@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix skos: . vivo:departmentOrSchool "Medicine, Faculty of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Guo, Ningning"@en ; dcterms:issued "2010-11-05T23:46:55Z"@en, "1991"@en ; vivo:relatedDegree "Master of Science - MSc"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description """The physiological significance of glutathione (GSH) in the mammalian central nervous system is still uncertain, although some evidence has indicated that GSH may play an important role in the CNS. To address the question of whether GSH may be a candidate for a neuropeptide in the CNS, one step is to establish that GSH receptors are present. In the present study, biotinyl-GSH was synthesized and purified to detect the GSH receptor in the CNS. Histochemical experiments showed that GSH binding sites appeared on the white matter ( such as cingulum, dorsal hippocampal commissure, cerebral peduncle, fasciculus retrbflexus, mammillothalamic tract etc.) of the rat brain. It thus suggested that the GSH receptors might be on astrocytes or oligodendrocytes. Radioactive receptor assays were performed on cultured astrocytocytes using [³⁵S]GSH. Scatchard analysis revealed two binding sites of K₁ = 4.67±0.75 nM, Bmax₂ =70±9.2 fmoles / 6.4 x10⁵ cells (or Bmax₁=6.6 x10⁴molecules / cell), Kd₂=35.14±2.1 nM, Bmax₂=260±12.77 fmole / 6.4 x10⁵ cell (or Bmax₂ = 2.4 x10⁵ molecules / cell). The association and dissociation kinetics studies gave a K₊₁ of 0.003nM⁻¹min¹, and a K₋₁ of 0.0168 min⁻¹for site I. These rate constants gave a K₁ of 5.6 nM, consistent with that from Scatchard analysis. Colloidal gold technique and immunofluorescence double staining also showed the GSH binding sites on cultured astrocytes, and suggested that the binding sites might be GSH receptors. The present study is the first to report the presence of GSH receptors on astrocytes. Based on receptor binding assays and cytochemical experiments, this study not only depicts the biochemical characteristics of GSH receptors in the brain, but also shows the receptor at the cellular level. These results support the view that GSH might be a neuroactively signal substance in the CNS."""@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/29856?expand=metadata"@en ; skos:note "GSH: A NEW CANDIDATE NEUROPEPTIDE IN THE CNS NINGNING GUO MS., NANJING UNIVERSITY, 1984 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES PROGRAM EN NEUROSCIENCE We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA March 1991 Q Ningning Guo In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of / Q - e ^ q x > S C l < g ^ K ^ g -The University of British Columbia Vancouver, Canada DE-6 (2/88) ABSTRACT The physiological significance of glutathione (GSH) in the mammalian central nervous system is still uncertain, although some evidence has indicated that GSH may play an important role in the CNS. To address the question of whether GSH may be a candidate for a neuropeptide in the CNS, one step is to establish that GSH receptors are present. In the present study, biotinyl-GSH was synthesized and purified to detect the GSH receptor in the CNS. Histochemical experiments showed that GSH binding sites appeared on the white matter ( such as cingulum, dorsal hippocampal commissure, cerebral peduncle, fasciculus retrbflexus, mammillothalamic tract etc.) of the rat brain. It thus suggested that the GSH receptors might be on astrocytes or oligodendrocytes. Radioactive receptor assays were performed on cultured astrocytocytes using [^5S]GSH. Scatchard analysis revealed two binding sites of Kd-| = 4.67±0.75 nM, Bmax2 =70±9.2 fmoles /6.4x105 cells (or Bmax-|=6.6x104molecules /cell), Kd2=35.14±2.1 nM, Bmax2 =260±12.77 fmole /6.4 x105 cell (or Bmax2 = 2.4 x10 5 molecules / cell). The association and dissociation kinetics studies gave a K+i of 0.003nM\"1min1, and a K-1 of 0.0168 min\"1 for site I. These rate constants gave a Kd-j of 5.6 nM, consistent with that from Scatchard analysis. Colloidal gold technique and immunofluorescence double staining also showed the GSH binding sites on cultured astrocytes, and suggested that the binding sites might be GSH receptors. The present study is the first to report the presence of GSH receptors on astrocytes. Based on receptor binding assays and cytochemical experiments, this study not only depicts the biochemical characteristics of GSH receptors in the brain, but also shows the receptor at the cellular level. These results support the view that GSH might be a neuroactively signal substance in the CNS. iii CONTENTS Abstract ii Table of contents iii List of tables iv List of figures v Acknowledgement vi I. Introduction 1 II. Experiments and Results 1. The biotinylated ligand as a probe for the related receptor Materials and Methods 6 Results and Discussion 7 2. Radioligand Receptor Assay M aterials and M ethods 1 2 Results and Discussion 1 3 3. The Colloidal Gold Technique For The Cellular Localization of GSH Binding Sites Materials and Methods 3 1 Results and Discussion 3 2 III. General Discussion 38 IV. References 50 iv LIST OF TABLES Table\"!. Kinetic and equilibrium constants for the specific binding of [ 3 5S]GSH to cultured astrocytes 25 Table 2. IC50S of different displacers for [3 5S]GSH binding 25 Table 3. Comparision of kinetic parameter of reactants for GSH 41 V LIST OF FIGURES Fig. 1. Illustration of reactions of glutathione S-conjugates in cells 2 Fig. 2. Mercapturic acid formation through conjugation with GSH 2 Fig. 3. The elution profile at 254 nm for the crude biotinylated reaction mixture on a Sephadex G-10 column 8 Fig. 4. Purification of biotinyl-GSH by reversed-phase HPLC 9 Fig. 5. Biotinyl-GSH binding sites in rat brain 11 Fig. 6. Effect of excess GSH on [35S]GSH binding 15 Fig. 7. Effect of cell concentration on [3 5S]GSH binding 16 Fig. 8. Time course of [3 5S]GSH to binding sites on astrocytes 17 Fig. 9. Binding of [3 5S]GSH to cultured astrocytes as a function of concentration 19 Fig. 10. Scatchard plot analysis of binding equilibrium of [3 5S]GSH to astrocytes 20 Fig.11. Rate of dissociation of [35S]GSH 21 Fig. 12. Site I dissociation rate constant plot analysis 23 Fig.13. Site I association rate constant plot analysis 24 Fig. 14. Comparision of the binding affinity of GSH, Cys, and Glu . .26 Fig. 15. Effect of -SH structure on [35S]GSH binding 29 Fig. 16. Effects of y-Glu anda-Glu structure on [3 5S]GSH binding 30 Fig.17. Distribution of GSH receptors on astrocytes 33 Fig. 18. Distribution of GSH receptors on astrocytes 35 Fig. 19. Distribution of GSH receptors on astrocytes 36 Fig.20. The growth direction of processes of an astrocyte tends toward that of another astrocyte 37 Fig.21. GSSG binding site of GSSG reductase (top) and the binding of S-(2,4-dinitropheny)glutathione (thick lines) at this site (bottom) 43 vi ACKNOWLEDGEMENT I'm very grateful to Dr. Christopher Mcintosh for his helpful comments, suggestions, and advice on the GSH receptor study, and for his generous help in purifying biotinyl-GSH. I'd like to thank Dr. Nelly Auersperg for the use of her Zeiss Axiphot microscope, Dr. Wayne Vogl for the use of his Pharmacia UV-1 protein detector and LKB autocollector, Dr. Steven Vincent for the use of his 1225 Sampling Manifold, Mr. Pan and for his help with fluorescence micrography, and Ms. Mala Glenwright for the supply of the primary cultures. I'm also grateful to my supervisor Dr. Chris Shaw for all his support and help, and to Dr. Christopher Mcintosh, Dr. John Church, and Dr. Terry Snutch for their helpful comments on this thesis. Without the generous assistance of these people this project would not have been possible. INTRODUCTION Glutathione (GSH, reduced form), an acidic tripeptide which has been studied for over a hundred years, is generally considered to be a powerful oxidant playing an important role in peripheral detoxification defense mechanisms (Chance et al, 1979; Mannervik et al, 1989; Ishikawa & Sies, 1989; Sies, 1988; Hinson & Kadlubar, 1988; Chasseaud, 1988). Glutathione is the major low-molecular weight soluble thiol present in mammalian cells. Its function as an antioxidant is linked to the sulfhydryl group. Biologically more important, many GSH-linked reactions are enzyme-catalyzed. The relationship between GSH and some enzymes is illustrated in Fig. 1. GSH, therefore, can be considered as a coenzyme or as an enzyme cofactor. One important role of GSH in the detoxification of reactive and toxic chemical compounds formed in the metabolism of oxygen is associated with GSH peroxidase, which reduces H2O2 and organic hydroperoxides with concomitant formation of glutathione disulfide, GSSG (Mannervik et al, 1988; Ishikawa & Sies, 1989; Chance et al 1979) (Fig. 1). Endogenous oxidative stress is a consequence of aerobic metabolism which in eucaryotes occurs mostly in the mitochondria. About 2% of mitochondrial O2 consumption generates H2O2. H202, if not reduced, can lead to the formation of the very reactive hydroxyl radical and will result in the formation of lipid hydroperoxides that can damage mitochondrial membranes and proteins and their functions. It has been found that mitochondria rely solely on GSH peroxidase to detoxify hydroperoxides (Chance et al, 1979). To maintain flux through the GSH peroxidase reaction, GSSG must be continuously reduced back to GSH by glutathione reductase at the expense of NADPH. The redox state of cellular glutathione is considered to affect major cellular functions including protein synthesis (Kosower & Kosower, 1978). Glutathione reductase, which catalyzes the ancillary reaction of the glutathione peroxidase system, plays a key role in maintaining GSH essentially completely in its reduced form (Mannervik et al, 1988; Ishikawa & Sies, 1989). The second major reactivity of GSH involved in detoxification function is the reaction with GSH S-transferases (Sies, 1988; Hinson & Kadlubar, 1988; Chasseaud, 1988). Glutathione S-transferases catalyze the reaction of potentially harmful lipophilic electrophiles with GSH to give a thioether, also 2 ( e«troc*MolO/SpOC* ) Fig. 1, Illustration of reactions of glutathione S-conjugates in cells. X, electrophile as substrate for glutathione S-transferase; X -SG, glutathione S-conjugates; R O O H , organic hydroperoxides; enzyme reactions or transfer steps inhibited by glutathione S-conjugates arc indicated by solid bars, (from H. Sies, 1988). R — X + H S C H , C H C O N H C H , C O O H I N H C 0 C H , C H , C H C 0 O H I N H , glutathione S-transfaraMa R — S C H , C H C O N H C H , C O O H I NHCOCH,CH,CHCOOH l N H , y-glutamyl-transpeptidaaa R — S C H , C H C O N H C H , C O O H N H , Cyateinyl-glycinaaa R — S C H , C H C O O H -I N H , rV-acetyle*« R — S C H , C H C O O H I N H C O C H , Fig. 2. Mercapturic acid formation through conjugation with GSH (from L.F. Chasseaud, 1988). 3 called S-conjugate, that is less toxic and more water soluble than the original compound (Fig.1). Glutathione conjugates are thought to be metabolized further by cleavage of the glutamate and glycine residues, followed by acetylation of the resultant free amino group of the cysteinyl residue, to produce the final product, a mercapturic acid ( Boyland & Chasseaud, 1969; Wood, 1970; Chasseaud, 1988) which is then excreted (Fig. 2). A new development quite unrelated to detoxification, emerged in 1979 when GSH conjugation was shown to participate in the biosynthesis of leukotrienes (Hammarstrom et al, 1979). Leukotrienes are biologically highly active mediators of profound physiological importance (Huber & Kippler, 1988). Enzymatic conjugation of leukotriene A4 with GSH generates leukotriene C4 (Hammastrom et al 1989), which plays a neuroendocrine role in luteinizing hormone secretion (Lindgren et al, 1984; Gerozissis et al, 1987; Samuelsson et al, 1987). Conversion of synthetic leukotriene A4 into leukotriene C4 has been shown in human polymophnuclear leukocytes (Radmark et al, 1980) and platelets (Pace-Asciak et al, 1986), in mouse mastocytoma cells (Dahinden et al, 1985; Soderstrom et al, 1988), and in homogenates of liver and lung from different species (Pace-Asciak et al, 1985; Wu, 1986). In the eye, a high concentration of GSH has been found in the lens, and the lens epithelium contains a concentration 5 times that of the cortex (Rathbun, 1989). One of the functions ocular GSH appears to be involved in is maintaining the Na + - , K+-ATPase activity at a normal level in situ ( Rathbun, 1989), and Na + - , K+-ATPase is quite critical to lens clarity. It has been suggested that GSH either is a cofactor of ionic transport or is directly involved in essential maintenance of Na +-, K+-ATPase activity. There is also evidence for another role of GSH in the maintenance of intracellular calcium ion homeostasis (Orrenius et al, 1983). Mammalian cells maintain the ionic calcium concentration in the cytosol (~ 10\"7 M) far below that in the extracellular medium ( ~10\"3 M). To maintain the cytosolic C a 2 + concentration, the mitochondria and the endoplasmic reticulum represent the predominant sites of C a 2 + sequestration (Claret-Berthon et al, 1977). The membrane C a 2 + pump plays a major role in the maintenance of the gradient existing between the extra- and intracellular environments by actively extruding C a 2 + from the cell (Schatzmann, 1982). GSH was found to protect against the inhibition of C a 2 + sequestration and exert a stimulatory effect on C a 2 + transport 4 in the plasma membrane (Orrenius, 1989). Evidence for an involvement of GSH in the control of C a 2 + transport though the ATP-dependent translocases in the plasma membrane and endoplasmic reticulum is that the activity of C a 2 + -ATPase could be restored by treatment of cells with GSH (Orrenius, 1989). However, GSH is also thought to affect non ATP-dependent Ca2+-transport control systems: (1) the GSH redox status may control the influx of C a 2 + through a voltage-dependent plasma membrane channel (Ammon & Mark, 1985), (2) the N a + / C a 2 + exchange system may also be under redox control (Reeves et al, 1986). Obviously, the concept of GSH function is no longer limited to detoxification of peripheral systems. Actually, research on glutathione has been markedly stimulated by the discovery of the involvement of this tripeptide in a number of important cell functions in addition to its well-characterized role in detoxification reactions. These functions include maintenance of membrane integrity and cytoskeletal organization, regulation of protein conformation and enzymatic activity (Larsson et al, 1983), and augmentation of glucose-induced release of insulin (Meister & Anderson, 1983). Although the physiological function of GSH in the central nervous system is still uncertain, the presence of large amounts of GSH in the rodent brain (Martin & Mcllwain, 1959; Reichelt & Fonnum, 1969) implies that GSH might also play an important role in the CNS. In whole brain, the concentration of total GSH (consisting of both reduced and oxidized forms) is 0.5-3.4 mmoles/gm tissue (Orlowski & Karkowsky, 1976), and GSH occurs primarily (97%) in its reduced form. GSH is synthesized rapidly in the brain. Intracisternal administration of [14C]glutamate to cats rapidly labels glutathione in the brainstem, cerebellum mesodiencephalon, and hippocampus (Berl & Purpura, 1966). In addition to these in vivo studies, in vitro experiments have shown incorporation of [2-14C]glycine and DL-[1-14C]glutamic acid into glutathione (Takahasi & Akabane, 1961). The first-order rate constant for GSH synthesis in the brain has been calculated as 0.17 x 103 min\"1. This value is higher than that calculated for human erythrocytes (0.12 x103 min\"1) and rabbit muscle (0.11 x 103 min\"1), but much lower than that calculated for liver (2.9 - 5.8 x 103 min\"1) and kidney (2.4 x 103 min\"1).These calculations assume a homogeneous pool of both precursors and products (Orlowski & Karkowsdy, 1976). 5 The level of GSH in the brain changes during development. In neonatal cat neocortex, the GSH concentration is about 80% of adult levels. The level markedly increases during the postnatal period and reaches adult levels by the end of the second week. The time course of this increase parallels the morphological development of neocortical elements in the cat (Berl & Purpura, 1963). GSH levels also increase in the hippocampus, brainstem, cerebellum, and mesodiencephelon during the developmental period (Berl & Purpura, 1966). Recently, Ogita et al (1986) found that GSH exerts a profound suppressive action on the Na+-dependent and -independent binding of glutamic acid. Later, Ogita and Yoneda (1987) reported the possible presence of GSH binding sites in synaptic membranes from rat brain. These results, especially the GSH pattern during development and the binding activity of GSH in the brain suggested that GSH might be a neurally active molecule in the CNS. To address the question of whether GSH may be a new candidate for a neuropeptide in the CNS, we attempted to determine if there are GSH receptors in the brain. 6 EXPERIMENTS AND RESULTS 1. The Biotinylated Ligand As A Probe For The Related Receptor The high affinity (K 90% ( McCarthy & De Vellis, 1980). It is thus credible that the receptor assays performed on the primary cultures reflect the receptor characteristics of astrocytes. Binding assay To examine the \"receptor hypothesis\", the first attempt was made to determine whether or not the astrocytes demonstrated a specific binding of [ 3 5S]GSH. A significant binding of [ 3 5S]GSH in cultured astrocytes was found in the present study (Fig. 6). The binding was tremendously inhibited by 0.3 mM GSH and independent of the incubation temperature: 97% inhibition at 37°C and 90.9% at 4°C. The specific binding was much higher at 37°C than at 4°C (t -test, P< 0.005), while the nonspecific binding, which was found in the presence of an excess amount of non-radioactive GSH, was not affected by the incubation temperature. In the no-cell control group, the [3^S]GSH binding, which was 79% of the nonspecific binding at 37°C and 81% at 4°C, reflected the adsorption of the isotope to glass filters. The specific binding linearly increased with increasing cell concentration at 37°C and 4°C (Fig.7). Because the specific binding was much higher at 37°C than at 4°C, other experiments including time course, saturation binding, dissociation analysis, and displacer competition etc. were all executed at 37°C. Fig.8 illustrates the data from experiments in which the time course of the binding was measured at an [ 3 5S]GSH concentration of 25nM, with the cell 15 H Total • 0.3 mM GSH H No-cells Fig. 6. Effect of excess GSH on [35S]GSH binding. Intact cultured astrocytes ( 6 x 10 5 /ml) were incubated with 3 nM [ 3 5S]GSH at 37 °C and 4°C for 60 min in the presence or absence of 0.3 mM unlabeled GSH. For both experiments, triplicate samples were performed. 16 2.5 3.5 4 4.5 Log Cells/ml 5.5 Fig. 7. Effect of cell concentration on [35S]GSH binding. Cells in various concentrations (6 x 10 2 - 6 x 10s / ml) were incubated with 3nM [ 3 5S]GSH at 37°C or 4°C for 60 min, Nonspecific binding was measured in the presence of 300 jiM unlabeled GSH. Each point was from the triplicate samples of each experiment. Fig. 8. Time course of pSJGSH to binding sites on astrocytes. Cells (6 x 10 5 / ml) were incubated with 25 nM [^SJGSH at 37°C for various incubation periods indicated. The association of [35S]GSH binding was completed after 60 min. The data were corrected for nonspecific binding, and each point was measured in triplicates of 300 al each. 18 concentration of 6 x 10^ cells/ml incubated at 37°C. The association of [ 3 5S]GSH binding was completed after 60 min. Equilibrium binding experiments showed that astrocytes bind [ 3 5S]GSH in a saturable manner. The concentration dependence of the binding is shown in Fig.9 over a wide range of [ 3 5S]GSH concentrations from 0.5 - 25 nM. The nonspecific binding was determined by adding an excess amount of unlabeled GSH at a concentration of 0.3 mM. Fig. 10 depicts the analysis of this binding data according to Scatchard (1949). A non-linear Scatchard plot was obtained that suggested two classes of binding sites on the cultured astrocytes: Kdi = 4.67+0.75 nM, Bmaxi = 70±9.2 fmole/ 6.4 x 105 cells or B m a X l =6.6 x104 molecules / cell, Kd 2 = 35.1412.17 nM, B m ax 2 = 260+12.77 fmole/ 6.4 x 105 cells or Bmax2 = 2 4 x 1 ° 5 molecules/ cell. Previous studies have found that the number of different receptors on different cell types varies from 103- 10 5 or even higher (Gong, 1985). For example, there are 8 x 104 EGF receptors per fibroblast cell, 1.6 x 105 atropine receptors per smooth muscle cell ( Hollenberg & Cuatrecasas, 1975). In the A875 melanoma cell line, NGF receptors are 7 x 105 / cell (Fabricant et al, 1977), and in A431 cell line, EGF receptors are 2 x 106 / cell (Carpenter, 1987). The GSH receptors shown in the Scatchard analysis are in the range of 104 -10 5 / cell. The numbers which we report here are in keeping with these previous studies and may imply that GSH receptors serve important physiological functions in astrocytes. Dissociation and association kinetics For one-site receptor binding, the binding reaches equilibrium following the kinetic equation: K+i R + L====RL [1]. K-1 The forward or association (K+i) and the reverse or dissociation (K-i) rate constants will give the kinetic determination of Kd, the equilibrium dissociation constant: K-1 K d = -\"-[2] K+ 1 (Bylund, 1980). 120 [ 3 5 S]GSH FREE ( nM ) Fig. 9. Binding of [3 5S]GSH to cultured astrocytes as a function of concentration. Cells (6.4 x 10 5/ ml) were incubated at 37 °C for 60 min with various concentration (0.5nM to 25nM) of [ 3 5S]GSH. Each value of total bound represents the mean±S.E. obtained from 3 separate experiments. Nonspecific binding of [ 3 5S]GSH was measured in the presence of 0.3 mM unlabeled GSH. 0 20 40 60 80 100 120 140 160 180 200 220 240 260 [ 3 5 S ] G S H B O U N D (fmoles/ 6.4 x 105 cells) Fig. 10. Scatchard plot analysis of binding equilibrium of [ 3 5 S ] G S H to astrocytes. A non-linear plot suggests that there are two classes of binding sites in the cells. Binding constants were calculated as shown. Each point represents the mean± S .E . obtained from 3 separate experiments. 120 < 20 T 1 1 1 ' 1 1 1 • 1 ' f-0 20 40 60 80 TIME (min) Fig.11. Rate of dissociation of [35S]GSH. Cells were preincubated with 3 nM [ 3 5S]GSH for 60 min at 37 °C. The dissociation of [ 3 5S]GSH was induced by the addition of 300 pM unlabeled GSH. The specific binding at T 0 and after different times of dissociation was measured in triplicate. The kinetic rate constants for the dissociation and association of [3bS]GSH from the higher affinity site I was determined by using [35S]GSH concentrations below or at the apparent K c •5 c 3 I a CO 8 — I 1 f— 20 40 60 80 100 120 140 160 180 TIME (min) Kob = 0.008 min \" 1 K*1 = 0.003 nlvf 1min\" 1 20 40 60 80 Time (min) 100 120 Fig. 13. Site I association rate constant plot analysis. A: Time course of association of [^SJGSH to binding sites on astrocytes. Cells (6.4x10 5 / ml) were incubated with 3 nM [^SJGSH at 37°C for various incubation times. B: Site I association rate constant plot according data from A. Table 1. Kinetic and equilibrium constants tor the specific binding of [J J JS]GSH to cultured astrocytes. Binding Constants Kinetic Analysis Scatchard Analysis K.i (min\"1) 0.0168 K+i (nM\"1 min\"1) 0.003 K d l (nM) 5.6 4.67 + 0.75 Bmax1 (fmoles/6.4x105 cells) 70 ±9 .21 (molecules / cell) 6.6 x 10 4 Kd2(nM) 35.14 ±2 .17 Bmax2 (fmole /6.4 x105 cell) 260 ± 12.77 (molecules / cell) 2.4 x 10 5 Table 2. IC50 of different displacers for [ 3 5S]GSH binding. Displacer IC50 GSH 10 nM GSSG 40 uM S-methyl-GSH 40 MM Cys 1 mM Y-GluGlu 1.5 mM y-D-Glu.Gly 6 mM y-Glu.Gly 10 mM S-hexyl-GSH 15 mM 26 • (Glu) • (Cys) • (GSH) 0 -9 -8 -7 -6 -5 -4 Log M Fig. 14 Compan'sion of the binding affinity of GSH, Cys, and Glu. The competitive inhibition of [ 3 5 S]GSH binding was measured by the addition of GSH, Cys, and Glu representively at various concentrations. Cells were incubated with 3nM [ 3 5 S]GSH at 37°C for 60 min. All experiments were performed in triplicate. IC 50 values: GSH, 10nM; Cys, 1 mM. Glu had almost no effect on [ 3 5S]GSH binding. Inhibition of f J O S]GSH binding and structure-selectivity Fig. 14 shows that GSH elicited a concentration-dependent inhibition of [3 5S]GSH binding in the cultured astrocytes. The IC50 of GSH for [35S]GSH binding is 10 nM. Because cysteine and its analogues have been reported to interact with excitatory amino acid binding sites (Pullan et al, 1987) and GSH contains a cysteine moiety, the relationship between cysteine and GSH receptor was studied by using cysteine as a displacer. A significant inhibition was obtained by the addition of cysteine although its IC50 was much higher than that of GSH (Table 2 and Fig.14). The affinity of cysteine for AP4 binding sites is 0.34 M M , and for NMDA sites is 170 ^ M (Pullan et al, 1987). However, the IC50 of cysteine for the GSH binding sites is 1mM (Fig. 14), suggesting that GSH receptors differ from these excitatory amino acid receptors. Glutamic acid only produced 15% inhibition even with the highest concentration (30 mM) of various concentrations examined (Fig.14). The failure of Glu to inhibit [ 3 5S]GSH binding reflects the specificity of the binding site for its ligand. The binding was also inhibited by GSSG (oxidized form of glutathione), S-methyl-GSH, and S-hexyl-GSH (Fig. 15). GSSG and S-methyl-GSH have almost the same IC50 (40MM), while S-hexyl-GSH exhibited a much higher IC50 of 15 mM (Table 2). The reason for this might be that the six-carbon side chain connected to GSH obstructs the specific binding of the S-hexyl-GSH molecule to the receptor. The fact that these GSH derivatives, which have no SH group but still maintain in different degrees the binding ability, clearly indicates that the binding in astrocytes is not due to a nonspecific binding of GSH to any membrane protein through an exchange of disulfide bond between GSH and membrane proteins, and also the -SH moiety is not necessarily crucial for the GSH molecule to bind its receptor. The importance of the y-Glu structure and its conformation in the ligand was studied. Peptides with the y-Glu structure in the L-conformation ( y-Glu-Glu, y-GluGly) and D-conformation (y-D-Glu-Gly) showed a significant inhibition of [ 3 5 S]GSH binding (Fig. 16), but obviously they had a much lower affinity to GSH receptors with 2-3 orders of magnitude increase in their IC50, compared to that of GSH (Table 2). When the y-Glu structure was substituted bya-Glu, it not only lost its ability to bind the receptor, but somehow greatly enhanced the [ 3 5 S]GSH binding as well (Fig.16). These results suggest that the y-Glu moiety is of importance, while the cysteine and glycine residues are also essential for the GSH molecule to maintain its binding affinity to the receptor. a-Glu-Glu, 28 which increased the [ 3 5S]GSH binding in astrocytes, might serve as a synergist of GSH in the binding reaction. 0 Log M Fig. 15. Effect of -SH structure on [ 3 5S]GSH binding. Cells were incubated with 3nM [ 3 5S]GSH at 37°C for 60 min in the presence of GSSG, S-methyl-GSH, and S-hexyl-GSH at various concentrations. Each experiment was performed in triplicate. IC50 values: GSSG, 40 uM; S-methyl-GSH, 40 uM; S-hexyl-GSH, 15mM. 30 o c o o c •5 c 00 o 00 8 200 f 180 160 140 120 100 it 80 + 60 40 20 0 -h 0 -7 -4 Log M -2 • (n-Glu.Glu) • (T-GIU.GIU) • (r-Glu.Gly) O (r-D-Glu.Gly) •1 Fig. 16. Effects of y-Glu and a-Glu structure on pSJGSH binding. Cells were incubated with 3 nM [35S]GSH at 37°C for 60 min in the presence of y-Glu.Glu, y-Glu.Gly, y-D-Glu.Gly, and a-Glu.Glu. All experiments were performed in triplicate. IC50 values: y-Glu.Glu, 1.5 mM; y-D-Glu.Gly, 6 mM; y-Glu.Gly, 10 mM. a-Glu.Glu did not compete with [ 3 5S]GSH but augmented its binding. 31 3. The Colloidal Gold Technique For The Cellular Localization Of GSH Binding Sites Although the results from the studies described above suggested that GSH receptors might exist on astrocytes, they did not allow a direct cellular resolution. If the conclusion based on these previous experiments is correct, i.e. there are GSH receptors on astrocytes, a positive result at the cellular level should be obtained with appropriate cytochemical techniques. In the present study, the colloidal gold method and fluorescence double staining were used to visualize the GSH receptor at the cellular level. Materials And Methods Materials Cultured astrocytes were prepared as described above. Streptavidin-gold (AuroProbe) and silver enhancer were from Amersham International Pic (Amersham, UK). Normal goat serum was from Vector Laboratories Inc. (Burlingame, CA). Rabbit immunoglobulins to cow glial fibrillary acidic protein (GFAP) was purchased from Dako Corporation (Santa Barbara, CA). Goat anti-rabbit-Rhodamine was from Cappel/Organon Teknika INC (Turnhout, Belgium). Colloidal gold method and double staining The colloidal gold method with silver enhancer (Owen et al, 1989) was used in the present study. Cultured cells were washed with CMF-PBS (Vale & Shooter, 1985) for 3 min x 2. Cells were incubated biotinyl-GSH (see \" synthesis and purification of biotinyl-GSH \" in part 1) for 60 min at room temperature. After the incubation, cells were washed with CMF-PBS for 5 min and with H2O for 5 min x 2. Streptavidin-gold (1:40 ) was added to cells at room temperature for 40 min. After washing with CMF-PBS for 5 min and with H2O for 5 min x 2, cells were fixed with methanol for 30 min at -20°C, and then washed with H2O for 1.5-2 hr. Freshly prepared silver enhancer was added and incubated with cells for 15-25 min at room temperature. After washing with H2O for 5 min x3, the cells were mounted with DPX for observation and microphotography. Different control experiments were performed with (1) biotinyl-GSH + GSH ( x 100) + silver, (2) strept-gold + silver, (3) biotinyl-GSH + silver, (4) silver only, and (5) no addition. A modified double staining method (Owen et al, 1989) was used in order to get a satisfactory result. Cells were fixed with methanol 30 min at -20°C after washing with CMF-PBS for 3 min x 2. After fixing, cells were washed with H2O for 1.5 hr. Normal goat serum ( 1:50 ) was added and incubated with cells for 30 min at room temperature. After washing with H2O for 10 min x 2, the primary antibody, rabbit immunoglobulins to cow GFAP, was added ( 1:100 ) and incubated with cells overnight at 4°C. The cells was washed with H2O for 5 min x 3, and then the second antibody, goat anti-rabbit IgG-Rhodamine (1:100 ), was added. After an incubation of 60 min at room temperature, the cells were washed with H2O for 10 min x 3. The colloidal gold method (as described above) was then performed. After double staining, the cells were finally mounted in a semipermanent mounting medium special for fluorescent antibody (Rodriguez & Deinhardt, 1960). Phase and immunofluorescence micrographs were recorded with a Zeiss Axiophot microscope. Results And Discussion Colloidal gold decoration Because only one biotin molecule could be conjugated to each GSH molecule, i.e. there is no amplification effect as with the biotinyl-protein / -large peptide, which has much more -NH2 groups for biotin to combine with, biotinyl-GSH itself did not give a satisfactory cellular resolution. This problem was solved by taking advantage of the colloidal gold technique. Colloidal gold -protein complexes have gained wide use as probes in electron and light microscopy procedures ( Roth, 1983; De Mey, 1984).The gold label is a discrete electron-dense, non-fading, red-colored marker capable of strong secondary electron emission, and the signal can be dramatically intensified by reacting with silver (Hacker et al, 1985; Jackson, 1987). In the present study, the colloidal gold-protein used was streptavidin-gold, which could bind to biotinyl-GSH molecules held by their receptors in the cell membrane. On addition of silver enhancer, precipitation of metallic silver occurs, which enlarges the colloidal gold labels normally visible only at the electron microscope level, yielding high-contrast signals visible by light microscopy. The distribution of GSH receptors on cultured astrocytes was thus visible by labeling cells with biotinyl-GSH. Colloidal gold decoration of biotin-GSH-GSH receptors complex revealed that these receptors were on both Fig. 17. Distribution of GSH receptors on astrocytes. Both cell bodies and their processes were labeled with biotinyl-GSH which is decorated with colloidal gold. astrocytic cell bodies and their processes (Fig. 17). This result from the intact cell staining also suggests that GSH receptors exist on the cell membrane of astrocytes. All control groups gave a negative reaction. GFAP double staining For primary culture, the results at confluence (2-3 weeks in culture) is a population of GFAP-positive cells in a monolayer which may comprise anything from 60-90% of the total number of cells present (Murphy & Pearce, 1987). Chief contaminants are macrophages and fibroblasts, followed by minor contaminants such as capillary endothelial cells, leptomeningeal cells, oligodendrocytes and neurons (usually in that order) (Murphy & Pearce, 1987). Although the primary cultures used in this study contained an enriched population of astrocytes up to 90% (McCarthy & De Vellis, 1980), it was necessary to identify whether these colloidal gold positive cells, which were labeled with biotinyl-GSH, were wholly astrocytes or included some contaminants such as oligodendrocytes and neurons. In vitro, the term\" astrocytes\" is synonymous with a CNS cell that labels with antibodies to glial fibrillary acid protein (GFAP) - the most commonly adopted criterion for positive astrocytes identification, because GFAP positive reaction (GFAP + ) is restricted in distribution to astroglial cells in the CNS (Bignami & Dahl, 1977). In the present study, Rhodamine-labeled secondary antibody was used to demonstrate the presence of GFAP + . Double staining with both GFAP/Rhodamine and biotinyl-GSH/colloidal gold showed that all cells which were stained by colloidal gold-silver were immunoactive to GFAP, i.e. those cells which possess GSH receptors were astrocytes (Fig. 18). Interestingly, not all the cultured astrocytes have GSH receptors. Fig. 19 shows that many astrocytes which are labeled with Rhodamine (GFAP + cells) are not stained with colloidal gold-silver. It seems that only morphologically mature astrocytes possessed GSH receptors and thus are decorated by colloidal gold. But many GFAP + cells, which look smaller and have not differentiated their processes, are colloidal gold negative, i.e. they do not have GSH receptors at that ontogenic stage. The double staining used in the present study also revealed a very interesting phenomenon: in primary cultures, the growth direction of processes of an astrocyte tend towards these of the other astrocytes (Fig. 20). 35 Fig. 18. Distribution of GSH receptors on astrocytes. A: Colloidal gold decoration of GSH receptor. biotinyl-GSH complex on primary culture cells. B:Cells in the same field as A were labeled with Rhodamine-antibody. These GFAP* cells confirm that GSH receptors exist on astrocytes. Fig. 19. Distribution of G S H receptors on astrocytes. A: Colloidal gold decoration of G S H receptors. B: Fluorescence double staining of the same field as A. Many G F A P + cells, which had no GSH receptors at that ontogenic stage, were Strept-gold\". Fig. 20. The growth direction of processes of an astrocyte tends toward that of another. A: Colloidal gold decoration. B: Fluorescence double staining (the same field as A). GENERAL DISCUSSION High levels of GSH have been found in the brain. A reasonable assumption would be that GSH in the CNS plays the same detoxification function as it does in the peripheral system. But most evidence indicates that detoxification of compounds by their conjugation with GSH occurs primarily in the kidney and liver, while little detoxification seems to take place in the brain (Orlowski & Karkowsky, 1976). In rat brain, little activity of GSH S-alkyltransferase (Johnson, 1966) and GSH S-transferase B (Orlowski & Karkowsky, 1976) has been found. Also, rat cerebrum, cerebellum, and brain stem contain little GSH S-arene oxide transferase activity (Haykawa et al, 1974). Additionally, the decrease in GSH during brain ischemia in vivo was not accompanied by any reciprocal increase in GSSG (Rehncrona et al, 1980). This result thus fails to support the hypothesis that peroxidative damage occurs during or following ischemia, but implies that GSH plays a role other than antioxidant in the brain. On the other hand, more and more evidence has shown that GSH may be involved in the functional processes of the CNS. The significance of GSH in brain function is suggested by the finding of mental retardation in patients with GSH metabolism problems in the nervous system (Orlowski & Karkowsky, 1976). Patients with GSH synthetase deficiency showed a gradual neurological deterioration of motor functions, retardation of movement, intention tremor and rigidity, and psychomotor retardation apparent in childhood (Jellum et al, 1983). Berl et al (1959) reported focal seizures in cat lesions associated with a significant decrease in GSH. It has also been reported that GSH is virtually absent in the nigra of patients with Parkinson's disease (Perry et al, 1982). Together with the finding of the different pattern of GSH levels in the brain during development (Berl & Purpura, 1963) and binding activity of GSH in the CNS (Ogita & Yoneda, 1987), available evidence suggests that GSH might be a neurallly active peptide, acting either as a neuromodulator or neurotransmitter in the brain. To address the question as to whether GSH is a neuropeptide in the CNS, one key point is to demonstrate whether there are GSH receptors in the CNS. By focussing on studies of the localization and characterization of GSH receptors in the brain, this thesis reaches a positive conclusion that GSH may indeed be a candidate for a neuropeptide in the CNS. Thhe evidence is summarized as follow: First, GSH binding sites in the brain were clearly revealed by binding of biotinyl-GSH (Fig. 5). The location of these binding sites suggested that GSH receptors might reside on glial cells in the CNS. Further experiments including radioligand receptor assays and cytochemical double staining confirmed that GSH receptors existed on astrocytes. The binding assays performed on cultured astrocytes showed the GSH binding sites displayed the characteristics of a receptor, such as reversible and saturable binding (Fig. 14 & Fig. 9 -10), high affinity (Table 1), and ligand specificity (Fig.14 -16, Table 2). These results suggest that there are GSH receptors on astrocytes. Results from the present study are also unlikely to be explained by binding of GSH with the receptors of another ligand in the CNS. Ogita et al ( 1986) reported that GSH produced a significant displacement of the Na+-dependent and -independent bindings of [3H]glutamic acid to the membrane preparation from rat brain. There was thus a possibility that the high GSH binding affinity demonstrated in this study might be due to the binding of GSH to Glu receptors on astrocytes. If this were the case, glutamic acid should show a strong displacement of [35S]GSH binding. In fact, glutamic acid did not displace [35S]GSH binding in astrocytes (Fig. 14). It is thus probable that GSH bind to its own, but not Glu receptors. A well known function of GSH in peripheral systems is to act as an antioxidant with the association of enzymes (Fig. 1). Is it possible that the specific binding presented in this study reflected the binding of GSH to those enzymes in astrocytes? Results from this study and others also suggest it is unlikely ( because GSH S-transferase and GSH peroxidase are the ones associated directly with GSH in detoxification mechanism, only these two enzymes are discussed here): (1) Glutathione S-transferases, a group of dimeric proteins involved in the detoxification of a broad spectrum of xenobiotics, catalyze the conjugation of these compounds with the -SH group of GSH, thereby neutralizing their electrophilic sites and rendering the products more water-soluble for further metabolism and then excretion (Habig et al, 1974; Chasseaud, 1988; Dekant et al, 1988) (Fig.1). The dissociation constant (K