Critical Role of Serine 465 in Isoflurane-induced Increase of Cell-surface Redistribution and Activity of Glutamate Transporter Type 3*

Glutamate transporters (also called excitatory amino acid transporters, EAATs) bind extracellular glutamate and transport it to intracellular space to regulate glutamate neurotransmission and to maintain extracellular glutamate concentrations below neurotoxic levels. We previously showed that isoflurane, a commonly used anesthetic, enhanced the activity of EAAT3, a major neuronal EAAT. This effect required a protein kinase C (PKC) α-dependent EAAT3 redistribution to the plasma membrane. In this study, we prepared COS7 cells stably expressing EAAT3 with or without mutations of potential PKC phosphorylation sites in the putative intracellular domains. Here we report that mutation of threonine 5 or threonine 498 to alanine did not affect the isoflurane effects on EAAT3. However, the mutation of serine 465 to alanine abolished isoflurane-induced increase of EAAT3 activity and redistribution to the plasma membrane. The mutation of serine 465 to aspartic acid increased the expression of EAAT3 in the plasma membrane and also abolished the isoflurane effects on EAAT3. These results suggest an essential role of serine 465 in the isoflurane-increased EAAT3 activity and redistribution and a direct effect of PKC on EAAT3. Consistent with these results, isoflurane induced an increase in phosphorylation of wild type, T5A, and T498A EAAT3, and this increase was absent in S465A and S465D. Our current results, together with our previous data that showed the involvement of PKCα in the isoflurane effects on EAAT3, suggest that the phosphorylation of serine 465 in EAAT3 by PKCα mediates the increased EAAT3 activity and redistribution to plasma membrane after isoflurane exposure.

Glutamate is the major excitatory neurotransmitter in the central nervous system. Similar to the case with many other neurotransmitters, there is no extracellular enzyme known to metabolize glutamate. Glutamate transporters, also called excitatory amino acid transporters (EAATs), 2 transport glutamate from extracellular space into cells (1,2). Thus, EAATs play a critical role in securing a high signal-to-noise ratio in glutamate neurotransmission and in preventing harmful over-stimulation of glutamate receptors under physiological conditions (2). Inhibition of EAATs has been shown to prolong the time course of glutamate neurotransmission (3), and decreased expression of EAATs is associated with neurodegeneration and increased infarct volume after brain ischemia (4,5). Five EAATs have been identified so far: EAAT1-5. They have about 520 -580 amino acids. In rats, EAAT1 and EAAT2 are expressed in glial cells, EAAT3 and EAAT4 are found in neurons and EAAT5 is located in the neurons and glial cells of retina (2). EAAT1, EAAT2, and EAAT3 are distributed in many brain regions including cerebral cortex, hippocampus, and cerebellum, whereas EAAT4 is predominantly expressed in the cerebellum (6). Thus, EAAT3 is the major neuronal EAAT in the central nervous system. All five EAATs are sodium co-transporters and require potassium coupling to complete the transporting cycling (7).
We have shown that isoflurane, a commonly used volatile anesthetic in clinical practice, increased EAAT3 activity (8,9). This increased activity required EAAT3 redistribution to the plasma membrane. These isoflurane effects were protein kinase C (PKC) ␣-dependent. The phosphorylation of EAAT3 appeared to increase after being exposed to isoflurane (9). Thus, we hypothesize that isoflurane-induced increase of EAAT3 activity and redistribution to the plasma membrane requires potential PKC phosphorylation sites in the EAAT3 molecule. To address this hypothesis, we performed site-directed mutagenesis of selective potential PKC phosphorylation sites and prepared cells stably expressing these mutants of EAAT3. Our results suggest that the serine 465 is essential for the isoflurane-increased EAAT3 activity and redistribution to the plasma membrane.

EXPERIMENTAL PROCEDURES
Materials-F-10 nutrient mixture (Ham's) and Opti-MEM I medium were from Invitrogen. Tissue culture flasks (25 cm 2 and 75 cm 2 ) and 6-well plates were manufactured by Corning (Corning, NY). L-[ 3 H]Glutamate (specific activity of 56 Ci/mM) was purchased from Amersham Biosciences. Wizard plus minipreps were from Promega (Madison, WI). QuikChange sitedirected mutagenesis kit was from Stratagene (La Jolla, CA). PIRES2-EGFP vector was from BD Bioscience. Lipofectamine 2000 was from Invitrogen. Protein A/G plus-agarose was purchased from Santa Cruz Biotechnology (Santa Cruz, CA).
Affinity-purified polyclonal rabbit anti-EAAT3 antibody raised against the C-terminal 14 amino acids of rat EAAT3 was from Alpha Diagnostics International (San Antonio, TX). Mouse anti-GFP antibody was purchased from BD Bioscience Clontech (Mountain View, CA). Sulfo-N-hydroxysulfosuccinimidobiotin, immunopure immobilized monomeric avidin, and Halt phosphatase inhibitor mixture were from Pierce. Fluorescent antibodies were purchased from Jackson Immuno-Research (West Grove, PA). Vectashield mounting medium was from Vector Laboratories (Burlingame, CA). Nitrocellulose membranes for Western blot were from Bio-Rad. Prolong antifade kit and Pro-Q Diamond dye were from Molecular Probes (Eugene, OR). Rat C6 glioma cells and COS7 cells were from American Type Culture Collection (Manassas, VA). G418 was from Calbiochem. Complete protease inhibitors (catalog number 1697498) were from Roche Diagnostics. Rat EAAT3 cDNA in BluescriptSKϪ was obtained from Dr. Mattias A. Hediger (Brigham and Women's Hospital, Harvard Institutes of Medicine). Other reagents were purchased from Sigma.
Preparation of COS7 Cells Stably Expressing EAAT3 or Its Mutants-As we described before (10), site-directed mutagenesis of EAAT3 was performed by using QuikChange site-directed mutagenesis kit with a pair of 28-base primers containing the desired mutations. The mutations were confirmed by sequencing ϳ500 bases that included the mutated sites. The wild type EAAT3 and EAAT3 mutants were inserted into the pIRES2-EGFP vector at the EcoRI and BamHI multiple cloning sites. This vector carries a kanamycin/neomycin resistance gene, a GFP gene and a cytomegalovirus promoter to control the expression of EAAT3. COS7 cells grown to 90% confluence in 6-well plates were transfected with 4 g of vector containing EAAT3 DNA in the presence of 10 l of Lipofectamine 2000 in 500 l of Opti-MEM I medium for 24 h. Three days after the transfection, COS7 cells were incubated with G418 500 g/ml for stringent selection. The surviving cells were subcultured in the presence of G418 500 g/ml, and the expression of GFP in the cells was observed under a fluorescent microscope. The expression of GFP and EAAT3 in these cells was confirmed by Western analysis.
Cell Culture-Rat C6 glioma cells that express endogenous EAAT3 were cultured in flasks in F-10 nutrient mixture (Ham's) containing 15% horse serum and 2.5% fetal bovine serum at 37°C in a 95% air-5% CO 2 incubator. When cells reached ϳ80% confluence, the culture medium was replaced by a serum-free medium (F-10 mixture only) for 24 h before isoflurane incubation.
COS7 cells were cultured in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum. They were exposed to isoflurane when they were at ϳ80% confluence and had been in the serum-free Dulbecco's modified Eagle's medium for 24 h.
Isoflurane Incubation-C6 cells and COS7 cells were incubated with isoflurane in an open system as follows. Fresh serum-free medium (50 -200 ml) that had been gassed with 95% air-5% CO 2 through or not through an isoflurane vaporizer at a flow rate 3 liters/min for 20 min was added to the cells for 5 min at 37°C. Preliminary experiments with gas chromatography showed that isoflurane concentrations in the medium reached equilibrium 5 min after the onset of gassing under the current experimental conditions. During the incubation the medium was continuously gassed with the carrier gases containing or not containing isoflurane to compensate for isoflurane loss from the solution to air.
Biotinylation-Biotinylation of cell surface proteins was performed as we described previously (9). After incubation with or without isoflurane, cells that were grown in 75-cm 2 tissue culture flasks were rinsed twice with warm phosphate-buffered saline (PBS) containing 0.1 mM calcium and 1.0 mM magnesium (PBS-Ca/Mg). The cells were then incubated with 2 ml of biotin solution (sulfo-N-hydroxysulfosuccinimidobiotin, 1 mg/ml in PBS-Ca/Mg) for 20 min at 4°C with gentle shaking. The biotinylation reaction was terminated by washing the cells three times with ice-cold PBS-Ca/Mg containing 100 mM glycine. After the cells were incubated in this wash solution for 45 min at 4°C with gentle agitation, they were then lysed in 2 ml of lysis buffer containing 100 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 1% sodium deoxycholate, 0.1% SDS, 1 g/ml leupeptin, 250 M phenylmethylsulfonyl fluoride, 1 g/ml aprotinin, 1 mg/ml trypsin inhibitor, and 1 mM iodoacetamide for 1 h at 4°C with vigorous shaking. The total lysates were centrifuged at 20,000 ϫ g for 20 min at 4°C to remove nuclei and debris. The resulting supernatants were incubated with equal volumes of suspension of avidin-conjugated beads (600 l of bead suspension to 600 l of lysates) for 1 h at room temperature with occasional stirring. The mixture was then centrifuged at 16,500 ϫ g for 15 min at 4°C. After being washed four times, each time with 1 ml of lysis buffer, the pellet that contained the biotinylated cell surface proteins was resuspended in 500 l of Laemmli buffer containing 62.5 mM Tris-HCl, pH 6.8, 2% SDS, 20% glycerol, and 5% 2-mercaptoethanol for 30 min to dissolve the biotinylated proteins. The mixture was centrifuged at 16,500 ϫ g for 10 min at 4°C and this third supernatant was kept for Western blot as the biotinylated fraction.
Western Blotting-After protein content in samples was quantitated by the Lowry assay using a protein assay kit, 25 g of protein per lane were subjected to Western analysis as described before (11,12). Briefly, proteins were separated with 10% SDS-PAGE and then were electrotransferred to nitrocellulose membranes. The protein bands were probed with primary antibodies (polyclonal rabbit anti-EAAT3 at 0.5 g/ml; polyclonal rabbit anti-GFP at 1:1000 dilution) and then a horseradish peroxidase-conjugated goat anti-rabbit or goat antimouse IgG secondary antibody (1:1000) and finally visualized by the enhanced chemiluminescence method with multiple exposures of films due to the limited linear range of intensity produced by this method. Quantitative analysis of the protein bands was performed using an ImageQuant 5.0 GE Healthcare Densitometer (GE Healthcare, Sunnyvale, CA). The relationship between the protein band signal and exposure time of the heaviest band on the films was established. Protein bands on a film where the intensity of the heaviest band was still within the linear range were measured to generate the data reported here.
Immunoprecipitation and Phosphoprotein Staining-As we described previously (9), C6 cells or COS7 cells cultured in 75-cm 2 dishes were lysed in 2 ml of buffer containing 50 mM Tris-HCl, 1% Nonidet P-40, 0.5% sodium deoxycholate, 50 mM NaCl, Halt phosphatase inhibitor mixture and complete protease inhibitors for 1 h at 4°C. The lysates were centrifuged at 14,000 ϫ g for 15 min to remove cell debris. The resulting supernatants were incubated overnight with 2 g of affinitypurified polyclonal rabbit anti-EAAT3 antibody at 4°C. The mixture was then incubated with 40 l of protein A/G plusagarose beads for 1 h at 4°C with gentle shaking. The sample was then centrifuged at 500 ϫ g for 2 min at 4°C. An aliquot of supernatant was saved for Western blot. The pellet containing bead-bound immune complexes was washed four times with the lysis buffer and the immune complexes were then eluted by incubation with 100 l of Laemmli buffer at 90 -95°C for 5 min. Control experiments using beads alone or rabbit IgG to replace the rabbit anti-EAAT3 antibody were performed to show the specificity of the anti-EAAT3 antibody. The prepared immunoprecipitates were separated with 10% SDS-PAGE and transferred to nitrocellulose membranes. The membranes were then stained with Pro-Q Diamond dye. The staining was visualized with a Typhoon 9400 Variable Mode Imager (Amersham Biosciences) at an excitation wavelength 546 nm and an emission wavelength 580 nm. This dye technology is very sensitive to stain phosphoserine-, phosphothreonine-, and phosphotyrosine-containing proteins.
Glutamate Uptake Assay-As described before (12,13), C6 cells or COS7 cells grown in 25-cm 2 flasks were washed twice with wash buffer containing 10 mM HEPES, 140 mM NaCl, 5 mM Tris base, 2.5 mM KCl, 2.5 mM CaCl 2 , 1.2 mM MgCl 2 , 1.2 mM K 2 HPO 4 , 10 mM dextrose, pH 7.2. They were then incubated with 10 M L-[ 3 H]glutamate in the wash buffer in the presence or absence of 2% isoflurane for 5 min at 37°C. The incubation was terminated by removing the incubation buffer and washing the cells three times with ice-cold wash buffer. The cells were lysed with 0.2 M NaOH, and radioactivity was measured in a liquid scintillation counter.
Statistical Analysis-The intensity of EAAT3 protein bands in the isoflurane treatment groups and EAAT3 mutant groups was normalized to that of EAAT3 bands from cells expressing wild type EAAT3 under control condition. The band intensity of phosphoproteins corresponding to EAAT3 was first normalized to that of total EAAT3 bands on Western blot. The results were then normalized to that from cells expressing wild type EAAT3 under control condition. Results are means Ϯ S.D. of the fold changes over the controls, with controls being set as 1.
Results of glutamate uptake assay are means Ϯ S.D. of the measured numbers in each sample. Statistical analysis was performed by unpaired t test or one-way analysis of variance followed by the Student-Newman-Keuls test for post hoc comparison as appropriate. A p Ͻ 0.05 was considered significant.

RESULTS
As we reported before (9,12), the anti-EAAT3 antibody detected a protein band with a mobility corresponding to ϳ60 kDa on SDS-PAGE in samples prepared from C6 cells and in COS7 cells transfected with EAAT3 DNA (Fig. 1). No protein band corresponding to EAAT3 was detected by the antibody in samples prepared from COS7 cells without transfection (data not shown). As predicted from the structure of the plasmid, the GFP and EAAT3 were not a fusion protein because no band was immunoreactive to both the anti-EAAT3 and anti-GFP antibodies in the Western analysis (data not shown).
Consistent with our previous results (9), a 5-min exposure of C6 cells to 2% isoflurane induced EAAT3 redistribution to the plasma membrane (Fig. 1). This isoflurane-induced redistribution was reproducible with wild type EAAT3 transfected in COS7 cells. To determine the involvement of potential PKC phosphorylation sites in the isoflurane effects on EAAT3, we performed site-directed mutagenesis on selective residues. PKCs phosphorylate serine/threonine residues in motif having basic residues both N-and C-terminal to the phosphorylation sites (14). According to this principle, we identified the three best potential PKC phosphorylation sites in the putative cytoplasmic domains of rat EAAT3 based on the primary sequence (15) and potential secondary structure. These three sites were: threonine 5, serine 465, and threonine 498. We mutated these residues to alanine, which cannot be phosphorylated by PKC, or to aspartic acid, which mimics the acidic charge of a phosphate. The mutation of threonine 5 or threonine 498 to alanine did not affect the isoflurane-induced EAAT3 redistribution to the plasma membrane. However, the mutation of serine 465 to alanine abolished isoflurane-induced EAAT3 redistribution (Fig.  1). The mutation of serine 465 to aspartic acid increased the expression of EAAT3 in the plasma membrane, although the total amount of this EAAT3 mutant expressed in the COS7 cells was similar to that of wild type EAAT3 in the COS7 cells (Fig. 2). Isoflurane exposure did not induce a redistribution of S465D EAAT3 to the plasma membrane (Fig. 1). These results suggest that the phosphorylation of serine 465 is critical for the isoflurane-induced EAAT3 redistribution.
We have shown previously that the isoflurane-induced increase in EAAT3 activity requires EAAT3 redistribution to the plasma membrane. Consistent with this notion, our glutamate transporter activity data from wild type and mutated EAAT3 (Fig. 3) mirrored the redistribution results, suggesting a critical role of serine 465 in isoflurane-induced increase of EAAT3 activity.
Our previous results suggest that isoflurane increases EAAT3 phosphorylation (9). Consistent with these results, isoflurane increased the expression of phosphoproteins that had the same mobility as EAAT3 in the anti-EAAT3 antibodyprepared immunoprecipitates from C6 cells or COS7 cells transfected to express wild type EAAT3, T5A, and T498A EAAT3 (Fig. 4). However, this isoflurane-induced increase of phosphoprotein was attenuated when immunoprecipitates were prepared from COS7 cells transfected with S465A or S465D EAAT3 (Fig. 4). These results suggest that the phosphorylation of serine 465 contributes to the isoflurane-induced increase of EAAT3 phosphorylation. Some phosphorylation staining of EAAT3 in control cells was observed. The reasons for this staining are not known, but it may represent phosphorylation at sites other than threonine 5, threonine 498, and serine 465 by various protein kinases.

DISCUSSION
Isoflurane has been a commonly used anesthetic for more than 20 years. However, its mechanism of action is not well defined. It has been proposed that anesthesia is induced by enhancing inhibitory neurotransmission and/or inhibiting excitatory neurotransmission (16). Available data strongly support enhancement of the inhibitory ␥-aminobutyric acid (GABA) neurotransmission in the anesthetic effects (17)(18)(19). However, anesthetics also inhibit glutamate neurotransmission and inhibition of glutamate receptors reduces volatile anesthetic requirements to render animals immobile (16,20). Thus, the effect of anesthetics on glutamate neurotransmission also contributes to anesthesia mechanism.
EAATs, via their glutamate transporting function, can regulate neurotransmission. Inhibition of EAAT activity prolonged glutamate-induced excitatory postsynaptic current (3). Inhibition of neuronal EAAT also decreased the inhibitory neurotransmitter GABA-mediated inhibitory postsynaptic current via a reduced synthesis of GABA because glutamate taken up by EAATs is a substrate for GABA synthesis (21,22). Our previous studies (8,9) showed that isoflurane as well as other volatile anesthetics halothane and sevoflurane increased activity of EAAT3, the major neuronal EAAT. Intrathecal injection of EAAT inhibitors in rats increased the concentration of isoflurane required to render immobility to these animals (23). Thus, the effects of volatile anesthetics on EAAT activity may contribute to their anesthesia mechanisms.
EAAT3 activity can be regulated by PKC. PKC␣ activation increased the EAAT3 activity and trafficking to the plasma membrane and PKC⑀ activation increased EAAT3 activity without increased EAAT3 trafficking to the plasma membrane (24). Our previous study showed that the effects of isoflurane  on EAAT3 were mediated by PKC␣ because isoflurane increased the EAAT3 activity and trafficking and PKC␣ downregulation abolished the isoflurane-increased EAAT3 activity and trafficking (9). To further understand the molecular basis for isoflurane regulation of EAAT3 activity, we mutated three potential PKC phosphorylation sites in EAAT3. Our results suggest that serine 465 plays a critical role in the isofluraneinduced increase in EAAT3 activity and in EAAT3 redistribution to the plasma membrane. These results, along with our previous results showing that the effects of isoflurane on EAAT3 activity and redistribution were PKC␣-dependent (9), suggest that the phosphorylation of serine 465 by PKC␣ is essential for isoflurane to increase EAAT3 activity and redistribution to the plasma membrane. Since isoflurane was shown to increase the amount of PKC␣ in the EAAT3 complex (9), our results also suggest that the effects of PKC␣ on EAAT3 may be direct and not through other intermediate proteins. All mutants maintained glutamate transporting functions, sug-gesting that these mutations do not change the binding of EAAT3 to glutamate, sodium, or potassium. The increased membrane localization and basal transporting activity of S465D compared with wild type EAAT3 suggest that serine 465 plays an important role in controlling the distribution of EAAT3 between the plasma membrane and the intracellular space.
A functional role of Ser-465 in EAAT3 has not been reported in the literature. However, a few studies have investigated other functional domains/sites in EAATs. Residues in C-terminal regions of EAATs contribute to substrate, inhibitor, and ion interactions (25)(26)(27)(28)(29)(30). For example, EAAT2 residues 396 -400 are critical for its transporting function and are the primary sodium binding sites (28,29). The same residues are at positions 365-369 in EAAT3 (31); however, the role of these residues in EAAT3 transporting functions has not been studied. It has been shown that arginine 445 controls the coupling between glutamate and cations in EAAT3 (32).
Phorbol 12-myristate 13-acetate (PMA), a PKC activator, has been shown to regulate the cell surface expression and activity of EAAT2 and EAAT3 but in apparent opposite directions: PMA increased the cell surface expression and activity of EAAT3 (8,9,33) and decreased the cell surface expression and activity of EAAT2 (10,34). The regulation of EAAT2 by PMA was mediated by a domain at residues 475-517. Mutation of all five serine residues simultaneously or Ser-486 singly to alanine in this domain partially reversed the PMA effect (34). A corresponding domain in EAAT3 at residues 444 -486 differs from the EAAT2 domain in 24 out of 43 amino acids. However, the role of this corresponding domain in PKC activation-induced EAAT3 redistribution to the plasma membrane could not be studied because the specific chimeras were not functional (34). Sequence analysis revealed that, except for Ser-465 that is found to mediate the isoflurane effects on EAAT3 in our study, the EAAT3 domain lacks the other 4 serine residues contained in the EAAT2 domain (15,35). A serine residue corresponding to the Ser-465 in EAAT3 exists in the other four EAATs as well. However, the Ser-465 in EAAT3 may be a better potential PKC phosphorylation site because it has a basic residue at both the N and C termini, and the serine residue in the other four EAATs has a basic residue only at the C terminus (15,(35)(36)(37)(38)(39). Mutation of the corresponding serine residue in EAAT2 (Ser-496) to alanine did not change the PMA effect on EAAT2 (34). We showed that isoflurane did not affect EAAT2 activity (10). Thus, phosphorylation of Ser-465 in EAAT3 may explain the specificity of isoflurane effects on this EAAT.
In a recent study, a short EAAT3 motif, 502 YVN 504 , was found to be necessary for platelet-derived growth factor-induced EAAT3 trafficking to the plasma membrane. This motif also contributed to PMA-induced EAAT3 trafficking. However, the Tyr residue in this motif was not phosphorylated under control and platelet-derived growth factor treatment conditions. Thus, the authors concluded that phosphorylation of Tyr in the motif was not a necessary step for EAAT3 trafficking induced by platelet-derived growth factor (40).
The PMA-induced increase in EAAT3 surface expression may be mediated by increased delivery of EAAT3 and/or reduced internalization of EAAT3. Our results suggest that EAAT3 phosphorylation is involved in these processes. How- . Isoflurane-induced EAAT3 phosphorylation. C6 glioma cells or COS7 cells transfected with wild type or mutated EAAT3 were incubated with 2% isoflurane for 5 min at 37°C. The total cell lysates were immunoprecipitated with an anti-EAAT3 antibody. The immunoprecipitates were separated with 10% SDS-PAGE and then were electrotransferred to nitrocellulose membranes that were stained with Pro-Q Diamond dye for phosphoproteins (top panel). The graphic presentation of the abundance of the phosphoproteins corresponding to EAAT3 as quantified by integrating the volume of autoradiograms from four separate experiments is shown in the bottom panel. Values in graphs are expressed as fold changes over the control and are presented as the means Ϯ S.D. *, p Ͻ 0.05 compared with the corresponding control. C, control; I, 2% isoflurane. ever, very little is known about the molecules involved in the cycling of neurotransmitter transporters and no study identifying the molecules required for EAAT cycling has been reported. Evidence has suggested that synataxin 1A may be required for delivery of the GABA transporter 1, serotonin transporters, and glycine transporters 1 and 2 to the plasma membrane (41)(42)(43) and that clathrin may be involved in the internalization of GABA transporter 1 (44). Biogenic amine transporters can be phosphorylated and the phosphorylation of transporters can change the functions of transporters (45,46). For example, phosphorylation at tyrosine 5 increased the transport function of the GABA transporter GAT1 (45). Future studies could be designed to identify which molecules may be involved in the EAAT cycling and how the phosphorylation of EAAT3 changes the interaction of EAAT3 with these molecules to change the dynamics of EAAT3 cycling.
Collectively, our studies suggest that isoflurane activates PKC␣ (9), which selectively phosphorylates EAAT3 at Ser-465 leading to EAAT3 translocation to the plasma membrane. PKC␣ also can induce AMPA receptor internalization (47), a process that is involved in synaptic plasticity (48). Since PKC␣ is implicated in the trafficking of both EAAT3 and AMPA receptors and isoflurane can affect synaptic plasticity (49), it is possible that PKC␣-mediated EAAT3 trafficking, too, have a role in synaptic plasticity. Consistent with this possibility, an earlier study showed that long term potentiation and contextual fear conditioning increased EAAT3 activity and trafficking to plasma membrane (50). Thus, our current results may be relevant to understanding the mechanisms of isoflurane-induced change of synaptic plasticity as well as isoflurane-induced anesthesia.