Up-regulation of glutamine synthesis in microglia activated with endotoxin

Neuroscience Letters 591 (2015) 99–104

Contents lists available at ScienceDirect

Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet

Research article

Up-regulation of glutamine synthesis in microglia activated with endotoxin Kazuyuki Nakajima a,∗ , Tomoyuki Kanamatsu b , Yosuke Takezawa a , Shinichi Kohsaka c a b c

Department of Bioinformatics, Faculty of Engineering, Soka University, Tokyo 192-8577, Japan Department of Environmental Engineering for Symbiosis, Faculty of Engineering, Soka University, Tokyo 192-8577, Japan Department of Neurochemistry, National Institute of Neuroscience, Tokyo 187-8502, Japan

h i g h l i g h t s • Microglia in vitro were found to express glutamine synthetase protein. • The microglia were shown to have glutamine synthase activity. • Endotoxin-activated microglia converted 13 C-glutamate to 13 C-glutamine.

a r t i c l e

i n f o

Article history: Received 8 November 2014 Received in revised form 31 January 2015 Accepted 10 February 2015 Available online 11 February 2015 Keywords: Microglia Glutamic acid Glutamine synthetase Nuclear magnetic resonance

a b s t r a c t We previously verified that newborn rat brain-derived microglia have the ability to uptake 14 C-glutamate (Glu) through glutamate transporter-1. A given amount of Glu incorporated into microglia was suspected to be metabolized to glutamine (Gln). However, the ability of microglia to do this had not been demonstrated. Thus, in the present study we examined the possibility that primary rat microglia metabolize Glu into Gln. Immunocytochemical and immunoblotting studies indicated that the microglia express glutamine synthetase (GS) protein. As expected from these results, GS activity was actually detected in microglia, although the specific activity was lower than that of astrocytes. Considering this microglial property, it seemed possible that the taken Glu is metabolized to Gln in the cells. To investigate this possibility, we exposed microglia to [13 C]Glu-containing medium and analyzed the change of Glu to Gln in a nuclear magnetic resonance examination. The results clarified that non-stimulated microglia hardly changed Glu to Gln, but when stimulated with lipopolysaccharide the microglia significantly metabolized [13 C]Glu to [13 C]Gln. Microglia were thus, strongly suggested to metabolize Glu to Gln via GS activity when activated in the inflammatory/pathological state of the nervous system. © 2015 Elsevier Ireland Ltd. All rights reserved.

1. Introduction The major excitatory neurotransmitter glutamic acid (Glu) is also known as an excitatory toxin in the nervous system because

Abbreviations: ATP, adenosine 5 -triphosphate; DMEM, Dulbecco’s modified essential medium; GFAP, glial fibrillary acidic protein; GLAST, glutamate–aspartate transporter; Gln, glutamine; GLT-1, glutamate transporter-1; Glu, glutamate; GS, glutamine synthetase; HRP, horseradish peroxidase; Iba1, ionized Ca2+ binding adapter molecule 1; LPS, lipopolysaccharide; M-CSF, macrophage colonystimulating factor; MP, myristoylated pseudosubstrate (20–28); NMR, nuclear magnetic resonance; PBS, phosphate-buffered saline; PKC␣, protein kinase C alpha; TSP, trimethylsilylpropionate. ∗ Corresponding author at: Department of Bioinformatics, Faculty of Engineering, Soka University, 1-236 Tangi-machi, Hachioji, Tokyo 192-8577, Japan. Tel.: +81 426 91 9370; fax: +81 426 91 9312. E-mail address: [email protected] (K. Nakajima). http://dx.doi.org/10.1016/j.neulet.2015.02.021 0304-3940/© 2015 Elsevier Ireland Ltd. All rights reserved.

it causes neuronal cell death through the abnormal excitation of Glu receptors [1,2]. To protect the neurons from the dangerous action of Glu in situ, glial type-Glu transporters such as glutamate transporter-1 (GLT-1) and glutamate–aspartate transporter (GLAST) take away excess amounts of Glu around the excitatory synapses [3,4]. In a previous study, we found that GLT-1 protein was highly enhanced in the activated microglia around injured motoneurons in the axotomized rat facial nucleus [5]. In other injured models [6–8] and infection models [9,10], microglia were similarly observed to increase GLT-1 levels. These findings suggest that the microglia scavenge dangerous Glu and protect excitatory neurons from Glu-induced damage in the pathological or inflammatory state. However, these results simultaneously raised a question about the metabolism of Glu taken into microglia. Astrocytes have been generally thought of as a major cell type that

100

K. Nakajima et al. / Neuroscience Letters 591 (2015) 99–104

converts the taken Glu to glutamine (Gln) in the nervous system [11], but the ability of microglia to change Glu to Gln has not been established. Therefore, in the present study, we sought to determine the ability of microglia to convert Glu into Gln. By using biochemical and nuclear magnetic resonance (NMR) methods, we analyzed primary rat microglia for Gln synthetase (GS) expression and for the metabolic change of [2-13 C]Glu to [2-13 C]Gln. 2. Materials and methods 2.1. Reagents and antibodies l-Glutamic acid ␥-monohydroxamate, l-glutamic acid (Glu), adenosine 5 -triphosphate disodium (ATP), and hydroxylamine were obtained from Sigma–Aldrich Japan (Tokyo). [2-13 C]Glu and deuterium oxide (99.9 atom%) were purchased from ISOTEC (Miamisburg, OH). Anti-glial fibrillary acidic protein (GFAP) antibody and antiglutamine synthase (GS) antibody were purchased from Millipore (Temecula, CA). Anti-ionized Ca2+ binding adapter molecule 1 (Iba1) antibody was supplied by Wako Pure Chemical Industries (Osaka, Japan). Anti-actin antibody was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Anti-glutamate–aspartate transporter (GLAST) and anti-glutamate transporter-1 (GLT-1) antibodies were obtained from CovalAb (Oullins Cedex, France). Alexa Fluor 488-conjugated anti-mouse IgG and Alexa Fluor 568-conjugated anti-rabbit IgG were from Life Technologies Japan. Horseradish peroxidase (HRP)-conjugated anti-mouse IgG, HRPconjugated anti-rabbit IgG, and HRP-conjugated anti-goat IgG were from Santa Cruz Biotechnology. 2.2. Animals To prepare the primary mother culture for obtaining astrocytes and microglia in vitro, pregnant Wistar rats were purchased from Clea Japan (Tokyo). These rats were kept on a 12-h daylight cycle with food and water ad libitum. All animal experiments were carried out in accordance with the guidelines laid down by the National Institutes of Health in the U.S. regarding the care and use of animals, and have been approved by the ethics committee of Soka University.

2.4. Immunocytochemistry Microglia or astrocytes seeded on coverslips were fixed with 3.7% formaldehyde/phosphate-buffered saline (PBS) for 10 min at room temperature, as described in [12]. For dual staining, the cells were initially incubated with anti-GS antibody (1:100) at 4 ◦ C for 16 h, and subsequently incubated with anti-Iba1 antibody (1:100) or anti-GFAP antibody (1:100) at 4 ◦ C for 16 h. After being rinsed with PBS, the cells were incubated with a mixture of secondary antibodies (Alexa Fluor 488-conjugated anti-mouse IgG and Alexa Fluor 568-conjugated anti-rabbit IgG). The cells were sufficiently rinsed and mounted with Perma-Flour (Shandon, Pittsburgh, PA). 2.5. Immunoblotting Immunoblotting was carried out essentially according to the method described in [12]. Cell homogenate was routinely subjected to sodium dodecyl sulfate polyacrylamide gel electrophoresis and transblotting. Briefly, the blotted Immobilon (Millipore) was incubated with primary antibody (1:1000 for GFAP, Iba1, GS, GLAST and GLT-1; 1:500 for actin) at 4 ◦ C for 16 h. After the membrane was rinsed, it was incubated with HRP-conjugated anti-rabbit IgG antibody (1:1000) or HRP-conjugated anti-mouse IgG antibody (1:1000) for 3 h at room temperature. The antigen–antibody complex on the membrane was detected by an enhanced chemiluminescence system. 2.5. Determination of GS activity GS (EC 6.3.1.2) catalyzes the formation of Gln from Glu and ammonia by using ATP. To measure the GS activity in the cultured cells, we used a hydroxamate method [14]. The principle of this method is based on the GS function, which adds hydroxylamine (instead of NH3 ) to Glu to produce ␥-glutamylhydroxamate (corresponding to Gln). The reaction mixture containing 50 mM imidazole (pH 7.2), 20 mM MgCl2 , 25 mM 2-mercaptoethanol, 50 mM l-Glu 10 mM ATP, 100 mM hydroxylamine and cell homogenate in a total volume of 500 ␮L was incubated at 37 ◦ C for 15 min. At the end of the incubation, we added 500 ␮L of cold 0.3 M Fe(NO3 )3 /20% trichloroacetic acid to stop the reaction and colorize the resultant ␥glutamylhydroxamate in the solution. The tubes were centrifuged at 8000 × g for 30 min, and the supernatant was measured for optical density at 500 nm.

2.3. Preparation of microglia and astrocytes

2.6. Nuclear magnetic resonance spectroscopy (NMR) analysis

Primary mother cultures were prepared from newborn rat brain, as described in [12]. Microglial cells were isolated by shaking the primary culture bottles [12]. The purity was over 99.9% by the assessment of Iba1 staining. The microglia were seeded onto the wells of 24-well plates (Nunclon, St. Louis, MO) at a density of 1.5 × 105 cells/well (for immunocytochemistry), onto 60-mm plastic dishes (Nunclon) at a density of 2 × 106 cells/dish (for immunoblotting), or onto 90-mm dishes (Nunclon) at a density of 5 × 106 cells/dish (for the GS activity assay and the analysis of the metabolic change of [2-13 C]Glu to [2-13 C]Gln). Astrocytes were prepared from the primary mother culture maintained for 3 weeks, as described in [13]. The astrocytes were subcultured onto the wells of 24-well plates (Nunclon) at a density of 5 × 104 cells/well (for immunocytochemistry), onto 60-mm dishes (Nunclon) at a density of 2 × 106 cells/dish (for immunoblotting), or onto 90-mm dishes (Nunclon) at a density of 5 × 106 cells/dish (for the GS activity assay and the analysis of the metabolic change of [2-13 C]Glu to [2-13 C]Gln). The cell purity was estimated to be 98% based on the staining with anti-GFAP antibody.

Nonstimulated microglia (5 × 106 /dish) were maintained with serum-free Dulbecco’s modified essential medium (DMEM) for 12 h, and then exposed to [2-13 C] Glu (0.8 mg/4 mL DMEM) for 24 h. For the activated state, microglia (5 × 106 /dish) were maintained with serum-free DMEM for 12 h, and then exposed to [2-13 C] Glu (0.8 mg/4 mL DMEM) in the presence of lipopolysaccharide (LPS) (0.5 ␮g/mL) for 24 h. After that, the microglia were rinsed with PBS twice and then collected with 75% ethanol. The microglia were sonicated and then centrifuged at 20,000 × g for 10 min, and the supernatants were recovered. To each supernatant, one volume of water-saturated chloroform was added and mixed. The resultant aqueous extracts were vacuum-dried. Similarly, nonstimulated astrocytes (5 × 106 /dish) in serum-free DMEM were exposed to [2-13 C] Glu (0.8 mg/4 mL DMEM) for 24 h and treated in the same way as the microglia. The extracted materials were vacuum-dried. The dried samples prepared as described above were dissolved in deuterium oxide and placed in an NMR tube (5 mm dia; PS-003, Shigemi, Hachioji, Japan). The [2-13 C] signals from astro-

K. Nakajima et al. / Neuroscience Letters 591 (2015) 99–104

101

cytic metabolites were measured with 1 H-decoupled 13 C NMR spectroscopy (EX-400, JEOL, Tokyo), essentially as described [15]. Spectra were accumulated under the following conditions: pulse width 45◦ ; spectral width 230 ppm; acquisition time 2.163 s; timedomain points 65,536; pulse interval 8.000 s; scans 3200. 13 C NMR spectra of microglial samples were recorded on a Bruker spectrometer (AV-800 with CryoProbe, Bruker, Billerica, MA) using the 1 H-decoupling method. Spectra were accumulated under the following conditions: pulse width 45◦ ; spectral width 230 ppm; acquisition time 10.0 s; time-domain points 131,072; pulse interval 8.580 s; scans 1024. Two milligrams of sodium trimethylsilyl propionate (TSP) was used as an external standard. We stimulated microglia (107 /dish) with macrophage colonystimulating factor (M-CSF; 20 ng/mL) and treated other microglia (107 /dish) with 5 ␮M protein kinase C alpha (PKC␣) inhibitor/LPS (0.5 ␮g/mL). We stimulated astrocytes (5 × 106 /dish) with LPS (0.5 ␮g/mL). In each group, the 13 C NMR spectra were recorded on a spectrometer (AVANCE III HD 600, Bruker). 2.7. Statistical analysis The data are presented as the means ± SD of three different experiments. We performed a one-way analysis of variance (ANOVA) followed by Tukey’s post hoc comparison or two-tailed unpaired Student’s t-test. Values of p < 0.05 were considered significant (*p < 0.05, **p < 0.01). 3. Results 3.1. Immunocytochemical detection of GS protein in microglia First, as a positive cell type for expressing GS protein, astrocytes were dually stained with anti-GS antibody and anti-GFAP antibody. The results indicated that GS protein is distributed to the cytoplasm in GFAP-positive astrocytes (Fig. 1A, left–right). Next, we investigated whether or not microglia express the GS protein. In the absence of primary antibody, fluorescence was not seen (data not shown), but in the presence of anti-GS antibody positive fluorescent staining was seen in microglia-like cells (Fig. 1B, left and center). The GS staining was almost consistent with Iba1positive microglia (Fig. 1B, right), although the intensity appeared to be weaker than that of the astrocytes. These immunocytochemical results strongly suggest that microglia express GS protein in their cytoplasm, as astrocytes do. 3.2. Detection of GS protein in immunoblotting Since the immunocytochemical study suggested the occurrence of GS protein in microglia, we examined the possibility by immunoblotting. High-purity microglia and astrocytes were prepared (Fig. 1C, GFAP and Iba1) and analyzed for GS protein. The results revealed that the astrocytes (Ast) expressed a relatively large amount of GS protein, and that the microglia (Mic) also definitively contain the protein (Fig. 1C, GS). In addition, the astrocytes were confirmed to express GLAST (Fig. 1C, GLAST), whereas the microglia expressed GLT-1 (Fig. 1C, GLT-1), as reported [16]. Thus, our immunoblotting demonstrated that microglia contain some amounts of GS protein. 3.3. GS activity in microglia It is important to confirm whether the GS protein in microglia is functionally active. Thus, we tried to determine the GS activity in microglia according to the hydroxamate method (Fig. 1D). As a positive cell type for exhibiting GS activity, we measured astrocytes.

Fig. 1. GS protein in microglia and astrocytes. (A) Immunocytochemistry for GS in astrocytes. Astrocytes were stained dually with anti-GS antibody and anti-GFAP antibody. GS was visualized by Alexa-488 (green; left) and GFAP by Alexa-568 (red; center). A merged image is shown on the right side. Scale bar = 50 ␮m. (B) Immunocytochemistry for GS in microglia. Microglia were stained dually with anti-GS antibody and anti-Iba1 antibody. GS was visualized by Alexa-488 (green; left) and Iba1 by Alexa-568 (red; center). A merged image is shown on the right side. Scale bar = 50 ␮m. (C) Immunoblotting for GS in microglia and astrocytes. Cell homogenates of astrocytes (Ast) and microglia (Mic) were subjected to immunoblotting for GFAP, Iba1, actin, GS, GLAST and GLT-1. (D) Determination of GS activity in microglia and astrocytes. Astrocytes (Ast) and microglia (Mic) were measured for GS activity, using a standard curve in which there was a clear linearity between ␥-glutamylhydroxamate (0, 5, 10, 20, 40, and 80 nmole) and absorbance at 500 nm. The values in astrocytes and microglia are expressed as the specific activity (nmole/min/mg protein). The data are means ± SDs of three separate experiments carried out using different lots of cells (one-way ANOVA followed by Tukey’s post hoc comparison; *p < 0.05, **p < 0.01). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

The specific activity was 68.2 ± 8.5 nmol/min/mg protein (n = 3). The activity was not detected in the absence of ATP, indicating an ATP-dependent reaction (Fig. 1D, Ast). In this assay system, the specific activity of microglia was 15.8 ± 2.7 nmol/min/mg protein (n = 3) (Fig. 1D, Mic), which is approx. one-fourth of the astrocytic value. Microglia were thus verified to exhibit GS activity, although the degree of activity is lower than that of astrocytes. 3.4. Conversion of Glu into Gln in microglia Given the existence of both GLT-1 [16] and GS (the present study) in microglia, we hypothesized that microglia metabolize Glu into Gln in the cells, and we attempted to detect [2-13 C]Gln changed from [2-13 C]Glu in these cells by means of NMR. First, as a positive control, we exposed nonstimulated astrocytes to [2-13 C]Glu for 24 h, and analyzed them for [2-13 C]Gln. The NMR spectra clarified the presence of a signal for [2-13 C]Gln (Fig. 2A), indicating that the Glu incorporated in astrocytes was metabolized to Gln by GS activity. We then examined the potency of nonstimulated microglia. In the microglia exposed to [2-13 C]Glu for 24 h, a significant signal of [2-13 C]Gln was not detected (Fig. 2B). [2-13 C]Gln was also not

102

K. Nakajima et al. / Neuroscience Letters 591 (2015) 99–104

3.5. GS activity/GS protein in LPS-stimulated microglia Since the above results suggested the possibility that LPSstimulated microglia enhance GS activity, we investigated the effects of LPS on the GS-specific activity in microglia in vitro. We found that the treatment of microglia with LPS did not promote the specific activity of GS, but rather reduced it (Fig. 3A). The amounts of GS protein tended to decrease under constant levels of Iba1 and actin as the LPS concentration increased (Fig. 3B). The levels of the GS band at 1.0 ␮g/mL LPS were estimated to be approx. one-fifth those of nonstimulated microglia (Fig. 3C). In contrast, GLT-1 levels were significantly upregulated by stimulation with LPS (Fig. 3B), and at 1.0 ␮g/mL LPS they were estimated to be approx. five-fold those of the nonstimulated microglia (Fig. 3D).

4. Discussion

Fig. 2. NMR analysis. (A) Analysis of astrocytes exposed to [2-13 C]Glu. Astrocytes exposed to [2-13 C]Glu for 24 h were subjected to the NMR analysis as described in the Materials and Methods. (B) Analysis of nonstimulated microglia exposed to [213 C]Glu. Nonstimulated microglia exposed to [2-13 C]Glu for 24 h were subjected to the NMR analysis. (C) Analysis of LPS-stimulated microglia exposed to [2-13 C]Glu. In the presence of LPS (0.5 ␮g/mL), microglia were exposed to [2-13 C]Glu for 24 h and then subjected to the NMR analysis as described above.

detected in the medium (data not shown). Thus, nonstimulated microglia were suggested to have a weak activity for changing Glu into Gln. Assuming the activated state, we exposed microglia to [2-13 C]Glu in the presence of LPS. The NMR spectra revealed the presence of a [2-13 C]Gln signal in addition to a [2-13 C]Glu signal, demonstrating that Glu taken into the cells is converted to Gln by GS activity (Fig. 2C). No metabolites other than Gln were detected in the cells (Fig. 2C) or in the medium (data not shown). To examine another activated state of microglia, we evaluated M-CSF-stimulated microglia since M-CSF is known to induce proliferation in microglia [17]. We found that the M-CSF-stimulated microglia did not facilitate the conversion of [2-13 C]Glu to [213 C]Gln (data not shown). In addition, astrocytes did not enhance the change of [2-13 C]Glu into [2-13 C]Gln when stimulated with LPS (data not shown). These data from NMR analysis demonstrated that LPS-stimulated microglia change Glu into Gln just like astrocytes.

We investigated the generation of [2-13 C]Gln from [2-13 C]Glu in microglia by conducting an NMR analysis. Although in light of the presence of GS activity/GS protein in nonstimulated microglia we expected the occurrence of a metabolic change of [2-13 C]Glu into [2-13 C]Gln no significant peak of [2-13 C]Gln was observed in the microglia (Fig. 2B). This nonstimulated state does not mean that the cells are unwell or dying, because they maintain the potential to respond to stimulators such as ATP [18] and M-CSF [19]. They are thought to be similar to resting microglia in vitro, keeping their energy metabolic state at low levels. This may be the reason why a significant ATP-dependent Gln synthesis reaction did not occur in nonstimulated microglia despite the presence of GS protein. When microglia were incubated with [2-13 C]Glu under stimulation with LPS, a definite signal of [2-13 C]Gln was observed in the NMR spectrum (Fig. 2C), indicating that microglia biochemically convert [2-13 C]Glu into [2-13 C]Gln. These LPS-stimulated microglia are known to enhance the production of tumor necrosis factor-alpha (TNF-␣) through the activation of PKC␣ and mitogenactivated protein kinase [20]. The activated microglia seem to be energy-metabolically active, and the conversion of Glu to Gln by GS protein could be driven by the active energy metabolism. Although the M-CSF-stimulated microglia were also predicted to show active energy metabolism, we observed no enhancement of the conversion of [2-13 C]Glu to [2-13 C]Gln in these cells (data not shown), suggesting that the role of Glu metabolism in proliferationtriggered microglia is different from its role in LPS-stimulated microglia. Given the difference in Gln synthesis between nonstimulated microglia and LPS-stimulated microglia, microglial GS activity might be significant in the infected state or an inflammatory situation. Supporting this view, GS protein has been detected in microglia in simian immunodeficiency virus-infected macaques [21] and in microglia in injured rat spinal cords [22]. In such disordered conditions, activated microglia would play a role as protective cells that metabolize dangerous Glu into safe Gln. To investigate the signaling mechanism by which the change of [2-13 C]Glu into [2-13 C]Gln was enhanced in LPS-stimulated microglia, we investigated the effects of a PKC␣ inhibitor, myristoylated pseudosubstrate (20–28) (MP), because PKC␣ is closely associated with some biochemical reactions in LPS-stimulated microglia [20]. The pretreatment with 5 ␮M MP prior to LPSstimulation did not significantly reduce LPS-dependent [2-13 C]Gln production (data not shown), suggesting that PKC␣ is not included in the signaling cascade leading to Glu/Gln conversion. Our finding that Glu was converted into Gln in LPS-stimulated microglia (Fig. 2) raised the possibility that the GS activity/GS protein level was upregulated in LPS-stimulated microglia. However, in response to LPS, the GS activity/GS protein level in the microglia

K. Nakajima et al. / Neuroscience Letters 591 (2015) 99–104

103

Fig. 3. GS activity and GS/GLT-1 proteins in LPS-stimulated microglia. (A) GS activity in LPS-stimulated microglia. Microglia were stimulated with LPS (0, 0.5, and 1.0 ␮g/mL) for 24 h, and their GS activities were determined. The data shown are means ± SDs from three independent experiments (*p < 0.05, **p < 0.01) (B) GS, GLT-1, Iba1, and actin protein levels in LPS-stimulated-microglia. Microglia stimulated with LPS (0, 0.5, and 1.0 ␮g/mL) were immunoblotted for GS, GLT-1, Iba1, and actin (Act). A typical profile is shown. (C) Effects of LPS on the GS levels. Microglia were stimulated with LPS (0, 0.5, and 1.0 ␮g/mL) for 24 h, and recovered. Their GS protein levels were analyzed by immunoblotting. The intensity is expressed as the value relative to that of control microglia (0.0 ␮g LPS/mL; defined as 1.0). The data are means ± SDs of three separate experiments (Student’s t-test; *p < 0.05, **p < 0.01). (D) Effects of LPS on the GLT-1 levels. The same samples as those shown in panel C were analyzed for GLT-1. The intensity is expressed as the value relative to that of LPS (1.0 ␮g/mL)-stimulated microglia (defined as 1.0). The data are means ± SDs of three separate experiments (Student’s t-test; *p < 0.05, **p < 0.01).

was significantly reduced (Fig. 3A–C), although the stimulation promoted GLT-1 levels (Fig. 3B and D). At the present moment, we cannot directly explain why the conversion of Glu to Gln was enhanced under reduced GS content in LPS-stimulated microglia. It is likely that the conversion of Glu to Gln in microglia is dependent not only on the amounts of GS, but also on the ATP level and/or NH3 level in addition to the levels of GLT-1. In conclusion, microglia in vitro were found to express GS protein/GS activity and to significantly metabolize [2-13 C]Glu into [2-13 C]Gln when stimulated with LPS. Activated microglia are strongly suggested to change Glu into Gln in inflammatory/pathological states of the nervous system. Conflict of interest The authors have no conflict of interest to declare. Acknowledgment We thank Yoko Tohyama for her excellent technical assistance. References [1] R. Sattler, M. Tymianski, Molecular mechanisms of glutamate receptor-mediated excitotoxic neuronal cell death, Mol. Neurobiol. 24 (2001) 107–129. [2] Y. Wang, Z.H. Qin, Molecular and cellular mechanisms of excitotoxic neuronal death, Apoptosis 15 (2010) 1382–1402. [3] J.H. Yi, A.S. Hazell, Excitotoxic mechanisms and the role of astrocytic glutamate transporters in traumatic brain injury, Neurochem. Int. 48 (2006) 394–403.

´ [4] L. Struzynska, A glutamatergic component of lead toxicity in adult brain: the role of astrocytic glutamate transporters, Neurochem. Int. 55 (2009) 151–156. [5] F. Lopéz-Redondo, K. Nakajima, S. Honda, S. Kohsaka, Glutamate transporter GLT-1 is highly expressed in activated microglia following facial nerve axotomy, Mol. Brain Res. 76 (2000) 429–435. [6] F.K. van Landeghem, J.F. Stover, I. Bechmann, W. Brück, A. Unterberg, C. Bührer, A. von Deimling, Early expression of glutamate transporter proteins in ramified microglia after controlled cortical impact injury in the rat, Glia 35 (2001) 167–179. [7] R. Beschorner, K. Dietz, N. Schauer, M. Mittelbronn, H.J.K. Schluesener, Trautmann, R. Meyermann, P. Simon, Expression of EAAT1 reflects a possible neuroprotective function of reactive astrocytes and activated microglia following human traumatic brain injury, Histol. Histopathol. 22 (2007) 515–526. [8] R. Beschorner, P. Simon, N. Schauer, M. Mittelbronn, H.J. Schluesener, K. Trautmann, K. Dietz, R. Meyermann, Reactive astrocytes and activated microglial cells express EAAT1 but not EAAT2, reflecting a neuroprotective potential following ischaemia, Histopathology 50 (2007) 897–910. [9] J. Jacobsson, M. Persson, E. Hansson, L. Rönnbäck, Corticosterone inhibits expression of the microglial glutamate transporter GLT-1 in vitro, Neuroscience 139 (2006) 475–483. [10] R.D. O’Shea, C.L. Lau, M.C. Farso, S. Diwakarla, C.J. Zagami, B.B. Svendsen, S.J. Feeney, J.K. Callaway, N.M. Jones, D.V. Pow, B. Danbolt, P.M. Beart, Effects of lipopolysaccharide on glial phenotype and activity of glutamate transporters: evidence for delayed up-regulation and redistribution of GLT-1, Neurochem. Int. 48 (2006) 604–610. [11] A. Schousboe, H.S. Waagepetersen, Role of astrocytes in glutamate homeostasis: implications for excitotoxicity, Neurotox. Res. 8 (2005) 221–225. [12] K. Nakajima, Y. Tohyama, S. Maeda, S. Kohsaka, T. Kurihara, Neuronal regulation by which microglia enhance the production of neurotrophic factors for GABAergic catecholaminergic, and cholinergic neurons, Neurochem. Int. 50 (2007) 807–820. [13] S. Maeda, K. Nakajima, Y. Tohyama, S. Kohsaka, Characteristic response of astrocytes to plasminogen/plasmin to upregulate transforming growth factor beta 3 (TGF␤3) production/secretion through proteinase-activated receptor-1

104

[14] [15]

[16]

[17]

[18]

K. Nakajima et al. / Neuroscience Letters 591 (2015) 99–104 (PAR-1) and the downstream phosphatidylinositol 3-kinase (PI3K)-Akt/PKB signaling cascade, Brain Res. 1305 (2009) 1–13. K. Iqbal, J.H. Ottaway, Glutamine synthetase in muscle and kidney, Biochem. J. 119 (1970) 145–156. T. Kanamatsu, Y. Tsukada, Effects of ammonia on the anaplerotic pathway and amino acid metabolism in the brain: an ex vivo 13 C NMR spectroscopic study of rats after administering [2-13 C] glucose with or without ammonium acetate, Brain Res. 841 (1999) 11–19. K. Nakajima, Y. Tohyama, S. Kohsaka, T. Kurihara, Ability of rat microglia to uptake extracellular glutamate, Neurosci. Lett. 307 (2001) 171–174. S. Yamamoto, K. Nakajima, S. Kohsaka, Macrophage-colony stimulating factor (M-CSF) as an inducer of microglial proliferation in axotomized rat facial nucleus, J. Neurochem. 115 (2010) 1057–1067. K. Inoue, K. Nakajima, T. Morimoto, Y. Kikuchi, S. Koizumi, P. Illes, S. Kohsaka, ATP stimulation of Ca2+ -dependent plasminogen release from cultured microglia, Br. J. Pharmacol. 123 (1998) 1304–1310.

[19] S. Yamamoto, S. Kohsaka, K. Nakajima, Role of cell cycle-associated proteins in microglial proliferation in the axotomized rat facial nucleus, Glia 60 (2012) 570–581. [20] K. Nakajima, Y. Tohyama, S. Kohsaka, T. Kurihara, Protein kinase C␣; requirement in the activation of p38 mitogen-activated protein kinase, which is linked to the induction of tumor necrosis factor alpha in lipopolysaccharide-stimulated microglia, Neurochem. Int. 44 (2004) 205–214. [21] F. Chrétien, A.V. Vallat-Decouvelaere, C. Bossuet, A.C. Rimaniol, R. Le Grand, G. Le Pavec, C. Créminon, D. Dormont, F. Gray, G. Gras, Expression of excitatory amino acid transporter-2 (EAAT-2) and glutamine synthetase (GS) in brain macrophages and microglia of SIVmac251-infected macaques, Neuropathol. Appl. Neurobiol. 28 (2002) 410–417. [22] C. Liu, W. Wu, B. Zhang, J. Xiang, J. Zou, Temporospatial expression and cellular localization of glutamine synthetase following traumatic spinal cord injury in adult rats, Mol. Med. Rep. 7 (2013) 1431–1436.