biochem Biophys 130:354-361. Chance B, Sies H, Boveris A (1979) Hydroperoxide metabolism in mammalian organs. Physiol Rev 59:527-605. Chapman D (1971) The role of fatty acids in myelin and other important brain structures. In: Lipids, malnutrition & the developing brain (Symposium ACF, ed), pp 31-58. Amsterdam: Elsevier Science. Charriaut-Marlangue C, Ben-Ari Y (1995) A cautionary note on the use of the TUNEL stain to determine apoptosis. Neuroreport 7:61-64. Chawla RK, Lewis FW, Kutner MH, Bate DM, Roy RG, Rudman D (1984) Plasma cysteine, cystine, and glutathione in cirrhosis. Gastroenterology 87:770-776. Chevion M (1988) A site-specific mechanism for free radical induced biological damage: the essential role of redox-active transition metals. Free Radic Biol Med 5:27-37. 106 References Choi DW (1987) Ionic dependence of glutamate neurotoxicity. J Neurosci 7:369-379. Choi YB, Lipton SA (2000) Redox modulation of the NMDA receptor. Cell Mol Life Sci 57:1535-1541. Clarke DD, Sokoloff L (1999) Circulation and energy metabolism of the brain. In: Basic neurochemistry: molecular, cellular and medical aspects, 6th Edition (Siegel GJ, Agranoff BW, Albers RW, Fisher SK, Uhler MD, eds), pp 637-669. Philadelphia: Lippincott-Raven. Coles JA (1995) Glial cells and the supply of substrates of energy. In: Neuroglia (Kettenmann H, Ransom BR, eds), pp 793-804. New York: Oxford. Collins AK, Makrigiorgos GM, Svensson GK (1994a) Coumarin chemical dosimeter for radiation therapy. Med Phys 21:1741-1747. Collins MK, Perkins GR, Rodriguez-Tarduchy G, Nieto MA, Lopez-Rivas A (1994b) Growth factors as survival factors: regulation of apoptosis. Bioessays 16:133-138. Conner EM, Grisham MB (1996) Inflammation, free radicals, and antioxidants. Nutrition 12:274-277. Cooper AJ, Kristal BS (1997) Multiple roles of glutathione in the central nervous system. Biol Chem 378:793-802. Corder EH, Saunders AM, Strittmatter WJ, Schmechel DE, Gaskell PC, Small GW, Roses AD, Haines JL, Pericak- Vance MA (1993) Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer's disease in late onset families. Science 261:921-923. Coyle JT, Puttfarcken P (1993) Oxidative stress, glutamate, and neurodegenerative disorders. Science 262:689-695. Cuajungco MP, Faget KY, Huang X, Tanzi RE, Bush AI (2000) Metal chelation as a potential therapy for Alzheimer's disease. Ann N Y Acad Sci 920:292-304. Culcasi M, Lafon-Cazal M, Pietri S, Bockaert J (1994) Glutamate receptors induce a burst of superoxide via activation of nitric oxide synthase in arginine-depleted neurons. J Biol Chem 269:12589-12593. Cutler RW, Spertell RB (1982) Cerebrospinal fluid: a selective review. Ann Neurol 11:1-10. D'Amico DJ, Klawans HL (1976) Direct microdetermination of manganese in normal serum and cerebrospinal fluid by flameless atomic absorption spectrophotometry. Anal Chem 48:1469-1472. Davies AM (1997) Neurotrophic factors and the regulation of neuronal survival in the developing peripheral nervous system. In: Molecular biology of the neuron. (Davies RW, Morris BJ, eds), pp 339-357. Oxford: BIOS Scientific. Demiryurek AT, Cakici I, Kanzik I (1998) Peroxynitrite: a putative cytotoxin. Pharmacol Toxicol 82:113-117. Denizot F, Lang R (1986) Rapid colorimetric assay for cell growth and survival. Modifications to the tetrazolium dye procedure giving improved sensitivity and reliability. J Immunol Methods 89:271-277. Desagher S, Glowinski J, Premont J (1996) Astrocytes protect neurons from hydrogen peroxide toxicity. J Neurosci 16:2553-2562. 107 References Desagher S, Glowinski J, Premont J (1997) Pyruvate protects neurons against hydrogen peroxide-induced toxicity. J Neurosci 17:9060-9067. Deschepper CF, Bouhnik J, Ganong WF (1986) Colocalization of angiotensinogen and glial fibrillary acidic protein in astrocytes in rat brain. Brain Res 374:195-198. Desjardins P, Ledoux S (1998) The role of apoptosis in neurodegenerative diseases. Metab Brain Dis 13:79-96. Dirnagl U, Iadecola C, Moskowitz MA (1999) Pathobiology of ischaemic stroke: an integrated view. Trends Neurosci 22:391-397. Dizdaroglu M (1991) Chemical determination of free radical-induced damage to DNA. Free Radic Biol Med 10:225-242. Dizdaroglu M, Aruoma 01, Halliwell B (1990) Modification of bases in DNA by copper ion-1,10-phenanthroline complexes. Biochemistry 29:8447-8451. Douki T, Ames BN (1994) An HPLC-EC assay for l,N2-propano adducts of 2'-deoxyguanosine with 4-hydroxynonenal and other alpha,beta-unsaturated aldehydes. Chem Res Toxicol 7:511-518. Douki T, Cadet J, Ames BN (1996) An adduct between peroxynitrite and 2'-deoxyguanosine: 4,5-dihydro-5- hydroxy-4-(nitrosooxy)-2'-deoxyguanosine. Chem Res Toxicol 9:3-7. Dringen R, Wiesinger H, Hamprecht B (1993) Uptake of L-lactate by cultured rat brain neurons. Neurosci Lett 163:5-7. Dringen R, Pfeiffer B, Hamprecht B (1999a) Synthesis of the antioxidant glutathione in neurons: supply by astrocytes of CysGly as precursor for neuronal glutathione. J Neurosci 19:562-569. Dringen R, Kussmaul L, Gutterer JM, Hirrlinger J, Hamprecht B (1999b) The glutathione system of peroxide detoxification is less efficient in neurons than in astroglial cells. J Neurochem 72:2523-2530. Drukarch B, Schepens E, Jongenelen CA, Stoof JC, Langeveld CH (1997) Astrocyte-mediated enhancement of neuronal survival is abolished by glutathione deficiency. Brain Res 770:123-130. Duffy S, So A, Murphy TH (1998) Activation of endogenous antioxidant defenses in neuronal cells prevents free radical-mediated damage. J Neurochem 71:69-77. Duncan MW, Steele JC, Kopin 1J, Markey SP (1990) 2-Amino-3-(methylamino)-propanoic acid (BMAA) in cycad flour: an unlikely cause of amyotrophic lateral sclerosis and parkinsonism- dementia of Guam. Neurology 40:767-772. Dunn MA, Blalock TL, Cousins RJ (1987) Metallothionein. Proc Soc Exp Biol Med 185:107-119. Echols H, Goodman MF (1991) Fidelity mechanisms in DNA replication. Annu Rev Biochem 60:477-511. Edstrom A, Ekstrom PA, Tonge D (1996) Axonal outgrowth and neuronal apoptosis in cultured adult mouse dorsal root ganglion preparations: effects of neurotrophins, of inhibition of neurotrophin actions and of prior axotomy. Neuroscience 75:1165-1174. Ellman GL (1959) Tissue sulfhydryl groups. Arch Biochem Biophys 82:70-77. 108 References Esterbauer H (1993) Cytotoxicity and genotoxicity of lipid-oxidation products. Am J Clin Nutr 57:779S-785S; discussion 785S-786S. Esterbauer H, Schaur RJ, Zollner H (1991) Chemistry and biochemistry of 4-hydroxynonenal, malonaldehyde and related aldehydes. Free Radic Biol Med 11:81-128. Estevez AG, Crow JP, Sampson JB, Reiter C, Zhuang Y, Richardson GJ, Tarpey MM, Barbeito L, Beckman JS (1999) Induction of nitric oxide-dependent apoptosis in motor neurons by zinc-deficient superoxide dismutase. Science 286:2498-2500. Everett SA, Dennis MF, Patel KB, Maddix S, Kundu SC, Willson RL (1996) Scavenging of nitrogen dioxide, thiyl, and sulfonyl free radicals by the nutritional antioxidant beta-carotene. J Biol Chem 271:3988-3994. Fairbairn JJ, Khan MW, Ward KJ, Loveridge BW, Fairbairn DW, O'Neill KL (1995) Induction of apoptotic cell DNA fragmentation in human cells after treatment with hyperthermia. Cancer Lett 89:183-188. Fariss MW, Reed DJ (1983) Measurement of glutathione and glutathione disulfide efflux from isolated rat hepatocytes. In: Isolation, characterization, and use of hepatocytes (Harris RA, Cornell NW, eds), pp 349-355. New York: Elsevier Biomedical. Fee JA, Valentine JS (1977) Chemical and physical properties of superoxide. In: Superoxide and superoxide dismutases (Michelson AM, McCord JM, Fridovich I, eds), pp 19-66. New York: Academic Press. Felig P, Wahren J, Ahlborg G (1973) Uptake of individual amino acids by the human brain. Proc Soc Exp Biol Med 142:230-231. Fenton HJH (1898) Note on the oxidation of certain acids in presence of iron. J Chem Soc Proc 14:119-120. Ferrari G, Yan CY, Greene LA (1995) N-acetylcysteine (D- and L-stereoisomers) prevents apoptotic death of neuronal cells. J Neurosci 15:2857-2866. Fisher AB (1987) Intracellular production of oxygen-derived free radicals. In: Oxygen radicals and tissue injury (Halliwell B, ed), pp 34-39. Bethesda: Upjohn. Fraga CG, Shigenaga MK, Park JW, Degan P, Ames BN (1990) Oxidative damage to DNA during aging: 8-hydroxy-2'-deoxyguanosine in rat organ DNA and urine. Proc Natl Acad Sci U S A 87:4533-4537. Franchi-Gazzola R, Gazzola GC, Dall'Asta V, Guidotti GG (1982) The transport of alanine, serine, and cysteine in cultured human fibroblasts. J Biol Chem 257:9582-9587. Fridovich I (1995) Superoxide radical and superoxide dismutases. Annu Rev Biochem 64:97-112. Furukawa S, Furukawa Y, Satoyoshi E, Hayashi K (1986) Synthesis and secretion of nerve growth factor by mouse astroglial cells in culture. Biochem Biophys Res Commun 136:57-63. 109 References Gabbianelli R, Ferri A, Rotilio G, Carri MT (1999) Aberrant copper chemistry as a major mediator of oxidative stress in a human cellular model of amyotrophic lateral sclerosis. J Neurochem 73:1175-1180. Gajewski E, Dizdaroglu M (1990) Hydroxyl radical induced cross-linking of cytosine and tyrosine in nucleohistone. Biochemistry 29:977-980. Garcia CK, Goldstein JL, Pathak RK, Anderson RG, Brown MS (1994) Molecular characterization of a membrane transporter for lactate, pyruvate, and other monocarboxylates: implications for the Cori cycle. Cell 76:865-873. Garcia-Ruiz C, Colell A, Morales A, KaplowitzN, Fernandez-Checa JC (1995) Role of oxidative stress generated from the mitochondrial electron transport chain and mitochondrial glutathione status in loss of mitochondrial function and activation of transcription factor nuclear factor-kappa B: studies with isolated mitochondria and rat hepatocytes. Mol Pharmacol 48:825-834. Gardner HW, Jursinic PA (1981) Degradation of linoleic acid hydroperoxides by a cysteine . FeC13 catalyst as a model for similar biochemical reactions. I. Study of oxygen requirement, catalyst and effect of pH. Biochim Biophys Acta 665:100-112. Gavrieli Y, Sherman Y, Ben-Sasson SA (1992) Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J Cell Biol 119:493-501. Giandomenico AR, Cerniglia GE, Biaglow JE, Stevens CW, Koch CJ (1997) The importance of sodium pyruvate in assessing damage produced by hydrogen peroxide. Free Radic Biol Med 23:426-434. Gilbert DL (1981) Oxygen and living processes: an inter-disciplinary approach. New York: Springer. Goldberg JL, Barres BA (2000) The relationship between neuronal survival and regeneration. Annu Rev Neurosci 23:579-612. Gomes FC, Maia CG, de Menezes JR, Neto VM (1999) Cerebellar astrocytes treated by thyroid hormone modulate neuronal proliferation. Glia 25:247-255. Goslin K, Banker G (1991) Rat hippocampal neurons in low-density culture. In: Culturing nerve cells (Banker G, Goslin K, eds), pp 251-281. Cambridge: MIT Press. Gotz ME, Kunig G, Riederer P, Youdim MB (1994) Oxidative stress: free radical production in neural degeneration. Pharmacol Ther 63:37-122. Gould E, McEwen BS (1993) Neuronal birth and death. Curr Opin Neurobiol 3:676-682. Gozlan H, Ben-Ari Y (1995) NMDA receptor redox sites: are they targets for selective neuronal protection? Trends Pharmacol Sci 16:368-374. Grasl-Kraupp B, Ruttkay-Nedecky B, Koudelka H, Bukowska K, Bursch W, Schulte-Hermann R (1995) In situ detection of fragmented DNA (TUNEL assay) fails to discriminate among apoptosis, necrosis, and autolytic cell death: a cautionary note. Hepatology 21:1465-1468. Green DR, Reed JC (1998) Mitochondria and apoptosis. Science 281:1309-1312. Grundman M (2000) Vitamin E and Alzheimer disease: the basis for additional clinical trials. Am J Clin Nutr71:630S-636S. 110 References Gurney ME, Pu H, Chiu AY, Dal Canto MC, Polchow CY, Alexander DD, Caliendo J, Hentati A, Kwon YW, Deng HX, et al. (1994) Motor neuron degeneration in mice that express a human Cu,Zn superoxide dismutase mutation. Science 264:1772-1775. Gutteridge JM, Smith A (1988) Antioxidant protection by haemopexin of haem-stimulated lipid peroxidation. Biochem J 256:861-865. Gutteridge JMC (1984) Copper-phenanthroline-induced site-specific oxygen-radical damage to DNA. Detection of loosely bound trace copper in biological fluids. Biochem J 218:983-985. Gutteridge JMC (1987) Lipid peroxidation: some problems and concepts. In: Oxygen radicals and tissue injury (Halliwell B, ed), pp 1-19. Bethesda: Upjohn. Hajimohammadreza I, Brammer M (1990) Brain membrane fluidity and lipid peroxidation in Alzheimer's disease. Neurosci Lett 112:333-337. Halliwell B (1978) Superoxide-dependent formation of hydroxyl radicals in the presence of iron chelates: is it a mechanism for hydroxyl radical production in biochemical systems? FEBS Lett 92:321-326. Halliwell B (1994) Free radicals and antioxidants: a personal view. Nutr Rev 52:253-265. Halliwell B (1996) Vitamin C: antioxidant or pro-oxidant in vivo? Free Radic Res 25:439-454. Halliwell B, Gutteridge JM (1984) Oxygen toxicity, oxygen radicals, transition metals and disease. Biochem J 219:1-14. Halliwell B, Gutteridge JMC (1985) Oxygen radicals and the nervous system. Trends Neurosci 8:22-26. Han SK, Mytilineou C, Cohen G (1996) L-DOPA up-regulates glutathione and protects mesencephalic cultures against oxidative stress. J Neurochem 66:501-510. Hanaki A, Kamide H (1975) Autoxidation of cysteine catalyzed by copper in glycylglycine buffer. Chem Pharm Bull (Tokyo) 23:1671-1676. Hanna PM, Mason RP (1992) Direct evidence for inhibition of free radical formation from Cu(I) and hydrogen peroxide by glutathione and other potential ligands using the EPR spin-trapping technique. Arch Biochem Biophys 295:205-213. Hartung HP, Toyka KV (1987) Phorbol diester TPA elicits prostaglandin E release from cultured rat astrocytes. Brain Res 417:347-349. Hatcher WH, Hill AC (1928) Hydrogen peroxide as an oxidizing agent in acid solution, IX. Trans R Soc Can 22:211-220. Hawkins RA, Williamson DH, Krebs HA (1971) Ketone-body utilization by adult and suckling rat brain in vivo. Biochem J 122:13-18. Hensley K, Robinson KA, Gabbita SP, Salsman S, Floyd RA (2000) Reactive oxygen species, cell signaling, and cell injury. Free Radic Biol Med 28:1456-1462. Hensley K, Hall N, Subramaniam R, Cole P, Harris M, Aksenov M, Aksenova M, Gabbita SP, Wu JF, Carney JM, et al. (1995) Brain regional correspondence between Alzheimer's disease histopathology and biomarkers of protein oxidation. J Neurochem 65:2146-2156. I l l References Heslop KE, Goss-Sampson MA, Muller DP, Curzon G (1996) Serotonin metabolism and release in frontal cortex of rats on a vitamin E-deficient diet. J Neurochem 66:860-864. Hidalgo J, Carrasco J (1998) Regulation of the synthesis of brain metallothioneins. Neurotoxicology 19:661-666. Hochman A, Sternin H, Gorodin S, Korsmeyer S, Ziv I, Melamed E, Offen D (1998) Enhanced oxidative stress and altered antioxidants in brains of Bcl-2- deficient mice. J Neurochem 71:741-748. Holleman AF (1904) Notice sur Taction de l'eau oxygenee sur les acides a-cetoniques et sur les dicetones 1.2. Rec Trav Chim Pays-Bas Belg 23:169-172. Holopainen I, Kontro P (1990) D-aspartate release from cerebellar astrocytes: modulation of the high K-induced release by neurotransmitter amino acids. Neuroscience 36:115-120. Hou JG, Cohen G, Mytilineou C (1997) Basic fibroblast growth factor stimulation of glial cells protects dopamine neurons from 6-hydroxydopamine toxicity: involvement of the glutathione system. J Neurochem 69:76-83. Houdou S, Kuruta H, Hasegawa M, Konomi H, Takashima S, Suzuki Y, Hashimoto T (1991) Developmental immunohistochemistry of catalase in the human brain. Brain Res 556:267-270. Hu ML (1994) Measurement of protein thiol groups and glutathione in plasma. Methods Enzymol 233:380-385. Huang X, Atwood CS, Hartshorn MA, Multhaup G, Goldstein LE, Scarpa RC, Cuajungco MP, Gray DN, Lim J, Moir RD, Tanzi RE, Bush AI (1999a) The A beta peptide of Alzheimer's disease directly produces hydrogen peroxide through metal ion reduction. Biochemistry 38:7609-7616. Huang X, Cuajungco MP, Atwood CS, Hartshorn MA, Tyndall JD, Hanson GR, Stokes KC, Leopold M, Multhaup G, Goldstein LE, Scarpa RC, Saunders AJ, Lim J, Moir RD, Glabe C, Bowden EF, Masters CL, Fairlie DP, Tanzi RE, Bush AI (1999b) Cu(II) potentiation of alzheimer abeta neurotoxicity. Correlation with cell-free hydrogen peroxide production and metal reduction. J Biol Chem 274:37111-37116. Hwang SM, Weiss S, Segal S (1980) Uptake of L-[35S]cystine by isolated rat brain capillaries. J Neurochem 35:417-424. Ingraham LL, Meyer DL (1985) Biochemistry of dioxygen. New York: Plenum Press. Iwata-Ichikawa E, Kondo Y, Miyazaki I, Asanuma M, Ogawa N (1999) Glial cells protect neurons against oxidative stress via transcriptional up-regulation of the glutathione synthesis. J Neurochem 72:2334-2344. Izumi Y, Benz AM, Zorumski CF, Olney JW (1994) Effects of lactate and pyruvate on glucose deprivation in rat hippocampal slices. Neuroreport 5:617-620. Jain A, Martensson J, Stole E, Auld PA, Meister A (1991) Glutathione deficiency leads to mitochondrial damage in brain. Proc Natl Acad Sci U S A 88:1913-1917. Janaky R, Varga V, Saransaari P, Oja SS (1993) Glutathione modulates the N-methyl-D-aspartate receptor-activated calcium influx into cultured rat cerebellar granule cells. Neurosci Lett 156:153-157. 112 References Janaky R, Ogita K, Pasqualotto BA, Bains JS, Oja SS, Yoneda Y, Shaw CA (1999) Glutathione and signal transduction in the mammalian CNS. J Neurochem 73:889-902. Jeremy J, Murphy S, Morrow C, Pearce B, Dandona P (1987) Phorbol ester stimulation of prostanoid synthesis by cultured astrocytes. Brain Res 419:364-368. Jimenez-Jimenez FJ, Molina JA, Aguilar MV, Meseguer I, Mateos-Vega CJ, Gonzalez-Munoz MJ, de Bustos F, Martinez-Salio A, Orti-Pareja M, Zurdo M, Martinez-Para MC (1998) Cerebrospinal fluid levels of transition metals in patients with Parkinson's disease. J Neural Transm 105:497-505. Jocelyn PC (1967) The standard redox potential of cysteine-cystine from the thiol- disulphide exchange reaction with glutathione and lipoic acid. Eur J Biochem 2:327-331. Jocelyn PC (1987) Chemical reduction of disulfides. Methods Enzymol 143:246-256. Joergstuerenburg H, Oechsner M, Schroeder S, Kunze K (1999) Determinants of the copper concentration in cerebrospinal fluid [letter]. J Neurol Neurosurg Psychiatry 67:252-253. Juurlink BH, Paterson PG (1998) Review of oxidative stress in brain and spinal cord injury: suggestions for pharmacological and nutritional management strategies. J Spinal Cord Med 21:309-334. Kachur AV, Koch CJ, Biaglow JE (1999) Mechanism of copper-catalyzed autoxidation of cysteine. Free Radic Res 31:23-34. Kadle R, Suksang C, Roberson ED, Fellows RE (1988) Identification of an insulin-like factor in astrocyte conditioned medium. Brain Res 460:60-67. Kafitz KW, Rose CR, Thoenen H, Konnerth A (1999) Neurotrophin-evoked rapid excitation through TrkB receptors. Nature 401:918-921. Kanai Y (1997) Family of neutral and acidic amino acid transporters: molecular biology, physiology and medical implications. Curr Opin Cell Biol 9:565-572. Kanfer S, Turro NJ (1981) Reactive forms of oxygen. In: Oxygen and living processes, a interdisciplinary approach (Gilbert DL, ed), pp 47-64. New York: Springer-Verlag. Kannan R, Kuhlenkamp JF, Jeandidier E, Trinh H, Ookhtens M, KaplowitzN (1990) Evidence for carrier-mediated transport of glutathione across the blood- brain barrier in the rat. J Clin Invest 85:2009-2013. Kapaki E, Zournas C, Kanias G, Zambelis T, Kakami A, Papageorgiou C (1997) Essential trace element alterations in amyotrophic lateral sclerosis. J Neurol Sci 147:171-175. Karlsen RL, Grofova I, Malthe-Sorenssen D, Fonnum F (1981) Morphological changes in rat brain induced by L-cysteine injection in newborn animals. Brain Res 208:167-180. Katoh-Semba R, Keino H, Kashiwamata S (1988) A possible contribution by glial cells to neuronal energy production: enzyme-histochemical studies in the developing rat cerebellum. Cell Tissue Res 252:133-139. Katsura K, Kristian T, Siesjo BK (1994) Energy metabolism, ion homeostasis, and cell damage in the brain. Biochem Soc Trans 22:991-996. 113 References Keller HJ, Do KQ, Zollinger M, Winterhalter KH, Cuenod M (1989) Cysteine: depolarization-induced release from rat brain in vitro. J Neurochem 52:1801-1806. Kerr JF, Wyllie AH, Currie AR (1972) Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics. Br J Cancer 26:239-257. Kilberg MS, Handlogten ME, Christensen HN (1981) Characteristics of system ASC for transport of neutral amino acids in the isolated rat hepatocyte. J Biol Chem 256:3304-3312. Kimelberg HK, Goderie SK, Higman S, Pang S, Waniewski RA (1990) Swelling-induced release of glutamate, aspartate, and taurine from astrocyte cultures. J Neurosci 10:1583-1591. Kirschner PB, Jenkins BG, Schulz JB, Finkelstein SP, Matthews RT, Rosen BR, Beal MF (1996) NGF, BDNF and NT-5, but not NT-3 protect against MPP+ toxicity and oxidative stress in neonatal animals. Brain Res 713:178-185. Kleinman WA, Richie JP, Jr. (2000) Status of glutathione and other thiols and disulfides in human plasma. Biochem Pharmacol 60:19-29. Klomp LW, Farhangrazi ZS, Dugan LL, Gitlin JD (1996) Ceruloplasmin gene expression in the murine central nervous system. J Clin Invest 98:207-215. Kobayashi H, Uchida Y, Ihara Y, Nakajima K, Kohsaka S, Miyatake T, Tsuji S (1993) Molecular cloning of rat growth inhibitory factor cDNA and the expression in the central nervous system. Brain Res Mol Brain Res 19:188-194. Kogawa S, Yasuda H, Terada M, Maeda K, Kikkawa R (2000) Apoptosis and impaired axonal regeneration of sensory neurons after nerve crush in diabetic rats. Neuroreport 11:663-667. Kohr G, Eckardt S, Luddens H, Monyer H, Seeburg PH (1994) NMDA receptor channels: subunit-specific potentiation by reducing agents. Neuron 12:1031-1040. Kosower NS, Kosower EM (1983) Glutathione and cell membrane thiol status. In: Functions of glutathione. Biochemical, physiological, toxicological and clinical aspects. (Larsson A, Orrenius S, Holmgren A, Mannervik B, eds), pp 307-316. New York: Raven Press. Kranich O, Hamprecht B, Dringen R (1996) Different preferences in the utilization of amino acids for glutathione synthesis in cultured neurons and astroglial cells derived from rat brain. Neurosci Lett 219:211-214. Kreutzberg GW (1996) Microglia: a sensor for pathological events in the CNS. Trends Neurosci 19:312-318. Kruman I, Bruce-Keller AJ, Bredesen D, Waeg G, Mattson MP (1997) Evidence that 4-hydroxynonenal mediates oxidative stress-induced neuronal apoptosis. J Neurosci 17:5089-5100. Lai CT, Yu PH (1997) Dopamine- and L-beta-3,4-dihydroxyphenylalanine hydrochloride (L-Dopa)-induced cytotoxicity towards catecholaminergic neuroblastoma SH-SY5Y cells. Effects of oxidative stress and antioxidative factors. Biochem Pharmacol 53:363-372. Landvik S (1997) Vitamin E. In: Oxidants,antioxidants, and free radicals. (Baskin SI, Salem H, eds), pp 79-94. Washington, DC: Taylor & Francis. 114 References Langeveld CH, Jongenelen CA, Schepens E, Stoof JC, Bast A, Drukarch B (1995) Cultured rat striatal and cortical astrocytes protect mesencephalic dopaminergic neurons against hydrogen peroxide toxicity independent of their effect on neuronal development. Neurosci Lett 192:13-16. Le Roux PD, Reh TA (1994) Regional differences in glial-derived factors that promote dendritic outgrowth from mouse cortical neurons in vitro. J Neurosci 14:4639-4655. Lebovitz RM, Zhang H, Vogel H, Cartwright J, Jr., Dionne L, Lu N, Huang S, Matzuk MM (1996) Neurodegeneration, myocardial injury, and perinatal death in mitochondrial superoxide dismutase-deficient mice. Proc Natl Acad Sci U S A 93:9782-9787. Lee SC, Liu W, Dickson DW, Brosnan CF, Berman JW (1993) Cytokine production by human fetal microglia and astrocytes. Differential induction by lipopolysaccharide and IL-1 beta. J Immunol 150:2659-2667. Lehmann A, Hagberg H, Orwar O, Sandberg M (1993) Cysteine sulphinate and cysteate: mediators of cysteine toxicity in the neonatal rat brain? Eur J Neurosci 5:1398-1412. Lehre KP, Levy LM, Ottersen OP, Storm-Mathisen J, Danbolt NC (1995) Differential expression of two glial glutamate transporters in the rat brain: quantitative and immunocytochemical observations. J Neurosci 15:1835-1853. Levander OA (1987) A global view of human selenium nutrition. Annu Rev Nutr 7:227-250. Levi G, Gallo V (1995) Release of neuroactive amino acids from glia. In: Neuroglia (Kettenmann H, Ransom BR, eds), pp 815-828. New York: Oxford. Levin LA, Geszvain KM (1998) Expression of ceruloplasmin in the retina: induction after optic nerve crush. Invest Ophthalmol Vis Sci 39:157-163. LeVine SM, Wulser MJ, Lynch SG (1998) Iron quantification in cerebrospinal fluid. Anal Biochem 265:74-78. Levison SW, McCarthy KD (1991) Astroglia in culture. In: Culturing nerve cells (Banker G, Goslin K, eds), pp 309-336. Cambridge: MIT Press. Lewen A, Matz P, Chan PH (2000) Free radical pathways in CNS injury. J Neurotrauma 17:871-890. Li X, Wallin C, Weber SG, Sandberg M (1999) Net efflux of cysteine, glutathione and related metabolites from rat hippocampal slices during oxygen/glucose deprivation: dependence on gamma-glutamyl transpeptidase. Brain Res 815:81-88. Lin LF, Doherty DH, Lile JD, Bektesh S, Collins F (1993) GDNF: a glial cell line-derived neurotrophic factor for midbrain dopaminergic neurons [see comments]. Science 260:1130-1132. Lindsay RM (1995) Neuron saving schemes. Nature 373:289-290. Liu X, Kim CN, Yang J, Jemmerson R, Wang X (1996) Induction of apoptotic program in cell-free extracts: requirement for dATP and cytochrome c. Cell 86:147-157. Lo SK, Janakidevi K, Lai L, Malik AB (1993) Hydrogen peroxide-induced increase in endothelial adhesiveness is dependent on ICAM-1 activation. Am J Physiol 264:L406-412. 115 References Loeffler DA, Le Witt PA, Juneau PL, Sima AA, Nguyen HU, DeMaggio AJ, Brickman CM, Brewer GJ, Dick RD, Troyer MD, Kanaley L (1996) Increased regional brain concentrations of ceruloplasmin in neurodegenerative disorders. Brain Res 738:265-274. Loudianos G, Gitlin JD (2000) Wilson's disease. Semin Liver Dis 20:353-364. Louvel E, Hugon J, Doble A (1997) Therapeutic advances in amyotrophic lateral sclerosis [see comments]. Trends Pharmacol Sci 18:196-203. Lovell MA, Ehmann WD, Butler SM, Markesbery WR (1995) Elevated thiobarbituric acid-reactive substances and antioxidant enzyme activity in the brain in Alzheimer's disease. Neurology 45:1594-1601. Lovell MA, Robertson JD, Teesdale WJ, Campbell JL, Markesbery WR (1998) Copper, iron and zinc in Alzheimer's disease senile plaques. J Neurol Sci 158:47-52. Lucius R, Sievers J (1996) Postnatal retinal ganglion cells in vitro: protection against reactive oxygen species (ROS)-induced axonal degeneration by cocultured astrocytes. Brain Res 743:56-62. Luetjens CM, Bui NT, Sengpiel B, Munstermann G, Poppe M, Krohn AJ, Bauerbach E, Krieglstein J, Prehn JH (2000) Delayed mitochondrial dysfunction in excitotoxic neuron death: cytochrome c release and a secondary increase in superoxide production. J Neurosci 20:5715-5723. Maeda Y, Matsumoto M, Hori O, Kuwabara K, Ogawa S, Yan SD, Ohtsuki T, Kinoshita T, Kamada T, Stern DM (1994) Hypoxia/reoxygenation-mediated induction of astrocyte interleukin 6: a paracrine mechanism potentially enhancing neuron survival. J Exp Med 180:2297-2308. Makar TK, Nedergaard M, Preuss A, Gelbard AS, Perumal AS, Cooper AJ (1994) Vitamin E, ascorbate, glutathione, glutathione disulfide, and enzymes of glutathione metabolism in cultures of chick astrocytes and neurons: evidence that astrocytes play an important role in antioxidative processes in the brain. J Neurochem 62:45-53. Marchetti P, Decaudin D, Macho A, Zamzami N, Hirsch T, Susin SA, Kroemer G (1997) Redox regulation of apoptosis: impact of thiol oxidation status on mitochondrial function. Eur J Immunol 27:289-296. Mark RJ, Lovell MA, Markesbery WR, Uchida K, Mattson MP (1997) A role for 4-hydroxynonenal, an aldehydic product of lipid peroxidation, in disruption of ion homeostasis and neuronal death induced by amyloid beta-peptide. J Neurochem 68:255-264. Markesbery WR (1992) Alzheimer's disease. In: Diseases of the nervous system: clinical neurobiology., 2nd Edition (Asbury AK, McKhann GM, McDonald WI, eds), pp 795-803. Philadelphia: W B Saunders. Markesbery WR (1997) Oxidative stress hypothesis in Alzheimer's disease. Free Radic Biol Med 23:134-147. Martin DL (1992) Synthesis and release of neuroactive substances by glial cells. Glia 5:81-94. Martin DL (1995) The role of glia in the inactivation of neurotransmitters. In: Neuroglia (Kettenmann H, Ransom BR, eds), pp 732-745. New York: Oxford. 116 References Martin JB (1999) Molecular basis of the neurodegenerative disorders. N Engl J Med 340:1970-1980. Martin RL, Lloyd HG, Cowan AI (1994) The early events of oxygen and glucose deprivation: setting the scene for neuronal death? Trends Neurosci 17:251-257. Masuda S, Okano M, Yamagishi K, Nagao M, Ueda M, Sasaki R (1994) A novel site of erythropoietin production. Oxygen-dependent production in cultured rat astrocytes. J Biol Chem 269:19488-19493. Matsumoto K, Yamada K, Kohmura E, Kinoshita A, Hayakawa T (1994) Role of pyruvate in ischaemia-like conditions on cultured neurons. Neurol Res 16:460-464. Mattson MP (1998) Modification of ion homeostasis by lipid peroxidation: roles in neuronal degeneration and adaptive plasticity. Trends Neurosci 21:53-57. McCaffery CA, Raju TR, Bennett MR (1984) Effects of cultured astroglia on the survival of neonatal rat retinal ganglion cells in vitro. Dev Biol 104:441-448. McCord JM, Day ED, Jr. (1978) Superoxide-dependent production of hydroxyl radical catalyzed by iron-EDTA complex. FEBS Lett 86:139-142. McGahon BM, Murray CA, Horrobin DF, Lynch (1999) Age-related changes in oxidative mechanisms and LTP are reversed by dietary manipulation. Neurobiol Aging 20:643-653. McGale EH, Pye IF, Stonier C, Hutchinson EC, Aber GM (1977) Studies of the inter-relationship between cerebrospinal fluid and plasma amino acid concentrations in normal individuals. J Neurochem 29:291-297. Mecocci P, MacGarvey U, Beal MF (1994) Oxidative damage to mitochondrial DNA is increased in Alzheimer's disease. Ann Neurol 36:747-751. Mecocci P, MacGarvey U, Kaufman AE, Koontz D, Shoffner JM, Wallace DC, Beal MF (1993) Oxidative damage to mitochondrial DNA shows marked age-dependent increases in human brain. Ann Neurol 34:609-616. Meister A (1989) Metabolism and function of glutathione. In: Glutathione (Dolphin A, Poulson R, Avramovic O, eds), pp 367-474. New York: John Wiley & Sons. Meister A (1992) On the antioxidant effects of ascorbic acid and glutathione. Biochem Pharmacol 44:1905-1915. Meister A (1995) Glutathione metabolism. Methods Enzymol 251:3-7. Meister A, Anderson ME (1983) Glutathione. Annu Rev Biochem 52:711-760. Mello Filho AC, Meneghini R (1984) In vivo formation of single-strand breaks in DNA by hydrogen peroxide is mediated by the Haber-Weiss reaction. Biochim Biophys Acta 781:56-63. Melner MH, Low KG, Allen RG, Nielsen CP, Young SL, Saneto RP (1990) The regulation of proenkephalin expression in a distinct population of glial cells. Embo J 9:791-796. Mena MA, Casarejos MJ, Carazo A, Paino CL, Garcia de Yebenes J (1997) Glia protect fetal midbrain dopamine neurons in culture from L-DOPA toxicity through multiple mechanisms. J Neural Transm 104:317-328. 1 1 7 References Michaels RL, Rothman SM (1990) Glutamate neurotoxicity in vitro: antagonist pharmacology and intracellular calcium concentrations. J Neurosci 10:283-292. Miller LP, Oldendorf WH (1986) Regional kinetic constants for blood-brain barrier pyruvic acid transport in conscious rats by the monocarboxylic acid carrier. J Neurochem 46:1412-1416. Miyata M, Smith JD (1996) Apolipoprotein E allele-specific antioxidant activity and effects on cytotoxicity by oxidative insults and beta-amyloid peptides. Nat Genet 14:55-61. Mojzsis SJ, Arrhenius G, McKeegan KD, Harrison TM, Nutman AP, Friend CR (1996) Evidence for life on Earth before 3,800 million years ago. Nature 384:55-59. Montine TJ, Huang DY, Valentine WM, Amarnath V, Saunders A, Weisgraber KH, Graham DG, Strittmatter WJ (1996) Crosslinking of apolipoprotein E by products of lipid peroxidation. J Neuropathol Exp Neurol 55:202-210. Moore P, Strauss BS (1979) Sites of inhibition of in vitro DNA synthesis in carcinogen- and UV- treated phi XI74 DNA. Nature 278:664-666. Morrison BM, Morrison JH (1999) Amyotrophic lateral sclerosis associated with mutations in superoxide dismutase: a putative mechanism of degeneration. Brain Res Brain Res Rev 29:121-135. Multhaup G, Ruppert T, Schlicksupp A, Hesse L, Bill E, Pipkorn R, Masters CL, Beyreuther K (1998) Copper-binding amyloid precursor protein undergoes a site-specific fragmentation in the reduction of hydrogen peroxide. Biochemistry 37:7224-7230. Munday R (1988) Dialuric acid autoxidation. Effects of transition metals on the reaction rate and on the generation of "active oxygen" species. Biochem Pharmacol 37:409-413. Munday R (1989) Toxicity of thiols and disulphides: involvement of free-radical species. Free Radic Biol Med 7:659-673. Murphy TH, Schnaar RL, Coyle JT (1990) Immature cortical neurons are uniquely sensitive to glutamate toxicity by inhibition of cystine uptake. Faseb J 4:1624-1633. Murray RK, Granner DK, Mayes PA, Rodwell VW (1993) Harper's Biochemistry, 23th Edition. Norwalk, CT: Appleton & Lange. Nagano S, Ogawa Y, Yanagihara T, Sakoda S (1999) Benefit of a combined treatment with trientine and ascorbate in familial amyotrophic lateral sclerosis model mice. Neurosci Lett 265:159-162. Nagayama T, Lan J, Henshall DC, Chen D, O'Horo C, Simon RP, Chen J (2000) Induction of oxidative DNA damage in the peri-infarct region after permanent focal cerebral ischemia. J Neurochem 75:1716-1728. Nair V, Cooper CS, Vietti DE, Turner GA (1986) The chemistry of lipid peroxidation metabolites: crosslinking reactions of malondialdehyde. Lipids 21:6-10. Nath KA, Salahudeen AK (1993) Autoxidation of cysteine generates hydrogen peroxide: cytotoxicity and attenuation by pyruvate. Am J Physiol 264:F306-314. Newman EA (1995) Glial cell regulation of extracellular potassium. In: Neuroglia (Kettenmann H, Ransom BR, eds), pp 717-731. New York: Oxford. 118 References Nishimoto I, Okamoto T, Giambarella U, Iwatsubo T (1997) Apoptosis in neurodegenerative diseases. Adv Pharmacol 41:337-368. Nishiyama M, Hori N, Watanabe T, Hori T, Ogata K, Watanabe K, Maru E, Shimizu T (1999) Lipid peroxides and neuronal plasticity. Adv Exp Med Biol 469:43-48. Noel F, Tofilon PJ (1998) Astrocytes protect against X-ray-induced neuronal toxicity in vitro. Neuroreport 9:1133-1137. Nunn JF (1997) Oxygen appeared in the Earth's atmosphere 3 million years ago. Eur J Anaesthesiol 14:228. O'Donnell-Tormey J, Nathan CF, Lanks K, DeBoer CJ, de la Harpe J (1987) Secretion of pyruvate. An antioxidant defense of mammalian cells. J Exp Med 165:500-514. Oestreicher AB (1986) Neuronal cell membranes and brain aging. In: Progress in Brain Research Vol. 70 (Swaab DF, Fliers E, Mirmiran M, Van Gool WA, Van Haaren F, eds), pp 239-254. Amsterdam: Elsevier Science. Ogihara H, Ogihara T, Miki M, Yasuda H, Mino M (1995) Plasma copper and antioxidant status in Wilson's disease. Pediatr Res 37:219-226. Oh LY, Yong VW (1996) Astrocytes promote process outgrowth by adult human oligodendrocytes in vitro through interaction between bFGF and astrocyte extracellular matrix. Glia 17:237-253. Ohno K, Suzumura A, Sawada M, Marunouchi T (1990) Production of granulocyte/macrophage colony-stimulating factor by cultured astrocytes. Biochem Biophys Res Commun 169:719-724. Olney JW, Ho OL (1970) Brain damage in infant mice following oral intake of glutamate, aspartate or cysteine. Nature 227:609-611. Olney JW, Ho OL, Rhee V (1971) Cytotoxic effects of acidic and sulphur containing amino acids on the infant mouse central nervous system. Exp Brain Res 14:61-76. Olney JW, Ho OL, Rhee V, Schainker B (1972) Cysteine-induced brain damage in infant and fetal rodents. Brain Res 45:309-313. Olney JW, Zorumski C, Price MT, Labruyere J (1990) L-cysteine, a bicarbonate-sensitive endogenous excitotoxin. Science 248:596-599. Orlowski M, Meister A (1970) The gamma-glutamyl cycle: a possible transport system for amino acids. Proc Natl Acad Sci U S A 67:1248-1255. Orte C, Lawrenson JG, Finn TM, Reid AR, Allt G (1999) A comparison of blood-brain barrier and blood-nerve barrier endothelial cell markers. Anat Embryol (Berl) 199:509-517. Packer L, Witt EH, Tritschler HJ (1995) alpha-Lipoic acid as a biological antioxidant. Free Radic Biol Med 19:227-250. Pahl HL, Baeuerle PA (1994) Oxygen and the control of gene expression. Bioessays 16:497-502. Palm R, Hallmans G (1982) Zinc concentrations in the cerebrospinal fluid of normal adults and patients with neurological diseases. J Neurol Neurosurg Psychiatry 45:685-690. Palmiter RD, Findley SD, Whitmore TE, Durnam DM (1992) MT-III, a brain-specific member of the metallothionein gene family. Proc Natl Acad Sci U S A 89:6333-6337. 119 References Pardo B, Paino CL, Casarejos MJ, Mena MA (1997) Neuronal-enriched cultures from embryonic rat ventral mesencephalon for pharmacological studies of dopamine neurons. Brain Res Brain Res Protoc 1:127-132. Pardo CA, Xu Z, Borchelt DR, Price DL, Sisodia SS, Cleveland DW (1995) Superoxide dismutase is an abundant component in cell bodies, dendrites, and axons of motor neurons and in a subset of other neurons. Proc Natl Acad Sci U S A 92:954-958. Pardridge WM, Oldendorf WH (1977) Transport of metabolic substrates through the blood-brain barrier. J Neurochem 28:5-12. Pean AR, Parsons RB, Waring RH, Williams AC, Ramsden DB (1995) Toxicity of sulphur-containing compounds to neuronal cell lines. J Neurol Sci 129 Suppl:107-108. Pellerin L, Magistretti PJ (1994) Glutamate uptake into astrocytes stimulates aerobic glycolysis: a mechanism coupling neuronal activity to glucose utilization. Proc Natl Acad Sci U S A 91:10625-10629. Perry TL, Hansen S, Kennedy J (1975) CSF amino acids and plasma--CSF amino acid ratios in adults. J Neurochem 24:587-589. Peters O, Back T, Lindauer U, Busch C, Megow D, Dreier J, Dirnagl U (1998) Increased formation of reactive oxygen species after permanent and reversible middle cerebral artery occlusion in the rat. J Cereb Blood Flow Metab 18:196-205. Phelps DT, Deneke SM, Daley DL, Fanburg BL (1992) Elevation of glutathione levels in bovine pulmonary artery endothelial cells by N-acetylcysteine. Am J Respir Cell Mol Biol 7:293-299. Poole RC, Halestrap AP (1993) Transport of lactate and other monocarboxylates across mammalian plasma membranes. Am J Physiol 264:C761-782. Pratico D, Delanty N (2000) Oxidative injury in diseases of the central nervous system: focus on alzheimer's disease. Am J Med 109:577-585. Prohaska JR (1997) Neurochemical roles of copper as antioxidant or prooxidant. In: Metals and oxidative damage in neurological disorders. (Connor JR, ed), pp 57-75. New York: Plenum. Pryor WA (1986) Oxy-radicals and related species: their formation, lifetimes, and reactions. Annu Rev Physiol 48:657-667. Puka-Sundvall M, Eriksson P, Nilsson M, Sandberg M, Lehmann A (1995) Neurotoxicity of cysteine: interaction with glutamate. Brain Res 705:65-70. Radi R, Beckman JS, Bush KM, Freeman BA (1991) Peroxynitrite oxidation of sulfhydryls. The cytotoxic potential of superoxide and nitric oxide. J Biol Chem 266:4244-4250. Rae TD, Schmidt PJ, Pufahl RA, Culotta VC, O'Halloran TV (1999) Undetectable intracellular free copper: the requirement of a copper chaperone for superoxide dismutase. Science 284:805-808. Raff MC, Barres BA, Burne JF, Coles HS, Ishizaki Y, Jacobson MD (1993) Programmed cell death and the control of cell survival: lessons from the nervous system. Science 262:695-700. 120 References Rakic P (1995) Radial glial cells: scaffolding for brain construction. In: Neuroglia (Kettenmann H, Ransom BR, eds), pp 746-762. New York: Oxford. Raps SP, Lai JC, Hertz L, Cooper AJ (1989) Glutathione is present in high concentrations in cultured astrocytes but not in cultured neurons. Brain Res 493:398-401. Ratan RR, Murphy TH, Baraban JM (1994) Macromolecular synthesis inhibitors prevent oxidative stress-induced apoptosis in embryonic cortical neurons by shunting cysteine from protein synthesis to glutathione. J Neurosci 14:4385-4392. Reaume AG, Elliott JL, Hoffman EK, Kowall NW, Ferrante RJ, Siwek DF, Wilcox HM, Flood DG, Beal MF, Brown RH, Scott RW, Snider WD (1996) Motor neurons in Cu/Zn superoxide dismutase-deficient mice develop normally but exhibit enhanced cell death after axonal injury. Nat Genet 13:43-47. Reed DJ, Savage MK (1995) Influence of metabolic inhibitors on mitochondrial permeability transition and glutathione status. Biochim Biophys Acta 1271:43-50. Reed DJ, Babson JR, Beatty PW, Brodie AE, Ellis WW, Potter DW (1980) High-performance liquid chromatography analysis of nanomole levels of glutathione, glutathione disulfide, and related thiols and disulfides. Anal Biochem 106:55-62. Regan RF, Guo Y (1999) Extracellular reduced glutathione increases neuronal vulnerability to combined chemical hypoxia and glucose deprivation. Brain Res 817:145-150. Reichenbach A, Robinson SR (1995) Ependymoglia and ependymoglia-like cells. In: Neuroglia (Kettenmann H, Ransom BR, eds), pp 58-84. New York, NY: Oxford University Press. Ripps ME, Huntley GW, Hof PR, Morrison JH, Gordon JW (1995) Transgenic mice expressing an altered murine superoxide dismutase gene provide an animal model of amyotrophic lateral sclerosis. Proc Natl Acad Sci U S A 92:689-693. Risau W, Dingier A, Albrecht U, Dehouck MP, Cecchelli R (1992) Blood-brain barrier pericytes are the main source of gamma- glutamyltranspeptidase activity in brain capillaries. J Neurochem 58:667-672. Roberfroid M, Calderon PB (1995) Free radicals and oxidation phenomena in biological systems. New York: Marcel Dekker. Romijn HJ, van Huizen F, Wolters PS (1984) Towards an improved serum-free, chemically defined medium for long-term culturing of cerebral cortex tissue. Neurosci Biobehav Rev 8:301-334. Rose RC (1993) Cerebral metabolism of oxidized ascorbate. Brain Res 628:49-55. Rosen DR, Siddique T, Patterson D, Figlewicz DA, Sapp P, Hentati A, Donaldson D, Goto J, O'Regan JP, Deng HX, et al. (1993) Mutations in Cu/Zn superoxide dismutase gene are associated with familial amyotrophic lateral sclerosis [published erratum appears in Nature 1993 Jul 22;364(6435):362] [see comments]. Nature 362:59-62. Rosen GM, Britigan BE, Halpern HJ, Pou S (1999) Free radicals: biology and detection by spin trapping. New York: Oxford. 121 References Rosenzweig AC, O'Halloran TV (2000) Structure and chemistry of the copper chaperone proteins. Curr Opin Chem Biol 4:140-147. Rothstein JD, Van Kammen M, Levey AI, Martin LJ, Kuncl RW (1995) Selective loss of glial glutamate transporter GLT-1 in amyotrophic lateral sclerosis. Ann Neurol 38:73-84. Rothstein JD, Martin L, Levey AI, Dykes-Hoberg M, Jin L, Wu D, Nash N, Kuncl RW (1994) Localization of neuronal and glial glutamate transporters. Neuron 13:713-725. Rothstein JD, Dykes-Hoberg M, Pardo CA, Bristol LA, Jin L, Kuncl RW, Kanai Y, Hediger MA, Wang Y, Schielke JP, Welty DF (1996) Knockout of glutamate transporters reveals a major role for astroglial transport in excitotoxicity and clearance of glutamate. Neuron 16:675-686. Rowland LP, Fink ME, Rubin L (1991) Cerebrospinal fluid: blood-brain barrier, brain edema, and hydrocephalus. In: Principles of neural science, 3rd Edition (Kandel ER, Schwartz JH, Jessell TM, eds), pp 1050-1060. Norwalk, CT: Appleton & Lange. Rowley DA, Halliwell B (1983) DNA damage by superoxide-generating systems in relation to the mechanism of action of the anti-tumour antibiotic adriamycin. Biochim Biophys Acta 761:86-93. Saez G, Thornalley PJ, Hill HA, Hems R, Bannister JV (1982) The production of free radicals during the autoxidation of cysteine and their effect on isolated rat hepatocytes. Biochim Biophys Acta 719:24-31. Sagara J, Makino N, Bannai S (1996) Glutathione efflux from cultured astrocytes. J Neurochem 66:1876-1881. Sagara JI, Miura K, Bannai S (1993) Maintenance of neuronal glutathione by glial cells. J Neurochem 61:1672-1676. Sakanaka M, Wen TC, Matsuda S, Masuda S, Morishita E, Nagao M, Sasaki R (1998) In vivo evidence that erythropoietin protects neurons from ischemic damage. Proc Natl Acad Sci U S A 95:4635-4640. Sass JB, Ang LC, Juurlink BH (1993) A simple, yet versatile, co-culture method for examining neuron-glia interactions. J Neurosci Methods 47:115-121. Sastry PS (1985) Lipids of nervous tissue: composition and metabolism. Prog Lipid Res 24:69-176. Saunders AM, Strittmatter WJ, Schmechel D, George-Hyslop PH, Pericak-Vance MA, Joo SH, Rosi BL, Gusella JF, Crapper-MacLachlan DR, Alberts MJ, et al. (1993) Association of apolipoprotein E allele epsilon 4 with late-onset familial and sporadic Alzheimer's disease. Neurology 43:1467-1472. Sayre LM, Perry G, Harris PL, Liu Y, Schubert KA, Smith MA (2000) In situ oxidative catalysis by neurofibrillary tangles and senile plaques in Alzheimer's disease: a central role for bound transition metals. J Neurochem 74:270-279. Schatzl HM (1995) Neurotrophic factors: ready to go? Trends Neurosci 18:463-464. Scheinberg IH, Sternlieb I (1983) Wilson's disease. Philadelphia, PA: WB Saunders. 122 References Schinder AF, Olson EC, SpitzerNC, Montal M (1996) Mitochondrial dysfunction is a primary event in glutamate neurotoxicity. J Neurosci 16:6125-6133. Schmalenbach C, Muller HW (1993) Astroglia-neuron interactions that promote long-term neuronal survival. J Chem Neuroanat 6:229-237. Schousboe A, Westergaard N, Waagepetersen HS, Larsson OM, Bakken IJ, Sonnewald U (1997) Trafficking between glia and neurons of TCA cycle intermediates and related metabolites. Glia 21:99-105. Schreck R, Rieber P, Baeuerle PA (1991) Reactive oxygen intermediates as apparently widely used messengers in the activation of the NF-kappa B transcription factor and HIV-1. Embo J 10:2247-2258. Schwartzman RA, Cidlowski JA (1994) Glucocorticoid-induced apoptosis of lymphoid cells. Int Arch Allergy Immunol 105:347-354. Selak I, Skaper SD, Varon S (1985) Pyruvate participation in the low molecular weight trophic activity for central nervous system neurons in glia-conditioned media. J Neurosci 5:23-28. Sen CK (1998) Redox signaling and the emerging therapeutic potential of thiol antioxidants. Biochem Pharmacol 55:1747-1758. Shafqat S, Tamarappoo BK, Kilberg MS, Puranam RS, McNamara JO, Guadano-Ferraz A, Fremeau RT, Jr. (1993) Cloning and expression of a novel Na(+)-dependent neutral amino acid transporter structurally related to mammalian Na+/glutamate cotransporters. J Biol Chem 268:15351-15355. Shen XM, Dryhurst G (1998) Iron- and manganese-catalyzed autoxidation of dopamine in the presence of L-cysteine: possible insights into iron- and manganese-mediated dopaminergic neurotoxicity. Chem Res Toxicol 11:824-837. Shibutani S, Takeshita M, Grollman AP (1991) Insertion of specific bases during DNA synthesis past the oxidation- damaged base 8-oxodG. Nature 349:431-434. Siddique T, Deng HX (1996) Genetics of amyotrophic lateral sclerosis. Hum Mol Genet 5:1465-1470. Slater AF, Nobel CS, Orrenius S (1995) The role of intracellular oxidants in apoptosis. Biochim Biophys Acta 1271:59-62. Slater TF, Sawyer B, Strauli U (1963) Studies on succinate-tetrazolium redictase systems. III. Points of coupling of four different tetrazolium salts. Biochim Biophs Acta 77:383-393. Slot JW, Geuze HJ, Freeman BA, Crapo JD (1986) Intracellular localization of the copper-zinc and manganese superoxide dismutases in rat liver parenchymal cells. Lab Invest 55:363-371. Smith CD, Carney JM, Starke-Reed PE, Oliver CN, Stadtman ER, Floyd RA, Markesbery WR (1991) Excess brain protein oxidation and enzyme dysfunction in normal aging and in Alzheimer disease. Proc Natl Acad Sci U S A 88:10540-10543. Smith MA, Sayre LM, Monnier VM, Perry G (1995) Radical AGEing in Alzheimer's disease. Trends Neurosci 18:172-176. 123 References Smith MA, Richey Harris PL, Sayre LM, Beckman JS, Perry G (1997) Widespread peroxynitrite-mediated damage in Alzheimer's disease. J Neurosci 17:2653-2657. Smith MA, Rottkamp CA, Nunomura A, Raina AK, Perry G (2000) Oxidative stress in Alzheimer's disease. Biochim Biophys Acta 1502:139-144. Soto-Otero R, Mendez-Alvarez E, Hermida-Ameijeiras A, Munoz-Patino AM, Labandeira-Garcia JL (2000) Autoxidation and neurotoxicity of 6-hydroxydopamine in the presence of some antioxidants: potential implication in relation to the pathogenesis of Parkinson's disease. J Neurochem 74:1605-1612. Spector R, Eells J (1984) Deoxynucleoside and vitamin transport into the central nervous system. Fed Proc 43:196-200. Spencer PS, Roy DN, Ludolph A, Hugon J, Dwivedi MP, Schaumburg HH (1986) Lathyrism: evidence for role of the neuroexcitatory aminoacid BOAA. Lancet 2:1066-1067. Spencer PS, Nunn PB, Hugon J, Ludolph AC, Ross SM, Roy DN, Robertson RC (1987) Guam amyotrophic lateral sclerosis-parkinsonism-dementia linked to a plant excitant neurotoxin. Science 237:517-522. Streit WJ (1995) Microglial cells. In: Neuroglia (Kettenmann H, Ransom BR, eds), pp 85-96. New York, NY: Oxford University Press. Stryer L (1988) Biochemistry, 3rd Edition. New York: W H Freeman. Stuerenburg HJ (2000) CSF copper concentrations, blood-brain barrier function, and coeruloplasmin synthesis during the treatment of Wilson's disease. J Neural Transm 107:321-329. Subbarao KV, Richardson JS, Ang LC (1990) Autopsy samples of Alzheimer's cortex show increased peroxidation in vitro. J Neurochem 55:342-345. Suberville C, Higueret P, Taruoura D, Garcin H, Higueret D (1987) Relative contribution of cysteine and methionine to glutathione content and thyroid hormone levels in the rat. Br J Nutr 58:105-111. SucherNJ, Lipton SA (1991) Redox modulatory site of the NMDA receptor-channel complex: regulation by oxidized glutathione. J Neurosci Res 30:582-591. Suzuki YJ, Forman HJ, Sevanian A (1997) Oxidants as stimulators of signal transduction. Free Radic Biol Med 22:269-285. Symons MCR (1991) Free radical in biological systems. In: Trace elements, micronutrients, and free radicals (Dreosti IE, ed), pp 1-24. Totowa, NJ: Humana Press. Szuchet S (1995) The morphology and ultrastructure of oligodendrocytes and their functional implications. In: Neuroglia (Kettenmann H, Ransom BR, eds), pp 23-43. New York, NY: Oxford University Press. Takahashi K, Avissar N, Whitin J, Cohen H (1987) Purification and characterization of human plasma glutathione peroxidase: a selenoglycoprotein distinct from the known cellular enzyme. Arch Biochem Biophys 256:677-686. 124 References Tanaka J, Toku K, Zhang B, Ishihara K, Sakanaka M, Maeda N (1999) Astrocytes prevent neuronal death induced by reactive oxygen and nitrogen species. Glia 28:85-96. Tanzi RE, Petrukhin K, Chernov I, Pellequer JL, Wasco W, Ross B, Romano DM, Parano E, Pavone L, Brzustowicz LM, et al. (1993) The Wilson disease gene is a copper transporting ATPase with homology to the Menkes disease gene. Nat Genet 5:344-350. Tate SS, Ross LL, Meister A (1973) The -glutamyl cycle in the choroid plexus: its possible function in amino acid transport. Proc Natl Acad Sci U S A 70:1447-1449. Tatton WG, Olanow CW (1999) Apoptosis in neurodegenerative diseases: the role of mitochondria. Biochim Biophys Acta 1410:195-213. Thoenen H (1995) Neurotrophins and neuronal plasticity. Science 270:593-598. Torreilles F, Salman-Tabcheh S, Guerin M, Torreilles J (1999) Neurodegenerative disorders: the role of peroxynitrite. Brain Res Brain Res Rev 30:153-163. Travis J (1994) Glia: the brain's other cells. Science 266:970-972. Trump BF, Berezesky IK (1995) Calcium-mediated cell injury and cell death. Faseb J 9:219-228. Tsacopoulos M, Magistretti PJ (1996) Metabolic coupling between glia and neurons. J Neurosci 16:877-885. Tymianski M, Wallace MC, Spigelman I, Uno M, Carlen PL, Tator CH, Charlton MP (1993) Cell-permeant Ca2+ chelators reduce early excitotoxic and ischemic neuronal injury in vitro and in vivo. Neuron 11:221-235. Uchida K, Stadtman ER (1992) Modification of histidine residues in proteins by reaction with 4-hydroxynonenal. Proc Natl Acad Sci U S A 89:4544-4548. Ursini F, Maiorino M, Gregolin C (1985) The selenoenzyme phospholipid hydroperoxide glutathione peroxidase. Biochim Biophys Acta 839:62-70. Utsunomiya-Tate N, Endou H, Kanai Y (1996) Cloning and functional characterization of a system ASC-like Na+- dependent neutral amino acid transporter. J Biol Chem 271:14883-14890. Vaca K, Wendt E (1992) Divergent effects of astroglial and microglial secretions on neuron growth and survival. Exp Neurol 118:62-72. Vannucci RC, Duffy TE (1976) Carbohydrate metabolism in fetal and neonatal rat brain during anoxia and recovery. Am J Physiol 230:1269-1275. Varon S, Skaper SD, Barbin G, Selak I, Manthorpe M (1984) Low molecular weight agents support survival of cultured neurons from the central nervous system. J Neurosci 4:654-658. Vernadakis A (1988) Neuron-glia interrelations. Int Rev Neurobiol 30:149-224. Vincent SR, Hope BT (1992) Neurons that say NO. Trends Neurosci 15:108-113. Vitek MP, Bhattacharya K, Glendening JM, Stopa E, Vlassara H, Bucala R, Manogue K, Cerami A (1994) Advanced glycation end products contribute to amyloidosis in Alzheimer disease. Proc Natl Acad SciUSA91:4766-4770. Von Korff RW (1969) Purity and stability of pyruvate and a-ketoglutarate. Methods Enzymol 13:519-523. 125 References Wade LA, Brady HM (1981) Cysteine and cystine transport at the blood-brain barrier. J Neurochem 37:730-734. Walsh E, Ueda Y, Nakanishi H, Yoshida K (1992) Neuronal survival and neurite extension supported by astrocytes co- cultured in transwells. Neurosci Lett 138:103-106. Walz W, Mukerji S (1988) Lactate production and release in cultured astrocytes. Neurosci Lett 86:296-300. Weiss JH, Choi DW (1988) Beta-N-methylamino-L-alanine neurotoxicity: requirement for bicarbonate as a cofactor. Science 241:973-975. White AR, Multhaup G, Maher F, Bellingham S, Camakaris J, Zheng H, Bush AI, Beyreuther K, Masters CL, Cappai R (1999) The Alzheimer's disease amyloid precursor protein modulates copper-induced toxicity and oxidative stress in primary neuronal cultures. J Neurosci 19:9170-9179. White RJ, Reynolds IJ (1996) Mitochondrial depolarization in glutamate-stimulated neurons: an early signal specific to excitotoxin exposure. J Neurosci 16:5688-5697. Wolburg H, Risau W (1995) Formation of the blood-brain barrier. In: Neuroglia (Kettenmann H, Ransom BR, eds), pp 763-776. New York: Oxford. Wullner U, Seyfried J, Groscurth P, Beinroth S, Winter S, Gleichmann M, Heneka M, Loschmann P, Schulz JB, Weller M, Klockgether T (1999) Glutathione depletion and neuronal cell death: the role of reactive oxygen intermediates and mitochondrial function. Brain Res 826:53-62. Wyllie AH, Kerr JF, Currie AR (1980) Cell death: the significance of apoptosis. Int Rev Cytol 68:251-306. Xiong J, Fischer WM, Inoue K, Nakahara M, Bauer CE (2000) Molecular evidence for the early evolution of photosynthesis. Science 289:1724-1730. Yamaguchi Y, Heiny ME, Gitlin JD (1993) Isolation and characterization of a human liver cDNA as a candidate gene for Wilson disease. Biochem Biophys Res Commun 197:271-277. Yan SD, Chen X, Schmidt AM, Brett J, Godman G, Zou YS, Scott CW, Caputo C, Frappier T, Smith MA, et al. (1994) Glycated tau protein in Alzheimer disease: a mechanism for induction of oxidant stress. Proc Natl Acad Sci U S A 91:7787-7791. Yoshida M, Saito H, Katsuki H (1995) Neurotrophic effects of conditioned media of astrocytes isolated from different brain regions on hippocampal and cortical neurons. Experientia 51:133-136. Yoshioka K, Nisimaru N, Yanai S, Shimoda H, Yamada K (2000) Characteristics of monocarboxylates as energy substrates other than glucose in rat brain slices and the effect of selective glial poisoning -a 3 IP NMR study. Neurosci Res 36:215-226. Yudkoff M, Pleasure D, Cregar L, Lin ZP, Nissim I, Stern J (1990) Glutathione turnover in cultured astrocytes: studies with [15N]glutamate. J Neurochem 55:137-145. Yuzaki M, Mikoshiba K, Kagawa Y (1993) Cerebellar astrocytes specifically support the survival of Purkinje cells in culture. Biochem Biophys Res Commun 197:123-129. Zangerle L, Cuenod M, Winterhalter KH, Do KQ (1992) Screening of thiol compounds: depolarization-induced release of glutathione and cysteine from rat brain slices. J Neurochem 59:181-189. 126 References Zhang HF, Ong WY, Leong SK, Laperche Y (1997) Species differences in the localisation of gamma-glutamyl transpeptidase immunopositive cells at the blood-brain interface. J Hirnforsch 38:323-330. 127 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items