l" J t‘L) \J NERSITY LIBRAR Milliillifinilltmwunmnuumh‘ll I 3 1293 01410 7357 This is to certify that the dissertation entitled MECHANISMS MEDIATING THE METHYLMERCURY-INDUCED ELEVATIONS IN THE INTRASYNAPTOSOMAL CONCENTRATIONS OF CALCIUM AND ZINC presented by Michael F. Denny has been accepted towards fulfillment of the requirements for Ph .D . Pharmacology and degree in Toxicology {gigs/Luv aWQW/K Major professor Date 9/22/95 MSUL: an Affirmative Action/Equal Opportunity Institution 0-12771 LIBRARY Michigan State University PLACE N RETURN BOX to romovo this chockout from your tocord. TO AVOID FINES rotum on or Moro duo duo. DATE DUE DATE DUE DATE DUE i—w—a MSU I. An Affitmatlw Action/E“ Opportunity lnotitttlon WWJ MECHANISMS MEDIATING THE METHYLMERCURY-INDUCED ELEVATIONS IN THE INTRASYNAPTOSOMAL CONCENTRATIONS OF CALCIUM AND ZINC By Michael F. Denny A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Pharmacology and Toxicology and Institute for Environmental Toxicology 1995 ABSTRACT MECHANISMS MEDIATING THE METHYLMERCURY-INDUCED ELEVATIONS IN THE INTRASYNAPTOSOMAL CONCENTRATIONS OF CALCIUM AND ZINC By Michael F. Denny Methylmercury (MeHg) is an environmental neurotoxicant upon chronic and acute exposure. The mechanisms underlying cytotoxicity remain unknown, but may be due to disruption of divalent cation homeostasis because MeHg disrupts Cay-dependent processes in several model systems. The effects of MeHg on divalent cation homeostasis were studied using isolated nerve temiinals from the rat brain (synaptosomes) loaded with the Caztselective fluorescent indicator fura-Z. MeHg (10-100 pM) caused an gradual increase in intrasynaptosomal Ca2+ concentration ([Ca2*]i) which was concentration- dependent with respect to both MeHg and extracellular Ca“ (Ca2+), and mediated by increased plasma membrane permeability. MeHg also caused an immediate elevation in the intrasynaptosomal concentration of an endogenous polyvalent cation other than Ca2+, which was independent of Ca2+e and maximal at 25 pM MeHg. The endogenous metal could not be identified with complete certainty, however, it was postulated to be Zn“, based on its interactions with fura-2. ”F-nuclear magnetic resonance (NMR) spectroscopy was performed on synaptosomes loaded with the polyvalent cation indicator SF-BAPTA to identify'the cation unequivocally. This technique allows for the identification and quantitation of + several cations, simultaneously. MeHg caused a three-fold elevation in [Zn2 ]i which was independent of Caz} but did not affect the concentration of cations other than Zn”. Endogenous sources of Zn“ in the nerve terminals include proteins, and synaptic vesicles. Pretreatment of fura-Z loaded synaptosomes with agents which mobilize synaptic vesicles did not attenuate the MeHg—induced elevations in [Zn2*]i, suggesting that synaptic vesicles do not contribute to the increases in [Zn2“]i. To determine if MeHg released Zn” from synaptosomal proteins, homogenates of synaptosomal suspensions were prepared, and fura-2 was added. Excitation scans were acquired before and after the addition of MeHg, and compared for effects consistent with release of Zn”. MeHg shifted fura-2 fluorescence in a manner consistent with release of Zn”, this activity was localized entirely to the soluble fraction. Anion exchange chromatography of this fraction resolved three peaks of MeHg-induced Zn2+ release, suggesting that more than one protein conuibuted to the elevation in [Zn2"]i. DEDICATION This dissertation is dedicated to the memory of my grandfather, Tony Crimarcki. iv ACKNOWLEDGEMENTS I would like to acknowledge the efforts of my guidance committee: Drs. Atchison, Bursian, Contreras, Galligan and Smith. I also would like to thank the many lab members who have been so helpful throughout my stay. In particular, the assistance of Michael, Jay, Yukun, Sue, Hong, Laura, Nana, Ravindra and Sandy is greatly appreciated. I want to express my gratitude to Lynne for her constant support and patience during this whole process, anyone else would have given up on me a long time ago. I need to thank my family, to whom the details of my research project have been a never- ending source of amusement. The support and friendship of all the graduate students is also greatly appreciated, I truly feel we did more than keep the drinking establishments in the Lansing area in business. I also want to thank the departmental office staff for keeping track of the endless stream of appointment papers, order forms, travel vouchers, etc. Lastly, I want to acknowledge my 12th grade English teacher, MS. Hnankovich, who stood me up in front of the class one day and told me I would never amount to anything, which may be right, but at least I’m not a 12th grade English teacher. TABLE OF CONTENTS LIST OF FIGURES LIST OF ABBREVIATIONS CHAPTER ONE: INTRODUCTION A. B. G. H. CHAPTER TWO: METHYLMERCURY ALTERS INTRASYNAPTOSOMAL General introduction Regulation of intracellular Ca2+ concentration ([Ca2*],) in neurons. Effects of MeHg on neuronal function as an index of disruption of cation regulation. Alterations in Ca2+ channel function by MeHg. . Effects of Mel-lg on presynaptic Ca2+ regulation. Mitochondrial Ca2+ regulation as a target of MeHg. Techniques for measuring [Caz*]i. Effects of Mel-lg on neuronal divalent cation homeostasis. viii 11 13 16 23 CONCENTRATIONS OF ENDOGENOUS POLYVALENT CATIONS 25 A. B. F3 .0 F“ Abstract Introduction Materials and Methods Results Discussion vi 26 27 30 37 68V CHAPTER THREE: METHYLMERCURY-INDUCED ELEVATIONS IN INTRASYNAPTOSOMAL ZINC CONCENTRATION: AN l9F-NMR STUDY A. Abstract B. Introduction C. Materials and Methods D. Results and Discussion CHAPTER FOUR: METHYLMERCURY CAUSES RELEASE OF ZINC FROM SOLUBLE SYNAPTOSOMAL PROTEINS A. Abstract B. Introduction C. Materials and Methods D. Results E. Discussion CHAPTER FIVE: CONCLUSIONS Summary of research. Relationship to previous work. .057”? Effects of MeHg on Ca2+ regulation. D. Effects of MeHg on Zn2+ regulation. BIBLIOGRAPHY vii 75 76 77 79 82 91 92 94 97 102 122 127 128 135 140 143 156 LIST OF FIGURES CHAPTER ONE 1.1. The chemical structure of the fluorescent cation indicator fura-2. 1.2. The chemical Structure of the fluorinated cation indicator SF-BAPTA. CHAPTER TWO 2.1. The effect of MeHg on the fluorescence properties of fura-2 in vitro. 2.2. The effect of MeHg on the 340/380 nm ratio in 200 pM Ca2"-HBS. 2.3. The effect of MeHg on the 340/380 nm ratio in 0.1 pM Ca2*-HBS. 2.4. Quench of fluorescence intensity in MeHg-treated, fura-2-loaded synaptosomal suspensions. 2.5. The effect of mitochondrial depolarization on the MeHg—induced elevations in the 340/380 nm ratio in 0.1 pM Ca2*-HBS. 2.6. The effect of ruthenium red on the MeHg—induced elevations in the 340/380 nm ratio in 0.1 pM Ca2"-HBS. 2.7. The effect of caffeine on the MeHg-induced elevations in the 340/380 nm ratio in 0.1 pM Ca2*-HBS. 2.8. The effect of thapsigargin on the MeHg-induced elevations in the 340/380 nm ratio in 0.1 pM Ca2*-HBS. 2.9. MeHg-induced alterations in Ca2*-insensitive fluorescence intensity and Mn2+ permeability of synaptosomes. 2.10. Effect of the cell-penneant heavy metal chelator TPEN on MeHg- induced alterations in Ca2*-insensitive fura-2 fluorescence. 2.11. Effect of the cell-impermeant heavy metal chelator DTPA on MeHg-induced alterations in Ca2*-insensitive fluorescence. 2.12. Effect of MeHg (0 - 25 pM) on the 340/380 nm ratio. CHAPTER THREE 3.1. Resonance positions of SF-BAPT A and various metal chelates in aqueous solution and 5F-BAPI‘A-loaded synaptosomes. viii 19 22 34 39 42 45 49 51 54 56 58 62 65 67' 84 3.2. Spectra of 5F-BAPTA-loaded in HBS containing 200 pM EGTA before and after MeHg treatment. 86 3.3. [Zn"”]i and [012*], in SF-BAPTA-loaded synaptosomes before and after acute exposure to MeHg. 88 CHAPTER FOUR 4.1. Fara-2 excitation profiles of Ca2+ or Zn”. 104 4.2. Effects of MeHg on the excitation profiles of fura-2 loaded synaptosomes following treatment with agents which mobilize synaptic vesicles. 108 4.3. Effect of MeHg on the excitation profile of fura-2 in solution with soluble synaptosomal proteins. 112 4.4. Anion exchange chromatography of soluble synaptosomal proteins. 114 4.5. SDS-PAGE analysis of anion exchange chromatography peaks. 116 4.6. Gel filtration chromatography of Peak 2 from anion exchange chromatography. 119 4.7. SDS-PAGE analysis of active peak obtained by gel filtration chromatography of anion exchange Peak 2. 121 ix AMPA ANOVA ATP °C [Ca2+]. Ca2+e [C321, CICR CNS dia DMSO DTPA EDDA EGTA EPP ER SF-BAPTA LIST OF ABBREVIATIONS optical absorbance at 280 nm acetylcholine acetoxymethylester a-amino-3-hydroxy-5-methylisoxazolepropionic acid analysis of variance adenosine triphosphate degrees centigrade intracellular or intrasynaptosomal Ca2+ concentration extracellular Ca2+ extracellular Ca2+ concentration Cay-induced Ca2+ release central nervous system diameter dimethyl sulfoxide diethylenetriaminepentaacetic acid ethylenediaminediacetic acid ethylene glycol bis-(B-aminoethyl ether)-N,N,N’N’-tetraacetic acid end-plate potential endoplasmic reticulum 1,2-bis(2-amino-5~fluorophenoxy)ethane-N,N,N’N’-tetraacetic acid HBS Hepes Mt. MeHg MEPP MHz min ml MPP” N G 108- 1 5 NGF nAChR nm NMDA NMR PC12 Hepes buffered saline gram force of gravity N-(2-hydroxyethyl)piperazine-N’-(2-ethanesulfonic acid) inositol-1,4,5-tris-phosphate dissociation constant kilodalton molar concentration megaohm methylmercury miniature end-plate potential megahertz minute milliliter 1-methyl-4-phenylpyridinium mouse neuroblastoma x rat glioma hybrid nerve growth factor nicotinic acetylcholine receptor nanometer N-methyl-D-aspartate nuclear magnetic resonance synaptosome enriched fraction rat clonal pheochromocytoma cell xi PKC PLA2 PLC PPm SDS-PAGE sec S.E.M. YSO35 [Zn2+]i protein kinase C phospholipase A2 phospholipase C parts per million pounds per square inch 340/380 nm ratio of fura-Z during Cazi-saturating conditions 340/380 nm ratio of fura-2 during Cap-free conditions sodium dodecylsulfate-polyacrylamide gel electrophoresis second standard error of mean emission intensity of fura-2 at 380 nm during Ca2*-saturating conditions emission intensity of fura-2 at 380 nm during Ca2*-free conditions N ,N,N’ ,N’-tetrakis-(2—pyridylmethyl)ethylene-diamine tetrodotoxin volume per volume dilution watt NJV-bis-(3,4-dimethoxyphenethyl)-N-methylamine intracellular or intrasynaptosomal Zn2+ concentration xii CHAPTER ONE INTRODUCTION A. General introduction. Methylmercury (MeHg) is an environmental contaminant produced by industrial processes as well as through the biomethylation of inorganic, divalent Hg“. MeHg disrupts nervous system function, and is neurotoxic upon acute or chronic exposure (T akeuchi et al., 1962; Bakir et al., 1973). There exists a considerable body of information regarding the toxic actions of MeHg. Manifestations of MeHg intoxication include impairments of vision, hearing and speech, sensory disturbances and weakness in the extremities (Chang, 1980). These effects are believed to be mediated by disruption of the central nervous system (CNS), particular the cerebellum (Chang, 1977). Although the mechanisms of MeHg-induced neurotoxicity are not known, they have been the topic of intense investigation (Atchison and Hare, 1994). One area that has received much attention is the ability of MeHg to disrupt neuronal divalent cation regulation (COOper et al., 1984). The purpose of this dissertation was to study the biochemical basis of MeHg- induced alterations in neuronal divalent cation homeostasis as a mediator of neurotoxicity. Much of the previous information available regarding this topic had been acquired using indirect techniques or through inference. As such, a goal this study was to measure more directly the disruption in divalent cation homeostasis caused by MeHg. B. Regulation of intracellular Co“ concentration (I Calm) in neurons. Ca2+ homeostasis in neurons is dependent upon the precise coordination of processes which elevate [Ca2+]i, such as influx of extracellular Ca“ (Cazv or release of Ca2+ sequestered within organelles, and those which lower [Ca2+]i, either by extrusion or sequestration of intracellular Ca2+ (Blaustein, 1988). Under resting conditions, neurons 3 normally maintain an [Ca2“]i of approximately 100 nM in the face of extracellular Ca2+ concentrations ([Ca2*],) of the order of I to 2 mM. The maintenance of the large inward electrochemical gradient for Ca2+ is essential for proper neuronal function, and relies upon many factors which are interdependent in some instances (Miller, 1988; Choi, 1988). In neurons, the primary route of influx of Ca2+, is through the activation of voltage-dependent or ligand-operated channels which are permeable to Ca2+ (Bertolino and Llinas, 1992). Activation of voltage-dependent Ca2+ channels, and the associated Ca2+ influx is required for the release of neurotransmitter from the presynaptic nerve terminal (Llinas et al., 1992). The [Ca2+]i may rise several-fold following the invasion of an action potential into the nerve terminal and subsequent Ca2+ influx. Activation of postsynaptic or presynaptic ligand-operated channels can also elicit an elevation in [Ca2+]i. Certain subtypes of neuronal nicotinic acetylcholine receptors (nAChR) (Vernino et al., 1992) and subtypes of glutamate receptors sensitive to N-methyl—d-aspartate (NMDA) are permeable to Ca“ following ligand binding (MacDermott et al., 1986). Some receptors elicit an elevation in [Ca2”]i by coupling to phospholipase C (PLC) via a membrane-associated guanine nucleotide-binding protein, or G-protein (Hokin, 1985). Ligand binding initiates a cascade of intracellular events in which the receptor-associated G-protein activates PLC. This stimulates the conversion of phoshotidylinositol into l,4,5-inositol-tris-phosphate (IP,) and diacylglycerol (Nahorski, 1988). IP3 is a second messenger which binds an intracellular receptor located on the endoplasmic reticulum (ER) to cause release of Ca2+ which was sequestered previously (Henzi and MacDermott, 1992). The plasma membrane is a permeability barrier which prevents direct influx of Caz“e down its large electrochemical gradient. An intact plasma membrane is essential 4 for preventing unregulated elevations in [Ca2+], (Kinter and Pritchard, 1977). Loss of plasma membrane integrity could cause unregulated influx of Ca2+” possibly leading to disruption of neuronal function or even cell death. Embedded within the plasma membrane are proteins which function to extrude intracellular Ca2+ (Nicotera et al., 1992). The most prominent of these are the Ca2*-ATPase and the Na“/Ca2+ exchanger. The Ca“- ATPase uses the energy stored in ATP to pump Ca2+ against its electrochemical gradient (Michaelis et al., 1987) while the Na“/Ca2+ exchanger lowers [Ca2+]i by facilitated diffusion (Sanchez-Armass and Blaustein, 1987). Here the energy associated with the influx of extracellular Na“ down its electrochemical gradient is used to drive Ca” efflux. Because the inward Na” gradient is maintained by the Na*/K+-ATPase, both of these integral membrane proteins ultimately require ATP for their activity. Toxicants which alter neuronal plasma membrane integrity can cause elevations in [Ca2+], either directly by increasing the Ca” permeability of the membrane, or indirectly such as by membrane depolarization which results in the activation of Ca2+ channels, or by decreasing ATP content within the neuron thereby interfering with the ATP-dependent processes responsible for extruding Ca“. Neurons also contain organelles which are capable of sequestering Ca2+ (Meldolesi et al., 1988; Blaustein et al., 1978). The most widely accepted are the ER and the mitochondria (Meldolesi et al., 1992). These organelles differ in their mechanisms of Ca2+ uptake as well as their role in cellular Ca2+ regulation. The ER represents a high affinity, low capacity Ca2+ sequestration site believed to be responsible for buffering [Ca2“]i in the physiological range. The ER membrane contains a Cazi-ATPase, distinct from that found in the plasma membrane, which transports Ca2+ from the cytosol to one 5 or more storage pools within the ER. This pool of sequestered Ca2+ in the ER can be mobilized by certain physiological or pharmacological stimuli, such as the production of +0! IP3 or the generation of so-called "trigger Ca2 necessary for Ca2*-induced Ca2+ release, or CICR (McPherson and Campbell, 1993). The mitochondria, on the other hand, are a high capacity, low affinity site of Ca2+ sequestration (Rahamimoff et al., 1975). The precise mechanism of Ca2+ uptake by the mitochondria is not clear, but may involve a ruthenium red-sensitive uniporter located on the inner mitochondrial membrane (Lehninger et al., 1978). Because mitochondria have a Km for Ca2+ uptake in the micromolar range, it is thought that their primary role in Ca2+ sequestration is to buffer abusive or pathological elevations in [Ca2+]i (Akemtan and Nicholls, 1983). However, sequestration of Ca2+ by the mitochondria has been observed under physiological conditions (Rizzuto et al., 1992; 1993). A substantial amount of Ca2+ within a neuron is associated with various neuronal Ca2+ binding proteins (Habennann and Richardt, 1986). The activity of some of these proteins, such as protein kinase C (PKC) and calcineurin, is regulated by the binding of Ca2+ (Persechini et al., 1989). Ca2+ also activates the important regulatory protein calmodulin. Neurons also contain other Ca2"-binding proteins, such as calbindin and parvalbumin (Baimbridge et al., 1992), which may act as a Ca2+ sink within the neuron. The localization of these proteins within the neuron may assist in buffering changes in [Ca2+]i both temporally as well as spatially. Thus, Ca2+ binding proteins are important both for lowering [Ca2+], and maintaining neuronal function. The regulation of [Ca2+], changes as a function of neuronal development and activity. Alterations in [Ca2+]i homeostasis have been observed during a variety of 6 physiological processes. As stated above, activation of certain types of neurotransmitter receptors elevates [Ca2+]i (Malinow et al., 1994), and the process of vesicular release of neurotransmitters is dependent upon Ca2+ influx (Miledi, 1973). Maturation of amphibian spinal cord neurons in culture is promoted by elevations in [Ca2+], in the cytosol and nucleus (Holliday et al., 1991). These increases in [Ca2+], are mediated by CICR and Ca2+ influx, and decline with neuronal maturation. Thus, the elevations in [Ca2"]i which drive cellular development are, in turn, regulated by developmental state. [Ca2+]i also regulates the outgrowth and morphology of the filopodia of neuronal growth cones (Rehder and Kater, 1992). Filopodia direct growth cones to their targets, and are essential for the development of the nervous system. In Helisoma neurons, experimentally-evoked elevations in [Cay]i cause filopodial elongation, followed by retraction. The extent of these changes in filopodia] morphology correlates closely with the peak elevation in [Ca2*]i, suggesting a possible role for regulation of [Ca2+], in the development of the nervous system. The regulation of [Ca2+]i can be altered by the mitotic state of the cell (Preston et al., 1991). In HeLa cells at interphase, Ca2+ influx is tightly coupled to release of Ca2+ from intracellular stores. However, during mitosis, this coupling of Ca2+ influx following depletion of Ca2+ stores ceases temporarily, demonstrating that regulation of [Ca2+], is itself a dynamic process. Because [Ca2+], plays such an integral part in cellular physiology, unregulated or prolonged alterations [Ca2+], regulation could result in the disruption of many cellular processes (Rossi et al., 1991; Nicotera et al., 1992). For example, agents which cause undesired elevations in [Ca2+], would cause aberrant activation of Ca2*-regulated proteins, possibly leading to cell death. Many toxic insults are believed to be mediated, at least 7 in part, by elevations in [Caz‘ji (Tsokos-Kuhn, 1989; Kass, et al., 1988). Hypoxia, induced by either KCN or reduction of 02, causes elevations in [Ca2+], in neuronal tissue, possibly due to the loss of cellular ATP (Johnson et al., 1994). Glutamate-induced excitotoxicity has also been suggested to be mediated by elevations in [Ca2*], resulting from abusive stimulation of postsynaptic N MDA receptors (Choi, 1987; Tymianski et al., 1993; Badar-Goffer et al., 1994). In some cell types, elevations in [Ca2+]i may also be associated with the initiation of programmed cell death (McCabe et al., 1992; Corcoran et al., 1994). C. Effects of MeHg on neuronal function as an index of disruption of cation regulation. The focus of initial experiments to examine potential mechanisms of neurotoxicity of MeHg was to characterize the changes in neuronal excitability and synaptic transmission following primarily acute exposure. These studies utilized electrophysiological techniques and suggested that MeHg acts at multiple sites to disrupt neuronal cation homeostasis. Owing to the importance of cation homeostasis in neurons, it was proposed that the neurotoxicity of MeHg was mediated, at least in part, by disruption of neuronal cation homeostasis (Juang, 1976a,b). Early studies focused on acute in vitro exposure to MeHg of frog sciatic nerve- sartorius muscle preparation (J uan g, 1976a). Muscle contractions mediated by stimulation of the sciatic nerve were blocked at lower concentrations than those elicited by direct electrical stimulation of the muscle, suggesting that MeHg disrupted neuromuscular transmission by a preferential action on motor nerves. Possible explanations for the disturbances in neuronal function following MeHg application included block of axonal 8 conduction, reduced depolarization-dependent Ca2+ influx into the nerve terminal, impaired docking or fusion of synaptic vesicles with the presynaptic membrane, or inhibition of postsynaptic nAChR. The possible neuronal sites of action of MeHg were explored initially using electrophysiological techniques (Juang, 1976b). These experiments yielded two key observations concerning the effects of MeHg on neuronal Ca2+ regulation. First, some of the effects of MeHg were consistent with a block of Ca2+ influx through voltage- dependent Ca2+ channels. Second, MeHg caused effects on neuronal function consistent with an alteration in Ca2+ homeostasis within the nerve terminal. The block of voltage- dependent Ca2+ channels and the increase in [Caz‘]i were manifested as a decrease in end- plate potential (EPP) amplitude and an increase in the frequency of miniature end-plate potentials (MEPP), respectively (Atchison and Spitsbergen, 1994). MeHg blocked nerve- evoked EPPs at the isolated nerve-muscle synapse by a presynaptic action because it did not alter the response to iontophoretically-applied ACh (Atchison and Narahashi, 1982). The decrease in EPP amplitude, but not the time to complete block of EPPs, was antagonized partially by increasing [Ca2+]e in the rat phrenic nerve-hemidiapharagm preparation (Atchison et al., 1986; Traxinger and Atchison, 1987). This observation was consistent with a noncompetitive block of voltage-dependent Ca2+ channels in the nerve terminal resulting in decreased Ca2+ influx following nerve stimulation. D. Alterations in Ca“ channel function by MeHg. Because of the indirect nature of investigations of Ca2+ channel function via examination of alterations in the electrophysiological properties of an isolated 9 neuromuscular junction, more direct methods of analysis were needed. The possibility that MeHg blocks depolarization-dependent influx of Ca2+ through voltage-dependent Ca2+ channels was investigated in several model systems, using a variety of techniques. In early studies, the effects of MeHg on Ca2+ influx were monitored by radiotracer flux analysis of depolarization-dependent uptake of “Ca2+ into synaptosomes isolated from rat cortex. Synaptosomes retain many of the structural and functional characteristics of an intact nerve terminal, and are a useful model system for studying presynaptic nerve terminal function (Gray and Whittaker, 1962). MeHg inhibited ”Ca2+ uptake elicited by K“ depolarization, consistent with a block of voltage-dependent Ca2+ channels (Atchison et al., 1986; Shafer and Atchison, 1989; Hewett and Atchison, 1992). The block of 45Ca“ uptake could not be reversed by 2“L, and more detailed kinetic analysis revealed that MeHg acted by a increasing [Ca noncompetitive mechanism (Hewett and Atchison, 1992). MeHg decreased the binding affinity of the dihydropyridine nitrendipine to synaptosomes, but effects of MeHg on total nitrendipine binding sites could not be established (Shafer and Atchison, 1989). In these studies, MeHg did not affect the synaptosomal ['”l]-conotoxin binding (Shafer et al., 1990). Thus, MeHg appears to alter voltage-dependent Ca2+ channel function in synaptosomes, possibly as a noncompetitive inhibitor of dihydropyridine-sensitive Ca2+ channels. Because the distribution of the different voltage-dependent Ca2+ channels in synaptosomes is not known, more detailed biochemical and electrophysiological analysis regarding the effects of MeHg on specific known Ca2+ channel subtypes were conducted using cells in culture as a model system. The advantage of using cells in culture is that 10 the effects of MeHg on specific Ca2+ channel subtypes could be studied directly using electrophysiological and biochemical techniques. These studies assessed the effects of MeHg on voltage-dependent Ca2+ channels expressed by a rat clonal pheochromocytoma cell line (PC12). PC12 cells grown in medium lacking nerve growth factor (NGF) morphologically resemble chromaffin cells of the adrenal medulla (Greene and Tischler, 1976). NGF causes these PC12 cells to differentiate into cells which resemble sympathetic neurons. Because differentiated PC12 cells possess fewer dihydropyridine- sensitive Ca2+ channels than undifferentiated cells, it is possible to alter the relative Ca2+- channel distribution in these cells by including NGF in the incubation medium (Shafer and Atchison, 1991a). MeHg blocked depolarization-dependent 45Ca2+ uptake into both undifferentiated and NGF-differentiated PC12 cells (Shafer et al., 1990). Electrophysiological analysis of Ba2+ currents in NGF—differentiated PC12 cells revealed inactivating and noninactivating components presumably mediated by N- and L-type Ca2+ channels, respectively (Shafer and Atchison, 1991b). In differentiated PC12 cells, MeHg decreased the peak Baz“ current, and inhibited completely the rapidly inactivating component of Ba2+ current, consistent with a block of N-type Ca2+ channels. MeHg also reduced the non-inactivating current in these cells, which was interpreted as a block of L-type Ca2+ channels. Thus, MeHg reduces Ca2+ currents in differentiated PC12 cell by inhibiting both N- and L-type Ca2+ channels. The actions of MeHg Ca2+ channel function have also been studied in dorsal root ganglion cells in culture (Arakawa et al., 1991). In this model system, MeHg inhibited total Ca2+ current and generated a slow inward current, possibly mediated by a nonspecific cation conductance. 11 E. Efi’ects of MeHg on presynaptic Ca“ regulation. In addition to inhibiting EPP amplitude in isolated nerve-muscle preparations. presumably due to decreased ACh release mediated by block of presynaptic voltage- dependent Ca2+ channels, MeHg also altered spontaneous release of neurotransmitter measured as changes in the frequency of occurrence of MEPPS (Juang, 1976b; Atchison and Narahashi, 1982; Atchison, 1986). MEPPS are thought to result from the spontaneous fusion of a single synaptic vesicle, or quantum, with the presynaptic plasma membrane, resulting in the discharge of vesicular contents into the synaptic cleft. Agents can alter MEPP characteristics in two distinct ways, assuming there are no postsynaptic effects of a compound. First, the amplitude of individual MEPPS could be increased or decreased by altering the number of neurotransmitter molecules packaged within each vesicle. Changes in MEPP amplitude imply a disruption of neurotransmitter synthesis, storage, or metabolism within the presynaptic nerve terminal. Vesamicol, which impedes transport of ACh into vesicles (Anderson et al., 1983), or hemicholinium-3, which impairs high affinity choline uptake from the synaptic cleft (Sacchi and Perri, 1973), both decrease MEPP amplitude. Another possible effect of toxicants is on the frequency of occurrence of MEPPs. Because the spontaneous fusion of a single vesicle with the presynaptic membrane is a probabilistic event, changes in the frequency of occurrence of MEPPS represent an alteration in the regulation of vesicular docking, fusion or release. Thus, it is possible to evaluate qualitatively the effects of toxicants on vesicular neurotransmitter content and release by monitoring changes in MEPP amplitude and frequency, respectively. 12 MeHg did not affect the amplitude of MEPPS, but MEPP frequency was first increased, then decreased (Juang and Yonemura, 1975; Juang, 1976b; Atchison and Narahashi, 1982), suggesting that MeHg did not alter ACh synthesis or metabolism within the nerve terminal, but did affect the process of transmitter release. These effects are consistent with an increase in resting Ca2+ concentration in the terminal (Miledi, 1973; Shalton and Wareham, 1979). The increases in MEPP frequency caused by MeHg were not affected by the voltage-dependent Nai channel blocker TTX or the voltage-dependent Ca2+ channel blocker Coz", either alone or in combination (Miyamoto, 1983). Therefore, the alterations in MEPP frequency could not be attributed to an unregulated increase in [Caz"]i mediated by activation of Na” or Ca2+ channels. This finding also suggested that MeHg does not need to enter the nerve terminal through these channels to elicit its effects. Thus, either the entry of MeHg is not essential, or MeHg is capable of entering the nerve terminal by a route other than the voltage-dependent Na+ or Ca” channels, possibly by diffusing across the plasma membrane due to its enhanced lipophilicity. Increasing the Ca2+ permeability at the nerve terminal hastened the time to onset of the MeHg-induced increase in MEPP frequency (Atchison, 1987). Pretreatment with solutions containing slightly elevated K“ concentration, or the voltage-dependent Na” channel activator veratridine, shortened the interval necessary for increased MEPP frequency. This suggested that the increased Ca2+ permeability facilitated the entry of MeHg into the terminal, or exacerbated the MeHg-induced elevations in [Ca2+]i. The increase in spontaneous release of neurotransmitter could be attenuated, but not blocked, by removal of Ca2+” suggesting a component of the MeHg-induced elevation in [Ca2+], was due to Ca2+ release from internal stores. Based on these studies it was proposed that 13 MeHg elevates resting [Ca2+]i by at least two mechanisms. MeHg increases the plasma membrane permeability to Ca”; and releases Ca2+ from an intracellular site. F. Mitochondrial Caz+ regulation as a target of MeHg. The observation that MeH g increased MEPP frequency in preparations bathed in Cay-free solutions led to investigations to identify the intracellular target of MeHg. Initial studies focused on the possible role of the mitochondria as the intracellular source of Ca”. Microscopic examination of brain tissue from subjects exposed to MeHg revealed altered mitochondrial morphology (Chang, 1980). Histochemical staining of brain slices for heavy metals demonstrated that mercury was associated with the mitochondrial membrane. The effects of MeHg on mitochondrial function have also been studied. MeHg depolarized the mitochondrial membrane (Hare and Atchison, 1992), inhibited ATP generation (Kauppinen et al., 1989) and disrupted mitochondrial respiration (Levesque and Atchison, 1991). As such, the mitochondria were investigated as the intracellular source of Ca2+. The effects of MeHg on spontaneous transmitter release were mimicked by several inhibitors of mitochondrial function such as dicoumarol, dinitrophenol, valinomycin and ruthenium red (Levesque and Atchison, 1987). Although these agents interfere with mitochondrial function through different mechanisms, they all increased MEPP frequency. Dicoumarol and dinitrophenol uncouple oxidative phosphorylation, valinomycin is a K” ionophore and ruthenium red is a putative inhibitor of the mitochondrial Ca2+ uniporter. Ruthenium red was unique in that application of MeHg to nerve-muscle preparations pretreated with ruthenium red did not elicit an increase in MEPP frequency, suggesting 14 that both agents acted on a common Ca2+ pool, possibly the mitochondria. This effect could not be attributed to a ruthenium red-induced depletion of vesicular stores or block of postsynaptic nAChR since La3+ increased MEPP frequency in ruthenium red-treated preparations. Another inhibitor of mitochondrial Ca2+ regulation, N,N—bis(3,4- dimethoxyphenethyl)-N—methylamine (YSO35), also blocked the effects of MeHg, but not La“, on MEPP frequency (Levesque and Atchison, 1988). These results suggested that release of Ca2+ previously sequestered in the mitochondria contributed to the MeHg- induced elevations in [Ca2+]i. The effects of MeHg on mitochondrial Ca2+ regulation were studied directly using intact mitochondria isolated from rat brain (Levesque and Atchison, 1991). Ca2+ uptake by isolated mitochondria occurs by two mechanisms which differ in magnitude and ATP requirement. “Caz“ uptake via the ATP-dependent mechanism is approximately an order of magnitude greater than the ATP-independent component. MeHg inhibited both components of mitochondrial Ca2+ uptake. The ability of MeHg to release Ca2+ sequestered within the mitochondria was studied by loadin g the mitochondria with ”Ca“, in the absence or presence of ATP, prior to MeHg exposure. MeHg reduced mitochondrial “’Caz” content which had previously been loaded into the mitochondria, regardless of ATP. The reduction in mitochondrial 45Ca2“ content could be due to stimulation of Ca“ efflux, or block of Ca“ uptake necessary for Ca2+ cycling. Ruthenium red also reduced the loading of 45Ca2+ into the mitochondria, in the absence or presence of ATP. Ruthenium red prevented the MeHg-induced reduction in mitochondrial 45Ca2+ content, but did not block the binding of Me[2°3]Hg to the mitochondria (Levesque and Atchison, 1991). Taken together, this evidence suggested that MeHg-induced disruption 15 mitochondria Ca2+ regulation contributed to the elevations in [Ca2+], in the nerve terminal, and that ruthenium red acted on a similar Ca2+ pool of mitochondrial Ca2+ as MeHg. While it was apparent that MeHg was capable of disrupting mitochondrial Ca2+ regulation, the relationship of this effect to the elevations in [Ca2+], and spontaneous release of neurotransmitter was unknown. The magnitude of MeHg—induced release of transmitter was quantified using synaptosomes preloaded with various radiolabelled neurotransmitters. MeHg caused release of dopamine, y«aminobutyric acid and ACh from nondepolarized synaptosomes prepared from striatum, cortex and hippocampus (Minnema et al., 1989). MeHg did not affect the release of these neurotransmitters elicited by depolarization with solutions containing elevated concentrations of K“. This suggested that, in synaptosomes, the block of voltage-dependent Ca2+ channel by MeHg was not sufficient to impede transmitter release. MeHg also released 45Ca2+ preloaded into isolated mitochondria, but did not release preloaded 45Ca2+ from intact synaptosomes. It was concluded that MeHg released of Ca2+ from the mitochondria, and that this Ca2+ remained within the cytosol of the intact synaptosomes to cause an elevation in [Ca2+]i. The amount of Ca2+ released from the mitochondria was sufficient to support spontaneous release of neurotransmitter from synaptosomes. However, the experimental design for loading the synaptosomes with 45Ca2+ may have allowed for exchange of the radiolabelled ”Ca2+ within the synaptosomes for nonradioactive Ca2+ in the extrasynaptosomal buffer. This could have caused depletion of 45Ca2" from the radiolabelled intracellular pool prior to the addition of MeHg. If remaining pools of “Ca” within the synaptosomes are not mobilized by MeHg, then these experiments would conclude incorrectly that MeHg did not affect [Ca2+]i. l6 Ruthenium red also increased spontaneous release of ACh from synaptosomes, although the amount of released ACh was not as great as that elicited by MeHg (Levesque et al., 1992). Pretreatment of synaptosomes with ruthenium red caused a concentration-dependent reduction in MeHg-induced ACh release in the presence or absence of added Ca2+, While this evidence suggested disruption of mitochondrial Caz“ regulation following MeHg exposure was sufficient to elicit spontaneous release of ACh, it was by no means conclusive. Moreover, MeHg-induced depolarization of the mitochondrial membrane was not inhibited by ruthenium red, suggesting that either the effects of ruthenium red or MeHg were not mediated by depolarization of the mitochondria. G. Techniques for measuring [Ca2+],». The limitation of previous experiments Studying the effects of MeHg on [Ca2*]i in the nerve terminal is that they relied upon indirect measurements, or inference, in most instances. The synthesis of indicators based on polyvalent cation chelators (Smith et al., 1983; Tsien et al., 1982; Grynkiewicz et al., 1985; Minta et al., 1989) allowed for the measurement of [Ca2+], in isolated neuronal preparations (Komulainen and Bondy, 1987a). These compounds had several significant advantages over their predecessors. Unlike the photoprotein aequorin and the fluorochrome arsenazo III, these new indicators were not structurally inactivated upon binding of Ca2+. They also interacted with Ca“ in a reversible manner, with dissociation constants for Ca“ in the physiological range, allowing changes in [Ca2+]i to be monitored more consistently over time. The new indicators could also be loaded into cells, and calibrated, relatively easily. Some l7 disadvantages of Ca2+ detection remained, however. The new compounds were not entirely Ca2+ specific (Grynkiewicz et al., 1985), and certain cells could actively transport the dyes from the cytosol (DiVirgilio et al., 1988). Even these problems could be circumvented by the use of cell-penneant heavy metal chelators (Arslan et al., 1985), and inhibitors of the organic anion transporter. The most commonly utilized structural design of these Ca2*-sensitive indicators is to position a fluorescent group near the Ca2*-binding region of the chelator (Fig. 1.1). The binding of Ca2+ distorts the electronic configuration of the nearby fluorescent group, resulting in a shift in the fluorescent properties of the dye in the Ca2*-free and Ca2*-bound form. The relatively high sensitivity of fluorometry permits excellent temporal and spatial resolution of [Ca2+],. Some Ca2*-sensitive fluorescent indicators have been designed for specific purposes or techniques. For example, fura-2 is used extensively in epifluorescence microscopy, indo-l is used primarily in continuous and stopped-flow cytometry, and fluo-3 is used primarily in fluorescence-activated cell sorting and confocal microscopy. In this way, a Cay-sensitive fluorescent dye exists for virtually every detection system. There are limitations to the use of fluorescent dyes, though. A major drawback lies in their affinity for cations other than Ca2+ (Grynkiewicz et al., 1985; Tomsig and Suszkiw, 1990; Hechtenberg and Beyersmann, 1993; Hinkle et al., 1992). The interaction of these indicators with cations other than Ca2+ confounds accurate determination of [Ca2+], (Arslan et al., 1985; Mason and Grinstein, 1990). Another problem is the intrinsic fluorescence of certain cells, such as red blood cells, makes fluorescence analysis difficult, if not impossible (Murphy et al., 1986). 18 FIG. 1.1. The chemical structure of the fluorescent cation indicator fura-2. The four carboxylic acid groups form the cation binding pocket, while the polycyclic ring system is fluorescent group. Binding of cations alters the electronic configuration of the fluorescent moiety, causing a shift in the fluorescence properties of the molecule. 19 Fig. 1.1. (-OOC-CH2)2N NICHz-COO'Iz O-CHz-CHz-O l Fura—2 20 Indicators resembling the fluorescent dyes have also been constructed for use with 19F-nuclear magnetic resonance (NMR) spectroscopy (Smith et al., 1983). The structure of these indicators contains equivalent fluorine nuclei instead of the fluorescent groups (Fig 1.2). When a cation binds to the indicator, it withdraws the electrons surrounding the fluorine nuclei, deshielding it from the applied magnetic field. This alters the resonance properties of the fluorine nuclei, and generates a resonance peak in the 19F spectrum. Since each cation deshields the fluorine nuclei to a different extent, their corresponding resonance peaks are distinct. '9F-NMR spectroscopy has several advantages over fluorescence-based systems (Bachelard and Badar-Goffer, 1993). First, nonspecificity is not a problem since the identity of the cation binding to the fluorinated indicator is readily apparent. Second, the changes in the intracellular concentrations of several cations can be monitored simultaneously. For example, using 19F-NMR spectroscopy of NG108-15 cells loaded with the fluorinated chelator SF-BAPT A, changes in the intracellular concentrations of Caz", Pb“ and Zn“ can be measured concurrently following Pb2+ exposure (Schanne et al., 1989). Third, in a single preparation, the changes in the cellular concentrations of polyvalent cations and high energy phosphates can be monitored by using ”P and 3‘P-NMR spectroscopy in tandem (Badar-Goffer et al., 1994). Fourth, because no endogenous fluorine is present in cell, there are no difficulties associated with intrinsic fluorine resonance. The main disadvantage of ”F- NMR specrroscopy is that its low sensitivity necessitates lengthy spectral acquisition intervals, reducing temporal resolution to tens of minutes, as opposed to the tenths of seconds resolution obtained with fluorescent indicators. Another drawback of this technique is its inability to generate useful spatial information at the cellular level. FIG 1.2. The chemical structure of the fluorinated cation indicator 5F-BAPTA. The four carboxylic acid groups form the cation binding pocket, while the resonance of the fluorine nuclei is observable using NMR spectroscopy. Binding of cations withdraws the electrons which surround the fluorine nuclei, altering its resonance properties. 22 Fig. 1.2. (-ooc-CH2I2N mom-coon, o-cH2-CH2-o o F F 5F-BAPTA 23 H. Efiects of MeHg on neuronal divalent cation homeostasis. Acute exposure of synaptosomes loaded with the fluorescent chelator fura-2 to MeHg caused a time- and concentration—dependent increase in [Ca2+]i (Komulainen and Bondy, 1987b). The elevations in [Ca2+], were also dependent on [Ca2+],, and were not reduced by the voltage-dependent Ca2+ channel blocker verapamil. This suggested MeHg did not cause Ca2+ influx through these channels, consistent with results gathered using radiotracer flux analysis and electrophysiological techniques. The MeHg-induced elevations in [Ca2+], were potentiated by ouabain or depletion of cellular ATP. Depolarization of the mitochondrial membrane with rotenone and oligomycin decreased these elevations in [Ca2+], providing additional suggestive evidence for MeHg-induced disruption of mitochondrial Ca2+ regulation. The concentration-dependence of the possible contributions of extracellular and mitochondrial Ca2+ sources to the MeHg-induced elevations in [Ca2+], was investigated in greater detail using fura-2 loaded synaptosomes (Kauppinen et al., 1989). MeHg concentrations below 30 pM increased [Ca2+],, but did not increase fura-2 leak from the synaptosomes, suggesting that MeHg acted at an intracellular target without significantly disrupting the plasma membrane. These concentrations of MeHg also altered mitochondrial function without affecting the plasma membrane potential. Higher concentrations of MeHg caused pronounced elevations in [Ca2+],, and increased an“ quenching of fura-2, which coincided with depolarization of the plasma membrane. This suggested that lower concentrations of MeHg increased [Ca2+], by an action at an intracellular Ca2+ store, possibly the mitochondria, while higher concentrations increased the plasma membrane permeability to Ca“, 24 Additional analysis of the effects of MeHg on neuronal divalent cation homeostasis revealed that in addition to elevating the [Ca2+],, MeHg also increased the intrasynaptosomal concentration Zn2+ ([Zn2*]i) (Denny et al., 1993; Denny and Atchison, 1994). The mechanisms mediating the elevations in [Ca2+]i and [Zn2‘“]i are distinct both temporally and mechanistically. The elevations in [Zn2"]i are immediate and precede the gradual increase in [Ca2+]i. The increases in [Zn2*]i are mediated by release of Zn2+ from several soluble synaptosomal proteins, while the elevations in [Cahji are due to increased plasma membrane permeability to Ca2+ leading to a Ca2*e-dependent elevation in [Ca2+]i (Denny et al., 1993, Denny and Atchison, 1995). The elevations in [Ca2+]i are believed to be responsible for alterations in spontaneous transmitter release observed following MeHg treatment. Similar findings have been obtained in other neuronal model systems. MeHg also elevates the intracellular concentrations of Ca2+ and another endogenous polyvalent cation, presumed to be Zn”, in neuroblastoma x glioma (NG108-15) hybrid cells (Hare et al., 1993) and isolated cerebellar granule cells in culture (Many et al., 1995). In NG108-15 cells, a component of the elevations in [Ca2+]i is mediated by release of Ca2+ which was sequestered previously in the IP3-sensitive Ca2+ pool of the ER (Hare and Atchison, 1995). However, MeHg mobilizes this Ca“ pool without generating IP3. In NG108-15 cells treated with MeHg these elevations in [Ca2+]i can be delayed by nifedipine or TTX (Hare and Atchison, 1996). This suggests voltagedependent Ca2+ or Na+ channels are involved in the MeHg-induced increases in plasma membrane permeability to Ca”, possibly by augmenting MeHg entry into the cells. CHAPTER TWO METHYLMERCURY ALTERS INTRASYNAPTOSOMAL CONCENTRATIONS OF ENDOGENOUS POLYVALENT CATIONS 25 26 ABSTRACT The effects of MeHg on intrasynaptosomal polyvalent cation concentrations were examined using fura-2. In the presence of Call, MeHg caused a concentration-dependent, biphasic elevation in the ratio of fluorescence intensity at the emission wavelength of 505 nm following excitation at 340 and 380 nm (340/380 nm ratio). The first phase was independent of Caz“e and complete within five sec. The second phase was dependent upon Caz“e and not complete within six min. MeHg increased the synaptosomal membrane permeability to Mn2+ suggesting that the second phase was due to influx of Caz} Ruthenium red (20 pM), mitochondrial depolarization (10 mM NaN3 plus 4 pg/ml oligomycin), thapsigargin (1 pM), or caffeine (40 mM) did not elevate [Ca2*]i or alter the response of the synaptosomes to MeHg. Upon closer inSpection, we noticed that MeHg simultaneously increased the fluorescence intensity at the excitation wavelengths of 340 and 380 nm, and at the Ca2*-insensitive excitation wavelength of 360 nm. Pretreatment of synaptosomes with the cell-permeant heavy metal chelator N,N,N’,N’-tetrakis-(2- pyridylmethyl)ethylene-diamine (TPEN, 50 pM) blocked the MeHg-induced elevations in 360 nm intensity and 340/380 nm ratio. TPEN given after MeHg, reversed the elevations in 360 nm intensity. The cell-impermeant heavy metal chelator diethylenetriaminepentaacetic acid (DTPA, 150 pM) had no effect. We conclude that MeHg disrupts polyvalent cation homeostasis by at least two mechanisms. The first involves release of endogenous non-Ca2+ polyvalent cations while the second is due to increased Ca2+ permeability of the plasma membrane. 27 INTRODUCTION Fluctuations in [Ca2+]i regulate many cellular processes, including neurotransmitter release, cell differentiation, growth cone elongation and excitation-contraction coupling (Bertolino and Llinas, 1992). Sustained or unregulated excessive elevations in [Ca2+], are toxic (Nicotera et al., 1992). CC14-induced hepatotoxicity (T sokos-Kuhn et al., 1986; Tsokos-Kuhn, 1989) and 1-methy1-4-phenylpyridinium (MPP*) toxicity in the substantia nigra (Kass et al., 1988) are both believed to be related to loss of Ca2*-regulating ability. Excitatory amino acid neurotoxicity is also thought to be associated with pathological elevations of [Ca2+], via activation of NMDA receptors (Choi, I987). Neurons are sensitive to sustained elevations in [Ca2*]i. Nerve terminals are particularly affected by impaired [Ca2+]i homeostasis, because of their large surface area to volume ratio, and the importance of tightly controlled Spatial and temporal changes in [Ca2+]i necessary for secretion (Blaustein, 1988). Changes in [Ca2+]i at the nerve terminal alter many processes, including synaptic transmission. MeHg, a known neurotoxicant, causes alterations in synaptic transmission which are characteristic of impairments in normal entry of Ca2+ during depolarization and sustained, unregulated elevations of [Ca2+], within the nerve terminal. Acute application of MeHg to isolated nerve terminals elevates [Ca2+li (Komulainen and Bondy, 1987a; 1987b; Kauppinen et al., 1989). Effects of MeHg on Ca2+ homeostasis are believed to be mediated in part by increased plasma membrane permeability to Ca2+ since reducing the [Ca2+], decreases the functional effects of MeHg at intact synapses (Atchison, 1986) and synaptosomes (Levesque et al., 1992). Removal of Caz“e decreases, but does not block the MeHg-induced elevations in [C112“]i in synaptosomes (Komulainen and Bondy, 28 1987b). Disruption of intracellular Ca2+ regulation is believed to mediate the effects of MeHg on [Ca2“]i and nerve terminal function following removal of Ca“, Because normal secretory function requires tight spatial and temporal regulation of [Ca2+]i, Ca2+ homeostasis within the nerve terminal is maintained by the concerted buffering action of Ca2+ binding proteins, endoplasmic reticulum (ER) and mitochondria (Blaustein et al., 1980; Akerman and Nicholls, 1983). The respective contributions of these storage pools varies according to the extent and duration of the Ca2+ load. Disruption of Ca“ regulating abilities results in responses typically associated with elevations in [Caz‘]. Several lines of evidence implicate disruption of intracellular Ca2+ regulation in the toxic actions of at least acute, in vitro application of MeHg. In isolated mitochondria, MeHg prevents uptake of ”Ca2+ and induces its release following preloading (Levesque and Atchison, 1991). Inhibition of Ca2+ release by YSO35 (Deana et al., 1984) and ruthenium red (Moore, 1971), a putative inhibitor of the mitochondrial Ca2+-uniporter, blocks the spontaneous release of transmitter induced by MeHg at intact peripheral synapses (Levesque and Atchison, 1987; 1988) and isolated central nerve terminals (Levesque et al., 1992). Ruthenium red also blocks MeHg-induced release of ”Ca” from preloaded mitochondria (Levesque and Atchison, 1991). Additionally, MeHg disrupts mitochondrial functions (Verity et al., 1975; Levesque and Atchison, 1991). Finally, inhibitors of oxidative phosphorylation such as dinitrophenol and dicoumarol cause effects similar to those of MeHg at intact synapses (Levesque and Atchison, 1987). The present study was initiated to determine the respective contributions of intracellular Ca2+ stores to the MeHg-induced elevations in [Ca2+]i. We focused initially on the mechanism of block by ruthenium red of MeHg-induced effects on mitochondrial 29 Ca2+ buffering and its potential relationship to the putative changes in [Ca2+],. We hypothesized that ruthenium red would prevent MeHg—induced elevations in [Ca2”]i by preventing release of mitochondrial Ca2+ Stores. Paradoxically, mitochondria did not contribute to changes in fura-2 fluorescence in synaptosomes. During the course of these experiments, we observed changes in the fura-Z fluorescence signals induced by MeHg which were inconsistent with changes in [Ca2+],. As such, we examined these changes in some detail with respect to their implications for the action of MeHg on nerve terminal cation homeostasis. 30 MATERIALS AND METHODS Chemicals and Solutions: Methylmercuric chloride (MeHg) was obtained from K and K Labs, Plainview, NY. Fura-Z pentapotassium salt, the acetoxymethyl ester derivative of fura-2 (fura-2AM) and TPEN were purchased from Molecular Probes, Eugene, OR. Thapsigargin was purchased from LC Services, Woburn, MA. Sodium azide was supplied by Aldrich, Milwaukee, WI. Caffeine, DTPA, oligomycin, ethyleneglycol-bis(B-aminoethyl ether)-N,N,N’,N'-tetraacetic acid (EGTA), Na”-Hepes, ruthenium red, Trizma base, and Trizma HCl were obtained from Sigma, St. Louis, MO. Dimethyl sulfoxide (DMSO) was obtained from IT. Baker, Phillipsburg, N]. All other chemicals were reagent grade. Hepes buffer (HBS) contained (mM): NaCl, 135; KC], 5; MgClz, 1; d—glucose, 10; and N-(2-hydroxyethyl)piperazine-N’-(2-ethanesulfonic acid) (NT-Hepes), 10. The pH at room temperature was adjusted to 7.4 with Trizma HCI. HBS contained 6 pM Ca2+ as measured by inductively coupled plasma spectroscopy (Atchison, 1986), and is hereafter referred to as 6 pM Ca2*-HBS. During loading of synaptosomes with fura-ZAM, 200 pM CaCl2 was present in the HBS (200 pM Ca2"—HBS). A 1 mM stock solution of fura-2AM was prepared twice a week in anhydrous DMSO. MeHg was prepared weekly as a 10 mM stock solution in appropriate HBS. Ruthenium red was prepared in 200 pM Ca2+- HBS and diluted six-fold upon addition to synaptosomes. NaN3 was prepared daily in deionized water as a 1 M stock solution. Oligomycin stock solutions (4 mg/ml) were dissolved in absolute ethanol. TPEN (2 mM) was dissolved in a solution of 50% (v/v) ethanol/deionized water which yielded a final ethanol concentration of 1.25% after addition to the synaptosomal suspension. EGTA (1 M) for RTnin determination was 31 dissolved in deionized water and the pH at room temperature adjusted with Trizma base to greater than 8. DTPA was prepared as a 15 mM stock solution in deionized water and dissolved by addition of NaOH. Caffeine was prepared as a 1.33 M stock solution and gently heated until dissolved. Thapsigargin was prepared as a 100 pM stock solution in 0.5% (v/v) DMSO. Control solutions contained appropriate concentrations of the solvents used. Preparation of Synaptosomes: Synaptosomes were prepared by discontinuous sucrose gradient centrifugation as described in detail by Shafer and Atchison (1989) with slight modifications. The sucrose buffers were prepared in 3 mM NazHPO4 buffer adjusted to pH 7.4 with HCl. The P2b fraction (Gray and Whittaker, 1964) was resuspended in 6 pM Ca2*-I-IBS, centrifuged at 10,000 x g for 10 min and resuspended in 10 ml 200 pM Ca2*—HBS. The synaptosomal preparation was then split into two 5 ml fractions; fura—2AM was added to one fraction (5 pM final concentration) while the other received DMSO alone. The final concentration of DMSO in both fractions was 0.5% (v/v). Both fractions were incubated at 30°C for 30 min to allow for uptake and hydrolysis of the fura—2AM. The fractions were then diluted six-fold by addition of 25 ml of 200 pM Cazi-HBS and incubated for an additional 15 min at 30°C to promote diffusion of partially hydrolyzed fura-ZAM from the synaptosomes. After centrifugation at 10,000 x g for 5 min, the pellets were resuspended in 10 ml 200 pM Ca2*-HBS, and kept on ice until needed, but no longer than three hr. rt? 1“. 4" 32 Determination of 340/380 nm ratio and [Ca2*l,-.' Although concentrations of MeHg up to 1 mM did not affect the fluorescence of 0.5 pM fura-2 pentapotassium salt when tested in a cuvette (Fig. 2.1), the addition of MeHg to fura-2 loaded synaptosomes changed fura-2 fluorescence in ways inconsistent with elevations in [Ca2+]i (see below). This effect was inhibited by a cell-permeant heavy metal chelator. Some concentrations of MeHg also caused leakage of fura-2 from the synaptosomes. For these reasons, the conversion of ratios to [Ca2+]i was precluded in most instances. Instead, the results are usually reported as changes in the ratio of 340 nm to 380 nm excitation intensity monitored at 505 nm (340/380 nm ratio). In certain instances, when low concentrations of MeHg were used and the increase in intracellular heavy metal concentration was prevented by a cell-permeant chelator, [Caz’fili determination could be made. Fura-2 loaded synaptosomes were centrifuged at 12,500 x g for 30 sec. The resulting pellet was resuspended in two ml of either 6 pM Cay-HBS (for studies involving [Ca2“]e of 6 pM or 0.1 pM) or 200 pM Cay-HBS (for studies involving 200 pM [Ca2+]e), transferred to a test tube, and incubated for 10 min at 37°C. The suspension was transferred to a polystyrene cuvette containing a magnetic stir bar and placed into a spectrofluorometer (SPEX Industries, Edison, NJ) equipped with a thermally jacketed cuvette holder at 37°C. The emission intensity was monitored at 505 nm (bandpass of 3.77 nm) following excitation (450 W xenon lamp) of the synaptosomal suspension at 340 and 380 nm, or 360 and 380 nm. Alternating excitation wavelengths were obtained using a beam chopper. Ratios and [Ca2+]i were corrected for intrinsic fluorescence using synaptosomes treated with 0.5% DMSO only. The intensity recordings of fura-2 loaded synaptosomes were also corrected for leakage of fura-2 from synaptosomes not exposed 33 TEE mo 85.585 858825 of Sofia 8: EU 28 ~ 9 a: wioE Co mcocmbcoocou .E: cow was on c3359 33? Ewco~o>a3 comes: 3 cocazoxo 3.36:8 Ewcofiéa coammEo 2: a 353:9: mm; x2295 858805E .332 21 cm .5 :OECB on. $2.: 3563 Luca new $2.: 333 888 @2860“ Eva mmmamu 2: c E NEE: Lo 28m cougoxm NEE Co moEomoa 8:08805c 2: :O quE Co Soto 2:. A...” .05 34 3.5 59.20.63 :ozozoxm oov own cum _ . _ (suun "IIQJBI MisueIUI é.“ .2”. 35 to MeHg. Mn2+ (final concentration 40 pM) was added to a fura—2 loaded synaptosomal suspension and the fluorescence intensity remaining two sec after addition was subtracted from the intensity prior to an” addition. This difference was due to extracellular fura-2 and was subtracted from all intensity recordings from that preparation. The sample used for determination of extracellular fura-2 was not used for any other experiments. Ratios were determined by dividing the corrected intensity recording from excitation at 340 nm by that of 380 nm excitation. When appropriate, [Ca2+]i was calculated from the equation of Grynkiewicz, et a1. (1985): [Ca2*]i=Kd[(R-Rmin)/(Rm,,-R)]x(S,2/S.,2): where Rm, is the 340/380 nm ratio during Ca2+ saturation, and R is the 340/380 nm ratio during Ca2*-free min conditions. S,2 and sz are the emission intensities at 380 nm excitation during Ca2*-free and Ca2*-saturating conditions, respectively. R was determined by lysing the max synaptosomes with 0.1% (v/v) TritonX-IOO in 2 mM Ca2+. R was determined on the min same sample by adding 50 mM EGTA (pH remained at 7.4). Synaptosomal preparations yielding Rm, values less than 3.00 were rejected due to inadequate fura-2AM hydrolysis. Assessment of synaptosomal permeability to divalent cations after M eHg treatment: Quenching of the fura-2 signal by Mn2+ was used to assess the permeability of the synaptosomal plasma membrane to divalent cations. Synaptosomal preparations were incubated with MeHg (0 - 100 pM) for six min. The fluorescence intensity at the Ca”- insensitive wavelength of 360 nm was monitored every 0.2 sec. After establishing a ten sec baseline, Mn2+ (40 11M final concentration) was added directly into the cuvette from an overhead injection port without interrupting data acquisition. The ability of the Mn2+ to quench total fura-2 fluorescence was monitored for four min. 36 To assess the immediate effects of MeHg on divalent cation permeability, Mn2+ was added prior to MeHg. Leak of fura-2 from the synaptosomes was determined by the subsequent addition of 150 pM DTPA to the synaptosomal suspension after a three min exposure to MeHg. If MeHg elevated extracellular fura-2 concentrations during the three min exposure, the immediate reversal in quenching produced by DTPA should exceed the initial quench produced by Mn2+ addition. Evaluation of non-Ca“ polyvalent cation contributions to fluorescence: Many cations interact with fura-2 (Grynkiewicz et al., 1985). Thus since some of the alterations in fura-2 fluorescence were inconsistent with changes in [Ca2+],, MeHg might alter the intracellular concentrations of cation(s) Other than Ca2+. To test this possibility, the effects of either the cell-permeant heavy metal chelator TPEN or the cell-impermeant heavy metal chelator DTPA on MeHg-induced elevations in 340/380 nm ratio and 360 nm excitation intensity were monitored. Statistical analysis: Elevations in 340/380 nm ratio were analyzed by a one-way block analysis of variance. Post-hoc analysis was performed using Tukey’s test (p<0.05) (Steel and Torrie, 1980). 37 RESULTS In 200 uM Ca2+-HBS, MeHg caused a biphasic increase in the 340/380 nm ratio that was concentration- and time-dependent (Fig. 2.2). The first phase increase was rapid and complete within five sec after MeHg addition. The magnitude of this phase was determined by subtracting the average ratio value ten sec prior to MeHg addition (50-60 sec) from that obtained ten sec following MeHg addition (62-72 sec) (inset, Fig. 2.2). Addition of 200 pM Ca2*-HBS caused a change in the resting 340/380 nm ratio of 0.06 :1: 0.01 (mean :1: standard error of the mean, S.E.M.) during this initial phase. MeHg concentrations of 10, 25, 50 and 100 uM caused 340/380 nm ratio changes of 0.18 :t 0.04, 0.25 :1: 0.06, 0.25 i 0.07 and 0.25 i 0.08, respectively. These changes were not statistically different from control (n=3, p>0.05), but suggest that the initial phase was concentration-dependent up to 25 pM MeHg. The rate of increase in 340/380 nm ratio during the second phase was gradual with no plateau occuning during the six min exposure period. Unlike the first phase, the second phase was not maximal at 25 pM MeHg. With the exception of 10 pM MeHg, the second phase elevations were several-fold greater than the initial elevations. The second phase was concentration-dependent with respect to both MeHg and Ca”; Lowering [Ca2+], to 6 pM reduced the second phase elevations relative to those at 200 pM [Ca2“]e for all MeHg concentrations tested (not shown). To determine the extent of Ca2+, dependence of the two phases, EGTA was added to the synaptosomal suspension to reduce [Ca2“_],3 further. Since [Caz“]i in nerve terminals ranges between 100 to 300 nM (Nachshen, 1985; Komulainen and Bondy, 1987a,b; Kauppinen et al., 1989), and the Ca2+ content of synaptosomes can be readily depleted 38 Am”: .modAav EEEHEES :EctE 2255.0 2: mo 0:02 .A<>OZ0.05). Therefore the apparent reduction in [Ca2+], caused by 20 uM EGTA did not alter the MeHg-induced immediate elevation in the 340/380 nm ratio. The second phase elevations observed in 6 or 200 uM Ca2*-HBS were reduced or abolished in 0.1 uM Ca2+-HBS, suggesting the second phase was primarily C212*e-dependent (compare Fig. 2.3 to Fig. 2.2). No significant difference between the first phase elevations were observed when MeHg was added one or four min following EGTA (not shown), thus depletion of Ca2+ from intracellular stores was unlikely. The Gaye-dependent second phase elevation in 340/380 nm ratio could be due to influx of extracellular Ca2+ and/or release of sequestered intracellular Ca2+. To identify the source, we monitored the ability of Mn“ to quench total fura-2 fluorescence of synaptosomes following MeHg treatment. Mn2+ has similar paths of entry in synaptosomes as does Ca“ (Drapeau and Nachshen, 1984; Nelson, 1986), but unlike Ca2+, Mn2+ quenches fura-2 fluorescence (Grynkiewicz et al., 1985). Mn2+ is commonly used in conjunction with fura-2 to assess the permeability of the plasma membrane to divalent 41 .3“: .modvnc _o:::8 E0: 8:282: 325%; 8286:: «M223: :8: 33:3,: :3 858:2: 83 29:3: :2: :6“: .m::oE::2:xo 82: :o: .E.m.m H SEE 2: E :3 coam .<>OZ< :83 $3.80 :3 :832: 22> 20552: 2:. A 2:sz :8 :888: m: :oEEbE: 203 :12): :3 :228 0:8 E: owm\o:m E 252220 BEBEE: 2C. ”:85 .mEoEcoaxo 8:: :0 :8E 2: 8:228: 8:: 85 .39: 8 :28::E 8 30:28.8: BEoman$ 2: o: :o::: :2: 225 onE :o mco:::::8:8 .3055 .<:.Om :o :o::::: wEBBB: EE .58 8.558: 28: E: emevm 2F .23. :o bemEExoaa: 9 9?:0. 82:2 9 <:.Om 21 ON :5 :28: 08>) mm:-+~:0 2: o E moEomoaaim :88— mésm .mm:-+~:U 21 :.o E 0:8 E: oonvm 2: :0 win: :0 88:0 2E. .m.~ .UE 42 2:5 2.: nunvr cow :2... 3:33 o m 3 o F . n-no one: u! abuauo 6.: .9: 03.0 Q WU 088/078 onu 43 cations (Merritt et al., 1989; Kass et al., 1990; Clementi er al., 1992). Because the concentration of Mn2+ within synaptosomes is minimal, decreases in fluorescence intensity result only from entry of extracellular Mn“. This provides a more critical evaluation of plasma membrane permeability than simply monitoring changes in 340/380 nm ratio. an" quenching of fura-2 fluorescence intensity occurs in two distinct phases. First, extracellular fura-2 is quenched immediately following Mn2+ addition. Second, intracellular fura-2 is quenched upon Mn“ entry into the synaptosome. Mn2+ quenching of fura-2 was monitored at the Cay-insensitive excitation wavelength of 360 nm. To determine the time-course of Mn2+ quenching of extracellular fura-2, 4O uM Mn2+ was added to a cuvette containing 0.5 pM fura-2 pentapotassium salt without synaptosomes. The 360 nm intensity decreased by 85% within 0.8 sec (Fig. 2.4, dashed line). If MeHg caused release of fura-2 from the synaptosomes during the six min incubation, the immediate quenching of the fluorescence intensity by Mn2+ would be enhanced. The second component of Mn2+ quenching would be expected to occur gradually as Mn2+ quenched intracellular fura-2. Increased Mn2+ permeability would. result in a greater rate of quenching during this phase. As expected, Mn2+ (4O uM) quenched the fura-2 signal in two distinct phases (Fig. 2.4). Immediately following addition of Mn2+ to fura-2 loaded synaptosomes, there was a rapid decrease in fluorescence intensity, similar to that observed in a cell-free system. The immediate quenching was quantitated as the percent difference between the average fluorescence intensity ten sec prior to the addition of Mn2+ (0 to 10 sec) and average fluorescence intensity over the two sec which followed the first full sec after Mn2+ addition (11 to 13 sec). The immediate quench in synaptosomes treated with up to 25 pM .:o:8:omo.a 28:68 9 233: :23?“ 2o? moon: £225 8:88.52: o5 .3 82? of. .oE: 38 :m oom Nd bo>o :83 2o? meEoUo: 56:25 MEoEComxo ooh: mo omEoZW o5 8:883: o:: Bow gum .mée §_::ooab:_ mo w:Eo:osc 9 on: box: 50:: $3 322:: 8:88:85 E o::oow 335:8 2C. .33. 2 9 :V 5235 A52 5:: oom :3 55 oz: 330:8 :oEB oom o): o5 :o>o Gov. 2 9 8 5:63 $52 9 St: oom :9 E8: 55:85 8:88:25 omEo>m E ommobo: EoBoa of m: no:§::a:v £3 w:Eo:o:U owns: EE .3385 8.». wd £5 mES: Uoiomno w:Eo:o:w :0 E295: o5 @3385 onE 2: oo— 8 Om 5:3 m:o_m:oam=m EEcmoaaim :o 5:83:— .:o:€w: $52 wEBo:8 oom wd :EE: 35:25 E o::oo: $me 22:88 33 $2.: 3563 moEomoamim 5923 mmI-+~mU .21 com S :8 8233885: N-E3 21 md mo w:Eo:o:c 2:. .335 333 E: com :0 Ew:o_o>w3 :ocmzoxo o>c_m:om:_-+~m0 o5 8 3:02:08 33 $52 2: ow 3 8:83:03: WEE no w:Eo:osc o5 Em .EE xi :8 :2: 8H - 8 332 £5 32335 8o? mcommfiamsm .mEomoamim .m:2m:o%=m EEomoamim 332 $83 .woaob-wmo2 E 56:25 8:83.52: :0 5:30 .vd .UE 45 2:5 25... .v m N _. O 02T=oo 41.1.; ml... 9 u m. 1.1 .A are: \l e 3 F J N or. «.1 mm o m “I s ( in .2“. ac: 46 MeHg was not different from control, but at higher concentrations (50 and 100 pM) there was a significant (p<0.05) increase in quench. This effect could be due either to increased fura-2 leakage from synaptosomes, lysis of a select p0pulation of synaptosomes or facilitation of quenching of intracellular fura—2. All concentrations of MeHg tested increased the quenching of intracellular fura-2 relative to controls (Fig. 2.4). Alternatively, the gradual decrease in fura-2 signal could reflect quenching of fura-2 leaking from damaged synaptosomes. In either case, however, Mel-lg altered the plasma membrane permeability which supports the notion that the second phase elevations in 340/380 nm ratio observed in Ca2*-HBS were due to increased Ca2+ flux across the plasma membrane, and not CICR. A few other conclusions can be made based on the Mn2+ study. First, quenching of extracellular fura-2 by Mn2+ in cell suspensions is essentially complete within one sec. Fura-2 which is quenched after one sec is most likely intracellular. Thus, the difference in intensity just prior to and immediately following Mn2+ addition yields a correction factor for extracellular fura-2 without significant quenching of intracellular fura-2. Second, because Mn2+ gradually enters the cytoplasm and quenches intracellular fura-2 fluorescence at 340 and 380 nm excitation unequally (Goldman and Blaustein, 1990), extended incubation with an+ precludes estimation of [Caz*]i. Since the initial elevation in 340/380 nm ratio was independent of [Ca2*],, experiments were undertaken to determine the intracellular source of this elevation. Previous studies at the neuromuscular junction have shown that some uncouplers of oxidative phosphorylation produce effects similar to those of MeHg (Levesque and Atchison, 1987; 1988). Mel-lg has a number of effects on mitochondrial function (Verity 47 et al., 1975; Cheung and Verity, 1981; Kauppenin et. al., 1989; Hare and Atchison, 1992), all of which have been shown to elevate [Ca2+], (Scott et al., 1980; Heinonen et al., 1984). Therefore, we sought to determine if the mitochondria were responsible for the immediate elevation in 340/380 nm ratio. In 0.1 pM Ca2*-I-IBS, depolarization of the mitochondrial membrane with 10 mM sodium azide plus 4 pg/ml oligomycin did not change the 340/380 nm ratio (Fig. 2.5). Subsequent addition of MeHg (0 - 100 pM) caused concentration-dependent increases in 340/380 nm ratio similar to those in the absence of mitochondrial inhibitors. Depolarization of synaptosomal mitochondria in 200 pM Ca2+— HBS also failed to elevate the 340/380 nm ratio, and did not alter the response to MeHg (not shown). Thus, depolarization of mitochondria per se was not sufficient to elevate [Ca2+]i in synaptosomes. Moreover, the intracellular actions of MeHg on fura-2 fluorescence did not involve alterations in mitochondrial Ca2+ buffering. Ruthenium red, a putative inhibitor of the mitochondrial Ca2+ uniporter (Moore, 1971), attenuates the spontaneous release of neurotransmitter induced by MeHg (Levesque and Atchison, 1987; Levesque et al., 1992). Ruthenium red also blocks MeHg-induced efflux of Ca” from preloaded isolated mitochondria (Levesque and Atchison, 1991). Since depolarization of the synaptosomal mitochondrial membrane did not alter the 340/380 nm ratio, experiments were performed to determine if the protective effects of ruthenium red on transmitter release or Ca2+ release were due to an action at another site. Ruthenium red shifted the 340/380 nm ratio to higher values due to preferential quenching of the 380 nm excitation intensity over 340 nm intensity (not shown). In 0.1 pM Ca“- HBS, pretreatment of synaptosomes with 10 or 20 pM ruthenium red for at least 30 min did not block the initial phase elevation in 340/380 nm ratio caused by MeHg (Fig. 2.6). 48 .2”: .modv3 32:8 50:: 325:: biocmcfim 885:5 «522.3, .A.2.m.m H :35: _ 25mg: 5 contomov m: 852522: 83 onZ :3 womsao 5:: E: owflovm 5 :5::>o_o 25:25:: 2: ”:35 2:25:35 8:5 :0 5:2: 2: 2:823: 85: zoom .21 _.0 32255855: 5 gang :o35 5 :26: £3 :2: ON: <:.Dm .mZoE :6 :5555: 5 :52: mez 25 O: :5: 523556 ESE v w5m= :5 S @3563: 2:3 28:52:: 35:05:25: 5:580:35? 2:. .mm:-+~mu 2: :d 5 5:8 5: oonvxm 5 25853 33:55:25 2: :0 5:35:63: 5.655055 :0 Soto 2:. .WN .UE 49 cow 0 m._‘ '9. m 0!}BJ WU 088/078 50 55:03.: 0:: :3 5:055:05 0:: Am": .modvav 85:00 50:: 50:55: 35535:; m05:> 08F: .A.2.m.m H ::05V _ 0.59:: 5 B5558: m: 58: 5: ng§ 5 :5::>0_0 0::6055: 0F: ”:03: .m::05::0ax0 00:5 00 ::05 0:: 50.650: 00:5 50$ .21 :d 5 44:8 :035 o: 3:02 o: :58 @050: 5:3 <._.Om .55 cm 5:5 :: :8 :0: mmmwflflv 21 com 5 :0: 55:05:: .2: 0m :53 8:355 0:03 5553?: 5585953 .mm:u+~wU 21 ~.O cm Oman: E: me\OVm E mCOCm>20 Uoosficmuwmoz 05 :0 U0: Eswcofite “—0 “00:0 OFF .©.N .mv—h 51 2:5 2.: m w v N o mI0E o R o F WI.» 9;... 4‘24V . ) m N t: 4 931443.. .5... om ..,9zé§.§§ £52 - oo: .3), . 4.. E... 3:2: 00— on an OF 0 w o u .3 a L a . «.o m. . m .. v... m. .0: .9”. '0. N D.” ‘9. v oneJ wu 088/078 52 In 200 pM Ca2*-HBS, ruthenium red pretreatment did not affect the second phase elevations in 340/380 nm ratio, but did block the elevation in ratio elicited by K“ depolarization (not shown). It is possible that the shift in ratio caused by ruthenium red makes direct quantitative comparisons of 340/380 nm ratio values inappropriate. Since neither the depolarization of mitochondria nor ruthenium red affected the MeHg-induced immediate elevations in 340/380 nm ratio, other potential intracellular sources for these increases were examined. The ER is thought to buffer Ca2+ in synaptosomes (McGraw et al., 1980; Rasgado-Flores and Blaustein, 1987). BR Ca“ can be mobilized by agents such as caffeine, 1,4,5-inositol—tris-phosphate, thapsigargin, or Ca2+ (Miller, 1988). In 0.1 pM Ca2+-HBS, caffeine (at concentrations up to 40 mM) was ineffective at increasing [Ca2+],, or preventing the MeHg-induced elevations of 340/380 nm ratio (Fig 2.7). Likewise, thapsigargin, an inhibitor of the ER Ca2*-ATPase, did not elevate [Ca2+], or block the effects of MeHg in 0.1 pM Ca2*-HBS (Fig 2.8). Apparently, the initial phase increase in 340/380 nm ratio was n0t due to Ca2+ release from the ER. In experiments originally designed to evaluate the temporal aspects of MeHg- induced increases in plasma membrane permeability, 40 pM an” was added prior to MeHg. We anticipated that the higher concentrations of MeHg (50 and 100 uM) might cause an enhanced rate of fura-2 signal quenching due to increased disruption of the plasma membrane. These experiments were conducted at the Ca2*-insensitive excitation wavelength of 360 nm (Grynkiewicz et al., 1985). Mn2+ caused in a biphasic decrease in fluorescence intensity (Fig. 2.9). Addition of MeHg two min after Mn2+ caused an immediate elevation in 360 nm fluorescence intensity which was maximal at 25 pM 53 5:520:06 05 E 50::555 0:: Am”: .36an 65:00 50:: ::0:0:::0 5:05:55: m05:> 0:05. .A.2.m.m M 5:05: _ 0.53.: 5 50:55:05 m: 05:: 5: owflcvm 5 55:53 555055: 05:. 50,3: 5:055:098 00:5 :0 ::05 05 50,0050: 00:5 :05: .55 03: :0: mmz-+~:0 21 :d 5 050:::0 25 0: 5:3 55:05 0:03 55509.5: 550,855.22: .8158 2.: S 5 2:2 E: 55:: 5 28:50 8035-502 2:: co 85:8 :0 some 2:. .2 .0: 54 00w AinymuzoSH 00.. on nN or one: u! aBueua m._. m. «a only wu oee/ove 55 50.2020 0:: :3 00:00:05 0:: Am”: .mOdVE 3:500 50: 50:0th 3:00:55; 002? 085. .32.m.m H 50:5 — 0::wE 5 0058:0000 m: 055: E: owQOVm 5 :0::>0_0 0520055 05. ”:35 0505:0008 00:5 :0 5:05 0;: 50850: 000:: :05 .55 00:5 :0: $5500 2: :.o 5 :mwswamwfi 21 _ :53 0050:: 0:03 5550000: :mEomoaaim .mm:-+~:u 21 :d 5 0:0: 5: oonYm 5 5555—0 00000500202 05 :0 5w5w5805 :0 000000 05. .w.~ .OE 56 or MN cm 00' A55: 05F :23 3:25 00 F on mu 0 F one: u! afiueqo O '0. F '9. N "2 n 092.! mu 088/078 57 500050080 5 050 0: 00:00.80 0:03 050% 05 :0 .0053 05505 00 :05000 05050005 00:5 00 0w0:0>0 05 8:000:00: 0:: :00m A20. om: <00b :002050 00:05 500: 5005005230 05 00 :05000 :3 :00. 55 00:5 0030200 55 03: :0 00000 003 w:02 .5: 00m 00 5000:0353 :00050x0 0555005550 05 :0 00:05:05 05.505 00:88:05.: 05 0:0 :05000 $002 0: :000 00500000050 00002 ~53 0: 00000 003 A20 000 +N:2 00500000053 00 50000500 +N:2 0:0 50:05: 00:83:05.: 355005-500 5 0:000:50 0000050202 .0.~ .GE 58 EE 2.: cow N O on 00 / or 0 0:: :: 0:: :00m $5505: 00:00:0:05: 5 0:00:05 0:: 00::0>0: 2mm; 20 Om :0 :05000 50000::0m .5: com :0 50005008 w530:0: 5: wow :0 5505 00:00:0:05: 0:: 00:0>0_0 55 5: :0 00:02 20 mm :0 00500< .55 05: :0 mm:-+~0U 20. :d 5 :050:0:00:>: 00000: N05: 0: 00000 :03 A20 02: <00b :0:0:0:0 5:05 500: ::005:005:-:00 0: 0. 00:00:0:05: 0>5::0:5-+~0U 5 ::0:0:0:_0 0000050202 :0 <0:.Q :0:0:0:0 5:05 :50: 5005005260 0:: :0 88:: .:.N .07. 65 :52: 2:: 2mm... 9 0:02 <05 : : : (.7: W31 .F flu 0...: o... m... o.N (Shun 'uqle) Mtsuetm 66 FIG. 2.12. Effect of MeHg (0 - 25 uM) on the 340/380 nm ratio. Each trace is the mean of three experiments. [Ca2+], is indicated on the right. Top, In 200 pM Ca2+-HBS, synaptosomes were exposed to 50 pM TPEN at 1 min followed by MeHg at 2 min. Bottom, In 0.1 uM Ca2+-HBS, synaptosomes were exposed to 50 pM TPEN at 1 min followed by 20 uM EGTA at 2 min. MeHg (0 - 25 uM) was then added at six min. 340/380 nm ratio 6.5 ' 5.0 3.5 4.0 ' 2.5 " 1.0 67 Fig. 2.12. —830 25 J'PEN MeHg —587 W - ., W W O "398 A E C V I—'-'u' + -398 Na 0 H —247 AN OUI 68 DISCUSSION Results of the present study confirm several earlier observations of MeHg-induced elevations in [Ca2+]i, (Komulainen and Bondy, 1987b; Kauppinen et al., 1989), but more importantly, have extended them in the following ways. First, MeHg appears to elevate the concentration of non-Ca2+ endogenous cations within the terminal. Second, ruthenium red did not prevent the intracellularly-mediated changes in fura-2 fluorescence induced by MeHg. Third, unlike mitochondria isolated from whole brain, intrasynaptosomal mitochondria did not appear to release appreciable Ca2+ in response to MeHg. Moreover, under the conditions of our experiments, intrasynaptosomal mitochondria did not release measurable amounts of Ca2+. Fourth, the temporal effects of MeHg on fura-Z fluorescence are complex and involve at least two phases: an immediate effect due to elevation of intra-synaptosomal non-Ca2+ polyvalent cation concentrations, and a gradual effect due to influx of Caz} Fifth, MeHg causes an immediate increase in plasma membrane permeability to divalent cations. This increased permeability mediates the second phase, extracellular Ca2*-dependent increases in fura-2 fluorescence. This paper is the first to report the disruption by MeHg of homeostasis of endogenous polyvalent cations other than Ca2+. Though this was not an original aim of the study, the observation was felt to be of sufficient importance to warrant attention. Though fura-2 is relatively selective for Ca“, several studies have reported interactions of non-Ca2+ polyvalent cations with fura-2 (Smith et al., 1989; Tomsig and Suszkiw, 1990) as well as other fluorescent Ca2+ indicators (Komulainen and Bondy, 1987a; Arslan et al., 1985; Mason and Grinstein, 1990). Whereas MeHg itself did not interact with fura-Z, it nevertheless altered the synaptosomal fura-2 fluorescence in ways inconsistent 69 with elevations in [Ca2+]? The cell-pemreant heavy metal chelator TPEN blocked the MeHg-induced immediate elevations in both 340/380 nm ratio and 360 nm fluorescence intensity. The cell-impermeant heavy metal chelator DTPA had no effect. This strongly suggests that the first phase elevations in 340/380 nm ratio arise from endogenous intrasynaptosomal cations. The cation(s) responsible must be present in synaptosomes in sufficient quantities and possess adequate affinity for fura-2. The endogenous cations which best fit these criteria are Fe“, an+ and Cu“. Aside from Ca2+ and Mg“, these are the most abundant polyvalent cations typically found in neurons. The affinity of fura-2 for Fe2+ and Zn2+ is 3-10x and 100x greater, respectively, than its affinity for Ca2+ (Grynkiewicz et al., 1985). Similarly, Cu2+ has an extremely high affinity for EGTA (Bartfai, 1979), the parent compound of fura-2. It is unlikely that the cation(s) was introduced during the synaptosomal preparation since the concentrations of iron, zinc and copper in our buffers were 229, 158 and 108 nM, respectively, as measured by inductively coupled plasma spectroscopy (unpublished results). The MeHg-induced changes were characterized by increased fluorescence intensity. Therefore, the cation is not likely to be Fe2+ or Cu“ because these ions quench fura-2 fluorescence (Grynkiewicz er al., 1985; unpublished results). However, Zn2+ possess a fura-2 spectrum similar to Ca2+ (Grynkiewicz et al., 1985), and is present in most nerve terminals in high concentrations, particularly in the forebrain (Frederickson er al., 1987). Within the mossy fibers of the hippocampus, Zn2+ is believed to be associated with the synaptic vesicles, and is mobilized upon electrophysiological stimulation possibly due to release of vesicular contents (Assaf and Chung, 1984). Given the effects of MeHg on the plasma and mitochondrial membranes 70 of nerve terminals (Kauppinen er al., 1989; Hare and Atchison, 1992), it is conceivable that MeHg is also capable of disrupting the vesicular membrane causing release of endogenous an” into the synaptosomal cytosol. The cation is not likely to be Hg2+ for several reasons. First, Hg2+ quenches fura-2 fluorescence (Vignes et al., 1993). Second, DTPA would prevent Hg“ effects on fura-2 fluorescence if the Hg2+ was added exogenously as a contaminant of MeHg. Third, demethylation of MeHg does not occur in brain tissue on the timescale of our measurements. A major impetus for the present study was to clarify the role of the mitochondria in MeHg-induced elevations of [Ca2+]i in the nerve terminal. The role of mitochondria in regulation of [Ca2+], is unsettled (Rahamimoff et al., 1975; Scott et al., 1980; Heinonen et al., 1984; Nachshen, 1985; Rasgado-Flores and Blaustein, 1987). There are conflicting reports of the relative importance of intraterminal mitochondria in buffering elevations in [Ca2+],, as well as the treatments required for causing Ca2+ release from mitochondria (Heinonen et al., 1984; Nachshen, 1985). Since Mel-lg releases Ca2+ preloaded into isolated mitochondria (Levesque and Atchison, 1991) we sought to determine whether intrasynaptosomal mitochondria release measurable amounts of Ca2+, and to evaluate the importance of mitochondrial Ca2+ stores with respect to elevations of [Ca2“]i by MeHg. In the absence of MeHg, depolarization of mitochondria caused no detectable changes in the 340/380 nm ratio. Either depolarization of the mitochondria alone is insufficient to elevate [Ca2+],, or in our experimental paradigm the mitochondria do not sequester substantial amounts of Ca2+ at rest. Depolarization of the mitochondria did not affect the response of the synaptosomes to subsequent addition of MeHg. 71 We hypothesized originally that ruthenium red would prevent the elevation in [Ca2+], induced by MeHg in the absence of Caz} Ruthenium red blocks the actions of MeHg at both intact synapses and synaptosomes (Levesque and Atchison, 1987; 1991; Levesque et al., 1992), possibly by preventing Ca2+ efflux from mitochondria by blocking the mitochondrial Ca2+ uniporter (Moore, 1971). Since the mitochondria could not be induced to release Ca2+ it is unlikely that effects of ruthenium red on synaptosomes are related to its actions on the mitochondrial Ca2+ uniporter. Since neurotransmitter release is dependent upon Ca2+, ruthenium red could act at another Ca2I-dependent site. Ruthenium red reduces the intrinsic fluorescence of the smooth muscle plasma membrane Ca2+ pump (Moutin et al., 1992) and blocks voltage-operated Ca” channels in synaptosomes (Goddard and Robinson, 1976; Tapia er al., 1985). MeHg entry into the terminal is not blocked since ruthenium red does not block the MeHg-induced depolarization of intraterminal mitochondria (Levesque er al., 1992). Ruthenium red did not alter either phase of MeHg-induced elevations in 340/380 nm ratio. However, because ruthenium red shifted the 340/380 nm ratio to higher values such comparisons between ratio values in the presence and absence of ruthenium red may not be valid. The disruption of non.Ca2+ cation buffering was the only intracellular effect of MeHg observed in the present study. Pretreatment with caffeine or thapsigargin did not elevate the 340/380 nm ratio, nor affect the immediate elevation in 340/380 nm ratio caused by MeHg. Either Ca2+ is not sequestered in the ER of the synaptosomes to. a significant extent, or this pool was insensitive to agents which cause release of Ca” in other systems. The regulation of the endogenous non-Ca2+ cation(s) is independent of 72 intracellular Ca2+ regulation since disruption of putative intracellular Ca” regulatory systems had no effect. Previous studies of the effect of MeHg on [Ca2+], have relied on static measurements -i.e. before and after treatment with MeH g (Komulainen and Bondy, 1987b; Kauppinen er al., 1989). Therefore we undertook a more thorough temporal characterization of the actions of MeHg. Due to this enhanced temporal sensitivity we observed a biphasic elevation in 340/380 nm ratio whose components were distinguishable by their temporal separation and ionic composition. The first phase was immediate and complete within five sec; its magnitude was concentration-dependent and maximal at 25 uM MeHg. This phase was due to elevations in the intracellular concentrations of non- Ca” cation(s). The second phase developed gradually; the rate of elevation of 340/380 nm ratio was concentration-dependent with respect to MeH g as well as Caz} The second phase was not apparent at MeHg concentrations less than 25 pM. TPEN enhanced the second phase possibly due to chelation of the interfering cytoplasmic non-Ca2+ cations. Because both [Ca2+]i and total cell Ca2+ are altered by manipulations of [Ca2+]e (Scott et al., 1980; Nachshen, 1985; Xiang er al., 1990), we could not establish unequivocally that the second phase was due to Ca2+ influx. For this reason, we used an+ to test the hypothesis that the second phase is mediated by an influx of Ca2+, (Merritt et al., 1989; Kass et al., 1990; Clementi er a1. 1992). All concentrations of MeHg increased the rate of Mn2+ quench of intracellular fura-2. Higher concentrations (50, 100 pM) of MeHg also increased Mn“ quench of extracellular fura-2. These observations confirm those of Kauppinen et al. (1989). MeHg increased immediately the 73 plasma membrane permeability to Mn“, suggesting the onset of the second phase was also immediate. The route of entry of Caz"c following MeHg treatment was not identified; possibilities include: direct compromise of the plasma membrane, entry through non- specific cation channels, receptor-operated ion channels or voltage-dependent Ca2+ channels. Since voltage—gated Ca2+ channels in synaptosomes and PC12 cells are blocked by the concentrations of MeHg used in the present study (Atchison et al., 1986; Shafer and Atchison, 1989; 1991; Shafer et al., 1990; Hewett and Atchison, 1992) it is unlikely these channels mediate the second phase elevations in 340/380 nm ratio. A non-specific cation conductance could not be distinguished in the present study, but has been demonstrated for high concentrations of MeHg in isolated dorsal root ganglion cells (Arakawa et al., 1991). Similarly, receptor-activated entry of Ca2+ cannot be excluded. Direct compromise of the membrane seems likely at high MeHg concentrations since fura-2 efflux is observed. However, lysis of synaptosomes is unlikely since concentrations of MeHg up to 50 pM do not cause lactate dehydrogenase release (Cheung and Verity, 1981). In conclusion, in vitro exposure of isolated nerve terminals to MeHg disrupts cation homeostasis. The effects are an intracellular release of an as-yet unidentified endogenous polyvalent cation and a plasma membrane action which increases Ca2+ entry. These effects may cause disturbances in signalling processes within the terminal. For instance, prolonged elevations in [Ca2“]i produce deficits in transmitter release such as decreased synchronous quantal release and increased asynchronous quantal release (Alnaes and Rahamimoff, 1975; Shalton and Wareham, 1979). Such changes are also observed 74 initially upon acute application of MeHg to an intact synapse. The elevations in non-Ca2+ polyvalent cations might represent actions of MeHg upon synaptic vesicles, potentially producing alterations in neurotransmission distinct from those due to elevations in [Ca2+]i. It is possible the gradual reduction in asynchronous quantal release may be due to destruction of vesicles in the nerve temtinal. These cations may only be markers of other intracellular effects, or they could possess additional toxicity. CHAPTER THREE METHYLMERCURY-INDUCED ELEVATIONS IN INTRASYNAPTOSOMAL ZINC CONCENTRATIONS: AN '9F-NMR STUDY 75 [\J l a. ti: 76 ABSTRACT MeHg increases the [Ca2"]i and another endogenous polyvalent cation in both synaptosomes (Denny et al., 1993) and NGIOS-IS cells (Hare et al., 1993). In synaptosomes, the elevation in [Cay]i was strictly dependent upon Ca2*,; similarly, in NG108-15 cells, a component of the elevations in [Ca2+]i was Ca2*,-dependent. The MeHg-induced elevations in endogenous polyvalent cation concentration were independent of Caz”e in synaptosomes and NGlO8-15 cells. The alterations in fura-2 fluorescence suggested the endogenous polyvalent cation may be Zn“. Using 19F-nuclear magnetic resonance (”F-NMR) spectroscopy of rat cortical synaptosomes loaded with the fluorinated chelator l,2-bis(2-amino-S-fluorophenoxy)ethane-N,N,N’,N’-tetraacetic acid (SF-BAPTA) we have determined unambiguously that MeHg increases the free intrasynaptosomal Zn2+ concentration ([Zn“]). In buffer containing 200 pM EGTA to prevent the Gaye-elevations in [Ca2+],, the [ZnB]i was 1.37 i 0.20 nM; following a 40 min exposure to MeHg-free buffer [Znh], was 1.88 i 0.53 nM. Treatment of synaptosomes for 40 min with 125 pM MeHg yielded [Zn2*]-, of 2.69 i 0.55 nM, while 250 uM MeHg significantly elevated [Zn2“]i to 3.99 :l: 0.68 nM. No Zn2+ peak was observed in synaptosomes treated with the cell-permeant heavy metal chelator TPEN (100 pM) following 250 pM MeHg exposure. [Ca2+], in buffer containing 200 pM EGTA was 338 i 26 nM, and 370 :l: 64 nM following an additional 40 min exposure to MeHg-free buffer. [Ca2"]i was 498 :t 28 nM or 492 :t 53 nM during a 40 min exposure to 125 or 250 pM MeHg, respectively. None of the values of [Ca2+]i differed significantly from either pretreatment levels or buffer-treated controls. 77 INTRODUCTION MeHg disrupts Ca2+ homeostasis within isolated central nerve terminals (Komulainen and Bondy, 1987; Kauppinen er al., 1989; Denny et al., 1993) as well as neuronal x glioma hybrid (NGIOS-IS) cells in culture (Hare et al., 1993). Recently, we described effects of MeHg on the fluorescence properties of the Ca2*-selective fluorescent indicator fura-2 which were inconsistent with an elevation in intracellular Ca2+ concentration ([Ca2"],) alone (Denny er al., 1993; Hare et al., 1993). These effects include increases in the fluorescence intensity at the Ca2*-insensitive excitation wavelength of 360 nm (Grynkiewicz er al., 1985), and could not be attributed to a direct interaction of MeHg with fura-2. This elevation in fluorescence intensity was inhibited or reversed by the cell- perrneant heavy metal chelator TPEN, but not by the cell-impermeant chelator DTPA. This suggested that MeHg also elevates the intracellular concentration of an endogenous polyvalent cation other than Ca2+. Unlike the increases in [Ca2+],, which were gradual and dependent upon Ca2+,” the elevations in endogenous heavy metal content were rapid and independent of Ca2+, Although the identity of the cation could not be established, we postulated it may be Zn2+ due to its presence within certain mammalian central nerve terminals (Frederickson et al., 1987) and its characteristic effects on fura-2 fluorescence (Grynkiewicz et al., 1985; Hechtenberg and Beyersmann, 1993). To identify and quantify intracellular polyvalent cation concentrations in synaptosomes exposed to MeHg, we performed l9F-NMR spectroscopy on synaptosomes loaded with the fluorinated chelator SF-BAPTA (Smith et al., 1983; Bachelard and Badar- Goffer, 1993). l9F-NMR spectroscopy of fluorine-labeled chelators has been used in many systems to monitor changes in [Ca2+], (Murphy et al., 1986; Levy et al., 1987; 78 Badar-Goffer et al., 1990; Dowd and Gupta, 1993). A major advantage of 19F-NMR spectroscopy over fluorescence analysis is that the identity of the cations bound to the indicator is apparent. l"F-NMR has been particular useful in the simultaneous measurement of [Ca2+], and intracellular sz” concentration following acute exposure of Pb2+ to isolated osteoblasts, NG108-15 cells or platelets (Schanne et al., 1989a,b; Dowd and Gupta, 1991). We monitored changes in the cytosolic concentrations of endogenous polyvalent cations by 19F--NMR spectroscopy of 5F-BAPT A loaded synaptosomes exposed to MeHg. A peak corresponding to Zn2+ was observed in 5F-BAP’I‘A loaded synaptosomes prior to MeHg exposure which increased following MeHg exposure. Thus, MeHg increased the free ionized intrasynaptosomal an“ concentration ([Zn2*]i). This is the first report of an effect of MeHg on [Zn2"]i. 79 MATERIALS AND METHODS Chemicals and Solutions: SF-BAPT A tetrapotassium salt was obtained from Teflabs (Austin, Tx). 5F—BAPTA acetoxymethyl ester (SF-BAPTA/AM) and TPEN were obtained from Molecular Probes (Eugene, OR). Methylmercuric chloride was obtained from K+K Labs (Plainview, NY). Pyrithione, ionomycin, 6-fluorotryptophan, EGTA, and Hepes were obtained from Sigma (St. Louis, MO). ZnCl2 was obtained from Aldrich (Milwaukee, WI). Deuterium oxide (D20) was obtained from lsotec (Miamisburg, OH). All other chemicals were reagent grade or better. Deionized water (18 M9) was used in all buffers. Sucrose buffers contained 3 mM NazHPO4 (pH 7.4). Hepes-buffered saline (HBS) contained (mM): 145 NaCl, 5 KCl, 1 MgClz, 10 d-glucose, 10 Hepes (pH 7.4 with Trizma base). For 19F-NMR analysis of 5F-BAPTA loaded synaptosomes, 200 pM EGTA was added to the HBS to reduce contaminating Ca2+ concentrations. NMR tubes (8" length, 5 mm dia) were obtained from Wilmad (Buena, NJ). ’9F-NMR Spectroscopy of 5F —BAPTA loaded Synaptosomes: Synaptosomes were prepared by discontinuous sucrose density gradient ultracentrifugation from the forebrains of male Sprague-Dawley rats (Harlan, 175-199 g) as described previously (Denny et al., 1993). The P2,, fraction (Gray and Whittaker, 1962) was resuspended in three volumes of HBS and centrifuged for 10 min at 10,000 x g. The synaptosomal pellet was resuspended by homogenization in 10 ml of HBS containing 31.3 pM 5F-BAPTA/AM and incubated for 30 min at 37°C. The synaptosomal suspensions were diluted by addition of 30 ml HBS, incubated for five min at 37°C, centrifuged at 10,000 x g for five min, 80 and resuspended in 1.5 ml HBS containing 10% (v/v) D20 and 200 pM EGTA. Aliquots (0.6 ml) were transferred to NMR tubes and kept on ice. A Varian VXR-SOO MHz NMR spectrometer equipped with a 5 mm ‘I-I/‘9F-NMR probe tuned to 470.3 MHz was used. Sample temperature was maintained at 30°C and samples were spun at 20 Hz throughout the experiment. The resonance peaks of SF- BAPT A loaded synaptosomes were compared to those of aqueous solutions of SF—BAPT A free acid and saturating concentrations of various metals relative to the reference standard 6-fluorotrypt0phan. The SF-BAPT A loaded synaptosomal samples were locked to internal deuterium. The 19F-NMR parameters used were: 23 usec 45° pulse width, 0.25 sec acquisition period followed by a 0.3 sec delay, and 32 kHz spectral width. Increasing the interpulse interval to 0.9 3 did not alter the spectral characteristics. A filter bandwidth of 7.7 kHz was used to reduce high frequency noise. Spectra consisted of 4,000 transients accumulated over 40 min. Data acquisition was paused after 2,000 transients, and the suspensions were mixed by inversion. Spectra of synaptosomal metal content before and after acute MeHg exposure resulted from a Fourier transformation of the transients using a line broadening factor of 40 Hz. [C321i and [Zn2“]i were calculated by the equation: [MM], = Km x [M"*25F- BAP’I‘A1/[5F-BAPTA1, where [M"*:5F-BAPTA]/[5F—BAPTA] is the ratio of the peak areas of the metal-bound peak to the free SF-BAP'I‘A peak (Smith et al., 1983). Kd values of SF-BAPTA for Ca“ and Zn“ were 500 nM and 7.9 nM, respectively (Schanne et al., 1989b; 1990). Peak areas were determined using Varian NMR software. 81 Protein Assays: Protein content was determined by the method of Lowry et al. (1951) using bovine serum albumin as a standard. Statistical Analysis: Comparisons of [Zn2‘“]i or [Ca2+], before and after MeHg treatment were made using a one-way ANOVA (p .<_ 0.05), and post-hoe analysis using the Bonferroni multiple comparison test. 82 RESULTS AND DISCUSSION SF-BAPT A in aqueous solution produced a single resonance peak approximately 1.0 ppm downfield of the reference standard 6-fluorotryptophan (Fig. 3.1). A single resonance peak at 1.0 ppm was also observed in solutions containing equimolar concentrations of SF-BAPTA and MeHg. The Zn“, Cd2+ and Ca2+ complexes of SF- BAPTA were observed 4.6, 6.0 and 6.8 ppm downfield, respectively (Fig. 3.1). 5F- BAPTA loaded synaptosomes treated with 250 uM Zn2+ and the Zn2+ ionophore pyrithione (50 pM) had a single peak at 4.5 ppm, consistent with an elevation in [Zn2+]i (Fig. 1). Similarly, 250 pM Ca2+ and the Ca2+ ionophore ionomycin (5 uM) produced a single peak at 6.7 ppm (Fig. 3.1). Several resonance peaks were observed in SF-BAPTA loaded synaptosomes in HBS which contained 200 pM EGTA (Fig. 3.2). Broad peaks corresponding to unbound and Cay-bound SF-BAPTA were observed at 1.2 and 6.7 ppm, respectively, as well as a peak of lower magnitude at 4.5 ppm corresponding to Zn“. [Zn2"]i was 1.37 :l: 0.20 nM (mean :t S.E.M., n=23) prior to MeHg exposure. Thus, the cytoplasm of synaptosomes contained detectable amounts of ionized an“ under resting conditions. Treatment of the same 5F-BAPTA loaded synaptosomal suspensions with MeHg-free buffer, 125 pM, or 250 pM MeHg for an additional 40 min yielded [Zn’*]i of 1.88 i 0.53 nM (n=11), 2.69 :1: 0.55 nM (n=4), or 3.99 i: 0.68 nM (n=8), respectively (Fig 3.3A). [Zn2*]i following exposure to 250 uM MeHg differed significantly from both the pretreatment and HBS-treated control value. No Zn“ peak was observed in synaptosomes treated with 100 pM TPEN after exposure to 250 pM MeHg (n=4, not shown). [Ca2+], prior to MeHg exposure was 338 :1: 26 nM (n=23), and 370 :t 64 nM (n=11) in HBS- treated controls (Fig. 3.2). [Cazili in synaptosomes exposed to 125 pM or 250 pM MeHg 83 FIG. 3.1. A: Resonance positions of free SF-BAPTA and various metal chelates in aqueous solution relative to the reference standard 6-fluorotryptophan (1.25 mM). The position of 2.5 mM 5F-BAPTA tetrapotassium salt was determined in metal-free buffer containing 150 mM KCl and 50 mM Hepes free acid (pH 7.1), and in solutions containing saturating concentrations of Zn“, Cd”, or Ca2+. The spectrum of 5F-BAPT A with 2.5 mM MeHg was indistinguishable from that of 5F-BAPTA alone. B: Resonance positions of 5F-BAPTA loaded synaptosomes during a 40 min exposure to 250 pM Zn2+ and 50 uM pyrithione (top), or 250 pM Ca2+ and 5 uM ionomycin (bottom). 84 A Fig. 3.1. Ca Cd Zn 5F-BAPTA 1.1 AJ_l__‘_Jl ll [ I [ I I? l l 9 8 7 6 5 4 3 2101 ppm downfield B Zn2*lpyrithione Ca’Vionomycin ”Wt/"MW“ WWWMM/"M 9 8 7 s 5 4 3 2 1 0 ppm 85 .5385 50550000 55 00 :0::0:0 m5:00 Am“: 55:08 0.302 20. Omm :0 AZ": 0:00.35 :0::0: 00::-wI02 0: 055005: :050:0:00:»: 050: 0:: :0 0:0:00x0 0: :0:0 55 G: :0: 0050000 :03 Ammu: .05: 505000: :055 :< .::0:0:000:0 55890050: 00000: <:.0 14.0). Upon completion of the pH gradient the column was washed with an additional 100 m1 of final buffer, regenerated with 250 ml 50 mM Hepes (pH 8.0) containing 1 M NaCl and equilibrated with at least 300 ml of 50 mM Hepes (pH 8.0). MeHg-induced Zn2+ release from soluble synaptosomal proteins was monitored qualitatively by fura-2 100 spectrofluorometry. Fura-2 pentapotassium salt was added to the sample to a final concentration of 0.5 uM. An initial fura-2 excitation scan was then acquired from the wavelengths of 320 to 400 nm at the emission wavelength of 505 nm. MeHg (50 pM final concentration) was then added to the cuvette and another fura-2 excitation scan was acquired. These two scans were compared for shifts in fura-2 fluorescence consistent with elevations in free Zn2+ concentration. Consecutive fractions which released an” upon MeHg addition were pooled, and frozen for subsequent analysis. Aliquots were withdrawn for protein determination and SDS-PAGE analysis. In some cases the peaks from the High Q column were further purified by gel filtration chromatography. The individual peaks were concentrated to approximately three ml by ultrafiltration as before, and applied to a Bio-Gel P-100 column (40 cm long x 2.5 cm dia) equilibrated with 50 mM Hepes (pH 8.0). A constant flow rate of 0.2 ml/min was maintained and fractions were collected every 10 min. Fractions which released Zn2+ upon addition of MeHg were pooled and saved for further analysis. Protein Assays: Protein content was determined by the method of Lowry et al. (1951) using bovine serum albumin as a standard. SDS-PAGE Analysis of Protein Fractions: Protein purity was evaluated by SDS- PAGE (Laemmli, 1970) using a Bio-Rad Protean II minigel electrophoresis system (Piscataway, NJ) according to manufacturer’s instructions. Protein samples were boiled for two min in sample buffer containing B-mercaptoethanol. Electrophoresis was performed at 200 V using 0.75 mm thick minigels containing 12% polyacrylamide. 101 Silver Staining of SOS-PAGE Minigels: Gels were stained using the ammoniacal silver method of Oakley et al. (1980). 102 RESULTS Because these studies relied heavily on monitoring MeHg-induced changes in fura- 2 fluorescence, the possibility that MeHg may interact directly with fura—2 was tested. MeHg concentrations up to 1 mM did not alter the fluorescence properties of fura-2 when added to a cuvette containing 0.5 uM fura-2 pentapotassium salt (not shown). The effects of Ca2+ on the fluorescence properties of fura-2 have been well defined (Grynkiewicz et al., 1985). The excitation maximum of fura-2 uncomplexed to Ca2+ is approximately 370 nm while that of Ca2*-complexed fura-2 is near 340 nm (Fig. 4.1A). Thus, increases in the fluorescence intensity at the excitation wavelength of 340 nm concurrent with decreases in intensity at 380 nm are consistent with an increase in Ca2+ concentration. An important consequence of this shift in the fluorescence properties of fura-2 caused by Ca2+ is the presence of an excitation wavelength at 360 nm which is insensitive to changes in Ca2+ concentration, and is known as the isosbestic point. The fluorescence intensity at the excitation wavelength of 360 nm remains constant at any Ca2+ concentration. Thus, any change in the fluorescence intensity at 360 nm cannot be attributed to a change in Ca2+ concentration. By monitoring the fluorescence intensity at the excitation wavelength of 360 nm it is possible to study the interaction of fura-2 with cations other than Ca2+. Zn2+ interacts with fura-2 with a higher affinity than Ca2+ (Grynkiewicz et al., 1985), but its effects on the fluorescence properties have not been studied systematically (Hechtenberg and Beyersmann, 1993). The effects of Zn“ on fura-2 fluorescence were studied in EDDA-buffered an“ solutions containing 0.5 uM fura-2 pentapotassium salt (Fig. 4.1B). Increasing the proportion of the Zn2*~containing buffer caused a gradual 103 FIG. 4.1. Fura-2 excitation profiles of Ca2+ or Zn“. Fura-2 (0.5 uM) fluorescence intensity was monitored at the emission wavelength of 505 nm. Points were acquired for one sec at integer excitation wavelength values. The Ca” or Zn2+ concentration was varied by altering the proportions of buffers which contained a metal chelator only and one with equimolar concentrations of chelator and Ca“ or Zn”. A: Fura-2 excitation profile in Cay-containing buffers. Buffers contained 1() mM EGTA and 10 mM EGTA with 10 mM Ca2+. B: Fura-2 excitation profile of Zn“. Buffers contained 10 mM EDDA and 10 mM EDDA with 10 mM Zn“. 104 Fig. 4.1. High Ca2+ \ \ \ , \ Ca2+—free K A Zn2+—free Intensity (arbi . units) 320 360 400 Excitation wavelength (nm) 105 increase in the fura-2 excitation spectrum. Unlike Ca2+ which shifted the excitation maximum of fura-2, an“ increased the fluorescence intensity at all excitation wavelengths. The fluorescence intensity at the excitation wavelengths between 340 and 380 nm was particularly responsive to elevations in Zn2+ concentration. There was no isosbestic point for Zn“ and fura-2. Based upon the profound difference in fura-2 spectra with Zn” or Ca2+, the effects of MeHg on Zn2+ concentration could be discerned from those on Ca“. MeHg increases [Zn2*]i by acting on some endogenous pool of Zn” present in the nerve terminal. Zn2+ is associated with the synaptic vesicles of certain glutamatergic nerve terminals. The possibility that vesicular Zn2+ contributed to the MeHg-induced elevations in [Zn2*]i was tested using fura-2 loaded synaptosomes pretreated with agents which mobilize synaptic vesicles. If a component of the MeHg-induced increases in [Zn2*]i was vesicular in origin, then elevations [Zn2*]i would be reduced by mobilization of synaptic vesicles prior to MeHg exposure. As such, the MeHg-induced elevations in [Zn2"]i were assessed following mobilization of synaptic vesicles by three different treatments. Elevating K“ concentration depolarizes the synaptosomal membrane, resulting in activation of voltage—dependent Ca2+ channels, Ca2+ influx, vesicle fusion and release of vesicular contents. Veratridine is a plant alkaloid which also mobilizes synaptic vesicles by a mechanism involving depolarization of the synaptosomal membrane and subsequent activation of Ca2+ channels. Veratridine prolongs the open time of voltage- dependent Na+ channels, ultimately leading to membrane depolarization due to increased Na+ influx. a-LTX is the primary active toxin of Black Widow spider venom. Although the precise mechanism of action of a-LTX is not clear, it is known to be dependent upon 106 binding to a synaptosomal membrane receptor. Following binding, Ot-LTX induces a massive release of vesicular stores which occurs in the presence or absence of Caz“e (Meldolesi et al., 1984). Changes in the excitation spectrum of fura—2 loaded synaptosomes were monitored following each of the different treatments (Fig. 4.2). A similar experimental paradigm was used for each of agents. After acquiring an initial excitation spectrum of fura-2 loaded synaptosomes, the synaptosomal suspensions were exposed to buffer, 45 mM K+, 6 nM a-LTX or 0.1 mM veratridine for five min. These concentrations have been shown to be sufficient to cause neurotransmitter release from synaptosomes (McMahon and Nicholls, 1993; Meldolesi et al., 1984; Richards et al., 1984). Following the five min interval for vesicle release, a second excitation spectrum was acquired. EGTA (20 pM) was added to the suspensions to reduce [Ca2+]c to approximately 0.1 nM, and a third excitation scan was acquired after a four min incubation period (Denny et al., 1993). A fourth excitation scan was acquired following addition of 25 uM MeHg to the synaptosomal suspension. The fifth scan was the acquired after addition of TPEN (5 uM) to verify the changes in fura-2 fluorescence were due to an increase in the intracellular concentration of an endogenous heavy metal. By analyzing the changes in fura-2 excitation spectra caused by addition of the various agents, it is possible to differentiate between changes in [Ca2”]i and [Zn2”]i. Elevating Ca2+ concentration increases fura-2 fluorescence intensity at 340 nm and decreases intensity at 380 nm, but does not alter the intensity at 360 nm. an” increases fura-2 intensity at all excitation wavelengths, particularly in the region between 340 and 380 nm. K*, a-LTX and veratridine all caused chan es in the fura-2 excitation s ectrum g P 107 0:00:00 5 00:::0::00 00050090 :00: :0 0w0:0>0 0;: m: 0::: :00m 07:05 5 000003 :0 0: 00:05:00 00 0:000 0053 520:5 8:003:02: m-0:0: 5 m0wc0s0 000005-w:02 0::: 8:00:05 00:0 00:80; 00:. 5:000:05 000000053: 5 m0wc0z0 0;: 00m:0>0: A2: 9 2m“: .+~:N 0035000: 0053 00:0:0 00005000 N-0:0: 0::: 5 m0wc0c0 00.0000 A20 mm: 3:02 .0700“: 000000: 0: 0050000000 0;: 0: 00000 0.03 :20. ON: <:.Om .93.: 50:03 05050:? 2:: :d :0 0&0: 80:03 XFAB 2.: 0 .0th as: D: 25 m0 5&0: «:0:: mmm “55 02: :0: 0::0w0 :00: w530:0: 0::: :0 0:0 0: 000090 :05 0:03 0008000000 005000600? 00030:: .5:: wow :0 ::w:0_0>03 00:00:80 0::: :0 005050 003 00580300.? 00000. ~-0:0: :0 000000 00005008 N-0:0: :055 0< 00:53:, 0000000 05:50:: 0023 300000 53 :005:00:: w530:0: 005000305? 00000: ~-0:0: :0 0050:: 00005000 0::: :0 $02 :0 m:00::m to...» .0: 108 .55 593.23: 5:03.806 09' 000 can 09! .«i 6E °.—. 'ttq-III) Kitsuuur (mun 109 consistent with an increase in [Ca2*]i (Fig. 4.2). Similarly, the effect of EGTA was consistent with a decrease in [Ca2+],. MeHg increased the fluorescence intensity in the region between 340 and 380 nm relative to the EGTA spectra. The MeHg-induced alterations in fura-2 fluorescence which are consistent with an increase in [Zn2“]i are indicated by the cross hatched areas. It is important to note, however, that these were not totally identical to an increase in [Zn2”]i alone. The increases observed in buffer-treated synaptosomal suspensions were of similar magnitude to those observed in synaptosomes treated with K*, a-LTX or veratridine. Thus, mobilization of synaptic vesicles did not reduce the MeHg-induced elevation in [Zn2"],. The effects of MeHg were reversed by the cell-permeant chelator TPEN. The soluble fraction of brain extracts contains several Zn2+ binding proteins (Itoh et al., 1983). It is possible the source of the MeHg-induced elevations in [Zn2"]i is one or more of these proteins. If this is the case, it may be possible to purify these proteins using protein chromatography. Fura-2 was used to monitor release of endogenous Zn“ from different protein fractions following MeHg exposure. Unfortunately, as a consequence, it is not possible to include other chelators such as ethylenediaminetetraacetic acid or EGTA in the buffers since they would effectively out-compete fura-2 for any released Zn“. Similarly, the use of Tris as a buffer was avoided because it also chelates Zn“. Synaptosomes were homogenized in 50 mM Hepes (pH 8.0), and centrifuged to separate the particulate and soluble fractions. The particulate fraction was resuspended in an equal volume of buffer. Fura-Z pentapotassium salt (0.5 pM) was added to 1.5 ml aliquots of each fraction, and an excitation spectrum was acquired. MeHg (50 pM) was 110 then added to the aliquots and another spectrum acquired. The two spectra were compared for changes consistent with a MeHg-induced elevation in Zn2+ concentration. An increase was observed in the soluble fraction (Fig. 4.3), but no activity was found in the particulate fraction (not shown). The soluble fraction was concentrated, applied to a High Q anion exchange column and proteins were eluted using a pH gradient. Three distinct peaks which release Zn2+ upon exposure to 50 uM MeHg were resolved by this procedure (Fig. 4.4). The first peak eluted in the second half of the flow through, the second peak eluted between pH 8.6 to 9.4, and the third peak eluted at pH 12.0 as indicated by the cross-hatched areas in the chromatograph. The protein components of these three peaks were analyzed by SDS-PAGE under reducing conditions using 12% polyacrylamide minigels. The gels did not stain adequately using the Coomassie blue method; therefore the gel were silver stained (Fig. 4.5). The protein content of Peak 1 was very complex indicating that little purification of this peak was achieved. Peak 2 and Peak 3 each contained a few major protein bands and several minor bands, consistent with significant purification of proteins from the crude soluble fraction. Because Peak 2 was more pure than Peak 1 and possessed greater total activity than Peak 3, it has been subjected to additional purifications steps. Peak 2 was concentrated to approximately five m1, and subjected to cation exchange chromatography using an Econo—prep High CM cartridge. A single active peak eluted in the flow through, no additional activity was observed following washing of the column with 0.1 or 1.0 M NaCl or IM (not shown). Thus, the protein(s) did not appear to have any affinity for the cation exchange support. The active peak obtained from the cation exchange column was 1 1 1 00000000000 +~=N 5 00000—0 00 :53 80:52:00 0: 000000000: N-0:0: 5 50.: 0000050202 0::. .05. 000000: 0050000 003 :00.» 00005008 05000 0:0 .00000 003 A20 Om: $02 .85. 0:00: 50 000 0:0 0mm 0003:00 0050000 0.03 50::0000 00005000 :0 0:0 .8.w :0: 0000:: 2:: Gm 053 03:50:50 00:000 0508:: 05000300.? 0:00:00 :0 0350:. :5 0.: 0 0: 00000 003 :20. We: :_0m 55000808000 N05,: 050:0:0 0508:0000: 0:00:00 :53 000300 5 m-0:0: :0 00:0:0 0000:5000 0::: :0 0.302 :0 :00::m .mé .UE 112 2E5 50:20:63 0030086 000 can own ova ON» 1 . : . _ . : . - o .5::— 'ttqm) Misuatur (snun 113 0.00.000 0:50: :0: 00>0m 000 0050500 0:03 +~cN 0000.0: 5053 00.000 .000.>.0:: 0000.0: +~0N 0:00005 3050:2000 0: 0000 0:03 005000500: N-0:0: 5 0:000:05:~ A005 00:30:: 3:02 :0 005000 0000 acN 00000.0: :053 00.000 00:5 05 00000:: 0.00.0.0 0;: 00:0:0000. 5050005050 0wc0c0x0 =0.:< ..:O0Z 2 . .0055000 00 5.3 :0::00 .055 :0 .5 0mm :0 00:50:00 :0::00 .05: 00: 0:0 .80 $0: 0000:: 25 on :0 .5 0mm 0050::00 :0::00 .055 05. A005 But .05. 005000: 050.00 0:: 50:: 0505:: 0:50 0: 0000 003 50.050~ :0 0 05.0000 0: 00520: 3:0. :0::0. A050 :0. .05. 0:00: 50500 50:50 :0 0005 :0 00 500005500 083005 003003: 5: owm :0 000000000 0::... 55:5 03: :0 0:0: 30.: 0 w5m0 55 005: 0:00 0050.0> .5 50 5 00:00:00 0:03 50000:”: 550.00 059.00 0050 0-500.: 080.8002 0 0:00 00000. 003 0505:: 0500050000 0.00.00 :0 25 8-2: 0.0500 < 0508:: 0500050000 0.00.00 :0 000500005050 05505 005< 0.0 .60: FR; 4L4. 114 14 12 8 — q q d ‘ 10: . l1 j W -..-- D...- P".'.. d" Peak 3 H l l 0 "O'OIOIOZOlO'O'O'O'O'O'O'O'O'O'O'O'O'O' ‘ O O O o ‘ . . Ogozo...0.0.0’0’0’0’0’0’ A 0 A 0 III-I-I.‘ d 4CHD 3C”) 0'0'0'0'0'0'0'00 °~’¢°o?e?e90?.?o9&0’o’o’e’o’o’e’e’o’e’4 2CHD 1130 we'ovo'c0'0'0'9'9'0'0'0‘0'01 ' . o..30‘9:0}.oto’o’o’e’o’e’o’e’e’o’o’o: - o o o o'9"0'0’0'o’o°o’o’o’o‘o°e°e’o’e’0’o’o°o’e°o’e’o°o’o’o°o‘ 0o.o.o.0.0.0.o...e.o.o.e.o.o.o.e.e.. .o.e.e.o.e.0.o.o.o.o.o.o.o.o.e.o.4 ‘..............' .‘..............I ‘ Peak 1 O O m“ o“eoueqmsqv (L5 Volume (ml) 115 FIG. 4.5. SDS-PAGE analysis of anion exchange chromatography peaks. Aliquots of the various fractions were analyzed by SDS-PAGE using 12% acrylamide minigels under reducing conditions. Lanes contained molecular weight standards (Marker), initial soluble fraction (1), Peak 1 (2), Peak 2 (3) or Peak 3 (4) obtained by anion exchange chromatography. Molecular weights (kDa) of the standards are indicated on the left. 116 Fig. 4.5. Marker 1 2 3 4 66.4 55.6 42.7 117 then applied to a Bio-Gel P-IOO column for further purification by gel filtration. MeHg- induced an” release was observed in a single region consistent with a minor protein species in the molecular weight range of 50 to 55 kDa, as indicated by the chromatograph (Fig. 4.6). SDS-PAGE analysis of the active fraction revealed it contained several minor protein components of a similar molecular weight (Fig. 4.7). Thus, the source of the MeHg-induced elevations in Zn2+ in Peak 2 has not yet been purified to homogeneity. The complete purification of this fraction should allow for identification by N-terminal sequence analysis and/or Western blotting. Future efforts will be directed toward purifying Peak 2 to homogeneity, and obtaining amino acid sequence information. Peak 1 consists of too many protein components to be purified effectively by gel filtration. It may be possible to use cation exchange chromatography to resolve this fraction into several protein components. Peak 3 is much less complex, however the total activity is very low. The protein may be denatured by the extremely high pH required for elution. Isolation of Peak 3 may require pooling the samples obtained from several isolations. 118 36. 0.0. 00. 0:50.00 00003089003. 00 000.: N 303 00.00 0x0000m0 0.000308%003. 0000 N €00 0000005000 .0 00080000505. 00.00 :0. 000 000.0000 3. m0. 3.30000 03000085003. 00.0w 90-00. 0.00. 00000000 €000 00:00:00 .0 No 3. 5.5000 0<0Q .00 3.0 0.0.0.: 0 :02 00:0 0.. 9N0 3.30.0. >000 200 300.8000 0000000007. 80:0 :00. 1.0.: 0000:. Z0Im-.000000 N00+ 00.00.00 200 000.0000 00.0.: 0:00.000. .0 300$ 20000000000. > 0.0.00 00500 0.. 00..<.Q 200 0000200 90.0000 0000:. 0.0.0 000.0 200. 000.00 000 m0<00 m2 002.000 000.000. Fig. 4.6. 119 O “m 082aoueqmosqv 0.02 ' 0 soo 4oo soo Volume (ml) 200 100 120 FIG. 4.7. SDS-PAGE analysis of active peak obtained by gel filtration chromatography of anion exchange Peak 2. Fractions were analyzed using 12% acrylamide minigels under reducing conditions. Lanes contained molecular weight standards (Marker), the initial soluble synaptosomal protein fraction (1), Peak 2 from anion exchange chromatography (2), active peak from subsequent cation exchange (3) and gel filtration chromatography (4). Two bands are present in Lane 3 between 50 and 55 kDa. 121 Fig. 4.7. Marker 1 2 3 4 122 DISCUSSION MeHg elevates [Zn2"Ji by an action at an endogenous source. The goal of this study was to determine if stores known to contain Zn2+ contributed to the MeHg-induced elevations in [Zn2*]i. The possible role of Zn” associated with the synaptic vesicles as well as synaptosomal proteins was investigated. Release of Zn“ from synaptic vesicles and particulate synaptosomal proteins does not contribute to the MeHg-induced elevations in [Zn2*]i; the source of the Zn2+ appears to be several soluble synaptosomal proteins. The possibility that the synaptic vesicles contributed to MeHg-induced elevations in [Zn""‘]i was investigated pharmacologically using fura-2 loaded synaptosomes treated with agents known to mobilize synaptic vesicles. If the Zn2+ was mobilized along with the vesicles, then pretreatment with the various pharmacological agents would reduce the MeHg-induced [Zn2*]i. Exposure of synaptosomal suspensions to 45 mM K*, 6 nM 0t- LTX or 0.1 mM veratridine for five min had no effect on the MeHg-induced elevations in [Zn2+]i. The results of the study involving mobilization of synaptic vesicles are consistent with vesicular Zn2+ not contributing to MeHg-induced elevations in [Zn2*]i. An alternative explanation is that vesicular Zn2+ contributes to the elevations, but the Zn" is not mobilized along with vesicular contents. Although this contingency was not tested directly, it is inconsistent with the work of others. Direct stimulation of Zn2*-containing hippocampal pathways in slice preparations increased the Zn2+ concentration in the bathing solution (Assaf and Chung, 1984; Howell et al., 1984). Furthermore, prolonged kainic acid-induced seizures reduce the amount of histochemically-stainable Zn“ associated with hippocampal mossy fiber boutons (Frederickson et al., 1988). These 123 studies provide evidence that vesicular Zn2+ is mobilized with transmitter following stimulation. The possibility that vesicular Zn2+ is released then resequestered rapidly is not consistent with the observation that cutting the fibers of a Zn2*-containing system causes a loss of histochemically stainable Zn2+ in the terminals within 12-24 hr (Hang et al., 1971), suggesting Zn2+ is taken in the soma and then transported to the terminal along the fibers. Another possibility is that Zn2+ is mobilized with vesicular contents, but the different agents did not mobilize sufficient quantities of synaptic vesicles to affect the MeHg-induced elevations in [Zn2*],. While transmitter release was not measured, all of these agents are capable of eliciting transmitter release (McMahon and Nicholls, 1993; Meldolesi et al., 1984; Richards et al., 1984). Significant mobilization of synaptic vesicles would be expected for the Ot-LTX treated synaptosomes since this toxin is active in the presence or absence of Caz”c (Meldolesi et al., 1984), and approximately ten min elapsed between the addition of a-LTX and MeHg. During this interval, a-LTX mobilized almost 50% of the radiolabeled norepinephrine loaded into guinea pig synaptosomes, with similar results for dopamine release from PC12 cells (Meldolesi et al., 1984). Based on this information, the elevations would be attenuated in a-LTX treated synaptosomes if the Zn2+ associated with synaptic vesicles contributed to the increase in [Zn2*],. The most plausible conclusion is that the Zn2+ associated with the synaptic vesicles does not contribute appreciably to the MeHg~induced elevations in [2112+]? Because vesicular Zn2+ did not appear to contribute to the MeHg-induced in [Zn2*]i, the next potential source investigated was the synaptosomal proteins. 124 Fractionation of homogenized synaptosomes into particulate and soluble components revealed that only the soluble fraction contained measurable MeHg-induced Zn2+ release. The protein components of the soluble fraction were resolved into three well defined peaks of MeHg-induced an“ release by anion exchange chromatography after elution by increasing pH. Peak 1 eluted in the flow-through and, therefore, is likely to be a basic protein. Peak 2 eluted between pH 8.6 and 9.4, and Peak 3 eluted near pH 12.0. Thus, it appears more than one protein is responsible for the MeHg-induced elevations in [Zn2+]i. Peak 2 was selected for further analysis by gel filtration chromatography because of its total activity and relative purity following anion exchange chromatography. Gel filtration chromatography of Peak 2 resolved the MeHg-induced Zn2+ release activity to a single region between 50 to 55 kD, which contained at least two proteins based on SDS-PAGE. Therefore, Peak 2 is likely to be composed of one or more proteins between the 50 and 55 kD. An additional purification step may resolve these two protein components. Once Peak 2 has been purified to homogeneity, it can be subjected to N- terminal sequence analysis. The sequence obtained can be used to search protein sequence data banks of N-terminal regions of known proteins. In addition to providing a means of identification of the protein present in Peak 2, this procedure may also assist in the identification of the other peaks as well. Because MeHg releases endogenous Zn2+ from proteins in all three peaks, it is possible the Zn2*-binding motifs of these proteins are similar. As such, it may be possible to use sequence information obtained from Peak 2 to search for similar Zn2*-binding structures in other proteins found in the nerve 125 terminal. This information would guide the purification of the remaining peaks, although this is entirely dependent upon the identification of the protein present in Peak 2. In the event that Peak 2 cannot be identified by sequence analysis, it may still be possible to isolate the remaining peaks using protein chromatography. Peak 1 does not bind to the anion exchange column, and thus is likely to be a basic protein. Cation exchange or hydroxyapatite chromatography may be suitable to purify Peak 1 further. Peak 3 may be more difficult to isolate due to the low total activity and high pH of the sample. It may be necessary to combine the samples from several isolations before proceeding to the next step. An area in need of refinement is the method of detection of MeHg-induced Zn” release. Currently, qualitative changes in fura-2 excitation spectra are used to measure activity. This is both a sensitive and rapid procedure to determine activity, but it lacks the potential for quantification due to the subtle changes in spectra observed in some samples. The use of SF-BAPTA and 19F-NMR spectroscopy (Denny et al., 1994), instead of fura-2 spectrofluorometry, would provide more quantitative information. At this time, attempts to quantify Zn2+ release using l9F-NMR spectrosc0py have been unsuccessful due to the low sensitivity of the procedure, but this can be overcome by increasing the protein concentration of the samples. Determination of the endogenous Zn2+ sources would be useful in the study of MeHg-induced neurotoxicity for several reasons. The loss of Zn“ from these protein may disrupt normal protein function, leading to toxic responses. This could be tested directly by measuring protein activity before and after MeHg exposure. Because synaptosomes are a heterogenous population, it is not known if these proteins, and the associated MeHg- 126 induced elevations in [Zn2*]i, are found in all synaptosomes, or only a sensitive suprpulation. This question could be addressed by immunocytochemistry of brain slices using antibodies directed against the different target proteins. Because an” is toxic to neurons, it is also possible a component of MeHg neurotoxicity is due to the elevation in [Zn2”]i. The nature and consequences of the elevation in [Zn2”], are the topics of future research efforts. This study addresses the endogenous sources of the MeHg-induced elevations in [Zn2*]i. The experiments focused on the two most logical potential sources of the Zn”: synaptic vesicle and soluble proteins. The Zn2+ associated with synaptic vesicles did not 2+1i. Prominent changes in fura—2 fluorescence contribute to the elevations in [Zn consistent with an increase in Zn2+ concentration were observed following exposure of soluble synaptosomal proteins to MeHg. Anion exchange chromatography of soluble synaptosomal proteins separated this fraction into three distinct peaks. Subsequent analysis of Peak 2 by gel filtration chromatography revealed the protein(s) was between 50 and 55 kDa. Purification of this fraction to homogeneity will require additional chromatographic techniques. Hopefully, the identification of this protein will assist in the purification of the remaining peaks. Once these proteins are identified, the information could be used to study possible role of elevations in [Zn2*]i in the mechanisms and specificity of MeHg neurotoxicity. CHAPTER FIVE CONCLUSIONS 127 128 A. Summary of research. Previous investigations using fura-2 loaded synaptosomes lacked sufficient temporal resolution to distinguish clearly the onset of the intracellular and extracellular components of MeHg-induced elevations in [Ca2+]i (Komulainen and Bondy, 1987b; Kauppinen et al., 1989). Additional studies using fura-2 loaded synaptosomes were designed to resolve the contributions of mitochondrial and Ca2+,3 to the MeHg-induced elevations in [Ca2+]i at the nerve terminal (Denny et al., 1993). MeHg caused a biphasic elevation in ratio of fluorescence intensity at the Cay-sensitive excitation wavelengths 340 and 380 nm (340/380 nm ratio). The initial phase occurred immediately, was independent of Caz“e and maximal at a MeHg concentration of 25 pM. The second phase was strictly dependent upon Ca2+e, was concentration-dependent with respect to both MeHg and Ca2+” and continued gradually over time. Using an+ as a surrogate for Ca“, it was determined that the second phase elevations in 340/380 nm ratio were due to increased synaptosomal plasma membrane permeability leading to influx of Caz} Disruption of mitochondrial Ca2+ regulation was examined as a possible cause of the immediate elevations in 340/380 nm ratio. In the absence of Caz’}, depolarization of synaptosomal mitochondria using NaN3 and oligomycin did not affect 340/380 nm ratio. Thus, either the mitochondria did not store measurable amounts of Ca2+ under resting conditions, or mitochondrial depolarization alone was insufficient to cause mitochondrial Ca2+ release. Predepolarization of the mitochondria did not alter the response of synaptosomes to subsequent addition of MeHg, suggesting that the immediate elevation in 340/380 nm ratio was not mediated by depolarization of the mitochondrial membrane. Ruthenium red was also ineffective at blocking the immediate elevations in 340/380 nm 129 ratio caused by MeHg. This suggested that these elevations were not due to MeHg- induced disruption of mitochondrial Ca2+ regulation. The ER was also explored as a possible source of the immediate elevations in 340/380 nm ratio. Neither thapsigargin, which inhibits the Ca2*-ATPase of the ER, nor caffeine, which facilitates CICR, increased the 340/380 nm ratio, and both were similarly ineffective at altering the response to subsequent addition of MeHg. Therefore, the initial elevation in 340/380 nm ratio could not be ascribed to release of Ca2+ from established intracellular stores. Experiments conducted at the Ca2*-insensitive excitation wavelength of 360 nm, designed originally to monitor the timecourse of the MeHg-induced increase in Mn2+ permeability, revealed a pronounced increase in fluorescence intensity immediately upon addition of MeHg. This elevation in intensity was not due to an increase in [Ca2+]., but rather an elevation in the intracellular concentration of some other cation which interacts with fura-2. The identification of this endogenous polyvalent cation as well as its intracellular source has been the topic of additional investigations (see below). Based on investigations using fura-2 loaded synaptosomes it was apparent that MeHg increased the plasma membrane permeability to divalent cations. These experiments also suggested that MeHg does not release Ca2+ which was previously sequestered into the mitochondria or ER. However, it is possible that the Ca2+ sequestering capacity of these organelles may have been compromised by the preparation of the synaptosomes. As such, potential contributions of Ca2+ from these sources were also been evaluated in individual intact cells. Changes in fluorescence were monitored in single fura-2 loaded neuroblastoma x glioma hybrid cells (NGlO8-15) using digital 130 imaging microscopy (Hare er al., 1993). Digital imaging microscopy also generates useful spatial information regarding the source of the elevations in [Ca2*]i. In the presence of Ca2*e, acute exposure to MeHg caused multiphasic alterations in fura-2 fluorescence. MeHg increased initially the ratio of fluorescence intensity at the excitation wavelengths of 340 and 380 nm. This was followed closely by a plateau phase, during which the ratio remained essentially constant, or declined slightly. A final phase consisting of a precipitous increase in 340/380 nm ratio occurred shortly thereafter. Increasing the concentration of MeHg from 2 to 5 uM did not alter the magnitude of these phases, but did hasten their onset. Closer inspection of the fluorescence intensity profiles revealed effects of MeHg on fura-2 fluorescence which were inconsistent with an elevation in [Ca2+], alone. The initial elevation in ratio consisted of an increase in fluorescence intensity at the excitation wavelength of 340 nm and a simultaneous decrease at 380 nm, consistent with an elevation in [Ca2+]. Following this initial elevation in [Ca2+]i (”first Ca2+ phase"), the fluorescence intensity at both 340 and 380 nm rose steadily in tandem, inconsistent with an increase in [Ca2+]i alone. The ionic basis for this effect of MeHg on fura-2 fluorescence cannot be attributed to an elevation in [Ca2+]i ("non-Caz“ phase"). This phase was followed by a drastic divergence of the fluorescence intensity at the excitation wavelengths of 340 and 380 nm, again consistent with an increase in [Cay]i ("second Ca2+ phase"). Reducing [C33]i to 0.1 pM using EGTA did not affect the first Ca2+ phase or the non-Ca2+ phase, but ablated the second Ca2+ phase. Thus, the first Ca2+ phase and the non—Ca2+ phase were independent of Ca2+” while the second Ca2+ phase was strictly dependent upon Caz} MeHg causes a similar multiphasic response in fura-2 loaded 131 cerebellar granule cells in culture, however these cells are approximately ten times more sensitive to MeHg than NGlOS—IS cells (Marty et al., 1995). This may account for the heightened sensitivity of this particular cell type following MeHg exposure (T akeuchi et aL,1962) The origin of the Ca2+ responsible for the first Ca2+ phase in NGlO8-15 cells has been examined in greater detail. The plasma membrane permeability to exogenously- administered Mn2+ was not increased during this phase, consistent with mobilization of Ca2+ from an intracellular source (Hare and Atchison, 1995). Predepolarization of the mitochondrial membrane did not elevate [Ca2+]i or alter the response to subsequent addition of MeHg either in the presence or absence of Caz”.3 (Hare et al., 1993). Apparently, depolarization of the mitochondria per se is not sufficient to elevate [Ca2*]i in these cells, and the elevations in [Ca2+], observed in the absence of Ca2+, can not be attributed to release of mitochondrial Ca2+ stores resultant from depolarization of the mitochondrial membrane. Because the mitochondria did not appear to contribute substantially to the first Ca“ phase, other intracellular Ca2+ stores were investigated. A prominent intracellular Ca2+ pool found in NGlO8-15 cells is located in the ER. The contents of this pool can be released by agents such as the nonapeptide bradykinin, which activates PLC to generate IP3 (Ogura et al., 1990). Mobilization of IP3-sensitive ER Ca2+ stores with bradykinin causes a rapid elevation in [Ca2+],. Upon depletion of Ca2+ stores, the ER is promptly refilled by a thapsigargin-sensitive Ca2*-ATPase located on the ER membrane (Lo et al., 1993). Pre-depletion of IP3-sensitive Ca2+ stores with bradykinin and thapsigargin in NGlOS-IS cells greatly attenuated the magnitude of the first Ca2+ phase upon subsequent exposure to MeHg, both in the presence and absence of 132 Caz“e (Hare and Atchison, 1995). Bradykinin did not elevate [Ca2+], in cells treated previously with MeHg, suggesting the first Ca2+ phase is mediated, at least in part, by mobilization of Ca2+ from an IP3-sensitive intracellular store. Possible mechanisms for this effect include elevating cellular IP3 content, or disruption of Ca2+ sequestration at the ER. MeHg did not increase cellular IP3 levels during the time course of the first Ca2+ phase, thus the MeHg-induced disruption of ER Ca2+ regulation was not mediated by elevations in IP3 content. However, similar concentrations of MeHg elevated IP3 levels in cultured cerebellar granule cells following a 30 min exposure (Sarafian, 1993), possibly due to a secondary activation of Cazi-regulated PLC isoforms mediated by MeHg—induced elevations in [Ca2+]i (Eberhard and H012, 1988). Caffeine, which facilitates CICR, also decreased the magnitude of the first Ca2+ phase, suggesting a contribution of CICR to first Ca2+ phase. Pretreatment of cells with the L-type voltage-dependent Ca2+ channel blocker nifedipine alone or in combination with the N-type Ca2+ channel blocker (o-conotoxin GVIA did not affect the first Ca2+ phase (Hare and Atchison, 1995). Thus, the first Ca2+ phase was due to mobilization of IP3-sensitive Ca2+ stores from the ER. The second Ca2+ phase has also been studied using digital imaging microscopy of fura—Z loaded N0108-15 cells (Hare et al., 1996). This phase was strictly dependent upon Ca2+ The time to onset of the second Ca2+ phase was delayed by the L-type Ca2+ channel antagonist nifedipine (0.1-10 pM) in a concentration-dependent manner. At these concentrations nifedipine is unlikely to be specific for L-type Ca2+ channels, suggesting the second Ca2+ phase is mediated by a nifedipine-sensitive pathway other than L-type Ca2+ channels. Indeed, these channels have been reported to be blocked by MeHg (Shafer and Atchison, 1987). The nonspecific Ca2+ channel blocker Ni2+ did not delay the second 133 Ca2+ phase, providing further evidence that this phase is not mediated by Ca2+ influx through voltage-dependent Ca2+ channels. The second Ca“ phase was also inhibited by TTX or reduction of extracellular Na+ concentration, suggesting a role for voltage- dependent Na“ channels. Both nifedipine and TTX also delayed the onset of the first Ca2+ phase and the non-Ca2+ phase. Because these phases are mediated by intracellular actions of MeHg, it has been proposed that nifedipine and TTX act by impeding MeHg uptake, thereby prolonging the time required to reach some critical toxic concentration within the cell. MeHg inhibits the regulatory systems for other cations. MeHg blocks mitochondrial ATP synthesis and reduces cellular ATP content (Verity et al., 1975; Kauppinen et al., 1989). Both of these effects could ultimately elevate [Ca2*]i. Decreases in cellular ATP content could reduce the activity of the plasma membrane Ca2*-ATPase, causing an elevation in [Ca2+]i. The Na*/K*-ATPase has a sensitivity to MeHg which is two-fold. First, the loss of cellular ATP reduces the activity of the Na+ +-ATPase. Second, the activity of brain microsomal Na"/K“-ATPase is inhibited directly by MeHg, possibly due to modification of a sulfhydryl group within the protein (Rajanna er al., 1990; Anner and Moosmayer, 1992). It is possible that a component of the elevations in [Ca2+]i is mediated by effects of mercurials on these cation regulatory proteins. The first study to suggest that MeHg altered Zn2+ homeostasis in the brain was conducted by Muto and coworkers (1991). Rats were administered a lethal dose of MeHg, and the metal content of tissue samples from brain, liver and kidney was analyzed by inductively coupled plasma atomic emission spectrometry. The primary effect of in vivo administration of MeHg was a disruption of brain Zn2+ and Ca2+ homeostasis. 134 Synaptosomal suspensions have been subjected to standard protein purification techniques to identify the protein(s) responsible for these elevations in [Zn’*],. Synaptosomal homogenates were separated into particulate and soluble fractions by centrifugation. The particulate fraction was resuspended in an equal volume of buffer, and fura-2 pentapotassium salt was added to the both fractions. Fura-2 excitation scans were acquired before and after addition of MeHg, then compared for MeHg-induced alterations in the fura-2 excitation profile consistent with an elevation in free an“ concentration (Hechtenberg and Beyersmann, 1993). MeHg altered the fura-2 spectrum of the soluble fraction in a manner consistent with Zn2+ release, but had no effect on the particulate fraction (Denny and Atchison, 1995). Thus, the source of the Zn” appears to be some soluble factor(s). Anion exchange chromatography of the soluble fraction resolved three distinct peaks of MeHg-induced Zn2+ release. Thus, more than one protein mediates the MeHg-induced elevations in [Zn2*]i. Although it is not known what role the elevations in [Zn2“]i play in MeHg-induced neurotoxicity, it is still possible to make some general speculations. Many proteins require Zn“ for enzymatic activity or to stabilize their three-dimensional structure. Thus, the loss of essential Zn2+ could result in an inactivation of the enzyme or improper protein folding. The lay-binding region of these proteins may actually have a higher affinity for MeHg than Zn”. Since several metalloproteins generally utilize a common metal-binding motif which depends on both the metal and the function of the pretein, such as the EF hand in Ca2*-activated proteins or the Zn2+ fingers in DNA binding proteins, the loss of endogenous Zn” caused by MeHg may occur in an entire class of metalloproteins which contain a particular Zn2*-binding m0tif. Another possible consequence in that the 135 elevation in [Zn2“]i itself could be toxic. While it would appear that the elevations in [Zn2+]i which occur in synaptosomes are quite modest, it nonetheless represents a three- fold elevation. Furthermore, because synaptosomes are a heterogenous population of nerve terminals, it is also possible that the elevations in [Zn2"]i occur in only a subpopulation of synaptosomes. If this is the case, the actual [Zn2*]i in this subpopulation would be considerably greater than the p0pulation average which would be reported by l"F-NMR spectroscopy. B. Relationship to previous work. Previous research suggested that a component of the MeHg-induced elevations in [Ca2+]i was due to disruption of mitochondrial Ca2+ regulation. Thus, the original goal of this work was to monitor changes in [Ca2+]i following acute exposure of synaptosomes loaded with a fluorescent indicator to MeHg, and to evaluate the contributions of intra- and extracellular sources of Ca2+. Substantial indirect evidence suggested that a component of the elevations in [Ca2*]i would be due to disruption of mitochondrial Ca2+ regulation. However, exposure of synaptosomes to MeHg caused elevations in [Cay]i which were strictly dependent upon Caz} Depolarization of the mitochondrial membrane with NaN3 and oligomycin was also ineffective at elevating [Ca2+]i and did not alter the effects of MeHg. Moreover, the mitochondrial Ca“ uniporter inhibitor ruthenium red did not block the effects of MeHg, suggesting that the effects of MeHg may be mediated by actions at other sites. While it is possible that the Ca2+ regulatory capacity of the mitochondria was compromised during synaptosomal preparation, it is unlikely, since similar findings were observed in NGlO8-15 cells in culture (Hare et al., 1993). 136 Subsequent investigations in this cell type, as well as isolated cerebellar granule cells, have shown that a component of the elevations in [Ca2+]i is mediated by release of Ca2+ previously sequestered in the ER. The ER pool is mobilized completely by IP,, and to a limited extent by caffeine. Therefore, the mitochondria did not appear to contribute significantly to the MeHg-induced elevations in [Ca2+]i. The results of this study are inconsistent with disruption of mitochondrial Ca2+ regulation contributing to the elevations in [Ca2+], following MeHg treatment. Also, the effects of MeHg on ER Ca2+ regulation and [Zn2"]i must now be considered. As such, it is possible the MeHg-induced increases in MEPP frequency observed in Ca”’-free solutions were actually mediated by release of Ca2+ from the ER, or possibly by elevations in [Zn2*],., but perhaps not by disruption of mitochondrial Ca2+ regulation. The inhibitors of mitochondrial function (dinitrophenol, dicoumarol and valinomycin) increased MEPP frequency after a twenty min delay. Pretreatment of hemidiaphragm preparations with these agents did not affect the subsequent elevations in MEPP frequency caused by MeHg. It is possible the 20 min delay preceding the elevations in MEPP frequency represented the time required for the mitochondrial inhibitors to reduce ATP levels in the nerve terminal. The reduction in cellular ATP content led to a secondary elevation in [Ca2+], and stimulation of spontaneous transmitter release, measured as an increase in MEPP frequency. Proper levels of ATP were then reinstated by increasing the rate of glycolysis, causing [Ca2+]i and MEPP frequency to fall back to control levels. Subsequent addition of MeHg to these preparation caused elevations in [Ca2+], which could not be regulated, and MEPP frequency once again increased. ‘- 137 Support for this mechanism of action of the mitochondrial inhibitors comes from their similarity to the Na*/K*-ATPase inhibitor ouabain. Depletion of cellular ATP causes a reduction in the activity of the Na*/I(*-ATPase; this is functionally equivalent to the action of ouabain. Loss of N a*/K*-ATPase activity causes a gradual depolarization of the nerve terminal which leads to activation of Ca2+ channels and an elevation in [Ca2+],. Ouabain increased MEPP frequency following a 50 min delay, with a maximum frequency similar to that of the mitochondrial inhibitors. Like the mitochondrial inhibitors, the effects of MeHg were not prevented by ouabain, suggesting an additional site of action is not affected by ouabain. The characteristics of the increases in MEPP frequency caused by ruthenium red were distinct, both qualitatively and quantitatively, from those of the mitochondrial inhibitors or ouabain. Unlike the mitochondrial inhibitors and ouabain, the increases in MEPP frequency elicited by ruthenium red were immediate. Also, the maximum MEPP frequency attained in ruthenium red-treated preparations was approximately one fifth that of those exposed to mitochondrial inhibitors or ouabain. However, neither MeHg nor dinitrophenol elevated MEPP frequency in preparations treated with ruthenium red. This effect could not attributed to depletion of vesicular contents from the nerve terminal since La3+ was capable of increasing MEPP frequency in ruthenium red-treated preparations. The effects of disruption of ER Ca2+ regulation on MEPP frequency have been studied using hemidiaphragm preparations treated with caffeine. Caffeine elevates [Ca2+], by facilitating CICR. Caffeine caused changes in MEPP frequency which resembled those of ruthenium red. MEPP frequency increased immediately to a maximum similar 138 to that of ruthenium red, but half that of the mitochondrial inhibitors. Unlike ruthenium red, caffeine did not block the effects of MeHg. The characteristics of the increases in MEPP frequency caused by caffeine and ruthenium red suggest they elevate [Ca2+]i by similar mechanisms. These mechanisms are not identical, though, since caffeine-treated preparations are sensitive to MeHg, but ruthenium red-treated preparations are not. A possible explanation for the differential sensitivity of these preparations to MeH g may be related to the extent of depletion of ER Ca2+ stores caused by caffeine and ruthenium red. It is possible that caffeine does not deplete the ER of Ca2+, but rather redistributes Ca2+ between the ER and cytosol; ruthenium red may completely mobilize the ER Ca2+ store. MeHg then mobilizes the remaining Ca2+ from the ER in caffeine-treated preparations, but has no effect following ruthenium red treatment. Depletion of ER Ca2+ stores with bradykinin and thapsigargin prior to MeHg or ruthenium red exposure may assist in determining the relative importance of the IP3-sensitive Ca2+ pool as a source of Ca2+. It is difficult to rationalize the ability of ruthenium red to inhibit the effects of dinitrophenol based solely on actions at the ER. The dinitrophenol-induced elevations in [Ca2+]i are presumably secondary to a reduction in cellular ATP levels. If ruthenium red blocks the actions of both MeHg and dinitrophenol by the same mechanism, it is unlikely to be due to effects exclusively at the ER. It is possible that ruthenium red acts as a permeability barrier to Ca2+ regulation at the ER and plasma membranes. Because the mechanisms which decrease [Ca2+]i tend to exceed those which increase [Caz*]i, thiseffect would cause [Ca2+]i to rise slightly, leading to an increase in MEPP frequency. The + elevations in [Ca2 ]i caused by MeHg or dinitrOphenol may be blocked by ruthenium red. .3“. 4 4.5 . 139 Another explanation is that ruthenium red inhibits Cay-mediated fusion of synaptic vesicles with the presynaptic membrane. If this is the case the vesicles would be rendered insensitive to changes in [Ca2*]i and therefore, no increases in MEPP frequency would occur following MEPP treatment. This also would require La3+ to work through a mechanism independent of elevations in [Ca2+],, possibly by causing vesicular fusion in a manner which is Ca2*-independent. The results concerning the effects of MeHg on [Ca2+]i in synaptosomes are largely consistent with the work of others (Komulainen and Bondy, 1987b; and Kauppinen et al., 1989). MeHg caused elevations in [Cazili which were time-dependent as well as concentration-dependent with respect to MeH g and Caz"c (Komulainen and Bondy, 1987b). The accuracy of the values derived for [Ca2+], may be questionable, though since 20 pM Mn2+ was added to quench extrasynaptosomal fura-2 throughout the experiments. For subsequent calibration of Km and Rm“, only 10 pM DTPA was added to chelate this exogenous Mn“, thus at least 10 pM Mn2+ was not chelated. This may actually represent the best-case scenario since the buffer itself contained 1.2 mM Ca2+; thus, some if not most, of the DTPA would be bound to Ca2+ rather than Mn“. In fact, in preliminary experiments using cell-free systems, I observed that 150 pM DTPA was necessary to reverse the quenching of 1 uM fura—2 caused by 40 pM Mn2+ in a buffer containing 200 pM Ca2+. Furthermore, some of the changes in fura-2 fluorescence could have arisen from release of endogenous Zn“. Unlike the present study, depolarization of the mitochondria was shown to decrease the elevations in [Ca2+]i, suggesting a component of the elevations in [Ca2+]i was from the mitochondrial Caz”. A possible explanation for the differences obtained is that rotenone was used in the previous work and NaN3 was used 140 here. Rotenone acts at Site 1 in the electron transport chain while NaN3 acts at Site 3. In mitochondria poisoned with rotenone, electron transport can still commence from Sites 2 and 3. Because NaN3 acts at Site 3, like NaCN, the transfer of electrons to molecular oxygen in inhibited; thus respiration ceases completely. Perhaps these subtle differences in the mechanism of block of electron transport may account for the differences reported in these studies concerning the role of mitochondria in MeHg-induced elevations in [Ca2+]i. C. Efiects of MeHg on Caz+ regulation. The information gathered in these studies, as well as others, clearly indicates that MeHg causes an elevation in [Ca2+],. In synaptosomes, the increases are mediated by influx of Ca2+” while in NGlO8-15 cells the elevations are mediated by release of Ca2+ from IP3-sensitive stores and influx of Caz}. This implies that MeHg disrupts plasma membrane integrity and interferes with Ca2+ regulation by the ER. Although the role of elevations in [Ca""‘]i in the neurotoxicity of MeHg is not known, based on the importance of Ca2+ as a second messenger it is reasonable to conclude that the disruption in Ca2+ homeostasis caused by MeHg has deleterious effects on cell function. Many neurotoxic insults have been suggested to be mediated, at least in part, by elevations in [Ca2+], (Choi, 1988). The exact mechanism underlying toxicity is not known, but may be due to aberrant regulation of Ca2+-sensitive proteins, and disruption of of Ca2*-regulated signalling pathways. Many proteins either require Ca2+ for activity, or are regulated allosterically by the binding of Ca2+ or Ca2*—activated regulatory proteins, such as calmodulin (Habennann and Richardt, 1986). Unregulated elevations in [Ca2+], 141 have been associated with changes in phosphorylation of proteins and with the degradation of proteins, lipid membranes and nucleic acids (Nicotera et al., 1992). It is believed that the increases in [Ca2+]i activate various Ca2*-regulated proteases, lipases and nucleases which lead to a cytotoxicity. Many cells, including neurons, contain a family Ca2*-activated neutral proteases called calpains (Melloni and Pontremoli, 1989). In their inactive form, calpains are normally found in the cytosol, but upon activation by Ca”, they associate with the plasma membrane and degrade cytoskeletal proteins. Calpain-mediated proteolysis can be inhibited by leupeptin, or by chelation of intra- or extracellular Ca2+ (Lee et al., 1991). It is conceivable the unregulated elevations in [Ca2i]i caused by MeHg could lead to an activation of calpain, and the destruction of cytoskeletal proteins, although this possibility has not been tested experimentally. Cells also contain several Cazi-activated lipases. The two most prominent and widely studied classes of lipases are PLC and PLAZ. PLC hydrolyzes phosphotidylinositol into IR, and diacylglycerol, which together activate PKC by releasing an IP3-sensitive Ca2+ store located in the ER, and by allosteric interactions with PKC, respectively. PLA2 cleaves arachidonic acid from plasma membrane phospholipids for the biosynthesis of eicosinoids, and protects the cell against free radical-induced damage by removing peroxy-fatty acids from the plasma membrane. Because Ca2+ increases the activity of both of these lipases, unregulated elevations in [Ca2+]i could cause extensive damage to the plasma membrane. The increase in PLC activity could also cause a secondary activation in PKC activity, further disrupting intracellular second messenger systems. 142 MeHg increases the activity of PLC and PLA2 in neurons in culture (Sarafian, 1993; Verity et al., 1994). A component of the MeHg-induced elevations in PLA2 activity is dependent upon [Ca2+] 6, suggesting that both intra- and extracellular sources of Ca2+ contribute to the increase in PLA2 activity. It is unlikely the increased PLA2 activity following MeHg treatment represented a physiological response to lipid peroxidation since or-tocopherol blocked lipid peroxidation (Sarafian and Verity, 1991), but did not prevent activation of PLAZ. The PLA2 inhibitor mepacrine reduced free arachidonic acid generation by one-half, suggesting that either a mepacrine-insensitive PLA2 existed, or MeHg also released arachidonic acid by some mechanism independent of PLAZ. The toxicological significance of the increases in PLA2 activity is uncertain since mepacrine did not reduce MeHg-induced toxicity, as determined by uptake of trypan blue, and release of lactate dehydrogenase (Verity et al., 1994). Activation of a Mg2*/Ca2*-dependent endonuclease is the hallmark of induction of programmed cell death. This endonuclease is activated by pathologically high concentrations of Ca2+, and is responsible for fragmentation of nuclear DNA (McCabe et al., 1992). MeHg induces programmed cell death in cerebellar neurons in culture based on changes in nuclear morphology and the fragmentation of nuclear DNA (Kunimoto, 1994). Transfection of the neuronal cell line GTl-7 with the prom-oncogene bcI—2 reduced the formation of reactive oxygen species in these cells and decreased the sensitivity of the cells to MeHg (Sarafian et al., 1994). It was suggested that bcI—2 attenuates MeHg toxicity by inhibiting formation of reactive oxygen species, thus preventing initiation of programmed cell death. 143 D. Effects of MeHg on Zn“ regulation. Acute exposure of synaptosomes (Denny et al., 1993), NGlOS-IS cells (Hare et al., 1993; Hare and Atchison, 1995), or cerebellar granule cells (Marty et al., 1995) elevated the fluorescence intensity of fura-2 at the Ca2+-insensitive wavelength of 360 nm. The increase was independent of Ca2+” and occurred prior to loss of plasma membrane integrity. In synaptosomes, the elevation was immediate and complete within a few sec; in NGlO8-15 and cerebellar granule cells the increase occurred gradually following Ca2+ release from the ER pool ("first Ca2+ phase") and before the massive influx of Ca2+e ("second Ca2+ phase"). Because fura-2 has an affinity for cations other than Ca2+, especially heavy metals (Grynkiewicz et al., 1985), it was proposed that the elevations in fluorescence intensity were due to an increase in heavy metal concentration. It was unlikely that this effect was due to a direct interaction of MeHg with fura-2 since these compounds do not interact in vitro. Furthermore, the increase in fluorescence intensity at 360 nm could not be attributed to demethylation of MeH g because Hg2+ decreases, rather than increases, fura-2 fluorescence. The cell-permeant heavy metal chelator TPEN reversed or inhibited these elevations, whereas the cell—impermeant chelator DTPA was ineffective. Thus, MeHg increased the intracellular concentration of a heavy metal whose source was likely to be endogenous. In synaptosomes, TPEN also prevented the immediate elevations in 340/380 nm ratio caused by MeHg, suggesting that this phase was due solely to an increase in the concentration of some other endogenous polyvalent cation. Due to the limitations of fluorometric analysis the cation could not be identified unequivocally, but was postulated 144 to be Zn2+ due to its characteristic effects on the fura—2 excitation spectrum (Grynkiewicz et al., 1985; Hectenberg and Beyersmann, 1993). To identify the endogenous heavy metal, changes in the 19F-nuclear magnetic resonance spectrum of synaptosomes loaded with the fluorinated chelator SF-BAPTA were evaluated following exposure to MeHg (Denny and Atchison, 1994). In the absence of Ca2+... synaptosomes had a detectable [Zn2*]i. MeHg caused a concentration-dependent elevation in [Zn2“]i which was reversed by TPEN, whereas [Ca2+]i was not affected. While the MeHg-induced elevations in [Zn2*]i were modest (1.37 nM prior to MeHg exposure versus 3.99 nM following addition of MeHg), they nonetheless represented a three-fold increase in [Zn2*]i. These results, along with previous results using fura-2, demonstrated that MeHg causes an immediate elevation in [Zn2*]i in synaptosomes, and possibly NGlOS-IS cells. Higher concentrations of MeHg were required to observe the elevations in [Zn2“]i in synaptosomes relative to those necessary in NGlOS-IS cells, possibly due to the higher protein content and associated nonspecific binding of MeHg in the synaptosomal suspensions. Alternatively, it is possible that the source of the MeHg-induced elevations in endogenous heavy metal content in NGlOS-IS cells is not identical to that which mediates the elevations in [Zn2*]i in synaptosomes, and that this pool in NGlO8-15 cells is more sensitive to MeHg. The observation that MeHg caused an elevation in [Zn2*]i was a novel and unexpected finding of this research. Zn2+ is an essential heavy metal required by numerous proteins. The implications of this newly-discovered effect are that a component of the neurotoxicity of MeHg could be due to disruption of Zn2+ regulation. Zn2+ is associated with the synaptic vesicles of certain glutamatergic neurons in the CNS 145 (Frederickson et al., 1987) and is released during neuronal activity (Assaf and Chung, 1984), suggesting that Zn2+ may be a modulator of postsynaptic glutamate receptors. In neuronal cells in culture, Zn“ inhibited responses mediated by glutamate receptors sensitive to NMDA (Westbrook and Mayer, 1987), but increased responses mediated by AMPA-sensitive glutamate receptors (Xie et al., 1993). Zn“ also decreases responses mediated by 'y-aminobutyric acidA receptors (Legendre and Westbrook, 1990), but potentiates ATP receptor-mediated responses (Li et al., 1993). While Zn2+ may act as a neuromodulator, high concentrations are neurotoxic (Yokoyama et al., 1986; Koh and Choi, 1994). Furthermore, Zn“ toxicity may be potentiated by excessive AMPA receptor activation (Choi et al., 1989; Weiss et al., 1993). Identification of the proteins responsible for the elevations in [Zn2“]i was given priority over examination of the toxic effects of Zn2+ for several reasons, some of which related to the use of synaptosomes as a model system. Synaptosomes derived from brain represent a heterogenous population of nerve terminal with respect to vesicular contents. It is possible that the proteins responsible for the MeHg-induced elevations in [Zn2*]i are present in only a subpopulation of nerve tenninals. The methods used to detect and quantify the elevations in [Zn2*]i are not capable of discriminating between identical increases in all synaptosomes, or greater increases in a subpopulation. In the latter case, the values of [Zn2*]i obtained using I9F-NMR spectroscopy of 5F-BAPT A loaded synaptosomes would underestimate the true [Zn2“]i in this subpopulation. There are several ways to address the issue of synaptosomal heterogeneity as it relate to the issue of MeHg-induced elevations in [Zn2*]i. One possible approach would be to test the effects of MeHg on synaptosomes derived from particular brain regions 146 which are enriched in nerve terminals containing a particular neurotransmitter. The advantage of this approach is that known targets of MeHg neurotoxicity could be tested directly, and correlations between target sites and elevations in [Zn2“]i may be possible. There are many disadvantages to this procedure, though. Synaptosomes would have to be prepared from several brain regions to ensure adequate sampling of nerve terminals containing different neurotransmitters, as well as target sites of MeHg. Analysis of brain regions would only be useful if the terminals containing these MeHg-sensitive Zn”- binding proteins were localized in certain brain regions. Thus, if the proteins are expressed in a subpopulation of nerve terminals which is distributed evenly throughout the brain, these experiments would cause us to conclude incorrectly that a sensitive subpopulation of synaptosomes does not exist. Another problem is that even in brain regions which are enriched in nerve terminals containing a particular neurotransmitter, other nerve terminals are also present. If the proteins are expressed in the associated nerve terminal type rather than the enriched nerve terminals, we would conclude the elevations in [Zn2*]i occur in the enriched population of nerve terminals when the changes are actually in the associated nerve terminals. Another contingency not addressed adequately by this method is that the elevations [Zn2“]i do occur in a subpopulation of synaptosome, but expression of the target proteins is not related to neurotransmitter content or regional localization. At best, the results from analysis of synaptosomes prepared from different brain regions might provide some information regarding the localization of these proteins in the brain, and possibly some assistance identifying the protein. However, it would still be necessary to purify and identify the proteins to verify that MeHg is capable of releasing the associated Zn“. 147 A common approach to addressing the population dynamics of a suspension of particles is the use of fluorescence-activated cell sorting. In general, this technique uses differences in fluorescence intensity of a particular marker, or markers, to discriminate between subpopulations of particles on a individual basis. For instance, it is possible to determine what percentage of cells in a suspension contain a particular receptor by staining the cells with a fluorescently-tagged antibody to the receptor. When the cells pass through the sorter, the cells expressing the receptor fluoresce more intensely than those lacking the receptor. It would be possible to evaluate the population characteristics of the MeHg-induced elevations in [Zn2*]i by loading synaptosomes with a Zn2*-sensitive indicator, and monitoring changes in the fluorescence intensity histogram before and after MeHg exposure. If MeHg elevated [Zn2”]i in the entire synaptosomal population, then the entire fluorescence intensity histogram would shift to slightly higher values. If, on the other hand, the elevations in [Zn2*]i were restricted to a subpopulation of synaptosomes, the histogram would show a two distinct peaks, one at the initial value corresponding to cells not affected by MeHg, and another at much higher intensity than the initial value representing the sensitive subpopulation. Another feature of this technique is that the subpopulations can be separated based on these differences in fluorescence intensity. A problem inherent with fluorescence-activated cell sorting is that because individual particles, or cells, are being analyzed they must be sufficiently labelled to allow for detection. In case of measuring changes in intracellular cation concentrations, the initial cell population also needs to be labelled uniformly. Otherwise, the peaks on the intensity histogram become so broad that it is not possible to evaluate changes in histogram characteristics. These problems limited the use of this technique in monitoring 148 the changes in [Zn2*]i following MeHg exposure. Due to their small size, the synaptosomes were difficult to detect and could not loaded uniformly with the indicator. Because the system utilizes an argon laser as the source of excitation light, it was not possible to use fura-2. Synaptosomes had to be loaded with flue-3 to monitor changes in [Zn2*]i. While flue-3 exhibits large changes in fluorescence intensity upon Ca2+ binding, Zn2+ is only 70% as effective as Ca2+. Thus, these experiments were hampered by the small size of the synaptosomes, and the lack of an ideal indicator. Nonetheless, synaptosomes could be loaded with the indicator and detected, but it was not realistic to draw conclusions about the population distribution of the elevations in [Zn’*]i based on the fluo-3 fluorescence intensity histograms. Perhaps when a more suitable Zn2+ indicator becomes available, it may be possible to use this technique to analyze and sort the synaptosomes. Purification of the proteins responsible for the elevation in [ani]i would assist in addressing the questions which arise due to the heterogeneous nature of synaptosomes. The purified proteins could possibly be identified based on sequence information. The distribution patterns of several proteins within the neuron are already known, making extrapolation of protein expression to the synaptosomal population relatively simple. In the event that a protein cannot be identified based on sequence information, or its cellular distribution is not known, it would still be possible to determine its localization within the synaptosomal population by immunocytochemistry. This would require generating antibodies to the protein and probing brain slices for immunoreactivity. If a particular brain region, or areas within the brain, are stained to a greater extent than others, it is likely the protein is found only in a particular subpopulation of synaptosomes. 149 Conversely, if the entire brain is stained to a similar extent, this suggests the particular protein is relatively evenly distributed, and the elevations in [Zn2*]i likely to occur in the entire synaptosomal population. Another advantage of purification of the soluble proteins from the total synaptosomal homogenate is that all of the proteins which release Zn2+ can be isolated at once. This means that should multiple subpopulations exist, they can be evaluated using the immunocytochemical procedure outlined above. Purification and identification of these Zn2"-releasing proteins also provides valuable structural information. Because Zn2+ is bound to proteins using certain motifs, and MeHg releases Zn2+ from more than one protein, it is reasonable to conclude that a common Zn2"-binding configuration is sensitive to MeHg. Thus, this may represent an effect of MeHg on an entire class of proteins. If the Zn2"-binding motif of these proteins can be identified, it should be possible to evaluate the effects of MeHg on other proteins which possess this particular configuration. The relevance of this effect of MeHg on Zn“ regulation to MeHg-induced neurotoxicity is not clear. Because release of Zn“ from soluble synaptosomal proteins mediates the elevations in [Zn2*]i, it is possible that the loss of Zn2+ from these proteins, or the increases in [Zn2*]i, is toxic. Thus far, no attempt has been made to correlate the elevations in [Zn2“]i with the toxicity of MeHg either in vitro or in vivo. It is clear that the MeHg-induced elevations in [ani]i are not an artifact due to synaptosomal preparation. Similar elevations have been observed in NG108-15 cells and isolated cerebellar granule cells in culture, but the effects of MeHg on [Zn2“]i in non-neuronal cells have not been evaluated. 150 Changes in [Zn2*]i have been reported for a number of physiological and toxic agents in a variety of systems. Exposure of NGlOB-IS cells to Pb2+ increases the intracellular concentrations of Ca2+, Pb2+ and Zn“ as determined by l9F-NMR spectroscopy (Schanne et al., 1989). Prolonged exposure of cerebrocortical slices to excitatory amino acids has also been demonstrated to cause in elevation in [Ca2+]‘ and [Zn2“]i using this technique (Badar—Goffer et al., 1994). Hypochlorous acid mobilizes intracellular Zn2+ in rat heart myocytes loaded with a Zn2*-sensitive fluorescent indicator TSQ (Tatsumi and Fliss, 1994). External application of Zn2+ causes increases in [Zn”“]i mediated by voltage-dependent Ca2+ channel in fura-Z loaded GH3 pituitary tumor cells and primary bovine chromaffin cells (Vega et al., 1994; Atar et al., 1995). Glucose causes a reduction in [Zn2*]i in isolated pancreatic islet cells loaded with the fluorescent indicator zinquin (Zalewski et al., 1994). The toxic consequences of disruption of Zn“ regulation as well as the toxicity of Zn“ itself have been studied to some extent, although most information remains phenomenological rather than mechanistic at this time (Vallee and Falchuk, 1993). Maternal Zn” deficiency causes pronounced developmental deficits in the offspring. Characteristics of developmental toxicity mediated by Zn2+ deficiency include a disruption of brain and bone formation. While the mechanisms responsible for these malformations are not known, they are believed to be due to lack of essential Zn2+ required by the DNA- binding proteins which regulate development. A bacterial DNA-binding protein which regulates expression of mercurial resistance genes contains a putative Zn2*-binding region also capable of binding Hg”. The binding of mercury is necessary for activation of 151 transcription of the mercury resistance genes. Thus, gene expression can be altered by changes in the Zn“ status of a regulatory protein. Zn2+ affects several voltage- and ligand-gated channels which are essential for neuronal function (Harrison and Gibbons, 1994). Like many Other divalent cations, external application of Zn2+ blocks voltage-dependent Ca2+ channels, but Zn2+ can permeate the channel (Blisselberg et al., 1994). Zn2+ reduces voltage-dependent Na” channel conductance, possibly by an action at the voltage sensor. In hippocampal neurons, the response of voltage-dependent K" channel A current to Zn2+ is dependent upon channel subtype and experimental parameters such as holding potential and voltage step. This has been interpreted as discrete actions of Zn2+ on both the activation and inactivation gates. Zn“ also affects the biophysical properties of many ligand-gated channels. The first class of neurotransmitter receptors demonstrated to be regulated by an+ were the Opioid receptors. The distribution patterns of enkephalins and Zn2+ within the brain are identical in some species. In these brain regions the binding of a opioid receptor ligand was inhibited by 10100 uM Zn”, a concentration that is likely to be encountered physiologically. Zn2+ also modulates the receptors for the amino acids GABA and glutamate. Zn2+ depresses Cl' conductance mediated by GABA, and blocks glutamate channels which are sensitive to NMDA by a mechanism that is distinCI from the voltage-dependent block caused by Mgz”. Zn2+ enhances the conductances of AMPA- sensitive glutamate channels as well as ATP receptors. It bears noting that in all instances listed above, the putative Zn2+ binding site is located on the extracellular surface of the channel protein. Thus, these effects are most likely due to conformational changes in the protein structures rather than intracellularly-mediated events. 152 Because MeHg causes elevations in [Zn2*],, it is likely the effects would be limited to disruption of intracellular processes. Exposure of mouse hemidiaphragm preparations to 50 or 100 pM an“ increased MEPP frequency in a concentration-dependent manner (Nishimura, 1988). Bathing solutions containing 2 mM Ca” and 10 mM K+ were less effective at increasing MEPP frequency in these Zn2+-treated preparations. Thus, either Zn2+ inhibited the transmitter release process, possibly by blocking voltage-dependent Ca2+ channels, or it caused depletion of vesicular stores. Disruption of ER Ca2+ regulation attenuated the increases in MEPP frequency induced by Zn“. Dantrolene sodium, which blocks release of Ca2+ sequestered in the ER (Statham and Duncan, 1976), or neomycin, which inhibits IP3 production (Schacht, 1976), both reduced the elevations in MEPP frequency caused by Zn”. This suggested that a component of the increases in MEPP frequency was mediated by mobilization of intracellular Ca2+ stores. However, the effect of neomycin may have been due to block of Zn“ influx through voltage-dependent Ca2+ channels (Atchison et al., 1988). More detailed analysis of the effects of Zn2+ on transmitter release suggested that Zn” enters the nerve terminal through voltage-dependent Ca2+ channels and acts as a partial agonist at Cay-sensitive release apparatus (Wang and Quastel, 1990). Zn2+ blocked Ca2"-stimulated transmitter release by inhibiting Ca2+ influx and causing irreversible alterations in the release mechanisms, rendering them insensitive to Ca”. The main difficulty in relating these studies of the effects of Zn2+ on spontaneous transmitter release to those observed following MeHg treatment is that the [.Zn2+]i necessary to elicit these changes is not known. Performing similar experiments except in the presence of Zn2+ and the Zn2+ ionophore pyrithione may assist in determining whether MeHg causes elevations in [Zn2*]i sufficient to disrupt the transmitter release 153 process directly. Similarly, the contribution of elevated [Zn2”]i to the MeHg-induced alterations in MEPP frequency could be evaluated using the cell-permeant chelator TPEN. Zn2+ has been shown to regulate the activity of many enzymes, and alter intracellular signalling pathways. The regulatory domain of PKC contains four Zn”- binding sites which are necessary for binding of phorbol esters as revealed by mutational analysis (Hubbard et al., 1991). This suggests that the conserved Zn2*-binding sites in PKC are responsible for the binding of diacyl glycerol, the physiological activator of PKC. The key Cay-binding regulatory protein calmodulin also binds Zn“, which is a partial agonist of calmodulin activity as measured by increases in calmodulin—dependent phosphodiesterase activity (Chao et al., 1984). an+ is a potent inhibitor of Ca”- dependent PLA2 isolated from snake venoms (Mezna et al., 1994). It is believe that Zn2+ occupies at least one Ca2*-binding site which is necessary for the activity of this enzyme. Zn2+ inhibits the endonuclease responsible for the nuclear dissolution observed during apoptosis in bovine liver (Lohmann and Beyersmann, 1994), and inhibits apoptosis induced by treatment of human lymphoma cells in culture with drugs which disrupt microtubules (Takano et al., 1993). However, initiation of apoptosis of peripheral blood lymphocytes caused by mitogen deprivation has been reported to be dependent upon Zn”, but not Ca2+ (Treves et al., 1994). The proteins responsible for the elevations in [Zn2“]i are not yet known, but a few candidates have been identified based on zinc-binding characteristics and molecular weight. Microtubules are composed of dimers of the 50 kDa acidic proteins a— and B- tubulin (Lemischka et al., 1981; Ginzburg et al., 1985). The presence of Zn“ influences the polymer characteristics and crystal structure of tubulin (Wolf et al., 1993). Neuron- 154 specific proteins possessing a similar molecular weight include an 53.5 kDa intermediate filament called peripherin (Thompson and Ziff, 1989), the 45 kDa microtubule-associated protein tau (Kosik et al., 1989), and the 53.1 kDa granule cell specific Ca2*/calrnodulin- dependent protein kinase IV (Ono and Means, 1989). The Zn2*-binding properties of these proteins have not been reported. The stromelysins are a family of metalloendoproteases which cleave collagen in the extracellular matrix upon activation, and are synthesized as inactive zymogens of approximately 54.2 kDa (Matrisian et al., 1986; Breathnach et al., 1987). These proteins contain an autoinhibitory region in which a cysteine residue sterically hinders the active site by interacting with a Zn” atom essential for catalysis (Kleiner and Stetler-Stevenson, 1993). Activation occurs when this steric hinderance is removed and the binding site is revealed. Normally the autoinhibitory region is cleaved from the protein, however organomercurials can also activate the stromelysins by interacting with the cysteine in the autoinhibitory region, causing it to move out of the active site. Because the Zn2+ in the binding site is essential for protease activity, the possible displacement of Zn“ by Mel-lg would be expected to inactivate the enzyme. The possible ramifications of the elevation in [Zn2*]i are difficult to deduce, because in many cases Zn2+ acts as a regulator of Ca2*-sensitive processes. Because elevations in [Ca2+]i occur along with the elevations in [Zn2*],, it will be difficult to establish clearly what role each cation plays in MeHg-induced neurotoxicity. While it is clear that MeHg disrupts divalent cation homeostasis, the toxicological significance of this effect is less certain. Very few studies have been designed specifically to evaluate the role of elevations in [Ca2+]i in MeHg-induced toxicity. As a rule, MeHg causes a myriad l” 155 of effects on the cellular and biochemical level including depolarization of the plasma and mitochondrial membranes (Hare and Atchison, 1992), depletion of cellular ATP (Kauppinen et al., 1989), block of protein (Cheung and Verity, 1981) and DNA synthesis (Gruenwedel and Cruikshank, 1979), changes in lipid metabolism (Sarafian, 1993; Verity et al., 1994), alterations in protein phosphorylation (Sarafian and Verity, 1990a; 1990b) and disruptions in microtubule structure (lmura et al., 1980). Due to the importance of Ca2+ as an intracellular second messenger, it is conceivable that some of these effects could be mediated, at least in part, by unregulated elevations in [Ca2+],. In summary, disruption of divalent cation homeostasis should be considered an a intermediate step in MeHg-induced toxicity rather than as a toxic consequence in and of itself. To this end, additional studies are needed to assess the impact this loss of divalent cation regulation has on neuronal structure and function on the both biochemical and cellular level. This may ultimately broaden our knowledge of both the mechanisms of MeHg-induced neurotoxicity, as well as the physiological regulators of neuronal development and function. BIBLIOGRAPHY 156 BIBLIOGRAPHY Akerman, K.E.O. and Nicholls, D.G. ( 1983) Physiological and bioenergetic aspects of mitochondrial calcium transport. Rev. Physiol. Biochem. Pharmacol. 95, 149-201. Alnaes, E. and Rahamimoff, R. (1975) On the role of mitochondria in transmitter release from motor nerve terminals. J. Physiol. (Land) 248, 255-278. Anderson, D.C., King SC, and Parsons, SM. (1983) Pharmacological characterization of the acetylcholine transport system in purified Torpedo electric organ synaptic vesicles. Mol. Pharmacol. 24, 48-54. Anner, B.M. and Moosmayer, M. (1992) Mercury inhibits Na-K-ATPase primarily at the cyt0plasmic side. Am. J. Physiol. 262, F843-F848. Arakawa, 0., Nakahiro, M., and Narahashi, T. (1991) Mercury modulation of GABA- activated chloride channels and non-specific cation channels in rat dorsal root ganglion neurons. Brain Res. 551, 58-63. Arslan, P., Di Virgilio, F., Beltrame, M., Tsien, R.Y., and Pozzan, T. (1985) Cytosolic Ca2+ homeostasis in Ehrlich and Yoshida carcinomas. J. Biol. Chem. 260, 2719- 2727. Assaf, S.Y. and Chung, SH. (1984) Release of endogenous Zn“ from brain tissue during activity. Nature (Lond.) 308, 734-736. Atar, D., Backx, P.H., Appel, M.M., Gao, W.D., and Marban, E. (1995) Excitation- transcription coupling mediated by zinc influx through voltage-dependent calcium channels. J. Biol. Chem. 270, 2473-2477. Atchison, W.D. (1986) Extracellular calcium-dependent and -independent effects of methylmercury on spontaneous and potassium-evoked release of acetylcholine at the neuromuscular junction. J. Pharmacol. Exp. Ther. 237, 672-680. Atchison, W.D. (1987) Effects of activation of sodium and calcium entry on spontaneous release of acetylcholine induced by methylmercury. J. Pharmacol. Exp. Ther. 241, 131-139. ‘ Atchison, W.D., Joshi, U., and Thomburg, J .E. (1986) Irreversible suppression of calcium entry in nerve terminals by methylmercury. J. Pharmacol. Exp. Ther. 238, 618- 648. 157 Atchison, W.D., Adgate, L., and Beaman, CM. (1988) Effects of antibiotics on uptake of calcium into isolated nerve terminals. J. Pharmacol. Exp. Ther. 245, 394-401. Atchison, W.D. and Hare, M.F. (1994) Mechanisms of methylmercury-induced neurotoxicity. FASEB J. 8, 622-629. Atchison, W.D. and Narahashi, T. (1982) Methylmercury-induced depression of neuromuscular transmission in the rat. NeuroToxicology 3, 37-50. Atchison, W.D. and Spitsbergen, J.M. (1994) The neuromuscular junction as a target for toxicity. In: Principles of Neurotoxicology, Chang LW, ed, New York, Marcel Dekker, Inc., 265-307. Bachelard, H. and Badar-Goffer, R. (1993) NMR spectroscopy in neurochemistry. J. Neurochem. 61, 412-429. Badar-Goffer, R.S., Ben-Yoseph, O., Dolin, S.J., Morris, P.G., Smith, G.A., and Bachelard, HS. (1990) Use of l,2-bis(2-amino-5-fluorophenoxy)ethane-NJV,N’ ,N’- tetraacetic acid (5FBAPT A) in the measurement of free intracellular calcium in brain by 19F—nuclear magnetic resonance spectroscopy. J. Neurochem. 55, 878- 884. Badar-Goffer, R.S., Morris, P., Thatcher, N., and Bachelard, H. (1994) Excitotoxic amino acids cause appearance of magnetic resonance spectroscopy-observable zinc in superfused cortical slices. J. Neurochem. 62, 2488-2491. Baimbridge, K.G., Celio, M.R., and Rogers, J.H. (1992) Calcium-binding proteins in the nervous system. Trends Neurosci. 15, 303-308. Bakir, F., Damluji, S.F., Amin-Zaki, L., Murtadha, M., Khalidi, A., Al-Rawi, N.Y., Tikriti, S., Dhahir, H.I., Clarkson, T.W., Smith, J.C., and Doherty, R.A. (1973) Methylmercury poisoning in Iraq. Science 181, 230-241. Bartfai, T. (1979) Preparation of metal-chelate complexes and the design of steady-state kinetic experiments involving metal nucleotide complexes. Adv. Cyc. Nucl. Res. 10, 219-242. Bertolino, M. and Llinas, R. (1992) The central role of voltage-dependent and receptor- operated calcium channels in neuronal cells. Ann. Rev. Pharmacol. Toxicol. 32, 399-421. Blaustein, MP. (1988) Calcium transport and buffering in neurons. Trends Neurosci. 10, 438-443. 158 Blaustein, M.P., Ratzlaff, R.W., and Schweitzer, ES. (1978) Calcium buffering in presynaptic nerve terminals: 11. Kinetic properties of the nonmitochondrial Ca sequestration mechanism. J. Gen. Physiol. 72, 43-66. Blaustein, M.F., Ratzlaff, R.W., and Schweitzer, ES. (1980) Control of intracellular calcium in presynaptic nerve terminals. Fed. Proc. 39, 2790-2795. Binah, 0., Meiri, U., and Rahamimoff, H. ( 1978) The effects of HgCl2 and mersalyl on the mechanisms regulating intracellular calcium and transmitter release. Eur. J. Pharmacol. 51, 453-457. Breathnach, R., Matrisian, L.M., Gesnel, M.-C., Staub, A., and Leroy, P. (1987) Sequences coding for part of oncogene-induced transin are highly conserved in a related rat gene. Nucleic Acids Res. 15, 1139-1151. Biisselberg, D., Evans, M.L., Rahmann, H., and Carpenter, DO. (1991) Effects of inorganic and triethyl lead and inorganic mercury on the voltage activated calcium channel of Aplysia neurons. NeuroToxicology 12, 733-744. Biisselberg, D., Platt, 8., Michael, D., Carpenter, DO, and Haas, H.L. (1994) Mammalian voltage-dependent calcium channel currents are blocked by Pb", Zn", and Al“. J. Neurophysiol. 71, 1491-1497. Chang, L.W. (1977) Neurotoxic effects of mercury- A review. Environ. Res. 14, 329- 373. Chang, L.W. (1980) Mercury. In Experimental and Clinical Neurotoxicology (Spencer RS. and Schaumburg H.H., eds.), The Williams and Wilkins Co. Chao, S.-H., Suzuki, Y., Zysk, J.R., and Cheung, W.Y. (1984) Activation of calmodulin by various metal cations as a function of ionic radius. Molec. Pharmacol. 26, 75- 82. Cheung, M. and Verity, M.A. (1981) Methyl mercury inhibition of synaptosome protein synthesis: role of mitochondrial dysfunction. Environ. Res. 24, 286-298. Choi, D.W. (1987) Ionic dependence of glutamate neurotoxicity in cortical cell culture. J. Neurosci. 7, 369-379. Choi, D.W. (1988) Calcium-mediated neurotoxicity: relationship to specific channel types and role in ischemic damage. Trends Neurosci. 11, 465-469. Choi, D.W., Weiss, J.H., Koh, 1., Christine, C.W., and Kurth, MC. (1989) Glutamate neurotoxicity, calcium and zinc. Ann. New York Acad. Sci. 568, 219-224. 159 Clarkson, T.W. (1986) Effects and general principles underlying the toxic action of metals. In Handbook of Toxicity of Metals, Second edition (Friberg L., Nordburg GE, and Vouk V., eds.), Elsevier Science Publ. Clementi, E., Scheer, H, Zacchetti, D., Fasolato, C., Pozzan, T., and Meldolesi, J. (1992) Receptor-activated Ca2+ influx. J. Biol. Chem. 267, 2164-2172. Coleman, J.E. (1992) Zinc proteins: enzymes, storage proteins, transcription factors, and replication proteins. Annu. Rev. Biochem. 61, 897-946. Cooper, G.P., Suszkiw, J.B., and Manalis, RS. (1984) Presynaptic effects of heavy metals. In: Cellular and Molecular Neurotoxicology, N arahashi T, ed, New York, Raven Press, pp 1-21. Corcoran, G.B., Fix, L., Jones, D.P., Moslen, M.T., Nicotera, P., Oberhammer, F.A., and Buttyan, R. (1994) Apoptosis: molecular control point in toxicity. Toxicol. Appl. Pharmacol. 128, 169-181. Deana, R., Panato, L., Cancellotti, F.M., Quadro, G., and Calzigna, L. (1984) Properties of a new calcium ion antagonist on cellular uptake and mitochondrial efflux of calcium ions. Biochem. J. 218, 899-905. Denny, M.F. and Atchison, W.D. (1994) Methylmercury-induced elevations in intrasynaptosomal zinc concentrations: an '9F-NMR study. J. Neurochem. 63, 383- 386. Denny, M.F. and Atchison, W.D. (1995) Methylmercury causes release of zinc from soluble synaptosomal proteins. The Toxicologist (abstr) 15, 14. Denny, M.F., Hare, M.F., and Atchison, W.D. (1993) Methylmercury alters intrasynaptosomal concentrations of endogenous polyvalent cations. Toxicol. Appl. Pharmacol. 122, 222-232. DiVirgilio, F., Fasolato, C., and Steinberg, TH. (1988) Inhibitors of membrane transport system for organic anions block fura-2 excretion from PC12 and N2A cells. Biochem. J. 256, 959-963. Drapeau, P. and Nachshen, DA. (1984) Manganese fluxes and manganese-dependent neurotransmitter release in presynaptic nerve endings isolated from rat brain. J. Physiol. (Lond.) 348, 493-510. ' Dowd, TL. and Gupta, R.I(. (1991) 19F study of the effect of lead on intracellular free calcium in human platelets. Biochim. Biophys. Acta 1092, 341-346. 160 Dowd, TL. and Gupta, R.K. (1993) NMR studies of the effect of hyperglycemia on intracellular cations in rat kidney. J. Biol. Chem. 268, 991-996. Eberhard, DA. and H012, R.W. (1988) Intracellular Ca2+ activates phospholipase C. Trends Neurosci. 11, 517-520. Frederickson, C.J., Kasarskis, E.J., Ringo, D., and Frederickson, RE. (1987) A quinoline fluorescence method for visualizing and assaying the histochemically reactive zinc (bouton zinc) in the brain. J. Neurosci. Meth. 20, 91-103. Frederickson, C.J., Hernandez, M.D., Goik, S.A., Morton, J .D., and McGinty, LE. (1988) Loss of zinc staining from hippocampal mossy fibers during kainic acid-induced seizures: a histofluorescence study. Brain Res. 446, 383-386. Ginzburg, I., Teichman, A., Dodemont, H.J., Behar, L., and Littauer, U.Z. (1985) Regulation of three beta-tubulin mRNAs during rat brain development. EMBO J. 4, 3667-3673. Goddard, GA. and Robinson, JD. (1976) Uptake and release of calcium by rat brain synaptosomes. Brain Res. 110, 331-350. Goldman, W.F. and Blaustein, M.F. (1990) Non-homogeneous an” quench of fura-2 signals in cultured arterial myocytes visualized with digital imaging. Biophys. J. 57,162a. Gray, EG. and Whittaker, VP. (1962) The isolation of nerve endings from brain: An electron-microscopic study of cell fragments derived by homogenization and centrifugation. J. Anat. 96, 79-87. Greene, LA. and Tischler, AS. (1976) Establishment of a noradrenergic clonal line of rat adrenal pheochromocytoma cells which respond to nerve growth factor. Proc. Natl. Acad. Sci. (USA) 73, 2424-2428. Gruenwedel, D.W. and Cruikshank, MK. (1979) Effect of methylmercuryGI) on the synthesis of deoxyribonucleic acid, ribonucleic acid and protein in HeLa S3 cells. Biochem. Pharmacol. 28, 651-655. Grynkiewicz, G., Poenie, M., and Tsien, R.Y. (1985) A new generation of Ca2+ indicators with greatly improved fluorescence properties. J. Biol. Chem. 260, 3440-3450. Habermann, E. and Richardt, G. (1986) Intracellular calcium binding proteins as targets for heavy metal ions. Trends Pharmacol. Sci. 7, 298-300. Hamer, DH. (1986) Metallothionein. Ann. Rev. Biochem. 55, 913-951. 161 Hare, M.F. and Atchison, W.D. (1992) Comparative aCtion of methylmercury and divalent inorganic mercury on nerve terminal and intraterminal mitochondrial membrane potentials. J. Pharmacol. Exp. Ther. 261, 166-172. Hare, M.F. and Atchison, W.D. (1995) Methylmercury mobilizes Ca++ from intracellular stores sensitive to inositol 1,4,5-trisphosphate in NG108-15 cells. J. Pharmacol. Exp. Ther. 272, 1016-1023. Hare, M.F., McGinnis, K.M., and Atchison, W.D. (1993) Methylmercury increases intracellular concentrations of Ca"+ and heavy metals in NG108-15 cells. J. Pharmacol. Exp. Ther. 266, 1626-1635. Hare, M.F. and Atchison, W.D. (1995) N ifedipine and tetrodotoxin ('I'I‘X) delay the onset of the methylmercury (MeHg)-induced Ca2+e-dependent increase in [Cazi]i in NG108-15 cells. Toxicol. Appl. Pharmacol. in press. Harrison, N.L. and Gibbons, SJ. (1994) Zn”: an endogenous modulator of ligand- and voltage-gated ion channels. Neuroplzarmacology 33, 935-952. Haug, F.M.S., Blackstad, T.W., Simonsen, A.H., and Zimmer, J. (1971) Timm’s sulphide silver reaction for zinc during experimental anterograde degeneration of hippocampal mossy fibers. J. Comp. Neurol. 142, 23-32. Hechtenberg, S. and Beyersmann, D. (1993) Differential control of free calcium and free zinc levels in isolated bovine liver nuclei. Biochem. J. 289, 757-760. Heinonen, E., Akerman, K.E.O., and Kaila, K. (1984) Depolarization of the mitochondrial membrane potential increases free cytosolic calcium in synaptosomes. Neurosci. Lett. 49, 33-37. Henzi, V. and MacDermott, AB. (1992) Characteristics and function of Ca2*- and inositol 1,4,5-trisphosphate-releasable stores of Ca2+ in neurons. Neuroscience 46, 251- 273. Hewett, S.J. and Atchison, W.D. (1992) Effects of charge and lipophilicity on mercurial- induced reduction of 45Ca2+ uptake in isolated nerve terminals of the rat. Toxicol. Appl. Pharmacol. 113, 267-273. Hinkle, F.M., Shanshala, ED, and Nelson, EJ. (1992) Measurement of intracellular cadmium with fluorescent dyes. Further evidence for the role of calcium channels in cadmium uptake. J. Biol. Chem. 267, 25553-25559. Hokin, LE. (1985) Receptors and phosphoinositide-generated second messengers. Annu. Rev. Biochem. 54, 205-235. 162 Holliday, J., Adams, R.H., Sejnowski, T.J., and Spitzer, NC. (1991) Calcium-induced release of calcium regulates differentiation of cultured spinal neurons. Neuron 7, 787-796. Howell, G.A., Welch, M.G., and Frederickson, C.J. (1984) Stimulation-induced uptake and release of zinc in hippocampal slices. Nature (Lond.) 8, 736—738. Hubbard, S.R., Bishop, W.R., Kirschmeier, P., George, S.J., Cramer, S.P., and Hendrickson, WA. (1991) Identification and characterization of zinc binding sites in protein kinase C. Science 254, 1776-1779. Hunt, J.B., Neece, S.H., Schachman, H.K., and Ginsburg, A. (1984) Mercurial-promoted Zn2+ release from Escherichia coli aspartate transcarbamoylase. J. Biol. Chem. 259, 14793-14803. Hunter, D. and Russell, D.S. (1954) Focal cerebral and cerebellar atrophy in a human subject due to organic mercury compounds. Neurol. Neurosurg. Psychiat. 17, 235-241. Imura, N., Miura, K., Inokawa, M., and Nakada, S. (1980) Mechanism of methylmercury cytotoxicity: by biochemical and morphological experiments using cultured cells. Toxicology 17, 241-254. Itoh, M., Ebadi, M., and Swanson, S. (1983) The presence of zinc-binding proteins in brain. J. Neurochem. 41, 823-829. Johnson, M.E., Gores, G.J., Uhl, CB, and Sill, J.C. ( 1994) Cytosolic free calcium and cell death during metabolic inhibition in a neuronal cell line. J. Neurosci. 14, 4040-4049. Juang, M.S. (1976a) Depression of frog muscle contraction by methylmercuric chloride and mercuric chloride. Toxicol. Appl. Pharmacol. 35, 183-185. Juang, M.S. (1976b) An electrophysiological study of the action of methylmercuric chloride and mercuric chloride on the sciatic nerve-sartorius muscle preparation of the frog. Toxicol. Appl. Pharmacol. 37, 339-348. ' Juang, MS. and Yonemura, K. (1975) Increased spontaneous transmitter release from presynaptic nerve terminal by methylmercuric chloride. Nature (Lond.) 256, 211- 213. Kass, G.E., Wright, J.M., Nicotera, P., and Orrenius S. (1988) The mechanism of 1- methyl-4-phenyl-l, 2, 3, 6-tetrahydropyridine toxicity: Role of intracellular calcium. Arch. Biochem. Biophys 260, 789-797. 163 Kass, G.E.N., Llopis, J., Chow, S.C., Duddy, SK, and Orrenius, S. (1990) Receptor— operated calcium influx in rat hepatocytes. J. Biol. Chem. 265, 17486-17492. Kauppinen, R.A., Komulainen, H., and Taipale, H. (1989) Cellular mechanisms underlying the increase in cytosolic free calcium concentration induced by methylmercury in cerebrocortical synaptosomes from guinea pig. Toxicol. Appl. Pharmacol. 248, 1248-1254. Kinter, W.B. and Pritchard, JB. (1977) Altered permeability of cell membranes. In: Handbook of Physiology, Lee DKA, Falk HL, Murphy SD, eds, Washington, DC, American Physiological Society, pp 563-576. Kleiner, D.E., Jr. and Stetler-Stevenson, W.G. (1993) Structural biochemistry and activation of matrix metalloproteases. Curr. Opin. Cell Biol. 5, 891-897. Komulainen, H. and Bondy, S.C. (1987a) The estimation of free calcium within synaptosomes and mitochondria with fura-2; comparison to quin-2. Neurochem. Int. 10, 55-64. Komulainen, H. and Bondy, S.C. (1987b) Increased free intrasynaptosomal Ca2+ by neurotoxic organometals: distinctive mechanisms. Toxicol. Appl. Pharmacol. 88, 77-86. Koh, J. and Choi, D.W. (1994) Zinc toxicity on cultured cortical neurons: involvement of N-methyl-D-aspartate receptors. Neuroscience 60, 1049-1057. Kunimoto, M. (1994) Methylmercury induces apoptosis of rat cerebellar neurons in primary culture. Biochem. Biophys. Res. Comm. 204, 310-317. Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (Lond.) 227, 680-685. Lee, K.S., Frank, S., Vanderklish, P., Arai, A., and Lynch, G. (1991) Inhibition of proteolysis protects hippocampal neurons from ischemia. Proc. Natl. Acad. Sci. USA 88, 7233-7237. Legendre, P. and Westbrook, G.L. (1990) The inhibition of single N-methyl-D-aspartate- activated channels by zinc ion on cultured rat neurones. J. Physiol. (Lond.) 429, 429-449. Lehninger, A.L., Reynafarje, B., Vereesi, A., and Tew, WP. (1978) Transport and accumulation of calcium in mitochondria. Ann. NY. Acad. Sci. 307, 160-176. 164 Lemischka, I.R., Farmer, S., Racaniello, V.R., and Sharp, RA. (1981) Nucleotide sequence and evolution of mammalian alpha-tubulin messenger RNA. J. Mol. Biol. 151, 101-120. Levesque, RC. and Atchison, W.D. (1987) Interactions of mitochondrial inhibitors with methylmercury on spontaneous quantal release of acetylcholine. Toxicol. Appl. Pharmacol. 87, 315-324. Levesque, RC. and Atchison, W.D. (1988) Effect of alteration of nerve terminal Ca2+ regulation on increased spontaneous quantal release of acetylcholine by methylmercury. Toxicol. Appl. Pharmacol. 94, 55-65. Levesque, RC. and Atchison, W.D. (1991) Disruption of brain mitochondrial calcium sequestration by methylmercury. J. Pharmacol. Exp. Ther. 256, 236-242. Levesque, P.C., Hare, M.F., and Atchison, W.D. (1992) Inhibition of mitochondrial Ca2+ release diminishes the effectiveness of methyl mercury to release acetylcholine from synaptosomes. Toxicol. Appl. Pharmacol. 115, 11-20. Levy, L.A., Murphy, E., and London, RE. (1987) Synthesis and characterization of ”F NMR chelators for measurement of cytosolic free Ca. Am. J. Physiol. 252, C441- C449. Li, C., Peoples, R.W., Li, 2., and Weight, ER (1993) Zn2+ potentiates excitatory action of ATP on mammalian neurons. Proc. Natl. Acad. Sci. USA 90, 8264-8267. Llinas, R.R., Sugimori, M., and Silver, RB. (1992) Microdomains of high calcium concentration in a presynaptic terminal. Science 256, 677-679. Lo., T-M. and Thayer, SA. (1993) Refilling the inositol 1,4,5-trisphosphate-sensitive Ca2+ store in neuroblastoma x glioma hybrid NG108-15 cells. Am. J. Physiol. 264, C641-C653. Lohmann, RD. and Beyersmann, D. ( 1994) Effects of zinc and cadmium on apoptotic DNA fragmentation in isolated bovine liver nuclei. Environ. Health Persp. 102, 269-271. Lowry, O.H., Rosenbrough, N.J., Farr, A.L., and Randall, R.J. (1951) Protein measurements with the folin phenol reagent. J. Biol. Chem. 193, 265-275. MacDermott, A.B., Mayer, M.L., Westbrook, G.L., Smith, S.J., and Barker, IL. (1986) NMDA-receptor activation increases cytoplasmic calcium concentration in cultured spinal cord neurones. Nature (Lond.) 321, 519-522. 165 Malinow, R., Otmakhov, N ., Blum, K.I., and Lisman, J. (1994) Visualization of hippocampal synaptic function by optical detection of Ca2+ entry through the N- methyl-D-aspartate channel. Proc. Natl. Acad. Sci. USA 91, 8170-8174. Marty, M.S., Hare, M.F., and Atchison, W.D. (1995) Methylmercury alters intracellular calcium concentration ([Ca2*]i) in rat cerebellar granule cell cultures. The Toxicologist (abstr) 15, 259. Mason, M.J. and Grinstein, S. (1990) Effect of cytoplasmic acidification on the membrane potential of T-lymphocytes: role of trace metals. J. Memb. Biol. 116, 139-148. Matrisian, L.M., Leroy, P., Ruhlmann, C., Gesnel, M.-C., and Breathnach, R. (1986) Isolation of the oncogene and epidermal growth factor-induced transin gene: complex control in rat fibroblasts. Mol. Cell. Biol. 6, 1679-1686. McCabe, M.J., Nicotera, P., and Orrenius, S. (1992) Calcium-dependent cell death. Role of the endonuclease, protein kinase C, and chromatin conformation. Ann. NY. Acad. Sci. 663, 269-278. McGraw, C.F., Somlyo, A.V., and Blaustein, M.F. (1980) Probing for calcium at presynaptic nerve terminals. Fed. Proc. 39, 2796-2801. McMahon, HT. and Nicholls, DO. (1993) Barium-evoked glutamate release from guinea- pig cerebrocortical synaptosomes. J. Neurochem. 61, 110-115. McPherson, RS. and Campbell, KR (1993) The ryanodine receptor/Ca2+ release channel. J. Biol. Chem. 268, 13765-13768. Meldolesi, J .C., Huttner, W.B., Tsien, R.Y., and Pozzan, T. (1984) Free cytoplasmic Ca2+ and neurotransmitter release: studies on PC12 cells and synaptosomes exposed to a-latrotoxin. Proc. Natl. Acad. Sci. USA 81, 620-624. Meldolesi, J., Villa, A., Podini, P., Clementi, E., Zacchetti, D., D’Andrea, P., Lorenzon, P., and Grohavaz, F. (1992) Intracellular Ca2+ stores in neurons. Identification and functional aspects. J. Physiol. (Lond.) 86, 23-30. Meldolesi, J., Volpe, P., and Pozzan, T. (1988) The intracellular distribution of calcium. Trends Neurosci. 11, 449-452. Meloni, E. and Pontremoli, S. (1989) The calpains. Trends Neurosci. 12, 438-444. Menza, M., Ahmad, T., Chettibi, S., Drainas, D, and Lawrence, A.J. (1994) Zinc and barium inhibit the phospholipase A2 from Naja naja atra by different mechanisms. Biochem. J. 301, 503-508. 166 Merritt, J.B., Jacob, R, and Hallam, T]. (1989) Use of manganese to discriminate between calcium influx and mobilization from internal stores in stimulated human neutrophils. J. Biol. Chem. 264, 1522-1527. Michaelis, M.L., Kitos, T.E., Nunley, E.W., Lecluyse, E., and Michaelis, E.K. (1987) Characteristics of Mg2*-dependent, ATP-acrivated Ca2+ transport in synaptic and microsomal membranes and in permeabilized synaptosomes. J. Biol. Chem. 262, 4182-4189. Miledi, R. (1973) Transmitter release induced by injection of calcium ions into nerve terminals. Proc. R. Soc. Lond. Ser. B 183, 421-425. Miller, R.J. (1988) Calcium signalling in neurons. Trends Neurosci. 11, 415-419. Minnema, D.J., Cooper, GP, and Greenland, RD. (1989) Effects of methylmercury on neurotransmitter release from rat brain synaptosomes. Toxicol. Appl. Pharmacol. 99, 510-521. Minta, A., Kao, J.P.Y., and Tsien, R.Y. (1989) Fluorescent indicators for cytosolic calcium based on rhodamine and fluorescein chromophores. J. Biol. Chem. 264, 8171-8178. Miyamoto, MD. (1983) ng“ causes neurotoxicity at an intracellular site following entry through Na and Ca channels. Brain Res. 267, 375-379. Moore, CL. (1971) Specific inhibition of mitochondrial Ca2+ transport by ruthenium red. Biochem. Biophys. Res. Comm. 42, 298-305. Moutin, M.J., Rapin, C., and Dupont, Y. (1992) Ruthenium red affects the intrinsic fluorescence of the calcium-ATPase of skeletal sarcoplasmic reticulum. Biochim. Biophys. Acta 1100, 321-328. Murphy, E., Levy, L., Berkowitz, L.R., Orringer, E.P., Gabel, S.A., and London, RE. (1986) Nuclear magnetic resonance measurement of cytosolic free calcium levels in human red blood cells. Am. J. Physiol. 251, C496-C504. Muto, M., Shinada, M., Tokuta, K., and Takizawa, Y. (1991) Rapid changes in concentrations of essential elements in organs of rats exposed to methylmercury chloride and mercuric chloride as shown by simultaneous multielemental analysis. Br. J. Ind. Med. 48, 382- 388. N achshen, DA. (1985) Regulation of cytosolic calcium concentration in presynaptic nerve endings isolated from rat brain. J. Physiol. (Lond.) 363, 87-101. 167 Nahorski, SR. (1988) Inositol polyphosphates and neuronal calcium homeostasis. Trends Neurosci. 11, 444-448. Nelson, MT (1986) Interactions of divalent cations with single calcium channels from rat brain synaptosomes. J. Gen. Physiol. 87, 201-222. Nicotera, P., Bellomo, G., and Orrenius, S. (1992) Calcium-mediated mechanisms in chemically induced cell death. Ann. Rev. Pharmacol. Toxicol. 32, 449-470. Nishimura, M. (1988) Zn2+ stimulates spontaneous transmitter release at mouse neuromuscular junctions. Br. J. Pharmacol. 93, 430-436. Oakley B., Kirsch D., and Morris N. (1980) A simplified ultrasensitive silver stain for detecting proteins in polyacrylamide gels. Anal. Biochem. 105, 361. Ogura, A., Myojo, Y., and Higashida, H. (1990) Bradykinin-evoked acetylcholine release via inositol trisphosphate-dependent elevation in free calcium in neuroblastoma x glioma hybrid NGlOS-IS cells. J. Biol. Chem. 265, 3577-3584. Ono, T. and Means, AR. (1989) Calspermin is a testis specific calmodulin-binding protein closely related to Ca2*/calmodulin-dependent protein kinases. Adv. Exp. Med. Biol. 255, 263-268. Persechini, A., Moncrief, ND, and Kretsinger, R.H. ( 1989) The EF-hand of calcium- modulated proteins. Trends Neurosci. 12, 462-467. Preston, S.F., Sha’afi, R.I., and Berlin, RD. (1991) Regulation of Ca2+ during mitosis: Ca2+ influx and depletion of intracellular Ca2+ stores are coupled in interphase but not mitosis. Cell Reg. 2, 915-925. Rahamimoff, R., Rahamimoff, I-l., Binah, O., and Meiri, U. (1975) Control of neurotransmitter release by calcium ions and the role of mitochondria. In: Calcium Transport in Contraction and Secretion, pp 253-269, ed. E. Carafoli, North-Holland Publishing Co. Rajanna, B., Chetty, CS, and Rajanna, S. (1990) Effect of mercuric chloride on the kinetics of cationic and substrate activation of the rat brain microsomal ATPase system. Biochem. Pharmacol. 39, 1935-1940. Rasgado-Flores, H. and Blaustein, MP. (1987) ATP-dependent regulation of cytoplasmic free calcium in nerve terminals. Am. J. Physiol. 252, C588-C594. Rehder, V. and Kater, SB. (1992) Regulation of neuronal growth cone filopodia by intracellular calcium. J. Neurosci. 12, 3175-3186. 168 Reuhl, K.R. and Chang, L.W. (1979) Effects of methylmercury on the development of the nervous system: a review. NeuroToxicology 1, 21-55. Richards, C.D., Metcalfe, J.C., Smith, G.A., and Hesketh, TR. (1984) Changes in free- calcium levels and pH in synaptosomes during transmitter release. Biochim. Biophys. Acta 803, 215-220. Rizzuto, R., Brini, M., Murgia, M., and Pozzan, T. (1993) Microdomains with high Ca2+ close to IP3-sensitive channels that are sensed by neighboring mitochondria. Science 262, 744-747. Rizzuto, R., Simpson, A.W.M., Brini, M., and Pozzan, T. (1992) Rapid changes of mitochondrial Ca2+ revealed by specifically targeted recombinant aequorin. Nature (Lond.) 358, 325-327. Rossi, A., Manzo, L., Orrenius, S., Vahter, M., and Nicotera, P. (1991) Modifications of cell signalling in the cytotoxicity of metals. Pharmacol. Toxicol. 68, 424-429. Sacchi, O. and Perri, V. (1973) Quantal mechanisms of transmitter release during progressive depletion of the presynaptic stores at ganglionic synapse. The action of hemicholinium-3 and thiamine depletion. J. Gen. Physiol. 61, 342-360. Sanchez-Armass, S. and Blaustein, MP. (1987) Role of sodium-calcium exchange in regulation of intracellular calcium in nerve terminals. Am. J. Physiol. 252, C595- C603. Sarafian, T.A. (1993) Methyl mercury increases intracellular Ca2+ and inositol phosphate levels in cultured cerebellar granule neurons. J. Neurochem. 61, 648-657. Sarafian, T.A., Vartavarian, L., Kane, D.J., Bredesen, D.E., and Verity, M.A. (1994) bcl-2 expression decreases methyl mercury-induced free-radical generation and cell killing in a neural cell line. Toxicol. Lett. 74, 149-155. Sarafian, T. and Verity, M.A. (1986) Mechanism of apparent transcription inhibition by methyl mercury in cerebellar neurons. J. Neurochem. 47, 625-631. Sarafian, T. and Verity, M.A. (1990a) Altered patterns of protein phosphorylation and synthesis caused by methyl mercury in cerebellar granule cell culture. J. Neurochem. 55, 922-929. Sarafian, T. and Verity, M.A. (1990b) Methyl mercury stimulates protein 32P phospholabeling in cerebellar granule cell culture. J. Neurochem. 55, 913-921. Sarafian, T.A. and Verity, M.A. (1991) Oxidative mechanisms underlying methyl mercury neurotoxicity. Int. J. Devl. Neurosci. 9, 147-153. 169 Schacht, J. (1976) Inhibition by neomycin of polyphosphoinositide turnover in subcellular fractions of guinea-pig cerebral cortex in vitro. J. Neurochem. 27, 1119-1124. Schanne, F.A.X., Dowd, T.L., Gupta, R.K., and Rosen, J.F. (1989a) Lead increases free Ca2+ concentration in cultured osteoblastic bone cells: Simultaneous detection of intracellular free Pb2+ by ”F NMR. Proc. Natl. Acad. Sci. 86, 5133-5135. Schanne, F.A.X., Moskal, J.R., and Gupta, R.K. (1989b) Effect of lead on intracellular free calcium ion concentration in a presynaptic neuronal model: ”F—NMR study of NG108-15 cells. Brain Res. 503, 308-311. Schanne, F.A.X., Dowd, T.L., Gupta, R.K., and Rosen, J.F. (1990) Development of ”F NMR for measurement of [Ca2+]i and [Pb2“]i in cultured osteoblastic bone cells. Environ. Health Perspect. 84, 99-106. Schanne, F.A.X., Moskal, J.R., and Gupta, R.K. (1989) Effect of lead on intracellular free calcium ion concentration in a presynaptic neuronal model: ”F-NMR study of NGlO8-15 cells. Brain Res. 503, 308-311. Scott, I.D., Akerman, K.E.O., and Nicholls, D.G. (1980) Calcium-ion transport by intact synaptosomes. Biochem. J. 192, 873-880. Shafer, T.J. and Atchison, W.D. (1989) Block of 45Ca uptake into synaptosomes by methylmercury: Ca“- and Na*-dependence. J. Pharmacol. Exp. Ther. 248, 696- 702. Shafer, T.J. and Atchison, W.D. (1991a) Transmitter, ion channel and receptor properties of pheochromocytoma (PC12) cells: a model for neurotoxicological studies. NeuroToxicology 12, 473-492. Shafer, T.J. and Atchison, W.D. (1991b) Methylmercury blocks N- and L-type Ca++ channels in nerve growth factor-differentiated pheochromocytoma (PC12) cells. J. Pharmacol. Exp. Ther. 258, 149-157. Shafer, T.J., Contreras, ML, and Atchison, W.D. (1990) Characterization of interactions of methylmercury with Ca2+ channels in synaptosomes and pheochromocytoma cells: radiotracer flux and binding studies. Mol. Pharmacol. 38, 102-113. Shalton, RM. and Wareham, AC. (1979) Calcium ionophore A-23187 and spontaneous miniature end-plate potentials of mammalian skeletal muscle. Exp. Neurol. 63, 379-387. Smith, G.A., Hesketh, R.T., Metcalfe, J.C., Feeney, J., and Morris, P.G. (1983) Intracellular calcium measurements by ”F NMR of fluorine-labeled chelators. Proc. Natl. Acad. Sci. 80, 7178-7182. 170 Smith, J.B., Dwyer, SD, and Smith, L. ( 1989) Cadmium evokes inositol polyphosphate formation and calcium mobilization. J. Biol. Chem. 264, 7115-7118. Springman, E.B., Angleton, E.L., Birkedal-Hansen, H., and Van Wart, HE. (1990) Multiple modes of activation of latent human fibroblast collagenase: evidence for the role of a Cys-73 active-site zinc complex in latency and a "cysteine switch" mechanism for activation. Proc. Natl. Acad. Sci. USA 87, 364-368. Steel, R.G.D. and Torrie, J .H. (1980) Principles and Procedures of Statistics, McGraw- Hill, New York. Statham, J.H. and Duncan, G]. (1976) Dantrolene and the neuromuscular junction: evidence for intracellular Ca stores. Eur. J. Pharmacol. 39, 143-152. Takano, Y., Okudaira, M., and Harmon, B.V. (1993) Apoptosis induced by microtubule disrupting drugs in cultured human lymphoma cells. Inhibitor effects of phorbol ester and zinc sulphate. Pathol. Res. Pract. 189, 197-203. Takeuchi, T., Kambara, T., Morikawa, N., Matsumoto, I-l., Shiraishi, Y., and Ito, H. (1959) Pathological observations of the Minamata disease. Acta Pathologica Japonica 9, 769-783. Takeuchi, T., Morikawa, N., Matsumoto, H., and Shiraishi, Y. (1962) A pathological study of Minamata Disease in Japan. Acta Neuropathol. 2, 40-57. Tapia, R., Sitges, M., and Morales, E. (1985) Mechanism of the calcium-dependent stimulation of transmitter release by 4-amin0pyridine in synaptosomes. Brain Res. 361, 373-382. Tatsumi, T. and Fliss, H. (1995) Hypochlorous acid mobilizes intracellular zinc in isolated rat heart myocytes. J. Mol. Cell. Cardiol. 26, 471-479. Thompson, M.A. and Ziff, EB. (1989) Structure of the gene encoding peripherin, an NGF-regulated neuronal-specific type III intermediate filament protein. Neuron 2, 1043-1053. Tomsig, J.L. and Suszkiw, J.B. (1990) Pb2*-induced secretion from bovine chromaffin cells: fura-2 as a probe for Pb“. Am. J. Physiol. 259, C762-C768. Traxinger, D.L. and Atchison, W.D. (1987) Reversal of methylmercury-induced block of nerve-evoked release of acetylcholine at the neuromuscular junction. Toxicol. Appl. Pharmacol. 90, 23-33. 171 Treves, S., Trentini, P.L., Ascanelli, M., Bucci, G., and DiVirgilio, F. (1994) Apoptosis is dependent on intracellular zinc and independent of intracellular calcium in lymphocytes. Exp. Cell Res. 211, 339-343. Tsien, R.Y., Pozzan, T., and Rink, T.J. (1982) Calcium homeostasis in intact lymphocytes: cytoplasmic free calcium monitored with a new intracellularly trapped fluorescent indicator. J. Cell. Biol. 94, 325-334. Tsokos-Kuhn, JD. (1989) Evidence in vivo for elevation of intracellular free Ca“ in the liver after diquat, acetaminophen and CC14. Biochem. Pharmacol. 38, 3061-3065. Tsokos-Kuhn, J.0., Smith, C.V., Mitchell, J.R., Tate, C.A., and Entman, ML. (1986) Evidence for increased membrane permeability of plasmalemmal vesicles from livers of phenobarbital-induced CCl4-intoxicated rats. Mol. Pharmacol. 30, 444- 451. Tymianski, M., Charlton, M.P., Carlen, PL, and Tator, CH. (1993) Source specificity of early calcium neurotoxicity in cultured embryonic spinal neurons. J. Neurosci. 13, 2085-2104. Vallee, BL. and Auld, D.S. (1990) Zinc coordination, function, and structure of zinc enzymes and Other proteins. Biochemistry 29, 5647-5659. Vallee, BL. and Falchuk, K.H. (1993) The biochemical basis of zinc physiology. Physiol. Rev. 73, 79-118. Vega, M.T., Villalobos, C., Garrido, B., Gandt’a, L., Bulbena, 0., Garcia-Sancho, J., Garcia, A.G., and Artalejo, AR. (1994) Permeation by zinc of bovine chromaffin cell calcium channels: relevance to secretion. Pflugers Arch. 429, 231-239. Verity, M.A., Brown, W.J., and Cheung M. (1975) Organic mercurial encephalopathy: in vivo and in vitro effects of methyl mercury on synaptosomal respiration. J. Neurochem. 25, 759-766. Verity, M.A., Sarafian, T., Pacifici, E.H.K., and Sevanian, A. (1994) Phospholipase A2 stimulation by methyl mercury in neuron culture. J. Neurochem. 62, 705-714. Vemino, S., Amador, M., Luetje, C.W., Patrick, J., and Dani, J.A. (1992) Calcium modulation and high calcium permeability of neuronal nicotinic acetylcholine receptors. Neuron 8, 1-20. ' Vignes, M., Guiramand, J., Sassetti, 1., and Récasens, M. (1993) Effect of thiol reagents on phosphoinositide hydrolysis in rat brain synaptoneurosomes. Eur. J. Neurosci. 5, 327-33. 172 Wang, Y.-X. and Quastel, D.M.J. (1990) Multiple actions of zinc on transmitter release at mouse end-plates. Pfliigers Arch. 415, 582-587. Weiss, J.H., Hartley, D.M., Koh, J., and Choi, D.W. (1993) AMPA receptor activation potentiates zinc neurotoxicity. Neuron 10, 43-49. Westbrook, G.L. and Mayer, ML. (1987) Micromolar concentrations of Zn2+ antagonize NMDA and GABA responses of hippocampal neurons. Nature (Lond.) 328, 640- 643. Wolf, S.G., Mosser, G., and Downing, K.H. (1993) Tubulin conformation in zinc-induced sheets and macrotubes. J. Struct. Biol. 111, 190-199. Xiang, J 2., Morton, J ., Brammer, M.J., and Campbell, 1.0 (1990) Regulation of calcium concentrations in synaptosomes: (X.,-adrenoceptors reduce free Ca2+ by closure of N-type Ca2+ channels. J. Neurochem. 55, 303-310. Xie X., Gerber U., Gahwiler B.H., and Smart TC. (1993) Interaction of zinc with ionotropic and metabotropic glutamate receptors in rat hippocampal slices. Neurosci. Lett. 159, 46-50. Yokoyama, M., Koh, J., and Choi, D.W. (1986) Brief exposure to zinc is toxic to cortical neurons. Neurosci. Lett. 71, 351-355. Zalewski, P.D., Millard, S.H., Forbes, I.J., Kapaniris, 0., Slavotinek, A., Betts, W.H., Ward, A.D., Lincoln, S.F., and Mahadevan, 1. (1994) Video image analysis of labile zinc in viable pancreatic islet cells using a specific fluorescent probe for zinc. J. Histochem. Cytochem. 42, 877-884. "‘llllllllllllllli