SEARCH

SEARCH BY CITATION

Keywords:

  • GLAST;
  • GLT-1;
  • Glutamate;
  • Rkt;
  • transport

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

In the nervous system, astrocytes express different ratios of the two glial glutamate transporters, glutamate transporter subtype 1 (GLT-1) and glutamate/aspartate transporter (GLAST), but little is known about the signaling pathways that independently regulate their expression. Treatment with dibutyryl-cAMP, epidermal growth factor (EGF) or other growth factors both induces expression of GLT-1 and increases expression of GLAST in astrocyte cultures. The induction of GLT-1 is correlated with morphological and biochemical changes that are consistent with astrocyte maturation. Pharmacological studies suggest that phosphatidylinositol 3-kinase (PI-3K) and the nuclear transcription factor-κB (NF-κB) may be involved in the induction of GLT-1 expression. In several signaling systems Akt, also known as protein kinase B (PKB), functions downstream of PI-3K. In these present studies we used lentiviral vectors engineered to express dominant-negative (DN), constitutively active (CA), or null variants of Akt to study the possible involvement of Akt in the regulation of GLT-1. Expression of DN-Akt attenuated the EGF-dependent induction of GLT-1. Expression of CA-Akt caused a dose- and time-dependent increase in GLT-1 protein, increased GLT-1 mRNA levels, increased dihydrokainate-sensitive (presumably GLT-1 mediated) transport activity, and caused a change in astrocyte morphology to a more stellate shape, but had no effect on GLAST protein levels. Finally, the expression of CA-Akt increased the expression of a reporter construct containing a putative promoter fragment from the human homolog of GLT-1, called EAAT2. From these studies, we conclude that Akt induces the expression of GLT-1 through increased transcription and that Akt can regulate GLT-1 expression without increasing GLAST expression in astrocytes.

Abbreviations used
CA-Akt

constitutively active Akt

dbcAMP

N6,2′-O-dibutyryladenosine 3′:5′-cyclic monophosphate

DHK

dihydrokainate

DMEM

Dulbecco's modified Eagle's medium

DN-Akt

dominant-negative Akt

EGF

epidermal growth factor

EGFP

GLT-1/EAAT2 promoter fragment with enhanced GFP

FBS

fetal bovine serum

GFAP

glial fibrillary acidic protein

GFP

green fluorescent protein

GLAST

glutamate/aspartate transporter

GLT-1

glutamate transporter subtype 1

GSK3β

glycogen synthase kinase 3β

HA

hemagglutinin

NF-κB

nuclear transcription factor κB

PACAP

pituitary adenylate cyclase-activating polypeptide

L-trans-PDC

L-trans-pyrrolidine-2,4-dicarboxylate

PI3-K

phosphatidylinositol 3-kinase

Glutamate is the predominant excitatory neurotransmitter in the mammalian CNS, but excessive activation of glutamate receptors can cause excitotoxicity (for a review, see Choi 1992). Extracellular concentrations of glutamate are controlled by a family of Na+-dependent transporters, including the astrocytic transporters GLT-1 (also called EAAT2) and GLAST (EAAT1), the neuronal transporters EAAC1 (EAAT3) and EAAT4, and the retinal transporter EAAT5. These transporters have critical roles in both controlling excitatory signaling and preventing an excitotoxic accumulation of extracellular glutamate (Sims and Robinson 1999; Danbolt 2001).

Several lines of evidence suggest that the glial glutamate transporters mediate the bulk of transport in the mammalian CNS. Mice deleted of GLT-1 display 5% of normal transport activity in synaptosomes prepared from cortex (Tanaka et al. 1997), antisense oligonucleotide knockdown of either GLT-1 or GLAST decreases the uptake activity in several brain regions (Rothstein et al. 1996), and GLT-1 has been estimated to represent up to 1% of total brain protein (Lehre and Danbolt 1998). There is evidence that expression of these transporters is differentially regulated during development (Furuta et al. 1997) and that astrocytes/glia express different ratios of these two transporters in the adult nervous system (Chaudhry et al. 1995). For example, GLAST expression is much higher in Bergman glia than in other populations of astrocytes (Rothstein et al. 1994). Expression of these transporters is altered in several animal models of neurodegenerative diseases, including traumatic brain injury (Rao et al. 1998), hypoxic/ischemic insults (Torp et al. 1995), epilepsy (Tanaka et al. 1997; Samuelsson et al. 2000), glioma (Ye et al. 1999) and amyotrophic lateral sclerosis (Trotti et al. 1999). The levels of GLT-1 and/or GLAST protein are also lower in brain tissue from patients with amyotrophic lateral sclerosis (Rothstein et al. 1995), epilepsy (Mathern et al. 1999), Alzheimer's disease (Li et al. 1997) and Huntington's disease (Lipton and Rosenberg 1994). However, the mechanisms underlying the dysfunction or loss of transporter in these diseases are poorly understood. Together, these studies suggest that the regulation of GLT-1 and GLAST will have both physiological and pathological consequences.

In previous studies, we and others found that astrocytes maintained in culture do not normally express GLT-1 (Swanson et al. 1997; Schlag et al. 1998). However, co-culturing neurons with astrocytes induces expression of GLT-1. This effect is mimicked by separating neurons from astrocytes either with a semipermeable membrane in a trans-well configuration (Schlag et al. 1998) or by treating astrocytes with neuron-conditioned medium (Gegelashvili et al. 1997; Zelenaia et al. 2000), suggesting that a secreted molecule contributes to the induction of GLT-1. This effect is also mimicked by treating astrocytes with dibutyryl-cAMP (dbcAMP), epidermal growth factor (EGF), or pituitary adenylate cyclase-activating polypeptide (PACAP) (Swanson et al. 1997; Schlag et al. 1998; Figiel and Engele 2000; Zelenaia et al. 2000). Many of these treatments increase GLT-1 mRNA levels and increase the expression of a reporter gene controlled by a fragment of the GLT-1 promoter (Zelenaia et al. 2000; Su et al. 2003; Sitcheran et al. 2005). Pharmacological evidence suggests that the effects of either neuron-conditioned medium, EGF, or dbcAMP depend on phosphatidylinositol 3-kinase (PI-3K) and the nuclear transcription factor-κB (NF-κB) (Zelenaia et al. 2000). NF-κB also directly interacts with and regulates a GLT-1 promoter (Sitcheran et al. 2005).

In several cellular systems, Akt (also called PKB, protein kinase B) is activated by growth factors, cytokines, hormones or neurotransmitters, and this activation frequently depends on PI-3K. Akt has been shown to increase NF-κB activity either by promoting the degradation of the IκBs that sequester NF-κB in the cytoplasm (Kane et al. 1999) or by activating the kinase that phosphorylates IκB (IκB kinase), leading to IκB degradation (Ozes et al. 1999).

Based on these observations, we hypothesized that Akt might regulate GLT-1 expression in astrocytes. We found that the expression of a dominant-negative variant of Akt (DN-Akt) reduced EGF-dependent increases in GLT-1 expression. In addition, a constitutively active variant of Akt (CA-Akt) induced expression of GLT-1. This effect was accompanied by an increase in the fraction of transport activity that was inhibited by the GLT-1 selective inhibitor, dihydrokainate. CA-Akt also increased GLT-1 mRNA and the expression of a reporter gene driven by a putative GLT-1 promoter fragment, providing strong evidence that these effects are, at least in part, a result of increased transcription. Unlike EGF and cAMP, which increase GLAST protein levels, CA-Akt had no effect on GLAST protein expression. The effects of CA-Akt were also associated with a change in astrocyte morphology from polygonal to stellate shape. Together, these studies identify Akt as a signaling molecule that may be important for cell-specific expression of GLT-1.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Materials

Fetal bovine serum (FBS) was obtained from Hyclone (Logan, UT, USA); all other cell culture reagents and TRIzol Reagent were from Invitrogen (Carlsbad, CA, USA). Poly-d-lysine, dbcAMP, the anti-glial fibrillary acidic protein (GFAP), and anti-actin antibodies were purchased from Sigma Chemical Co. (St Louis, MO, USA). l-[3H]Glutamate and [α-32P]deoxycytidine 5′-triphosphate were purchased from DuPont/NEN (Boston, MA, USA). Donkey anti-rabbit horseradish peroxidase, IgG, rainbow molecular weight markers, Ready-To-Go DNA Labeling Beads (-dCTP), and enhanced chemiluminescence (ECL) kits were obtained from Amersham (Arlington Heights, IL, USA). Positively charged nylon membranes were from Roche (Indianapolis, IN, USA). Immobilon-P membranes were from Millipore (Bedford, MA, USA). Dihydrokainate (DHK) was from Tocris (Ellisville, MO, USA). Mouse recombinant EGF was obtained from Collaborative Biomedical Products (Bedford, MA, USA) and dissolved in sterile de-ionized water. Vectashield Mounting Medium was obtained from Vector laboratories (Burlingame, CA, USA). The anti-GLAST and anti-GLT-1 antibodies, both directed against the carboxyl termini, were previously described (Rothstein et al. 1994; Furuta et al. 1997). The anti-Akt antibodies were from Cell Signaling Technology (Beverly, MA, USA). Living color A.V. peptide antibody against GFP was from BD Bioscience (Palo Alto, CA, USA). Anti-mouse IgG and IgM-fluorescein and anti-rabbit IgG-rhodamine conjugates were obtained from Jackson ImmunoResearch (West Grove, PA, USA). A2B5 hybridoma supernatant was a generous gift of Dr Judy Grinspan (Children's Hospital of Philadelphia, Philadelphia, PA, USA). Rabbit complement was purchased from ICN Biomedicals (Aurora, OH, USA).

The GLT-1 cDNA in pBluescript SK– was a generous gift of Dr Baruch Kanner (Hebrew University, Jerusalem, Israel). The GLAST cDNA was generated by reverse transcription polymerase chain reaction with specific primers and cloned into pBluescript SK– (Zelenaia et al. 2000). All Akt constructs were from Dr Morris Birnbaum (University of Pennsylvania, Philadelphia, PA, USA) and were engineered to express a hemagglutinin (HA) epitope tag. CA-Akt lacks the pleckstrin homology domain (Δ 4–129) and contains a Src myristoylation signal sequence, which results in constitutive activation upon expression in mammalian cells. The null variant of Akt lacks the pleckstrin homology domain and does not contain the Src myristoylation signal sequence (Kohn et al. 1996). The dominant-negative variant of Akt has three substitutions (K179A, T308A and S473A) and results in a phosphorylation-deficient and kinase-inactive protein (Wang et al. 1999). A 2.7-kb EcoRI fragment of the human EAAT2 promoter was used to control the expression of enhanced GFP, termed GFP (Rothstein et al. 2005). The Akt constructs and green fluorescent protein (GFP) were each subcloned into the lentiviral transfer plasmid pTY-CMV; the GLT-1/EAAT2 promoter fragment with enhanced GFP (EGFP) were subcloned into the same vector without the cytomegalovirus (CMV) promoter.

Preparation of lentiviral vectors

Lentiviral vectors were produced using modifications of recently published procedures (Karolewski et al. 2003; Watson and Wolfe 2003 Susarla et al. 2004). HEK-293T cells were maintained in poly-d-lysine-coated 15-cm diameter plates and triple-transfected with: a packaging plasmid pCMVΔ 8.2 (16.8 µg), an envelope plasmid JS-86 (5.6 µg), and the transfer plasmid (pTY-CMV) containing the transgene of interest (22.5 µg) using a Ca2+-phosphate transfection kit (BD Bioscience). After approximately 8–12 h, the cell culture medium was replaced. For the next 2–3 days, the medium containing virus was collected, filtered through a 0.45-µm filter, and concentrated by centrifugation at 50 000 g for 2 h at 4°C. The virus-containing pellet was resuspended in 1/36th of the original volume of ice-cold growth medium and stored in aliquots at −80°C until being used later in the experiment (the pellet was stored for less than three months). Expression of DN-Akt was verified by western blot of cell lysates from HEK293 cells used to produce the lentiviral vector, but expression of transgene in astrocytes was not consistently observed. Each lot of CA- or null Akt virus was directly titered by infection of astrocytes with various dilutions of concentrated virus, and protein expression was measured by western blot. From these studies, we found that a 1× dilution of CA-Akt (equivalent to diluting 1 mL of concentrated viral stock into 36 mL of cell culture medium) generally increased the total Akt immunoreactivity by 2–3-fold above that observed in untreated astrocytes. Higher concentrations of the supernatant containing the lentivirus engineered to express the null variant of Akt (5–10×) were consistently required to achieve comparable levels of either Akt or HA immunoreactivity.

Preparation of astrocyte cultures

Astrocytes were prepared from the cortices of neonatal rats (1–2-days old) as previously described with minor modifications (Zelenaia et al. 2000). Cells were plated at a density of 4.5 × 104 cells/cm2 onto 10-cm diameter culture dishes and maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% FBS, 10% Ham's F-12, 100 U of penicillin/mL and 100 µg of streptomycin/mL at 37°C with 5% of CO2. The growth medium was completely exchanged with fresh medium every 3–4 days until they were nearly 100% confluent (12–14 days). Although these cultures are enriched in GFAP-positive cells, they also contain A2B5 positive oligodendrocyte precursors that express GLT-1 (Zelenaia et al. 2000). To eliminate the A2B5-positive cells, cultures were treated with an anti-A2B5 antibody and rabbit complement for 45 min as previously described (Zelenaia et al. 2000). The cultures were washed three times with HEPES-buffered saline solution, incubated for 24 h in growth medium and then replated onto either 6-well or 12-well plates with a slight dilution (∼10–20%) based on surface area. Between 1 and 2 days later, astrocytes were either infected with lentiviral vectors or treated with other compounds. To infect astrocytes, virus was thawed on ice, mixed with growth medium, and gently added into the astrocyte cultures. After an overnight incubation, cells were rinsed once with plain DMEM, and then fresh medium was added. Cells were fed with fresh medium every 3–4 days until completion of the experiment.

Western blot analyses

Cells were lysed and harvested as previously described (Zelenaia et al. 2000). Protein was measured using the Lowry protein assay (Lowry et al. 1951) and equal quantities of protein were loaded onto 10% sodium dodecyl sulfate (SDS)/polyacrylamide gels. Immobilon P membranes were probed with antibodies directed against the carboxyl terminus of GLT-1 (1 : 10 000), the carboxyl terminus of GLAST (1 : 75), the carboxyl terminus of Akt (1 : 1000), the phosphorylated form (Ser473) of Akt (1 : 1000), actin (1 : 5000), GFP (1 : 100) or in some cases combinations of these antibodies. As was previously reported (Haugeto et al. 1996), glutamate transporter immunoreactivity was generally observed at molecular weights that were consistent with monomers and multimers. The blots were visualized with enhanced chemiluminescence. The density of immunoreactive bands was quantified with NIH Image 1.62. The effects of treatments on actin immunoreactivity were examined in every experiment, and there was no evidence for effects of either EGF or lentiviral vectors on actin levels. Therefore, all data were normalized to actin as a loading control. This had no effect on the conclusions, other than slightly reducing the variance in some experiments.

Measurement of l-[3H]glutamate transport

Sodium-dependent transport of l-[3H]glutamate was measured as previously described (Garlin et al. 1995). Briefly, triplicate assays were performed in either Na+-containing or choline-containing buffers in a 12-well plate. Na+-dependent transport activity was calculated as the difference in radioactivity accumulated in the presence and absence of Na+.

Northern blot analyses

Total RNA was extracted from astrocytes using the TRIzol reagent, following the protocol provided by the manufacturer (Invitrogen). RNA samples were separated on 1% agarose/6.6% formaldehyde gels in 1× 3-(N-morpholino) propanesulfonic acid buffer. RNA was transferred to a positively charged nylon membrane and immobilized by baking at 80°C for 2 h. After prehybridization for 2 h at 65°C, membranes were incubated with the specific cDNA probes at 65°C for 16–20 h. The EcoRI/XbaI fragment of the GLT-1 cDNA clone (1.8 kb), the XhoI/XbaI fragment of the GLAST cDNA clone (1.6 kb), and the BamHI–HindIII fragment of rat cyclophilin cDNA (0.7 kb) were used as specific probes for the corresponding mRNAs. cDNA probes were radiolabeled with [α-32P]deoxycytidine 5′-triphosphate by random priming with Ready-To-Go DNA Labeling Beads (–dCTP) kit. After sequential washes in 2× standard saline citrate to 0.1× standard saline citrate, membranes were exposed to a Phosphoimager screen or to film for 12–36 h. Radioactivity was quantified with a Phosphoimager SI with the ImageQuant analysis program (Molecular Dynamics, Sunnyvale, CA, USA). Data were normalized to cyclophilin mRNA.

Immunocytochemistry

Immunocytochemistry was performed as previously described (Zelenaia et al. 2000). Cells plated on sterile glass coverslips coated with poly-d-lysine (50 µg/mL) were fixed with 4% paraformaldehyde in phosphate-buffered saline (PBS). After incubation with primary antibodies (mouse monoclonal anti-GFAP 1 : 100 and rabbit anti-GLT-1 1 : 100) overnight at 4°C, coverslips were washed with blocking buffer and incubated with anti-mouse IgG-fluorescein and anti-rabbit IgG rhodamine conjugates and then subjected to confocal microscope analysis.

Results

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Lentiviral transduction of astrocytes

In previous studies, we tested a number of transfection procedures to introduce foreign genes into astrocytes and found that no more than 10–20% of cells expressed reporter proteins (GFP; Zelenaia et al. 2000). In the present study, we examined the efficiency of transduction of astrocytes using a lentiviral vector expressing GFP (1×) and found that at least 80% (generally closer to 90%) expressed robust green fluorescence within 3–4 days and that this expression did not diminish for as long as the astrocytes were maintained in culture (for at least two weeks, data not shown, replicated in at least three independent experiments). Furthermore, there was no evidence of astrocyte toxicity with the highest dose tested (10×); the lentiviral vector engineered to express GFP did neither cause any obvious changes in astrocyte morphology nor result in the appearance of cellular debris in the medium. Although not rigorously tested, these observations suggest that the lentivirus did not cause robust astrocyte death under these conditions.

Effects of DN-Akt on EGF- or dbcAMP-dependent regulation of GLT-1

As was previously observed (Schlag et al. 1998; Zelenaia et al. 2000), we found no significant expression of GLT-1 in untreated astrocytes, and EGF consistently induced GLT-1 expression (Fig. 1a). EGF is known to activate Akt in several systems (for a review, see Datta et al. 1999). To determine whether the effects of EGF on GLT-1 expression might be dependent on Akt activation, astrocytes were either infected with a lentiviral vector engineered to express DN-Akt or infected with a lentiviral vector engineered to express the null variant of Akt. Two days later either uninfected or infected astrocytes were treated with EGF for two weeks. In astrocytes infected with lentivirus engineered to express either DN-Akt or the null variant of Akt, there was no evidence of GLT-1 expression, but the DN-Akt attenuated the EGF-dependent induction of GLT-1. This effect was dose-dependent, but at the highest dose tested, DN-Akt only reduced the EGF-dependent induction of GLT-1 to approximately 60% of that observed in uninfected EGF-treated astrocytes (Figs 1a and b). Akt phosphorylates glycogen synthase kinase 3β (GSK3β) in many cellular systems (for example, see Cross et al. 1995). Therefore, in these same experiments, the levels of phospho-GSK3β were examined. EGF increased the level of phospho-GSK3β in uninfected astrocytes, and this increase in phospho-GSK3β was attenuated in astrocytes infected with the lentivirus engineered to express DN-Akt (Fig. 1a, lower panel). Expression of the null variant of Akt (a control vector) had no effect on EGF-dependent increases in either GLT-1 or phospho-GSK3β.

image

Figure 1.  Effects of DN-Akt on either EGF- or dbcAMP-dependent induction of GLT-1 expression. (a) Astrocytes were infected with different doses of lentiviral vectors engineered to express DN or null variants of Akt. Two days later they were either treated with vehicle or with EGF (30 ng/mL). After 14 days, GLT-1, actin (not shown) and phospho-GSK3β (P-GSK) were examined by western blot. In this and subsequent figures, transporter immunoreactivity is observed as both monomers and multimers. Expression of the null variant of Akt had no effect on P-GSK3β (data not shown). (b) Summary of results from between six and eight independent experiments with either DN-Akt or the highest dose of the null variant of Akt. Effects of EGF were compared in uninfected astrocytes (Untr.) and in astrocytes infected with lentiviral vector engineered to express DN-Akt. **p < 0.01, ***p < 0.001 compared with EGF-treated, uninfected astrocytes. (c) In parallel experiments, astrocytes were infected with lentiviral vectors engineered to express the dominant-negative variant of Akt. Two days later they were either treated with vehicle or dbcAMP (250 µm). After 14 days, GLT-1 was examined by western blot. Western blot shows the effects of DN-Akt. This experiment was reproduced in four independent experiments.

Download figure to PowerPoint

There is evidence that the effects of EGF and dbcAMP on GLT-1 expression may be dependent upon activation of at least some of the same signaling molecules, including both PI3-K and NF-κB (Zelenaia et al. 2000; Su et al. 2003). To determine whether the effect of DN-Akt on EGF-dependent increases in GLT-1 expression extends to the effects of dbcAMP, astrocytes infected with DN-Akt lentivirus were treated with dbcAMP in a parallel set of experiments. In these studies, expression of DN-Akt had no effect on dbcAMP-dependent induction of GLT-1 (Fig. 1c). Together, these data suggest that the effects of EGF on GLT-1 expression are at least in part dependent upon Akt activation and that the effects of dbcAMP are independent of Akt.

Although the DN-Akt lentiviral vector had effects on both the induction of GLT-1 and on the phosphorylation of a down-stream target (GSK3β), we were unable to consistently document the expression of DN-Akt protein in astrocytes. We recently used this same DN-Akt construct in a different set of studies to examine the role of Akt in trafficking of EAAC1 (Krizman-Genda et al. 2005). In these studies, we had no difficulty observing protein by western blot using an antibody against the HA epitope tag after transient transfection in C6 glioma. To rule out the possibility that mutations were introduced during subcloning into the lentiviral vector, both sequences were compared and found to be identical. As another control, robust immunoreactive bands were also consistently observed in lysates from the HEK293 cells used to produce the lentiviral vector (data not shown). Finally, we considered the possibility that expression of DN-Akt in HEK293 cells might reduce the efficiency of virus production; HIV-1 p24 coat protein was measured in the resultant viral stocks by ELISA, and comparable levels of p24 were observed with the three lentiviral vectors. Therefore, it appears that DN-Akt protein may be less stable than either CA-Akt or null Akt in astrocytes, but this was not pursued further.

Effects of CA-Akt on both GLT-1 and GLAST protein

To determine whether activated Akt is sufficient to induce GLT-1 expression, astrocytes were infected with a lentivirus engineered to express either CA-Akt or a null variant of Akt, and GLT-1 protein was examined by western blot. In these studies, expression of exogenous Akt was verified by western blot. As predicted, both variants of Akt migrated as protein bands of either ∼45 kDa or ∼15 kDa smaller than wild-type Akt (Fig. 2a, middle panel; Fig. 2b, lower panel). The effect of EGF was mimicked by infection of astrocytes with CA-Akt, but the induction observed after one week was somewhat modest compared with the effects of EGF (Fig. 2a, upper panel). As viral infection can influence astrocytic expression of glutamate transporters (Yu et al. 1998; Wang et al. 2003), astrocytes were also infected with two control lentiviral vectors, one engineered to express a null variant of Akt and one engineered to express GFP; neither of these vectors resulted in increased expression of GLT-1 (Fig. 2a). In these experiments, phosphorylated Akt, a surrogate measure of Akt activation, was only observed in cells infected with lentiviral vector that expresses CA-Akt; there was no evidence of activation/phosphorylation of endogenous Akt after 7 days (note the absence of an immunoreactive band at ∼60 kDa in the lower panel of Fig. 2a).

image

Figure 2.  Effects of CA-Akt on GLT-1 expression. (a) Astrocytes were infected with the lentivirus engineered to express CA-Akt (1× dose), null-Akt (3× dose) or GFP (1× dose). Uninfected astrocytes were also treated with either dbcAMP (250 µm) or EGF (30 ng/mL). After seven days, expression of GLT-1, exogenous Akt (Exo Akt), or phospho-Akt (p-Akt) were analyzed by western blot (40 µg of cell lysate protein per lane; *except 10× were 20 µg of cell lysate was loaded). These results are representative of three independent experiments. (b) Representative western blot showing the effects of increasing doses of CA-Akt on GLT-1 expression. Astrocytes were either infected with increasing doses of lentivirus engineered to express CA-Akt or treated with vehicle or EGF (30 ng/mL). After 7 days, GLT-1, actin or Akt immunoreactivity were examined by western blot (35 µg of cell lysate protein per lane). Note that the anti-Akt antibody recognizes both the endogenous Akt (End.-Akt) (∼60 kDa) and CA-Akt (∼45 kDa). (c) Summary of the effects of increasing doses of CA-Akt on GLT-1 expression. GLT-1 immunoreactivity was normalized to actin and expressed as a fraction of that observed in EGF-treated astrocytes. Data are mean ± SEM of four independent observations, except 10×, which is the mean of two observations that were within 15% of the mean. (d) Western blot showing no effect of increasing dose of virus engineered to express GFP on GLT-1 expression. Astrocytes were either infected with increasing doses of lentivirus engineered to express GFP or treated with vehicle (Untr.) or EGF (30 ng/mL). After 7 days, either GLT-1 or GFP immunoreactivity were examined by western blot (35 µg of cell lysate protein per lane). Data are representative of two independent experiments.

Download figure to PowerPoint

When astrocytes were infected with increasing quantities of the lentiviral vectors engineered to express CA-Akt, a dose-dependent increase in the amount of CA-Akt immunoreactivity was observed 1 week after infection (Fig. 2b, lower panel). This effect was associated with an increase in the quantity of GLT-1 protein expression (Fig. 2b, upper panel and Fig. 2c). Similar experiments were also conducted using the lentivirus engineered to express GFP. In these studies, there was a dose-dependent increase in GFP expression (see Fig. 2d, lower panel), and there was no evidence of GLT-1 expression (Fig. 2d, upper panel). At the highest dose of CA-Akt, astrocyte morphology changed faster than that observed with either lower doses of virus or with EGF treatment (data not shown). After about 10 days, the cell bodies and processes of astrocytes infected with 10× CA-Akt swelled, and within a few days many of the cells died. At lower doses of CA-Akt, there was no evidence of astrocyte death based on either the presence of floating cellular debris or the quantity of protein observed after cell harvest. No cell death was observed with the GFP-expressing vector, suggesting that expression of CA-Akt, rather than infection with lentivirus, can cause astrocyte death at the highest dose.

The time-dependence for the effect of CA-Akt on GLT-1 expression was examined at a 1× dose of CA-Akt; the effects of the null variant of Akt (5–10×) were examined in parallel (Fig. 3). Under these conditions, the levels of exogenous Akt were comparable in astrocytes infected with the CA or null variants of Akt (exogenous) and approximately 2–3-fold higher than the levels of endogenous Akt (Fig. 3a, middle panel). Although the levels of exogenous Akt were stable throughout the experiment, the levels of phospho-Akt were maximal at one week, declining thereafter (Fig. 3a, lower panel). In contrast, the levels of GLT-1 protein continued to increase at the longest time point examined (3 weeks). This suggests that the induction of GLT-1 expression is markedly delayed relative to Akt activation. The effects of EGF on GLT-1 protein were also somewhat delayed with protein levels still increasing at 3 weeks. In cells expressing the null variant of Akt, there was no significant expression of GLT-1. This provides strong evidence that the effects of CA-Akt cannot simply be attributed to a non-specific effect of infection of astrocytes with lentivirus.

image

Figure 3.  Time-dependence of effects of CA-Akt on expression of GLT-1 in astrocytes. (a) Representative western blot of effects of CA- or null-Akt on GLT-1 protein (upper panel), endogenous (End.) or exogenous (Exo.) Akt (middle panel), or phospho-Akt (p-Akt) (lower panel). Astrocytes were infected with lentivirus expressing either CA-Akt (1×) or null-Akt (5–10×) and maintained in culture for varying lengths of time. Sister cultures were also either left untreated (Untr.) or incubated with EGF (30 ng/mL). Twenty µg of total cell lysate was loaded in each lane. (b) Summary of effects of CA-Akt on GLT-1 expression. GLT-1 protein levels were normalized to actin and expressed as a fraction of the level of immunoreactivity observed in cells infected with CA-Akt for three weeks. Data are the mean ± SEM of four independent observations.

Download figure to PowerPoint

In previous studies, induction of GLT-1 expression with dbcAMP, with EGF or by co-culturing with neurons was accompanied by a change in astrocyte morphology from a polygonal shape to a complex process-bearing morphology, which can be easily visualized with GFAP immunohistochemistry (Swanson et al. 1997; Schlag et al. 1998; Zelenaia et al. 2000). In either untreated astrocytes or astrocytes infected with lentivirus engineered to express the null variant of Akt, GFAP immunoreactivity displayed a diffuse pattern, and only background levels of GLT-1 immunoreactivity were observed (Fig. 4, data not shown for null variant of Akt). Expression of CA-Akt for two weeks was accompanied by a dramatic change in cellular morphology, similar to that previously observed with EGF treatment (Zelenaia et al. 2000), and GLT-1 immunoreactivity was co-localized with many but not all GFAP-positive cells.

image

Figure 4.  Double-label immunohistochemistry of GFAP and GLT-1 in astrocytes expressing CA-Akt. Astrocyte cultures plated on glass coverslips were infected with 1× CA-Akt and maintained in culture for 2 weeks. After staining, immunoreactivity was examined in optical cross-sections of cells using confocal microscopy. These results were reproduced in three independent experiments.

Download figure to PowerPoint

In earlier studies, treatment of astrocytes with EGF, dbcAMP or PACAP resulted in both an induction of GLT-1 expression and increased GLAST expression (Schlag et al. 1998; Figiel and Engele 2000; Zelenaia et al. 2000). To determine whether the effects of CA-Akt are generalized to GLAST, the same astrocyte samples that had been examined for the effects of CA-Akt on GLT-1 expression were tested for the effects on GLAST protein expression by western blot (Fig. 5). In these studies, transduction of astrocytes with a 1× dose of CA-Akt had no affect on GLAST expression even after three weeks (Figs 5a and b). As a positive control, the effects of EGF on GLAST protein were examined. Similarly, increasing levels of CA-Akt expression had no effect on GLAST protein levels (Fig. 5c). As has been previously reported, no EAAC1 protein was observed in untreated cultures, and CA-Akt did not induce EAAC1 expression (data not shown, n = 2, using brain tissue as a positive control for the western blot). These studies provide strong evidence that CA-Akt has selective effects on GLT-1 expression without affecting GLAST protein levels.

image

Figure 5.  Effects of CA-Akt on expression of GLAST. (a) Representative western blot of effects of CA- or null-Akt on GLAST protein. Astrocytes were infected with lentivirus expressing either CA-Akt (1×) or null-Akt (5–10×) and maintained in culture for varying lengths of time. Sister cultures were also either left untreated (Untr.) or treated with EGF (30 ng/mL). Twenty µg of total cell lysate was loaded in each lane. (b) Summary of effects of CA-Akt on GLAST expression. GLAST protein levels were normalized to actin and expressed as a fraction of the amount of immunoreactivity observed in cells infected with CA-Akt for three weeks. Data are the mean ± SEM of four independent observations. (c) Representative western blot showing the lack of effect of increasing doses of CA-Akt on GLAST protein levels. Thirty five µg of total cell lysate was loaded in all lanes except with 10× virus (*20 µg of lysate was loaded). All of the measures of GLAST expression were performed using the same samples as those analyzed for GLT-1 expression in Fig. 3.

Download figure to PowerPoint

Effects of CA-Akt on glutamate transport activity

To determine whether the effects of CA-Akt on GLT-1 expression were accompanied by functional changes in transport activity, Na+-dependent accumulation of glutamate was measured in untreated astrocytes and in astrocytes transduced to express CA-Akt. Given the fact that GLT-1 protein increased by CA-Akt and GLAST protein levels are unaffected, we predicted that CA-Akt would be likely to increase the total transport activity and would increase the sensitivity of transport activity to pharmacological agents that selectively inhibit GLT-1. Surprisingly, transduction with CA-Akt (1× for 2 weeks) reduced transport activity to ∼50% of control levels measured either at low (0.5 µm, 46 ± 9%, n = 5) or high (300 µm, 53 ± 11%, mean n = 3) concentrations of glutamate. As the Km value for glutamate transport activity in control astrocytes is typically ∼50 µm (Garlin et al. 1995; Schlag et al. 1998; Zelenaia et al. 2000), these data suggest that expression of CA-Akt reduces the Vmax for transport activity. Because the cell surface availability of many transporters is regulated by a variety of signaling pathways, the effects of CA-Akt on biotinylated GLT-1 and GLAST were examined as previously described (Kalandadze et al. 2004; Susarla et al. 2004). In these studies, at least 85% of GLAST immunoreactivity was found in the biotinylated/cell-surface fraction, and the quantities of protein in either the biotinylated or non-biotinylated fractions did not differ between either control astrocytes or astrocytes expressing CA-Akt (data not shown, two independent experiments; actin was also examined as a control to ensure that cells were not lysed in these experiments). Similarly, greater than 80% of the GLT-1 immunoreactivity was observed in the biotinylated fraction in astrocytes expressing CA-Akt. These studies suggest that most of the GLAST and GLT-1 found in lysates in these experiments are at the plasma membrane.

The glutamate analog DHK selectively inhibits GLT-1-mediated transport activity with an IC50 value of 20–100 µm, whereas GLAST-mediated transport activity is essentially unaffected by concentrations of up to 10 mm (Arriza et al. 1994; Garlin et al. 1995; Tan et al. 1999). Although 100 µm DHK had no effect on transport activity in untreated cultures, it significantly inhibited the transport activity measured in astrocytes expressing CA-Akt (Fig. 6). To rule out the possibility that this increased sensitivity to DHK is caused by a generalized increase in the sensitivity of transport activity to all inhibitors, the effects of a non-selective inhibitor of Na+-dependent glutamate transport activity, l-trans-pyrrolidine-2,4-dicarboxylate (l-trans-PDC), were compared both in untreated astrocytes and in astrocytes transduced with CA-Akt. The effects of PDC in these two types of cultures were not significantly different. Together, these studies show that CA-Akt reduces transport activity in astrocytes, but at the same time increases the fraction of transport that is sensitive to DHK, consistent with induction of GLT-1 protein observed.

image

Figure 6.  Effects of CA-Akt on the pharmacological properties of Na+-dependent transport activity. Na+-dependent glutamate (0.5 µm) transport activity was measured in astrocytes 2 weeks either after infection with lentivirus expressing CA-Akt (1×) or in untreated astrocytes. Transport activity was measured in the absence (open bars) or presence (solid bars) of either a selective inhibitor of GLT-1-mediated activity, DHK (100 µm), or a general inhibitor of the various Na+-dependent glutamate transport systems, l-trans-PDC (100 µm). Data are presented as mean ± SEM of three independent experiments. *p < 0.05 compared with the corresponding control by one sample two-tailed t-test; †p = 0.05 compared with untreated astrocytes incubated with DHK, by paired two-tailed t-test. The effects of l-trans-PDC were not significantly different in untreated astrocytes and astrocytes expressing CA-Akt.

Download figure to PowerPoint

Effects of CA-Akt on GLT-1 and GLAST mRNA levels

In previous studies the induction of GLT-1 protein in astrocytes by EGF, dbcAMP or PACAP on GLT-1 was accompanied by an increase in GLT-1 mRNA (Schlag et al. 1998; Figiel and Engele 2000; Zelenaia et al. 2000). Northern blot analyses were used to examine the steady state levels of GLT-1 and GLAST mRNA two weeks after transduction of astrocytes with CA-Akt. As previously reported (Zelenaia et al. 2000), hybridization with specific cDNA probes revealed single mRNA bands for both GLT-1 and GLAST with apparent sizes of ∼11 and 3.9 kb, respectively (Fig. 7). In untreated astrocytes, the levels of GLT-1 mRNA were at the limit of detection, consistent with earlier studies (Swanson et al. 1997; Schlag et al. 1998). Two weeks after infection of astrocytes with lentivirus engineered to express 1× CA-Akt, GLT-1 mRNA was increased 10 times compared with uninfected astrocytes (Figs 7a and b), which was parallel to the changes in the protein expression of GLT-1. In contrast, GLAST mRNA was significantly reduced by CA-Akt to 38% of that observed in uninfected astrocytes (Fig. 7c). These studies suggest that the effects of CA-Akt might be related to increased transcription of GLT-1.

image

Figure 7.  Effects of CA-Akt on GLT-1 and GLAST mRNA levels. Astrocytes were either infected with CA-Akt lentivirus (1×) or treated with EGF (30 ng/mL) for two weeks. (a) Representative northern blot for GLT-1, GLAST or cyclophilin mRNA. Thirty µg of total RNA was loaded for each sample. The figures were from the same blot that was first hybridized with the GLT-1 probe, then stripped, and rehybridized with the GLAST and cyclophilin probes. (b) Summary of three independent experiments for GLT-1 mRNA. The levels of GLT-1 mRNA were normalized to cyclophilin mRNA and expressed relative to the level of corresponding mRNA in untreated cells. Data were mean ± SEM and were compared to 1 using a one-sample two-tailed t-test. *p < 0.05 compared with untreated (Untr.) astrocytes. (c) Summary of three independent experiments for GLAST mRNA. The level of GLAST mRNA was normalized to cyclophilin mRNA and expressed relative to the level of corresponding mRNA in untreated cells. Data are mean ± SEM and were compared to 1 using a one-sample two-tailed t-test. **p = 0.01 compared with untreated (Untr.) astrocytes.

Download figure to PowerPoint

Effects of CA-Akt on activity of a putative GLT-1 promoter

In recent studies, a putative promoter fragment for the human variant of GLT-1 (also called EAAT2) was isolated and characterized (Rothstein et al. 2005). We subcloned this GLT-1 promoter-reporter fragment into the lentiviral transfer plasmid. Astrocytes were infected with either the promoter-reporter lentiviral vector (E2P-GFP) alone or in combination with the lentivirus engineered to express CA-Akt. In astrocytes infected with the E2P-GFP lentiviral vector alone, the level of GFP was quite low when analyzed by western blot (Fig. 8). Treatment of these astrocytes with EGF significantly increased the expression of GFP, consistent with previous suggestions that EGF increases GLT-1 expression at least in part by increasing transcription. Co-infection of astrocytes with CA-Akt resulted in a dramatic increase in EGFP expression, providing further evidence that CA-Akt can increase transcription through the GLT-1 promoter.

image

Figure 8.  Effects of CA-Akt on the expression of a reporter protein (enhanced GFP, EGFP) from a putative EAAT2/GLT-1 promoter. Astrocytes were infected with lentivirus (3×) engineered to express GFP under the control of the putative human GLT-1/EAAT2 promoter fragment (E2P) either alone or in combination with infection with lentivirus expressing CA-Akt virus (1×). Astrocytes transduced with E2P-GFP were also treated with EGF (30 ng/mL). After 2 weeks, the astrocytes were harvested and analyzed for GFP expression levels by western blot. (a) A representative western blot showing expression of GFP driven either by the EAAT2/GLT-1 promoter. As a control, lysatses from HEK cells used to make the E2P-GFP virus were also analyzed for GFP expression. (b) Summary of between three and five independent experiments for the assay of the promoter activity in either astrocytes treated with EGF or in astrocytes expressing CA-Akt. *p < 0.05 compared with untreated astrocytes transduced with the reporter construct. **p < 0.01 compared with untreated astrocytes transduced with the reporter construct.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

In previous studies, we and others have shown that astrocytes grown in culture generally do not express the GLT-1/EAAT2 subtype of glutamate transporter. Treatment with different growth factors, with dbcAMP, or by co-culturing with neurons induces expression of GLT-1 in astrocytes (Swanson et al. 1997; Schlag et al. 1998; Figiel and Engele 2000; Zelenaia et al. 2000). All of these treatments are associated with a change in astrocyte morphology from polygonal shape to a process-bearing stellate shape that has been thought to reflect a more differentiated state. By either using pharmacological inhibitors in these studies or examining reporter activity from a putative promoter fragment, both PI3-K and the nuclear transcription factor, NF-κB have been implicated in the effects of EGF and dbcAMP (Zelenaia et al. 2000; Su et al. 2003; Sitcheran et al. 2005). In several systems, PI3-K activates Akt (for a review, see Chan et al. 1999), and Akt can activate NF-κB (Ozes et al. 1999; for a recent schematic of the signaling pathways, see Sitcheran et al. 2005).

In the present study, variants of Akt were introduced into astrocytes using a lentiviral vector system that consistently transduces a very high percentage (> 80–90%) of cells to determine whether Akt might be involved in the regulation of GLT-1 expression. We found that DN-Akt reduced the EGF-dependent induction of GLT-1 expression, but did not reduce the dbcAMP-dependent induction of GLT-1 expression. This effect of DN-Akt was not mimicked by the expression of a null variant of Akt. Although expression of DN-Akt had effects on both induction of GLT-1 and on phosphorylation of a down-stream target (GSK3β), we were unable to document DN-Akt protein expression in astrocytes. We also found that the expression of CA-Akt increased the GLT-1 protein levels in a dose- and time-dependent fashion; these effects were not mimicked by lentiviral infection and expression of two different control proteins (either null Akt or GFP). In these same experiments, expression of CA-Akt caused a robust change in astrocyte morphology, which resembled that observed with EGF. The effects of CA-Akt on GLT-1 protein were associated with an increase in the steady state levels of GLT-1 mRNA. A putative promoter fragment from the human GLT-1 (EAAT2) gene was isolated and subcloned together with GFP as a reporter into the same lentiviral vector. Using this construct, it was demonstrated that either EGF or CA-Akt increased the expression of reporter protein. Together with the evidence for increased mRNA expression, these data provide strong evidence that the effects of CA-Akt on GLT-1 mRNA are likely to be related to increased transcription. In contrast to the effects of either growth factors or dbcAMP, expression of CA-Akt did not increase GLAST protein levels and resulted in decreases in GLAST mRNA levels. These studies identify Akt as a signaling molecule that can increase GLT-1 protein expression without increasing GLAST protein levels.

The glutamate analog, DHK, is frequently used to identify changes in GLT-1-mediated transport activity. In fact, treatment of astrocytes with dbcAMP does not result in detectible DHK-sensitive transport, even though it causes large increases in GLT-1 immunoreactivity (Swanson et al. 1997; Schlag et al. 1998). Although DHK sensitivity is observed in EGF-treated astrocytes, the inhibition is quite modest, ∼9% at 60 µm (Zelenaia et al. 2000). It is possible that the proportion of transport mediated by GLT-1 is low relative to that mediated by GLAST, and that DHK-sensitive transport is obscured. Alternatively, GLT-1 may not be fully functional after these treatments. Recently, Vermeiren and colleagues showed that astrocytes maintained in a cocktail of growth factors express GLT-1 but have no DHK-sensitive transport activity unless protein kinase C is activated, either directly with phorbol esters or indirectly with mGluR5 agonists (Vermeiren et al. 2005). This apparent activation occurs within minutes and is independent of changes in the numbers of transporters present on the plasma membrane, suggesting that non-functional GLT-1 can be expressed under certain circumstances and subsequently activated by protein kinase C. Unlike other treatments that induce GLT-1 expression in astrocytes, the effects of CA-Akt on GLT-1 protein were associated with an increase in the sensitivity of Na+-dependent glutamate uptake to concentrations of DHK that should selectively reduce GLT-1-mediated transport activity. In fact, based on the published IC50 values of between 20 and 100 µm for the inhibition of GLT-1 (Arriza et al. 1994; Tan et al. 1999), 100 µm DHK should inhibit GLT-1-mediated transport activity between 50% and 83% (assuming simple competitive inhibition). GLAST-mediated transport activity is not affected by concentrations of DHK of up to 10 mm (Arriza et al. 1994; Garlin et al. 1995). EAAC1 is inhibited by DHK with an IC50 value of ∼1 mm (Dowd et al. 1996). As we find no evidence of EAAC1 expression, the fact that expression of CA-Akt results in a significant increase in DHK-sensitive transport activity is consistent with the notion that GLT-1 is functional under these conditions.

We also noted that the expression of CA-Akt reduced total Na+-dependent transport activity. This decrease in transporter activity was observed at either low or high concentrations of glutamate relative to its Km value (Garlin et al. 1995; Schlag et al. 1998; Zelenaia et al. 2000), suggesting that this effect is the result of a decrease in Vmax. Examination of the cell surface expression of either GLT-1 or GLAST using biotinylation provided strong evidence that this decrease in transport activity cannot be attributed to an Akt induced internalization of either GLAST or GLT-1. This might suggest that Akt decreases the catalytic efficiency (turnover number) of GLAST, however, Boehmer and colleagues have co-expressed GLAST and a different variant of constitutively active Akt in Xenopus oocytes and found that it had no effect on GLAST-mediated transport and actually reversed the inhibition of GLAST caused by an ubiquitin ligase, Nedd4–2 (Boehmer et al. 2003). Although the simplest explanation is that expression of CA-Akt may either directly or indirectly decrease the intrinsic activity (catalytic efficiency) of GLAST, this possible regulation of GLAST by Akt has not been pursued further in the present study.

Expression of CA-Akt also induced a change in astrocyte morphology that is similar to that previously observed with the various treatments that have been shown to induce GLT-1 expression and that is thought to resemble astrocyte differentiation. Previous studies have suggested that Akt activation is important for the differentiation of several different cell types, including adipocytes, neurons and astrocytes (for discussions, see Bang et al. 2001; Hermanson et al. 2002; Bae et al. 2003; Vojtek et al. 2003; Setoguchi and Kondo 2004). The differential timing of both GLT-1 and GLAST mRNA and protein expression during development and their differential distribution in tissues from adult animals suggests that the expression of both GLT-1 and GLAST are under independent control (Rothstein et al. 1994; Sutherland et al. 1996; Furuta et al. 1997). For example, in rodent forebrain GLAST mRNA levels are high during embryonic development and decline during the early postnatal period, whereas the levels of GLT-1 mRNA and protein increase dramatically during the period of synaptogenesis (postnatal days 7–21). In addition, GLAST expression is enriched in specialized glia throughout the nervous system, such as Bergmann glia of the cerebellum, Müller cells of the retina, and supporting glia in the vestibular end organ (for a review, see Danbolt 2001). Although EGF has been implicated in astrocyte differentiation and induces expression of GLT-1 in astrocyte cultures, it also increases GLAST expression (Zelenaia et al. 2000). Similarly, dbcAMP induces GLT-1 expression and increases GLAST expression (Schlag et al. 1998). PACAP was shown to selectively induce GLT-1 expression at lower concentrations, but also increased GLAST expression at 10-fold higher concentrations than those required to induce GLT-1 expression (Figiel and Engele 2000). In the present study, expression of CA-Akt induced expression of GLT-1 and did not increase GLAST protein levels. In fact, expression of CA-Akt reduced GLAST mRNA levels. These studies provide evidence that CA-Akt can selectively induce GLT-1 expression and identify Akt as a signal that could selectively control GLT-1 expression in vivo.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

The authors would like to thank Dr Morris Birnbaum (University of Pennsylvania) for the Akt-1 constructs and Dr Judy Grinspan (Children's Hospital of Philadelphia) for providing the A2B5 hybridoma supernatant. We would also like to thank the members of the Robinson laboratory for their advice with this project. This work was supported by NIH grant NS36465 to MBR and JDR and NS38690 to JHW.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  • Arriza J. L., Fairman W. A., Wadiche J. I., Murdoch G. H., Kavanaugh M. P. and Amara S. G. (1994) Functional comparisons of three glutamate transporter subtypes cloned from human motor cortex. J. Neurosci. 14, 55595569.
  • Bae S. S., Cho H., Mu J. and Birnbaum M. J. (2003) Isoform-specific regulation of insulin-dependent glucose uptake by Akt/protein kinase B. J. Biol. Chem. 278, 49 53049 536.
  • Bang O. S., Park E. K., Yang S. I., Lee S. R., Franke T. F. and Kang S. S. (2001) Overexpression of Akt inhibits NGF-induced growth arrest and neuronal differentiation of PC12 cells. J. Cell Sci. 114, 8188.
  • Boehmer C., Henke G., Schniepp R., Palmada M., Rothstein J. D., Broer S. and Lang F. (2003) Regulation of the glutamate transporter EAAT1 by the ubiquitin ligase Nedd4–2 and the serum and glucocorticoid-inducible kinase isoforms SGK1/3 and protein kinase B. J. Neurochem. 86, 11811188.
  • Chan T. O., Rittenhouse S. E. and Tsichlis P. N. (1999) AKT/PKB and other D3 phosphoinositide-regulated kinases: kinase activation by phosphoinositide-dependent phosphorylation. Annu. Rev. Biochem. 68, 9651014.
  • Chaudhry F. A., Lehre K. P., Campagne M. V. L., Ottersen O. P., Danbolt N. C. and Storm-Mathisen J. (1995) Glutamate transporters in glial plasma membranes: Highly differentiated localizations revealed by quantitative ultrastructural immunocytochemistry. Neuron 15, 711720.
  • Choi D. W. (1992) Excitotoxic cell death. J. Neurobiol. 23, 12611276.
  • Cross D. A., Alessi D. R., Cohen P., Andjelkovich M. and Hemmings B. A. (1995) Inhibition of glycogen synthase kinase-3 by insulin mediated by protein kinase B. Nature 378, 785789.
  • Danbolt N. C. (2001) Glutamate uptake. Prog. Neurobiol. 65, 1105.
  • Datta S. R., Brunet A. and Greenberg M. E. (1999) Cellular Survival: a play in three Akts. Genes Dev. 13, 29052927.
  • Dowd L. A., Coyle A. J., Rothstein J. D., Pritchett D. B. and Robinson M. B. (1996) Comparison of Na+-dependent glutamate transport activity in synaptosomes, C6 glioma, and Xenopus Oocytes expressing excitatory amino acid carrier 1 (EAAC1). Mol. Pharmacol. 49, 465473.
  • Figiel M. and Engele J. (2000) Pituitary adenylate cyclase-activating polypeptide (PACAP), a neuron-derived peptide regulating glial glutamate transport and metabolism. J. Neurosci. 20, 35963605.
  • Furuta A., Rothstein J. D. and Martin L. J. (1997) Glutamate transporter protein subtypes are expressed differentially during rat central nervous system development. J. Neurosci. 17, 83638375.
  • Garlin A. B., Sinor A. D., Sinor J. D., Jee S. H., Grinspan J. B. and Robinson M. B. (1995) Pharmacology of sodium-dependent high-affinity L-[3H]glutamate transport in glial cultures. J. Neurochem. 64, 25722580.
  • Gegelashvili G., Danbolt N. C. and Schousboe A. (1997) Neuronal soluble factors differentially regulate the expression of the GLT1 and GLAST glutamate transporters in cultured astroglia. J. Neurochem. 69, 26122615.
  • Haugeto Ø., Ullensveng K., Levy L. M., Chaudhry F. A., Honore T., Neilsen M., Lehre K. P. and Danbolt N. C. (1996) Brain glutamate transporter proteins form homomultimers. J. Biol. Chem. 271, 27 71527 722.
  • Hermanson O., Jepsen K. and Rosenfeld M. G. (2002) N-CoR controls differentiation of neural stem cells into astrocytes. Nature 419, 934939.
  • Kalandadze A., Wu Y., Fournier K. and Robinson M. B. (2004) Identification of motifs involved in endoplasmic reticulum retention-forward trafficking of the GLT-1 subtype of glutamate transporter. J. Neurosci. 24, 51835192.
  • Kane L. P., Shapiro V. S., Stokoe D. and Weiss A. (1999) Induction of NF-κB by the Akt/PKB kinase. Curr. Biol. 9, 601605.
  • Karolewski B. A., Watson D. J., Parente M. K. and Wolfe J. H. (2003) Comparison of transfection conditions for a lentivirus vector produced in large Volumes. Hum. Gene Ther. 14, 12871296.
  • Kohn A. D., Summers S. A., Birnbaum M. J. and Roth R. A. (1996) Expression of a constitutively active Akt Ser/Thr kinase in 3T3-L1 adipocytes stimulates glucose uptake and glucose transporter 4 translocation. J. Biol. Chem. 271, 31 37231 378.
  • Krizman-Genda E., González M. I., Zelenaia O. and Robinson M. B. (2005) Evidence that Akt mediates platelet-derived growth factor-dependent increases in activity and surface expression of the neuronal glutamate transporter. Neuropharmacol. 49, 872882.
  • Lehre K. P. and Danbolt N. C. (1998) The number of glutamate transporter subtype molecules at glutamatergic synapses: chemical and stereological quantification in young adult rat brain. J. Neurosci. 18, 87518757.
  • Li S., Mallory M., Alford M., Tanaka S. and Masliah E. (1997) Glutamate transporter alterations in Alzheimer disease are possibly associated with abnormal APP expression. J. Neuropathol. Exp. Neurol. 56, 901911.
  • Lipton S. A. and Rosenberg P. A. (1994) Excitatory amino acids as a final common pathway for neurological disorders. N. Engl. J. Med. 330, 613621.
  • Lowry O. H., Rosenberg N. J., Farr A. L. and Randall R. J. (1951) Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265275.
  • Mathern G. W., Mendoza D., Lozada A., et al. (1999) Hippocampal GABA and glutamate transporter immunoreactivity in patients with temporal lobe epilepsy. Neurology 52, 453472.
  • Ozes O. N., Mayo L. D., Gustin J. A., Pfeffer S. R., Pfeffer L. M. and Donner D. B. (1999) NF-κB activation by tumor necrosis factor requires the Akt serineñthreonine kinase. Nature 401, 8285.
  • Rao V. L. R., Baskaya M. K., Dogan A., Rothstein J. D. and Dempsey R. J. (1998) Traumatic brain injury down-regulates glial glutamate transporter (GLT-1 and GLAST) protein in rat brain. J. Neurochem. 70, 20202027.
  • Rothstein J. D., Martin L., Levey A. I., Dykes-Hoberg M., Jin L., Wu D., Nash N. and Kuncl R. W. (1994) Localization of neuronal and glial glutamate transporters. Neuron 13, 713725.
  • Rothstein J. D., Van Kammen M., Levey A. I., Martin L. J. and Kuncl R. W. (1995) Selective loss of glial glutamate transporter GLT-1 in amyotrophic lateral sclerosis. Ann. Neurol. 38, 7384.
  • Rothstein J. D., Dykes-Hoberg M., Pardo C. A., et al. (1996) Knockout of glutamate transporters reveals a major role for astroglial transport in excitotoxicity and clearance of glutamate. Neuron 16, 675686.
  • Rothstein J. D., Patel S., Regan M. R., et al. (2005) Beta-lactam antibiotics offer neuroprotection by increasing glutamate transporter expression. Nature 433, 7377.
  • Samuelsson C., Kumlien E., Flink R., Lindholm D. and Ronne-Engstrom E. (2000) Decreased cortical levels of astrocytic glutamate transport protein GLT-1 in a rat model of posttraumatic epilepsy. Neurosci. Lett. 289, 185188.
  • Schlag B. D., Vondrasek J. R., Munir M., Kalandadze A., Zelenaia O. A., Rothstein J. D. and Robinson M. B. (1998) Regulation of the glial Na+-dependent glutamate transporters by cyclic AMP analogs and neurons. Mol. Pharmacol. 53, 355369.
  • Setoguchi T. and Kondo T. (2004) Nuclear export of OLIG2 in neural stem cells is essential for ciliary neurotrophic factor-induced astrocyte differentiation. J. Cell Biol. 166, 963968.
  • Sims K. D. and Robinson M. B. (1999) Expression patterns and regulation of glutamate transporters in the developing and adult nervous system. Crit. Rev. Neurobiol. 13, 169197.
  • Sitcheran R., Gupta P., Fisher P. B. and Baldwin A. S. (2005) Positive and negative regulation of EAAT2 by NF-kappaB: a role for N-myc in TNFalpha-controlled repression. EMBO J. 24, 510520.
  • Su Z. Z., Leszczyniecka M., Kang D. C., Sarkar D., Chao W., Volsky D. J. and Fisher P. B. (2003) Insights into glutamate transport regulation in human astrocytes: cloning of the promoter for excitatory amino acid transporter 2 (EAAT2). Proc. Natl Acad. Sci. USA 100, 19551960.
  • Susarla B. T. S., Seal R. P., Zelenaia O., Watson D. J., Wolfe J. H., Amara S. G. and Robinson M. B. (2004) Differential regulation of GLAST immunoreactivity and activity by protein kinase C: evidence for modification of amino and carboxyl termini. J. Neurochem. 91, 11511163.
  • Sutherland M. L., Delaney T. A. and Noebels J. L. (1996) Glutamate transporter mRNA expression in proliferative zones of the developing and adult murine CNS. J. Neurosci. 16, 21912207.
  • Swanson R. A., Liu J., Miller J. W., Rothstein J. D., Farrell K., Stein B. A. and Longuemare M. C. (1997) Neuronal regulation of glutamate transporter subtype expression in astrocytes. J. Neurosci. 17, 932940.
  • Tan J., Zelenaia O., Rothstein J. D. and Robinson M. B. (1999) Expression of the GLT-1 subtype of Na+-dependent glutamate transporter: Pharmacological characterization and lack of regulation by protein kinase C. J. Pharmacol. Exp. Ther. 289, 16001610.
  • Tanaka K., Watase K., Manabe T., et al. (1997) Epilepsy and exacerbation of brain injury in mice lacking the glutamate transporter GLT-1. Science 276, 16991702.
  • Torp R., Lekieffre D., Levy L. M., Haug F. M., Danbolt N. C., Meldrum B. S. and Ottersen O. P. (1995) Reduced postischemic expression of a glial glutamate transporter, GLT-1, in the rat hippocampus. Exp. Brain Res. 103, 5158.
  • Trotti D., Rolfs A., Danbolt N. C., Brown R. H. and Hediger M. A. (1999) SOD1 mutants linked to amyptrophic lateral sclerosis selectively inactivate a glial glutamate transporter. Nat. Neurosci. 2, 848.
  • Vermeiren C., Najimi M., Vanhoutte N., Tilleux S., Hemptinne I. D., Maloteauz J.-M. and Hermans E. (2005) Acute up-regulation of glutamate uptake mediated by mGluR5a in reactive astrocytes. J. Neurochem. 94, 405416.
  • Vojtek A. B., Taylor J., DeRuiter S. L., Yu J. Y., Figueroa C., Kwok R. P. and Turner D. L. (2003) Akt regulates basic helix-loop-helix transcription factor-coactivator complex formation and activity during neuronal differentiation. Mol. Cell. Biol. 23, 44174427.
  • Wang Q., Somwar R., Bilan P. J., Liu Z., Jin J., Woodgett J. R. and Klip A. (1999) Protein Kinase B/Akt Participates in GLUT4 Translocation by Insulin in L6 Myoblasts. Mol. Cell. Biol. 19, 40084018.
  • Wang Z., Pekarskaya O., Bencheikh M., Chao W., Gelbard H. A., Ghorpade A., Rothstein J. D. and Volsky D. J. (2003) Reduced expression of glutamate transporter EAAT2 and impaired glutamate transport in human primary astrocytes exposed to HIV-1 or gp120. Virology 312, 6073.
  • Watson D. J. and Wolfe J. H. (2003) Lentiviral vectors for gene transfer to the central nervous system. Applications in lysosomal storage disease animal models. Meth Mol Med. 76, 383403.
  • Ye Z., Rothstein J. D. and Sontheimer H. (1999) Compromised glutamate transport in human glioma cells: reduction-mislocalization of sodium-dependent glutamate transporters and enhanced activity of cystine-glutamate exchange. J. Neurosci. 19, 10 76710 777.
  • Yu N., Billaud J. N. and Phillips T. R. (1998) Effects of feline immunodeficiency virus on astrocyte glutamate uptake: implications for lentivirus-induced central nervous system diseases. Proc. Natl. Acad. Sci. USA 95, 26242629.
  • Zelenaia O., Schlag B. D., Gochenauer G. E., Ganel R., Song W., Beesley J. S., Grinspan J. B., Rothstein J. D. and Robinson M. B. (2000) Epidermal growth factor receptor agonists increase expression of glutamate transporter GLT-1 in astrocytes through pathways dependent on phosphatidylinositol 3-kinase and transcription factor NF-κB. Mol. Pharmacol. 57, 667678.