Abstract
Zinc ions (Zn2+) are stored in synaptic vesicles with glutamate in a number of regions of the brain. When released into the synapse, Zn2+ modulates the activity of various receptors and ion channels. Excitatory amino acid transporters (EAATs) maintain extracellular glutamate concentrations below toxic levels and regulate the kinetics of glutamate receptor activation. We have investigated the actions of Zn2+ on two of the most abundant human excitatory amino acid transporters, EAAT1 and EAAT2. Zn2+is a noncompetitive, partial inhibitor of glutamate transport by EAAT1 with an IC50 value of 9.9 ± 2.3 μm and has no effect on glutamate transport by EAAT2 at concentrations up to 300 μm. Glutamate and aspartate transport by EAAT1 are associated with an uncoupled chloride conductance, but Zn2+selectively inhibits transport and increases the relative chloride flux through the transporter. We have investigated the molecular basis for differential inhibition of EAAT1 and EAAT2 by Zn2+ using site-directed mutagenesis and demonstrate that histidine residues of EAAT1 at positions 146 and 156 form part of the Zn2+binding site. EAAT2 contains a histidine residue at the position corresponding to histidine 146 of EAAT1, but at the position corresponding to histidine 156 of EAAT1, EAAT2 has a glycine residue. Mutation of this glycine residue in EAAT2 to histidine generates a Zn2+ sensitive transporter, further confirming the role of this residue in conferring differential Zn2+ sensitivity.
Zn2+is found in many parts of the brain and has been reported to modulate glutamate receptors (Peters et al., 1987; Westbrook and Mayer, 1987), γ-aminobutyric acid and glycine receptors (Smart and Constanini, 1983; Laube et al., 1995), glutamate transporters (Balcar and Johnston, 1972; Gabrielsson et al., 1986; Spiridon et al., 1998), and the Na+K+ATPase (Hexum, 1974); thus, it may modulate neurotransmission in a number of different ways (reviewed by Frederickson, 1986). In particular regions of the brain, most notably in the mossy fibers of the hippocampus, Zn2+ is co-localized with glutamate in synaptic vesicles and is released into the synapse in a calcium-dependent manner; it may reach concentrations of up to 118 μm(Howell et al., 1984; Spiridon et al., 1998). The different glutamate receptor subtypes show differential sensitivity to Zn2+. The NMDA receptor subtype is potently inhibited by Zn2+ in a rapid and reversible manner, yet the kainate and quisqualate receptor subtypes are relatively insensitive to Zn2+ (Westbrook and Maher, 1987). Furthermore, Zn2+ has differential effects on different subtypes of NMDA receptors (Chen et al., 1997).
Balcar and Johnston (1972) and Gabrielsson et al. (1986)reported inhibition of [3H]glutamate transport in brain slices and synaptosomes, respectively, by moderate concentrations of Zn2+ (30–100 μm). However, in these experimental systems, Zn2+ may have a number of effects, including inhibition of Na+K+ATPase activity, which may indirectly reduce glutamate transport activity by disrupting the Na+ and K+gradients required for glutamate uptake. More recently, Zn2+ has been demonstrated to inhibit glutamate transport in Müller cells and cone cells of the salamander retina (Spiridon et al., 1998). The effects of Zn2+ in this study were rapid in onset and fully reversible, which suggests a direct role in transport inhibition. Zn2+ does not seem to compete with glutamate, sodium, potassium or protons binding to the transporter, which implies that Zn2+ allosterically modulates transporter function. In addition to inhibiting glutamate transport, Zn2+ seems to have opposing effects on the chloride conductance associated with the glutamate transporter in the two cell types investigated in the salamander retina. In Müller cells, the chloride conductance is increased by Zn2+; in the cone cells, however, the chloride conductance is inhibited by Zn2+. The transporter in Müller cells is likely to be similar in properties to EAAT1/GLAST1, whereas the cone-cell glutamate transporter is most similar to EAAT5 (Eliasof et al., 1997).
We have investigated the effects of Zn2+ on two of the most abundant human glutamate transporter subtypes, EAAT1 and EAAT2 (Arriza et al., 1994) expressed in Xenopus laevis oocytes. Zn2+ is a noncompetitive inhibitor of glutamate transport by EAAT1, but has no effect on transport by EAAT2. Zn2+ seems to cause an increase in the relative chloride component of the EAAT1 transport current while inhibiting the transport component. In addition, analysis of the Zn2+ sensitivity of point mutated transporters has been used to characterize the molecular basis for differential Zn2+ inhibition of glutamate transport by EAAT1 compared with EAAT2.
Materials and Methods
Chemicals.
l-glutamate-Na salt,d-aspartic acid, HEPES-hemi Na salt, gluconate salts and methanesulfonic acid were obtained from Sigma/Aldrich, Sydney, Australia. Zinc sulfate (analytical grade) was obtained from Standard Laboratories, Melbourne, Australia. Stock solutions of zinc sulfate in frog Ringer buffer were made up fresh on the day of use. All other buffer components were of analytical grade or acceptable for use in high performance liquid chromatography.
Expression of transporters and electrophysiological recording.
cDNAs encoding the human glutamate transporters, EAAT1 and EAAT2, were subcloned into pOTV for synthesis of RNA and expression in X. laevis oocytes as described previously (Arrizaet al., 1994). Site-directed mutagenesis was carried out using the Altered Sites kit (Stratagene, La Jolla, CA) as described in the manufacturer’s instructions. Two to seven days after RNA injection, current recordings were made at 22° with a Geneclamp 500 interfaced with an IBM-compatible computer using a Digidata 1200 A/D controlled by pCLAMP software (version 6.0.2; Axon Instruments). The standard frog Ringer recording solution contained 96 mmNaCl, 2 mm KCl, 1 mmMgCl2, 1.8 mmCaCl2, and 5 mm HEPES, pH 7.55. In experiments in which chloride was removed, the oocytes were dialyzed overnight in a chloride-free frog Ringer solution in which all the chloride was replaced with gluconate (96 mm Na-gluconate, 2 mm K-gluconate, 1 mm Mg (gluconate)2, 1.8 mm Ca (gluconate)2, and 5 mm HEPES, pH 7.55). It has been estimated that this procedure reduces the intracellular chloride concentration from 40 mm to 4 mm (Wadiche et al., 1995). The next day, recordings were made using a chloride-free Ringer solution in which the chloride was replaced with either gluconate, as described above, or methanesulfonate (96 mm methanesulfonic acid, 2 mm KOH, 1 mm MgSO4, 1.8 mm Ca-acetate, 5 mm HEPES and the pH adjusted with NaOH to pH 7.55). When recording using the chloride-free solutions, junction potentials were minimized by the use of a 3m KCl-agar bridge from the recording chamber to a reservoir containing 3 m KCl and Ag/AgCl electrodes connected to a bath clamp headstage. The oocytes were voltage clamped at −30 mV and the current-voltage relations were determined by subtraction of steady state current measurements in the absence of substrate, obtained during 200-msec voltage pulses to potentials between −100 mV and +50 mV, from corresponding current measurements in the presence ofl-glutamate or d-aspartate. In experiments concerned with the pH dependence of Zn2+inhibition of glutamate transport, the pH of the bath solution was adjusted with HCl. In some oocytes, Zn2+ blocked a current that was present in uninjected oocytes, water-injected oocytes, and also oocytes injected with either EAAT1 or EAAT2. Therefore, all base-line current measurements were made in the presence of the appropriate Zn2+ concentration so that only glutamate transport currents were measured.
Analysis of kinetic data.
Zn2+chelates glutamate; therefore, to estimate the free Zn2+ and free glutamate concentrations, we have used the correction methods described by Dawson et al.(1986) and also employed by Spiridon et al. (1998). When calculating values for IC50, EC50, and %Inhibition, we have used the free Zn2+ and free glutamate concentrations. Current (I) as a function of l-glutamate concentration ([S]) was fitted by least squares to I = Imax · [S]/(EC50 + [S]), where Imax is the maximal current and EC50 is the concentration ofl-glutamate generating half the maximal current. Imax and EC50 values are expressed as mean ± standard error and were determined by fitting data from individual oocytes. For Zn2+ dose responses, current (I) as a function of Zn2+concentration ([Zn2+]) was fitted by least squares to I = Imax(Z) − [(Imax(Z) · [Zn2+])/(IC50 + [Zn2+])] + Ir, where Imax(Z) is the maximal current generated byl-glutamate that is inhibited by Zn2+, IC50 is the concentration of Zn2+ that reduced Imax(Z) by 50% and Ir is the residual glutamate transport current in the presence of a maximal dose of Zn2+. %Inhibition = Imax(Z)/(Imax(Z)+Ir). IC50 and %Inhibition values are expressed as mean ± standard error.
Results
Application of l-glutamate to X. laevisoocytes expressing the human glutamate transporters, EAAT1 or EAAT2, generates dose-dependent inward currents when voltage clamped at −60 mV (Arriza et al., 1994). Co-application of 100 μm Zn2+ with 30 μmglutamate reduced the EAAT1 transport current but had no effect on EAAT2 (Fig. 1A). The inhibition of transport currents in EAAT1 are rapid in onset and fully reversible, which indicates a direct action of Zn2+ on the transporter. EC50 values for glutamate transport measured in the presence and absence of 100 μmZn2+ were similar (Table1) but the maximal transport current was decreased by approximately 50% in the presence of 100 μmZn2+ compared with its absence (Fig. 1B). This suggests that Zn2+ is acting as a noncompetitive inhibitor of glutamate transport. Inhibition of EAAT1 transport currents by Zn2+ is dose dependent with an IC50 value of 9.9 ± 2.3 μmand maximal inhibition of 59 ± 4% (Fig. 1D). In contrast, Zn2+, at concentrations up to 300 μm, had little or no effect on glutamate transport currents for oocytes expressing EAAT2 (Fig. 1, C andD). Thus, the actions of Zn2+ on EAAT1 seem to be similar (although less potent for EAAT1) to that observed for glutamate transporters of salamander retina Müller cells.
Glutamate transport is coupled to 2–3 Na+, 1 H+ (or possibly counter-transport of 1 OH−) and the counter-transport of 1 K+ (Kanner and Sharon, 1978; Barbour et al., 1988; Zerangue and Kavanaugh, 1996). The different transporter subtypes also allow a varying degree of chloride ion flux, which is thermodynamically uncoupled to the rate of glutamate transport (Fairman et al., 1995; Wadiche et al., 1995;Billups et al., 1996; Eliasof and Jahr, 1996). In salamander-retina Müller cells, Zn2+ seems to increase the chloride conductance activated by glutamate transport (Spiridon et al., 1998). We investigated the effects of Zn2+ on the chloride conductance activated by transport by EAAT1. d-aspartate transport by EAAT1 activates a significantly greater chloride conductance thanl-glutamate transport (Wadiche et al., 1995), which allows more accurate measures of the chloride conductance. Thed-aspartate transport current reverses direction at 3 ± 2 mV (n = 4), with the outward current caused by the uncoupled chloride flux (Wadiche et al., 1995). The membrane potential at which the current reverses direction may be used as an indicator of the chloride conductance relative to the transport component of the current. In the presence of 100 μmZn2+ the d-aspartate transport current reverses direction at −14 ± 4 mV (n = 4). This shift in transport current reversal potential toward the chloride reversal potential for X. laevis oocytes (∼ −20 mV; Barish, 1983; Fairman et al., 1995; Wadiche et al., 1995) suggests that in the presence of Zn2+, chloride ions contribute a larger proportion of the transport current compared with conditions in which Zn2+ is absent. At membrane potentials greater than the reversal potentials for d-aspartate, the net transport currents in the presence and absence of Zn2+ converge (Fig.2). This suggests that although the chloride conductance relative to substrate flux has increased, the absolute chloride conductance has not increased significantly.
The voltage dependence of glutamate transport block by Zn2+ was also measured. For these experiments, we used oocytes that had been dialyzed in chloride-free Ringer solution (gluconate substituted for chloride; see Materials and Methods for buffer details) to avoid chloride effects on the transport currents, which would otherwise create an apparent voltage dependence, separate from any Zn2+ effects. Zn2+inhibition of glutamate transport by EAAT1 was then measured using a chloride-free Ringer solution in which the chloride was replaced with either gluconate or methanesulfonate. After taking into account the chelation of Zn2+ by gluconate, as described byDawson et al. (1986), and Spiridon et al. (1998), the IC50 for Zn2+inhibition of 1 mm glutamate transport currents, measured using gluconate-substituted buffer solutions, was at least an order of magnitude greater than similar measurements using the chloride-containing frog Ringer solution. In contrast, the IC50 value measured using a buffer with methanesulfonate substituted for chloride was similar to IC50 measurements made in the presences of chloride. Therefore, we used the methanesulfonate-substituted frog Ringer solution to measure the voltage dependence of Zn2+ inhibition of glutamate transport. The IC50 for Zn2+ inhibition of glutamate transport was relatively constant over the membrane potential range of −100 mV to +50 mV, varying from 10.8 ± 2.6 μm at −100 mV to 15.7 ± 6.0 μm at +50 mV (Fig. 3). The %Inhibition was also relatively constant over this membrane potential range at around 60%. This suggests that the Zn2+ binding site on the transporter is not influenced by the electric field of the membrane and is likely to be near the external surface of the transporter. Again, similar voltage dependence of Zn2+inhibition of glutamate transport by Müller cells was observed, consistent with the identification of the transporter in this cell type as similar to EAAT1.
It has been demonstrated that a reduction in pH reduces Zn2+ inhibition of transport in Müller cells and that this is unlikely to reflect competition between Zn2+ and protons for a site on the transporter (Spiridon et al., 1998). Zn2+inhibition of glutamate transport currents for EAAT1 was similarly abolished at pH 6.0 (Fig. 4). This suggests that the protonation state of the Zn2+binding site on EAAT1 is altered at pH 6.0 compared with pH 7.5, which indicates the role of histidine and also possibly cysteine residues in forming part of the Zn2+ binding site (e.g., seeWang et al., 1995). To identify amino acid residues that may form part of a Zn2+ binding site on glutamate transporters and also explain the differential Zn2+ sensitivity of EAAT1 compared with EAAT2, we targeted histidine, cysteine, and negatively charged amino acid residues for mutagenesis using the following criterion: residues located in regions likely to be present within extracellular domains and regions that are distinct from the putative pore region of the transporters (carboxyl-terminal domain; Pines et al., 1995;Vandenberg et al., 1995; Kavanaugh et al., 1996;Mitrovic et al., 1998).
Histidine 146, of EAAT1, is conserved among all the cloned glutamate transporters, except EAAT3, and is located at the extracellular edge of the putative transmembrane domain 3 (Seal et al., 1996;Arriza et al., 1997; Wahle and Stoffel, 1997). Application of glutamate to oocytes expressing the EAAT1-H146A mutant generates dose dependent transport currents of similar magnitude and with an EC50 value similar to that of oocytes expressing wild type EAAT1. However, the EAAT1-H146A mutant shows marked reduction in sensitivity to Zn2+, with concentrations up to 1 mm required to cause any significant inhibition of the glutamate transport currents (Figs. 5 and6; Table 1). Thus, the H146A mutation selectively alters the Zn2+ sensitivity without affecting glutamate transport. Therefore, it is likely that in the wild type EAAT1 transporter, Zn2+ interacts with this histidine residue. Further evidence in favor of this conclusion comes from analysis of Zn2+ inhibition of an additional mutant, in which the histidine residue has been changed to an aspartic acid residue. Aspartic acid residues have also been identified at the Zn2+ binding sites of various zinc finger proteins; therefore, it was expected that this mutant might yield more detailed information about the Zn2+ binding site. The H146D mutant is significantly less sensitive to Zn2+ (IC50 = 199 ± 60 μm) than the wild type (IC50 = 9.9 ± 2.3 μm). These results demonstrate that subtle changes in the side chain of the amino acid residue at position 146 are sufficient to significantly alter the Zn2+ sensitivity of EAAT1.
The amino acid residue in EAAT2 at the position corresponding to histidine 146 of EAAT1 is also a histidine; therefore, the presence of this histidine residue alone cannot explain the differential Zn2+ sensitivity of EAAT1 compared with EAAT2. An additional EAAT1 mutant, H156A, showed reduced Zn2+ sensitivity with an IC50 value of 237 ± 37 μm(Figs. 5 and 6; Table 1), which is approximately 24-fold higher than for the wild type EAAT1. The maximal glutamate transport currents and the EC50 for glutamate of this mutant were not significantly different from that of wild type EAAT1 (Fig. 6; Table 1), which suggests that the mutation has specifically changed the Zn2+ binding site without altering the overall structure or expression of the transporter. Alignment of the amino acid sequences of the glutamate transporters shows some distinct differences between subtypes corresponding to this histidine residue (Arrizaet al., 1997). The amino acid residue in EAAT2 that corresponds to histidine 156 of EAAT1 is glycine; as such, the presence or absence of a histidine residue at this position may explain the differential Zn2+ sensitivity of glutamate transporter subtypes. We tested this hypothesis by introducing a histidine into EAAT2 at position 154 in place of the glycine residue, expecting to cause an increase in Zn2+sensitivity compared with wild type EAAT2. As predicted, Zn2+ inhibits glutamate transport of the EAAT2 mutant G154H (Fig. 5), with an IC50 value of 44 ± 7 μm and a maximal inhibition of 59 ± 3% (Fig. 6; Table 1). Further, Zn2+ (at 100 μm) decreases the Imax for glutamate transport currents without affecting the EC50 value for glutamate (Table 1), as observed for EAAT1. Thus, it may be concluded that the Zn2+ binding site on EAAT1 includes two histidine residues and the differential inhibition of glutamate transporter subtypes may be explained by the presence or absence of a histidine residue corresponding to position 156 of EAAT1. Additional mutations in EAAT1 that did not affect Zn2+ sensitivity include E86Q (extracellular loop 1); E184Q, C186A (extracellular loop 2); E303A (extracellular loop 3) (Table 1).
Discussion
This study has demonstrated that Zn2+is a noncompetitive, partial inhibitor of glutamate transport by EAAT1 and has no effect on glutamate transport by EAAT2. In addition, although Zn2+ inhibits transport by EAAT1, it seems to increase the relative chloride conductance of EAAT1. The results observed for EAAT1 are similar in many respects to the effects of Zn2+ on the salamander-retina Müller cell glutamate transporter (Spiridon et al., 1998). The major difference is that Zn2+ is less potent on EAAT1 than the Müller cell transporter. The cone cell glutamate transporter does show some differences with respect to the effects on the chloride conductance. The chloride conductance activated by glutamate transport in cone cells is inhibited whereas for EAAT1 and the Müller cell transporter the chloride conductance is increased. This discrimination by Zn2+ in modulation of the different transporter functions is particularly interesting. An explanation suggested by Spiridon et al.(1998) for the different effects on transport versus chloride conductance activation, is that Zn2+ may stabilize a conformation of the transporter that slows the passage of glutamate through the transporter. With glutamate being bound to the transporter for a longer period a more efficient activation of the chloride conductance may be achieved.
Molecular basis for Zn2+ modulation of EAAT1.
To better understand the molecular basis for the mechanism of action of Zn2+ on glutamate transporters, we have begun to characterize the Zn2+ binding site on EAAT1. In this study, we demonstrated that two histidine residues of EAAT1 at positions 146 and 156 are likely to form part of the Zn2+ binding site. Although the histidine residue at position 146 is conserved between glutamate transporter subtypes (except EAAT3), a histidine residue corresponding to position 156 of EAAT1 is found in only EAAT1 and EAAT4. We postulate that the nature of the amino acid residue at this position is an important determinant of Zn2+ sensitivity. Consistent with this suggestion is that a glycine-to-histidine substitution at this position in EAAT2 generates a Zn2+ sensitive EAAT2 mutant. EAAT5 has a glutamate residue at the position corresponding to histidine 156 of EAAT1; because glutamate residues have also been identified as forming Zn2+ binding sites on other proteins (e.g., Vazeux et al., 1996) it may be expected that Zn2+ would also interact with EAAT5. It would be of interest to confirm this prediction, because the glutamate transporter found in cone cells of the salamander retina is inhibited by Zn2+ (Spiridon et al., 1998) and is likely to be closely related to EAAT5 (Eliasof et al., 1997). Although this study has identified two amino acid residues required for the formation of the Zn2+ binding site on EAAT1 and also the molecular basis for differential sensitivity between transporter subtypes, it is anticipated, based on the structures of other zinc binding sites on other proteins (see Branden and Tooze, 1991) that at least one other amino acid residue is required to bind Zn2+.
Although the roles of individual amino acid residues involved in forming the pore region of the transporters are not well understood, the highly conserved carboxyl-terminal domain of glutamate transporters [residues 354–499 of EAAT1 (Arriza et al., 1997)] has been suggested to form a pore structure, containing binding sites for glutamate, sodium ions, potassium ions, and protons (Pines et al., 1995; Kavanaugh et al., 1996; Mitrovic et al., 1998). From the one-dimensional structure derived from the amino acid sequence, the Zn2+ binding site is distinct from the putative pore region of the transporter. In some respects this would be expected, given that Zn2+ is a noncompetitive modulator of transporter function. It will be of considerable interest to understand how the various regions of the transporters fold up in the three-dimensional structure such that occupation of an apparently distal site by Zn2+ is able to modulate the properties of the pore. One possibility is that although the third transmembrane domain may not be directly involved in forming the pore of the transporter, it might be closely associated with other domains that do form the glutamate, sodium ion, potassium ion, or proton binding sites. The binding of Zn2+ to the extracellular edge of the third transmembrane domain may alter the conformation of this transmembrane region, which may in turn alter the conformation of the pore to regulate substrate and ion passage through the transporter.
Physiological implications of the differential modulation of glutamate transporter subtypes by Zn2+.
The modulation of neurotransmission by zinc ions has been most extensively studied in the CA3 region of the rat hippocampus (e.g., Assaf and Chung, 1984;Frederickson, 1986). Although both GLAST1 (the rat equivalent of EAAT1) and glutamate transporter1 (GLT-1; rat equivalent of EAAT2) are expressed in the CA3 region of hippocampus, GLT-1 is the more abundant transporter (Rothstein et al., 1994; Chaudhry et al., 1995). Thus, the differential modulation of glutamate transport by Zn2+ ions in the CA3 region is unlikely to play a major role in shaping normal glutamatergic neurotransmission.
In the retina, however, GLAST1 (or EAAT1) is the most abundant transporter and is expressed in Müller cells, whereas GLT1 (or EAAT2) is present at much lower levels and seems to be expressed in bipolar cells (Rauen et al., 1996). Although zinc is found presynaptically in photoreceptors of the retina, the concentrations of extracellular zinc achieved are not well established (Wu et al., 1993; Spiridon et al., 1998). If extracellular zinc concentrations achieved in the retina are similar to those achieved for the CA3 region of the hippocampus, the differential sensitivity of glutamate transporters to inhibition by zinc ions may provide a very important role in modulating retinal neurotransmission and may also provide a novel mechanism for controlling or regulating the various intracellular pools of glutamate. Glutamate transport by Müller cells provides one of the major mechanisms for clearance of extracellular glutamate in the retina and inhibition of this process would alter the time course of glutamate within the synapse, which would presumably alter the kinetics of glutamate receptor activation. Glutamate taken up into Müller cells is converted to glutamine, which is then transported back to the neurons and converted back to glutamate. This pathway seems to be a major source of glutamate for bipolar and ganglion cells (Pow and Robinson, 1994). Therefore, the inhibition of EAAT1 (expressed by Müller cells) by Zn2+ will reduce uptake of glutamate into Müller cells and reduce the amount of glutamine that may be transported back into the neurons for subsequent conversion back to glutamate. This process would result in more glutamate uptake by other cells containing Zn2+-insensitive transporters. Thus, Zn2+ may provide an important regulator of the spatial regulation of both synaptic glutamate concentrations and intracellular pools of glutamate.
The inhibition of glutamate transport by Zn2+under pathological conditions may have a number of effects on glutamatergic neurotransmission. High concentrations of Zn2+, which would be generated under conditions such as ischemia, would decrease the rate of clearance of glutamate, which may be expected to potentiate glutamate excitotoxicity (Choiet al., 1987). However, these effects would have to be counterbalanced with the inhibition of NMDA receptors by Zn2+ (Westbrook and Mayer, 1987) and also the inhibition of reverse glutamate transport by Zn2+(Spiridon et al., 1998). Thus, a possible scenario is that, under mild ischemic conditions, although Zn2+ may prolong the clearance rate of glutamate from the synapse, the inhibition of NMDA receptors and the block of reverse glutamate transporter operation may limit the excitotoxic damage that may otherwise occur.
Acknowledgments
This article is dedicated to the memory of an inspirational teacher, the late Dr. Gregory Ralston. We are grateful to David Attwell for access to unpublished results. We thank Melanie Scott and Karin Aubrey for assistance in conducting various experiments, Macdonald Christie for helpful comments in the preparation of this manuscript, and Kong Li and Suzanne Habjan for expert assistance.
Footnotes
- Received January 16, 1998.
- Accepted April 2, 1998.
-
Send reprint requests to: Dr. Robert Vandenberg, Department of Pharmacology, University of Sydney, Sydney, NSW, 2006, Australia. E-mail: robv{at}pharmacol.usyd.edu.au
-
This work was supported by the Australian National Health and Medical Research Council and The Ramaciotti Foundation for Medical Research.
Abbreviations
- EAAT1
- excitatory amino acid transporter 1, EAAT2
- excitatory amino acid transporter 2
- HEPES, 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid
- NMDA
- N-methyl-d-aspartate
- GLAST
- glutamate aspartate transporter
- The American Society for Pharmacology and Experimental Therapeutics