SEARCH

SEARCH BY CITATION

Keywords:

  • AMPAR;
  • amyotrophic lateral sclerosis;
  • calcium-permeable;
  • excitotoxicity;
  • trafficking;
  • tumour necrosis factor alpha

Abstract

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

J. Neurochem. (2010) 113, 692–703.

Abstract

The α-amino-3-hydroxyl-5-methyl-4-isoxazole-propionate receptor (AMPAR) subunit GluR2, which regulates excitotoxicity and the inflammatory cytokine tumour necrosis factor alpha (TNFα) have both been implicated in motor neurone vulnerability in amyotrophic lateral sclerosis/motor neurone disease. TNFα has been reported to increase cell surface expression of AMPAR subunits to increase synaptic strength and enhance excitotoxicity, but whether this mechanism occurs in motor neurones is unknown. We used primary cultures of mouse motor neurones and cortical neurones to examine the interaction between TNFα receptor activation, GluR2 availability, AMPAR-mediated calcium entry and susceptibility to excitotoxicity. Short exposure to a physiologically relevant concentration of TNFα (10 ng/mL, 15 min) caused a marked redistribution of both GluR1 and GluR2 to the cell surface as determined by cell surface biotinylation and immunofluorescence. Using fura-2-acetoxymethyl ester microfluorimetry, we showed that exposure to TNFα caused a rapid reduction in the peak amplitude of AMPA-mediated calcium entry in a PI3-kinase and p38 kinase-dependent manner, consistent with increased insertion of GluR2-containing AMPAR into the plasma membrane. This resulted in a protection of motor neurones against kainate-induced cell death. Our data therefore, suggest that TNFα acts primarily as a physiological regulator of synaptic activity in motor neurones rather than a pathological drive in amyotrophic lateral sclerosis.

Abbreviations used:
ALS

amyotrophic lateral sclerosis

AMPA

α-amino-3-hydroxyl-5-methyl-4-isoxazole-propionate

AMPAR

AMPA receptor

ECL

enhanced chemiluminescence

HBM

HEPES buffered medium

PBS

phosphate-buffered saline

PI3K

PI3-kinase

SDS

sodium dodecyl sulfate

TNFR

tumor necrosis factor receptor

TNFα

tumour necrosis factor-alpha

In amyotrophic lateral sclerosis (ALS)/motor neurone disease there is substantial evidence implicating glutamate-mediated excitotoxicity (Rattray and Bendotti 2006). CSF from ALS patients is toxic to cultured neurones and this toxicity is blocked by α-amino-3-hydroxyl-5-methyl-4-isoxazole-propionate (AMPA) receptor (AMPAR) antagonists (Couratier et al. 1993). AMPAR antagonists are neuroprotective in transgenic mice carrying the familial ALS-linked mutation SOD1G93A (Canton et al. 2001; Van Damme et al. 2003) and motor neurone vulnerability may be because of high levels of calcium-permeable AMPARs lacking the GluR2 subunit (Roy et al. 1998; Carriedo et al. 2000; Vandenberghe et al. 2000; Van Damme et al. 2002; Tateno et al. 2004).

Inflammation is also part of the pathogenic process in ALS (Lobsiger and Cleveland 2007). Pro-inflammatory cytokines, including tumour necrosis factor-alpha (TNFα), are up-regulated in spinal cords of ALS patients, increase pre-symptomatically in SOD1G93A mice (Elliott 2001; Nguyen et al. 2001; Hensley et al. 2002; McGeer and McGeer 2002; Yoshihara et al. 2002) and stimulate p38 kinase phosphorylation (Veglianese et al. 2006). TNFα is a cell death signal in developing and mature motor neurones (Robertson et al. 2001; He et al. 2002; Sedel et al. 2004; Wen et al. 2006; Leonoudakis et al. 2008) and produces changes to neuronal mitochondria similar to those observed in ALS (Stommel et al. 2007). Blocking TNFα signalling, or knockout of TNF receptor (TNFR)1 and TNFR2, rescues motor neurones from degeneration in Wobbler mice and protects motor neurones against experimental injury (Terrado et al. 2000; Raivich et al. 2002; Bigini et al. 2008). Despite this evidence, controversy exists as to whether TNFα is neuroprotective or neurotoxic (Ghezzi and Mennini 2001; Chadwick et al. 2008). While TNFα can potentiate excitotoxicity in many types of neurone (e.g. Chao and Hu 1994; Yu et al. 2002; Ferguson et al. 2008), it can also be protective (Barger et al. 1995; Cheng et al. 1994; Carlson et al. 1998, 1999; Marchetti et al. 2004; Dolga et al. 2008), or both (Bernardino et al. 2005). Transgenic mice lacking TNFRs are more sensitive to excitotoxicity (Bruce et al. 1996) and knockout of TNF in SOD1G93A mice does not delay disease progression (Gowing et al. 2006). There is little evidence that addresses the mechanisms underlying these effects in motor neurones.

TNFα may control sensitivity to excitotoxicity in motor neurones by modulating the membrane levels of glutamate receptor subunits. TNFα directly increases the surface expression of the AMPAR subunit, GluR1, in hippocampal neurones (Beattie et al. 2002; De et al. 2003; Ogoshi et al. 2005), corresponding with alterations in synaptic strength. The key mediator of excitotoxicity is the GluR2 subunit, which determines AMPAR calcium permeability (Hollmann et al. 1991), and is also regulated by TNFα (Ogoshi et al. 2005; Leonoudakis et al. 2008). TNFα modulation of GluR2 could determine motor neurone susceptibility to excitotoxicity. We have directly tested the interaction between TNFR activation, GluR2 availability, AMPAR function and excitotoxicity in motor neurones. Here, we report that TNFα causes a rapid reduction in AMPAR-evoked calcium entry into motor neurones coincident with increased levels of GluR2 at the cell surface and this confers protection against excitotoxicity.

Materials and methods

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

All reagents were from Sigma-Aldrich (Poole, UK) unless otherwise stated.

Primary cultures of motor neurones and cortical neurones

All reagents for cell culture were obtained from Invitrogen (Paisley, UK) unless otherwise stated. Pregnant NIH Swiss mice, obtained from Harlan, UK, were killed by CO2 inhalation in accordance with UK Home Office Regulations. Motor neurone-enriched cultures from E15 embryos were prepared by a modification of published methods (Camu and Henderson 1992; Duong et al. 1999). Spinal cords were cut into approximately 15 pieces and trypsinised for 15 min at 37°C, before resuspension in Neurobasal medium supplemented with 10% heat-inactivated horse serum, 2 mM l-glutamine, 1 ng/mL BDNF (Sigma-Aldrich), 0.5% penicillin/streptomycin, 16.5 mM glucose (Sigma-Aldrich), 10% mixed hormone preparation (100 μg/mL apo-transferrin, 25 μg/mL insulin, 60 μM putrescine, 20 μM progesterone, and 30 μM sodium selenite; Sigma-Aldrich), and further treatment with DNase (Lorne Laboratories, Reading, UK). The cell suspension was layered onto a 4% bovine serum albumin cushion and centrifuged at 25°C for 10 min at 300 g and the remaining pellet resuspended in 1 mL of supplemented Neurobasal medium and layered onto a 10.4% (w/v) solution of iodixanol (Optiprep; Axis-Shield, Dundee, UK) in L15 medium before centrifugation at 25°C for 15 min at 500 g without braking. Cells from the interface were collected by centrifugation through a 4% bovine serum albumin cushion and maintained at 37°C in an atmosphere of 5% CO2 on poly-ornithine/laminin coated plates. Neurones were grown on glass coverslips, coated with 15 μg/mL poly-ornithine and 3 μg/mL laminin (Sigma-Aldrich), in 24 well tissue culture plates for 7–14 days, or in poly-ornithine coated 90 mm diameter culture dishes for 12–14 days (unless otherwise stated). Primary cortical neurones from E15 embryos were prepared as previously described in detail by Schroeter et al. 2007;. Primary cortical neurones were used for biochemical studies including RT-PCR and western blotting, as described previously by Tortarolo et al. 2004.

Immunofluorescence

Cells grown on glass coverslips were washed twice with phosphate-buffered saline (PBS) then fixed in 4% paraformaldehyde (30 min, 25°C), washed with PBS then permeabilised and blocked with 0.2% Triton X-100, 1% normal goat serum, in PBS (30 min) prior to overnight incubation at 4°C with primary antibody. The antibodies used were as follows: affinity purified C-terminal directed rabbit polyclonals against GluR1 (1 : 200), GluR2 (1 : 500) or, the AMPAR subunit GluR4 (1 : 100) or a mouse monoclonal against the AMPAR subunit GluR3 (1 : 250) (all from Millipore, Billerica, MA, USA); mouse anti-non-phosphoneurofilament H monoclonal antibody (SMI-32; 1 : 1000; Covance, Princeton, NJ, USA); or rabbit anti-β-Tubulin III antibody (1 : 250; Sigma-Aldrich). Coverslips were washed thrice in PBS before incubation for 90 min at 25°C with secondary antibody (Alexa Fluor 488 goat anti-rabbit IgG, or Alexa Fluor 488 goat anti-mouse IgG; 1 : 500; Invitrogen). Finally, cells were washed thrice for 5 min in PBS before mounting in mowiol containing Hoechst 33 342 (Invitrogen). For analysis of cell surface AMPARs, cells were treated as above but without permeabilisation by omission of Triton X-100. Extracellular N-terminal epitope directed primary antibodies were used; rabbit anti-GluR1 (1 : 10; Calbiochem, San Diego, CA, USA), or mouse anti-GluR2 (1 : 100; Invitrogen). Cells were visualised and images captured using a Carl Zeiss Axioplan 2 microscope rig with AxioVision Imaging software (Carl Zeiss MicroImaging Inc., Thornwood, NY, USA).

RT-PCR

Total RNA was extracted from 8- and 13-day-old primary mouse cortical neurones, grown in tissue culture dishes, using the RNeasy Mini Kit (Qiagen, Crawley, UK). RNA (3 μg) was reverse transcribed for 60 min at 42°C followed by 5 min at 99°C, using 2.5 U/μL M-MLV reverse transcriptase in the presence of 5 mM MgCl2, 12.5 μg/mL oligo(dT), 1 U/μL RNasin in 1× RT buffer (all from Promega, Southampton, UK) and 1 mM dNTPs (GE Healthcare Life Sciences, Little Chalfont, UK). Oligonucleotide primers used to amplify specific regions of the GluR1, GluR2, GluR3 and GluR4 genes were designed using Primer3 and checked for hairpins and dimers using NetPrimer. The primers used were as follows: GluR1 forward: TCCGCAAGATTGGTTACTGG; GluR1 reverse: CAGATCTCGTAGGCCAAAGG; GluR2 forward: AATAGAAAGGGCCCTCAAGC; GluR2 reverse: ATTCCAAGGCTCATGAATGG; GluR3 forward: CCCTTATGAGTGGCACTTGG; GluR3 reverse: TGCAATTTCAGTCTGCTTGG; GluR4 forward: GAAGCACGTCAAAGGCTACC; GluR4 reverse: TTCCAATAGCCAACCTTTCG. PCR amplification was carried out in a 25 μL reaction containing 0.1 μM forward primer, 0.1 μM reverse primer, 2.5 μL cDNA, 0.025 U/μL Taq DNA polymerase (Promega), 0.05 mM dNTPs and 1.5 mM MgCl2 in 1× buffer (Promega) for 40 cycles. PCR products were separated by 1% agarose gel electrophoresis and visualised by SYBR Safe (Invitrogen) staining under UV transillumination. Images were recorded using the GeneSnap SYNGENE system (Syngene, Cambridge, UK).

Western blotting

Primary mouse cortical neurones grown in tissue culture dishes were subjected to lysis and western blot analysis as described in detail previously by Tortarolo et al. 2004. Blots were incubated overnight at 4°C with primary antibody: rabbit anti-GluR1 (1 : 400), rabbit anti-GluR2 (1 : 500), mouse anti-GluR3 (1 : 500), or rabbit anti-GluR4 (1 : 400) (all C-terminal directed antibodies from Millipore). Blots were incubated for 45 min at 25°C with secondary antibodies: either goat anti-rabbit peroxidase-conjugated IgG (1 : 1000; Sigma-Aldrich), or affinity purified goat anti-mouse peroxidase-conjugated IgG (1: 2000; Millipore). Protein bands were detected using enhanced chemiluminescence (ECL) western blotting detection reagents with Hyperfilm-ECL (GE Healthcare Life Sciences), and analysed using ImageJ software (Wayne Rasband, National Institutes of Health, Bethesda, MD, USA). Statistical analysis of mean band optical densities was carried out using GraphPad Prism software (GraphPad Software, San Diego, CA, USA).

AMPA receptor cell surface biotinylation

To determine the amount of AMPAR subunit cell surface localisation in cortical neurones, cell-surface biotinylation experiments were performed essentially as described by Peacey et al. 2009. Primary mouse cortical neurones grown in tissue culture dishes were washed twice with HEPES-buffered medium (HBM) [20 mM HEPES (Invitrogen), 140 mM NaCl, 5 mM KCl, 5 mM NaHCO3, 1.2 mM Na2HPO4, 5.5 mM glucose, 1.2 mM CaCl2, pH 7.4], then incubated in HBM ± 10 ng/mL recombinant mouse TNFα (R & D Systems, Minneapolis, MN, USA) for 15 min at 37°C. The cells were rinsed twice at 37°C with PBS containing calcium and magnesium (138 mM NaCl, 2.7 mM KCl, 1.5 mM KH2PO4, 9.6 mM Na2HPO4, 1 mM MgCl2, and 0.1 mM CaCl2, pH 7.4), before being incubated with 3 mL of a solution containing 1 mg/mL sulfo-NHS biotin (Pierce, Rockford, IL, USA) for 20 min at 4°C with gentle shaking. The biotinylation solution was removed by two washes in PBS containing 100 mM glycine prior to further incubation in PBS/glycine for 45 min at 4°C with gentle shaking. The cells were then lysed in radioimmunoprecipitation assay buffer [100 mM Tris–Cl, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 1% sodium deoxycholate, 0.1% sodium dodecyl sulfate (SDS), 1 μg/mL leupeptin, 1 μg/mL aprotinin, and 250 μM phenylmethylsulfonyl fluoride] at 4°C for 60 min. The cell lysates were centrifuged at 18 000 g at 4°C for 20 min. Of the supernatant fractions, 200 μL were retained (total cell lysate) and immediately frozen, and 300 μL were incubated with an equal volume of immunopure immobilised monomeric avidin beads (Pierce) at 25°C for 60 min. Following avidin incubation, samples were centrifuged at 17 860 g at 4°C for 5 min, after which 300 μL of the supernatants (intracellular fraction) were retained and immediately frozen. The beads were then washed thrice with radioimmunoprecipitation assay buffer, and adsorbed proteins were eluted with SDS sample buffer [62.5 mM Tris, pH 6.8, 2% (w/v) SDS, 5% (v/v) 2-mercaptoethanol, 10% (v/v) glycerol, and 0.0025% (w/v) bromophenol blue] at 25°C for 30 min (biotinylated fraction). The total cell lysate, intracellular fraction, and biotinylated fraction were analysed by western blotting as described above. Samples were probed for GluR1 or GluR2 with C-terminal directed antibodies from Millipore. Protein bands were detected using ECL western blotting detection reagents with Hyperfilm-ECL (GE Healthcare Life Sciences), and analysed using ImageJ software (Wayne Rasband, National Institutes of Health). Band optical densities were normalised as follows: the cell surface membrane (M) fraction and intracellular/cytosolic (C) fraction were calculated as M/(M + C) and C/(M + C) respectively. Data were plotted as mean ± SEM from three independent experiments, and statistically analysed using GraphPad Prism software.

Microfluorometric imaging of intracellular calcium

Primary mouse cortical or motor neurones grown on coverslips were loaded with fura-2-acetoxymethyl ester (5 μM; Invitrogen) in the presence of 1 mM probenecid, at 37°C for 45 min. Dye loading and subsequent experiments (unless otherwise stated) were performed in a sodium-based assay buffer (140 mM NaCl, 5 mM KCl, 10 mM glucose, 10 mM HEPES, 2 mM CaCl2, and 1 mM MgCl2, pH 7.4) containing 50 μM cyclothiazide (Tocris Bioscience, Bristol, UK). In some cases, as described in the text, experiments were carried out in a sodium-free assay buffer (140 mM KCl, 10 mM glucose, 10 mM HEPES, 2 mM CaCl2, and 1 mM MgCl2, pH 7.4), also containing 50 μM cyclothiazide. The potassium-containing buffer is used as an equimolar replacement for sodium. The purpose of the buffer is to prevent the possibility of AMPA-receptor mediated sodium entry into cells causing calcium changes through, for example Sodium–calcium exchange.

Recombinant mouse TNFα (final concentration 10 or 150 ng/mL), was added directly into wells for the final 15 min of the dye loading period. When cells were pre-treated for 24 h, TNFα (final concentration 10 or 150 ng/mL) was added directly into culture medium for 23 h and 15 min and was present for the 45 min duration of the dye loading period. When used, the PI3-kinase (PI3K)inhibitor wortmannin (150 nM; Tocris Bioscience), or the p38 kinase inhibitor SB203580 (500 nM; Tocris Bioscience) was added directly into wells for the final 25 min of the dye loading period. Compounds [50 μM AMPA (Tocris Bioscience), 50 mM KCl] were applied to cells at 25°C by local microperfusion. In experiments where voltage-gated calcium channel blockers were used [final concentration; 10 μM Nimodipine (Tocris Bioscience), 100 nM ω-Conotoxin GVIA, 100 nM ω-Conotoxin MVIIC, 100 nM ω-Agatoxin IVA, 100 nM SNX-482 (Alomone Labs, Jerusalem, Israel)], AMPA (final concentration 50 μM) was applied locally without microperfusion. Images of individual cells (typically 15–20 per field of view for motor neurones and 30–40 per field of view for cortical neurones) were captured every 2 s at 340 and 380 nm excitation wavelengths, with emission measured at 520 nm, using a microscope-based imaging system (Photon Technology International, Birmingham, NJ, USA). Analyses of emission intensity ratios at 340/380 nm were performed with the ImageMaster suite software (Photon Technology International). Data were statistically analysed using GraphPad Prism software.

Motor neurone viability

Immunofluorescence was used to determine whether TNFα pre-treatment sensitises primary mouse motor neurones to kainate-induced excitotoxicity. Cells were grown on coverslips for 7 days then washed twice with HBM prior to administration of 10 ng/mL recombinant mouse TNFα or control for 15 min at 37°C. Medium was replaced and kainate (Tocris Bioscience) added directly to all wells (except control) to give a final concentration of 100 or 300 μM. Cells were incubated for 16 h prior to fixation and staining for SMI-32 by immunofluorescence as described above. The viability of motor neurones was assayed as follows: SMI-32-positive cells, with typical morphology were counted, at a magnification of 40×, following the width of the cover slip in non-overlapping pathways. Cells with a pyramidal cell body > 20 μm in diameter, and at least one long process were included. The number of cells counted in treated wells was calculated as a percentage of mean SMI-32-positive cells counted in untreated control wells within each separate experiment. Data were analysed using GraphPad Prism software.

Results

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

GluR2 and GluR1 AMPA subunits are expressed in motor neurones and are rapidly recruited to the cell surface following 15 min exposure to a physiologically relevant concentration of TNFα

We first characterised the expression of AMPAR subunits in mouse motor neurone cultures, using cortical neurones as a comparator. In motor neurones, using immunofluorescence with antibodies directed against intracellular domains, we detected abundant GluR1 and GluR2 with little observable GluR3 and GluR4 (Fig. 1), this distribution was similar in cultures grown for 6 days up to 14 days in vitro (data not shown). We also determined the subunit expression in cortical neurone cultures derived from the same embryos as motor neurones because it is possible to generate larger numbers of cortical neurones compared with motor neurones to carry out biochemical analysis. Immunofluorescence showed a similar expression of glutamate receptor subunits to motor neurones in that GluR1 and GluR2 are expressed abundantly with lower levels of expression of GluR3 and GluR4 (data not shown), this was confirmed by western blotting and RT-PCR (Fig. 1). As cortical neurones have similar AMPAR subunit composition to motor neurones, and to determine whether, in our hands, TNFα can cause redistribution of glutamate receptor subunits, we first used cell-surface biotinylation and chose a physiologically relevant, low concentration of recombinant mouse TNFα. TNFα (10 ng/mL = 588 pM) was applied for 15 min to the cultures before application of the sulphobiotin reagent and cell lysis. TNFα increased the levels of GluR1 and GluR2 at the cell surface of cortical neurones with a corresponding decrease in intracellular levels (Fig. 2). For GluR1 cell surface levels expressed as a proportion of total immunoreactivity in the cell extracts was increased from 85 ± 1.5% to 95 ± 0.62% (= 3, < 0.01), and for GluR2 from 48 ± 1.6% to 60 ± 4.1% (= 3, < 0.05). These data are consistent with published data on TNFα in hippocampal neurones showing enhanced cell surface GluR1 and GluR2 (Beattie et al. 2002; Ogoshi et al. 2005; Stellwagen et al. 2005; Leonoudakis et al. 2008). However, the effects on GluR2 reported here are more rapid than those described by Leonoudakis et al. 2008.

image

Figure 1.  Motor neurones and cortical neurones express α-amino-3-hydroxyl-5-methyl-4-isoxazole-propionate receptor subunits. Representative immunofluorescence micrographs of motor neurones expressing SMI-32, β-III tubulin, GluR1, GluR2, GluR3 and GluR4 (green) are shown. Hoechst 33 342-stained nuclei appear blue. Scale bar, 50 μm, except SMI-32 where it represents 70 μm. RT-PCR shows expression of GluR1, GluR2, GluR3 and GluR4 subunit mRNA in 8 (a) and 13 (b) div cortical neurones. Bioline hyperladders, with band sizes in base pairs, are shown. Representative immunoblots demonstrate expression of GluR1 (106 kDa), GluR2 (108 kDa), GluR3 (110 kDa), and GluR4 (101 kDa) subunits in cortical neurones.

Download figure to PowerPoint

image

Figure 2.  Tumour necrosis factor-α (TNFα) induces rapid increases in cell surface expression of GluR1 and GluR2. (a, b) Representative immunoblots from cell surface biotinylation experiments, probed for GluR1 (a) or GluR2 (b). The 14 DIV cortical neurone sister cultures were either treated with TNFα for 15 min or left untreated (control) before undergoing biotinylation. C denotes cytosolic and M denotes cell surface membrane fractions. (c, d) Quantification of all experiments for the levels of cytosolic (C) and cell surface membrane (M) GluR1 (c) and GluR2 (d). Cytosolic and membrane receptor levels are represented graphically as a percentage of total band intensity (where total is equal to cytosolic plus membrane band intensity for each receptor treatment group). Error bars indicate SEM. *< 0.05; **< 0.01 compared with relative control groups; Student’s unpaired t-test; = 3 separate experiments.

Download figure to PowerPoint

As cell surface biotinylation was not possible for motor neurone cultures, to discover whether AMPARs are recruited to the cell surface of motor neurones after TNFα treatment, we carried out immunofluorescent analysis using antibodies directed against the extracellular N-terminal domains of the GluR1 and GluR2 subunits which under the non-permeabilisation conditions employed are indicative of subunit expression at the cell surface (see Lu et al. 2001). The staining pattern confirmed that GluR1 and GluR2 were expressed at the cell surface. A short exposure to TNFα (10 ng/mL, 15 min) caused a marked redistribution of GluR1 and GluR2 in many motor neurones, so that there was increased levels of immunoreactivity which, for GluR2, was particularly prominent in the motor neurone processes (Fig. 3). The results suggest that, in a similar way to cortical neurones, GluR1 and GluR2 are rapidly increased at the cell surface of motor neurones following a short exposure to TNFα. The rapid nature of this response excludes the possibility that GluR1 and GluR2 surface alterations represent de novo protein synthesis.

image

Figure 3.  Tumour necrosis factor-α (TNFα) induces rapid increases in cell surface expression of GluR1 and GluR2 in motor neurones as shown by representative immunofluorescence micrographs. The 7 DIV motor neurones were either left untreated or incubated with TNFα for 15 min immediately prior to fixation. Neurones were processed without membrane permeabilisation for indirect immunofluorescence using antibodies directed against extracellular epitopes of GluR1 and GluR2 (green). Hoechst 33 342-stained nuclei appear blue. Exposure matched images were captured using a Carl Zeiss Axioplan 2 microscope with AxioVision Imaging software. Arrows indicate punctate staining. Scale bar, 50 μm.

Download figure to PowerPoint

TNFα rapidly reduces AMPA receptor-mediated calcium entry into motor neurones in a PI3-kinase- and p38 kinase-dependent manner

The subunit composition of AMPARs determines their ion channel properties, receptors that possess the edited GluR2 subunit show low calcium permeability in comparison to receptors without GluR2 which have high calcium permeability (Hollmann et al. 1991). Motor neurones have previously been shown to express high levels of calcium-permeable AMPARs (Carriedo et al. 1996). To determine whether changes in cell surface levels of GluR1 and GluR2 subunits resulted in altered AMPAR-mediated increases in intracellular calcium, we used fura-2-acetoxymethyl ester microfluorimetry. First, we measured the intracellular calcium response to AMPA in motor neurones and cortical neurones. AMPA (50 μM) was applied for 25 s followed by superfusion in the absence of agonist. AMPA caused a rapid increase in intracellular calcium (Fig. 4, see Figure S1 for examples of individual responses from cortical neurones). The peak calcium increase in motor neurones was greater than in cortical neurones, suggesting higher levels of calcium-permeable AMPA subunits in motor neurones compared with cortical neurones. For both types of neurone the response to AMPA was completely blocked by pre-incubation with the selective non-competitive AMPAR antagonist GYKI-52 466 (50 μM) or the competitive antagonist 2,3-dihydroxy-6-nitro-7-sulfamoyl-benzo[f]quinoxaline-2,3-dione (10 μM) (data not shown). For clarity, Figs 4 and 5 show mean responses from four to six separate experiments, where each experiment is itself the pooled response of multiple neurones. Within each experiment, there is a wide range of individual neuronal responses (results not shown), showing a degree of biological variation in the density of calcium-permeable AMPA receptors present fon cultured neurones. To analyse such variation between individual neurones is beyond the scope of this study.

image

Figure 4.  α-Amino-3-hydroxyl-5-methyl-4-isoxazole-propionate (AMPA) evokes a rapid increase in intracellular calcium, as shown by Fura-2-acetoxymethyl ester (fura-2AM) microfluorimetry. (a, b) Representative traces from Fura-2AM loaded motor neurones (a) or cortical neurones (b) superfused with AMPA (50 μM) and KCl (50 mM) in a Na+ containing buffer. (c, d) AMPA (50 μM) was administered locally to fura-2AM-loaded motor neurones incubated in Na+-free buffer containing a cocktail of voltage-gated calcium channel blockers and cyclothiazide (50 μM). The AMPA-induced intracellular calcium rise (c) was antagonised by 50 μM GYKI-52 466 (d). fura-2AM fluorescence was measured using a microscope-based imaging system, analysed with ImageMaster suite software and is represented graphically as the average 340 nm/380 nm ratio of 15–40 cells monitored individually. TNFα, Tumour necrosis factor-α.

Download figure to PowerPoint

image

Figure 5.  Tumour necrosis factor-α (TNFα) rapidly reduces α-amino-3-hydroxyl-5-methyl-4-isoxazole-propionate (AMPA) receptor-mediated calcium entry into motor neurones. (a, b) Representative traces from fura-2-gacetoxymethyl ester(fura-2AM)-loaded motor neurones. Cells were left untreated (a) or pre-treated with TNFα (b) for 15 min prior to local administration of AMPA (50 μM) in Na+ free buffer containing a cocktail of voltage-gated calcium channel blockers and cyclothiazide (50 μM). (c) The average change in AMPA-evoked 340 nm/380 nm ratio was decreased following TNFα incubation in both Na+-free and Na+-containing buffer with cyclothiazide (50 μM), in the presence and absence respectively of a cocktail of voltage-gated calcium channel blockers. Error bars indicate SEM. **< 0.01; Student’s unpaired t-test; ≥ 6 separate experiments where each n represents 15–19 cells monitored individually. (d) TNFα applied alone in either buffer did not affect intracellular calcium levels. Fura-2AM fluorescence was measured using a microscope-based imaging system, analysed with ImageMaster suite software and is represented graphically (a, b, d) as the average 340 nm/380 nm ratio of 15–22 cells monitored individually.

Download figure to PowerPoint

As calcium increases in neurones in response to AMPA may be indirect, i.e. following depolarisation-induced opening of voltage-gated calcium channels and removal of voltage-dependent magnesium block from NMDA receptors, measurements were made in a sodium-free buffer containing the calcium channel blockers Nimodipine (L-type channel blocker), ω-Conotoxin GVIA (N-type), ω-Conotoxin MVIIC (Q-type), ω-Agatoxin IVA (P/Q-type), and SNX-482 (R type). To eliminate the confounding effects of desensitisation and to measure the maximum AMPA-evoked calcium increase, the AMPAR desensitisation inhibitor cyclothiazide (50 μM) was also included (Williams and Glowinski 1996; Perkinton et al. 1999). In motor neurones, AMPA (50 μM) caused a rapid increase in intracellular calcium under these conditions (Fig. 4c) that was completely abolished by the AMPAR-selective antagonist GYKI 52 466 (50 μM) (Fig. 4d). These data confirm that motor neurones express high levels of GluR2-lacking AMPARs that are directly permeable to calcium (Carriedo et al. 1996, 2000; Roy et al. 1998; Van Damme et al. 2002).

To determine whether TNFα altered the AMPAR-mediated increase in intracellular calcium, motor neurones were pre-incubated with TNFα (10 ng/mL) for 15 min. The treatment caused a significant reduction in peak amplitude, accompanied by an alteration in kinetics from a sustained to a transient response (Fig. 5). The peak amplitude was reduced by 30 ± 8.5% (= 6, < 0.01), showing that TNFα reduced AMPAR-mediated calcium entry. To determine whether this effect was present under more physiologically relevant conditions, the experiments were repeated in a sodium-containing buffer without calcium channel blockers. Under these conditions, TNFα (10 ng/mL, 15 min) also caused a significant reduction in peak amplitude (28 ± 9.3%, = 6, < 0.01). TNFα alone had no effect on calcium levels under any conditions tested (Fig. 5d) which is in contrast to a previous study where TNFα caused calcium influx and potentiated NMDA receptor-dependent calcium influx (Jara et al. 2007). Our data show that TNFα induces increased levels of GluR2 at the cell surface and this correlates with reduced calcium entry in response to AMPA.

TNFα effects can be related to concentration and the duration of exposure (e.g. Bernardino et al. 2005; Leonoudakis et al. 2008). Since much published data reporting TNFα effects on AMPARs has used supramaximal concentrations of TNFα (e.g. Beattie et al. 2002; Stellwagen et al. 2005; Stellwagen and Malenka 2006; Leonoudakis et al. 2008) we compared the effects of incubation with TNFα at 10 ng/mL with that of a higher concentration (150 ng/mL). In experiments designed to test whether the inhibitory effect of TNFα was time- and concentration-dependent, the change in 340 nm/380 nm ratio induced by AMPA in the absence of TNFα was 2.65 ± 0.18 (= 6). Administration of TNFα 10 ng/mL, 15 min caused a highly significant 31 ± 10% reduction in peak calcium levels. Changing the time of TNFα exposure or the concentration of TNFα did not alter TNFα-induced reductions in AMPA responses: TNFα 10 ng/mL for 24 h caused a 32 ± 5% reduction (= 3), TNFα 150 ng/mL for 15 min caused a 20 ± 10% reduction (= 3), TNFα 150 ng/mL for 24 h caused a 26 ± 8% reduction (= 3).

It has previously been shown that TNFα-induced changes in the membrane levels of GluR1 are dependent on PI3-kinase (Stellwagen et al. 2005). The PI3-kinase inhibitor, wortmannin (150 nM) completely abolished the TNFα-mediated reduction of intracellular calcium, demonstrating complete PI3-kinase dependency of the TNFα effect (Fig. 6). Wortmannin also reduced AMPA-mediated increases of intracellular calcium, showing a PI3-kinase involvement consistent with PI3-kinase dependency of AMPAR signalling and trafficking in neurones (Perkinton et al. 1999; Man et al. 2003). Effects of TNFα receptor signalling can be mediated through p38 kinase (Yuasa et al. 1998) and the potent and selective p38 kinase inhibitor SB203580 (500 nM) completely abolished the TNFα-mediated reduction of intracellular calcium, demonstrating complete dependence of the TNFα effect on p38 kinase, it did not however, alter the AMPA response alone (Fig. 6).

image

Figure 6.  Tumour necrosis factor-α (TNFα) induced decreases in α-amino-3-hydroxyl-5-methyl-4-isoxazole-propionate (AMPA) receptor-mediated calcium entry are PI3-kinase and p38 kinase dependent, as shown by Fura-2-acetoxymethyl ester microfluorimetry. Prior to application of AMPA (50 μM), motor neurones were left untreated or pre-treated with TNFα alone for 15 min, or in combination with the PI3-kinase inhibitor wortmannin (150 nM) or the p38 kinase inhibitor SB203580 (500 nM). AMPA was administered in Na+-free buffer containing a cocktail of voltage-gated calcium channel blockers and cyclothiazide (50 μM). Fura-2-acetoxymethyl ester fluorescence was measured using a microscope-based imaging system, analysed with ImageMaster suite software and is represented graphically as the average change in 340 nm/380 nm ratio of cells monitored individually. Error bars indicate SEM. *< 0.05; **< 0.01; One-way anova with Bonferroni post hoc test; = 3 separate experiments where each ≥ 15 cells monitored individually.

Download figure to PowerPoint

Acute and chronic TNFα partially protects motor neurones against kainate-induced excitotoxicity

It is controversial as to whether TNFα acts to confer neuroprotection or enhance neurotoxicity. In motor neurones TNFα regulation of AMPAR-mediated calcium entry may modify vulnerability to excitotoxic insults. Therefore, we directly assessed the ability of TNFα to alter susceptibility of motor neurones to an excitotoxic insult. To assess excitotoxic cell de ath, we applied kainate, which acts as a non-desensitising AMPAR agonist, to motor neurones for 16 h. Kainate produced a concentration-dependent cell death with concentrations of 100 and 300 μM producing 55% and 83% cell death respectively, allowing either neuroprotective or neurotoxic effects of TNFα to be assessed. TNFα (10 ng/mL) was applied to cells for 15 min then removed prior to exposure to kainate (100, 300 μM) or a vehicle control for a further 16 h before counting SMI-32-positive motor neurones. TNFα pre-treatment caused a small but statistically significant protection of motor neurones against cell death induced by 300 μM kainate (Figure S2). 29 ± 4.7% of motor neurones survived when pre-treated with TNFα, compared with a cell survival of 17 ± 2.6% without pre-treatment (= 6, < 0.05). Similar protective effects of TNFα were observed when TNFα (10 ng/mL) was co-incubated with kainate for 16 h (data not shown). In no case did TNFα enhance excitotoxicity caused by kainate.

Discussion

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

TNFα is often regarded as a mediator of neuronal damage in neurodegenerative disorders that have an inflammatory component, although it is becoming increasingly clear that it is also a potent and rapid regulator of synaptic plasticity. TNFα has previously been shown to alter synaptic strength in hippocampal neurones by regulating the trafficking of AMPAR subunits to the cell surface (Beattie et al. 2002; Stellwagen et al. 2005; Stellwagen and Malenka 2006). In this study, we show for the first time, using a combination of functional and biochemical assays, that acute (15 min) or chronic (24 h) exposure to a physiologically relevant concentration of TNFαreduces AMPAR-evoked calcium rises in motor neurones and this reduction correlates with increases in the levels of both GluR1 and GluR2 at the cell surface. Our findings lend strong support to the underlying concept that TNFα induces rapid trafficking of calcium impermeable, i.e. GluR2 containing AMPARs to the cell surface, and show that motor neurone function is exquisitely responsive to TNFα over a relatively short time scale.

Motor neurones express high levels of calcium-permeable AMPARs (Carriedo et al. 1996, 2000; Roy et al. 1998; Van Damme et al. 2002) and the limited calcium buffering capacity of motor neurones makes this cell type particularly susceptible to calcium-dependent cell death (e.g. Carriedo et al. 2000). Thus, in this cell type, changes in cell surface GluR2 are likely to result in profound alterations in the levels of intracellular calcium following AMPAR activation, as we demonstrate here. Through increasing the cell surface expression of GluR1 and therefore enhancing the surface levels of GluR2-lacking AMPARs, TNFα has been proposed to exacerbate neuronal injury in hippocampal neurones (Ogoshi et al. 2005; Leonoudakis et al. 2008) and following spinal cord injury (Ferguson et al. 2008), by leading to more calcium-permeable AMPARs, and to an enhancement of AMPA-mediated excitotoxicity. However, and in striking contrast to this hypothesis, in isolated motor neurones and cortical neurones, TNFα induces a rapid increase in the cell surface levels of both GluR1 and GluR2 resulting in decreased calcium-permeable AMPAR activity and no potentiation of excitotoxicity. The most likely explanation for this is a net increase in synaptic GluR1/GluR2 heterodimers at the expense of GluR1 homodimers, similar to the well-established model whereby activity at calcium-permeable AMPAR causes rapid incorporation of GluR2-containing AMPARs to reduce calcium permeability (Liu and Cull-Candy 2000). We have not yet established whether the newly incorporated receptor assemblies are synaptic or extrasynaptic but irrespective of their subcellular localisation the lateral mobility of GluR2 subunits enables rapid diffusion and exchange with other subunits within the membrane to acutely regulate calcium handling and synaptic strength (Newpher and Ehlers 2008). Our observed reduction in AMPA-evoked calcium rises concomitant with enhanced cell surface GluR2 supports this. Under more pathological condition, or following microinjection of TNFα into spinal cord, the incorporation of GluR2 containing receptors appears to be lost and the result is a net increase in GluR2 lacking AMPA receptors (Ferguson et al. 2008) perhaps resulting in an increase in homomeric GluR1 receptors. The underlying reason for the loss of GluR2 containing assemblies under these conditions is unknown although it would appear that this is mediated by a preferential down-regulation of both synaptic and extrasynaptic GluR2 subunits.

TNFα has been proposed to either enhance or inhibit neuronal cell death depending on the concentration of TNFα used, the time of exposure to TNFα, and the receptor subtype involved (Bernardino et al. 2005; Jara et al. 2007; Chadwick et al. 2008; Dolga et al. 2008). Indeed, there can be fundamental differences between cell types in their TNF response, depending on their complement of particular intracellular signalling molecules or TNF receptor interacting proteins (e.g. Gustin et al. 2004).

Our data show, for the first time, that in cultured motor neurones the predominant signalling of TNF receptors is not to potentiate excitotoxicity but to activate a putative neuroprotective mechanism; namely, through increasing insertion of GluR2 into the cell membrane, AMPAR-mediated calcium entry is substantially reduced. The lack of neurotoxic effect of TNFα is striking and robust. Indeed, our data support the hypothesis that reduced calcium entry is translated into a modest but significant protection of motor neurones against excitotoxicity at high kainate concentrations. We note that there are other instances where neuroprotective actions of TNFα predominate. For example, in hippocampal neurones, TNFα over a wide range of concentrations which activate both TNF receptor types (TNFR1 and TNFR2) can confer significant neuroprotection against glucose-deprivation-induced injury, which correlates with a TNFα-induced reduction in intracellular calcium levels (Cheng et al. 1994). Motor neurones express both TNFR1 and TNFR2 (Veglianese et al. 2006) and we have used concentrations of TNFα which activate both receptor subtypes (see Yang et al. 2002), yet our data show negligible activation of the cell death activity associated with TNFR1 signalling (Tartaglia et al. 1993; Fontaine et al. 2002; Yang et al. 2002). Because these effects in motor neurones are found after short (15 min) exposures to TNFα they are unlikely to be accounted for by induction of the time-dependent Akt- and NFκB-dependent neuroprotective mechanisms which can occur several hours after TNFα exposure (Dolga et al. 2008). As in our motor neurone cultures there are astrocytes present, it is conceivable that the motor neurones are ‘primed’ through astrocytic release of TNFα into the culture medium: this is an area worthy of further investigation.

Under the conditions used in this study, TNFα does not enhance AMPA-mediated calcium entry into motor neurones, even at that higher concentrations of TNFα that were reported to enhance excitotoxicity in hippocampal neurones through up-regulation of cell surface GluR1 (Leonoudakis et al. 2008). Our data therefore suggest that it is the mechanisms linking TNF receptor activation to GluR2 membrane insertion in motor neurones which accounts for this apparent difference to other neuronal types. The GluR2 appearance on the cell surface following TNF receptor stimulation is rapid, i.e. within 15 min of TNFα exposure, whereas other groups report that GluR2 recruitment to the membrane following TNFα occurs over a longer time scale than GluR1 (Leonoudakis et al. 2008). This enhanced sensitivity of GluR2 to regulation in motor neurones may reflect the basal activity of PI3K and p38MAPK that provides an environment permissive for rapid recruitment of GluR2 to the cell surface.

AMPAR–PI3K complexes regulate the insertion of AMPAR at activated CA1 synapses (Man et al. 2003) and PI3K activity is involved in the co-ordination of signalling events downstream of AMPARs in synaptic plasticity (Perkinton et al. 1999), long term potentiation (LTP) Sanna et al. (2002), and in central sensitisation in the spinal cord (Pezet et al. 2008). In this study, we show that AMPA-evoked calcium rises in motor neurones are also regulated by PI3K which could be linked either to activity-dependent AMPAR trafficking or to altered channel properties following glutamate receptor phosphorylation. PI3K mediates rapid insertion of AMPAR into the plasma membrane (Man et al. 2003) and this probably accounts for the reduction in AMPA-evoked calcium rises seen in the presence of the PI3K inhibitor wortmannin. Previous studies have shown that TNFα-induced changes in the membrane levels of GluR1 are also dependent on PI3K and in general agreement with this we found that the ability of TNFα to reduce AMPA-induced intracellular calcium levels was abolished in the presence of wortmannin. It is known that the p85 subunit of PI3K binds directly to a membrane proximal region of the C-terminus common to all glutamate receptors, so our data are consistent with PI3K-dependent trafficking of both GluR1 and GluR2 in motor neurones. Under disease conditions, PI3K activity is up-regulated by neurotrophins and other inputs as part of a coordinated cytoprotective response to increase Akt activity and it is possible that this could also impact on GluR subunit insertion to influence synaptic excitability and calcium levels within motor neurones. This needs to be determined but our findings suggest at least two levels of PI3K regulation of AMPA receptor function.

In addition to the observed PI3K sensitivity, we have also demonstrated a role for p38 kinase in mediating the acute inhibitory effect of TNFα on AMPA-induced calcium rises, although interestingly, p38 kinase inhibition did not influence the effect of AMPA alone. Many of the cellular effects of TNFα are mediated through activation of p38 kinase (Fiers et al. 1995), and activation of this kinase has been associated with motor neurone death in some studies (Raoul et al. 2006) but not others (Veglianese et al. 2006). It is not yet clear however, if p38 kinase sits within the same pathway as PI3K or whether it represents a parallel signalling pathway for AMPAR regulation. Signalling through p38 kinase has been implicated in the regulation of glutamate receptor internalisation and long term potentiation (LTP) (Pickering et al. 2005), but has not previously been shown to influence AMPA-induced calcium rises and receptor function, and to our knowledge this is the first report of p38 kinase regulation of AMPAR function in motor neurones.

It has been assumed that under pathological conditions such as those in ALS where there are very high levels of TNFα, physiological control over glutamate receptor trafficking might be lost in motor neurones and cell death exacerbated. Although we cannot exclude the possibility that very long exposures to TNFα would result in diminished neuroprotective signalling and enhance cell death, overall our data strongly suggest that TNFα acts as a physiological regulator of synaptic activity in motor neurones rather than a pathological drive in ALS. Our data therefore suggest that approaches to treating ALS by sequestration of TNFα should be approached with caution: a view reinforced by a recent case report of the rapid appearance of ALS symptoms in a patient treated with an anti-TNFα therapy (Loustau et al. 2009).

Acknowledgements

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

This work was supported by the Motor Neurone Disease Association through a PhD studentship award to SRS in memory of Tim Bridgman. We thank Stuart Bevan for advice and help with calcium imaging. The authors declare no financial or other conflict of interest.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information
  • Barger S. W., Hörster D., Furukawa K., Goodman Y., Krieglstein J. and Mattson M. P. (1995) Tumor necrosis factors alpha and beta protect neurons against amyloid beta-peptide toxicity: evidence for involvement of a kappa B-binding factor and attenuation of peroxide and Ca2+ accumulation. Proc. Natl Acad. Sci. USA 92, 93289332.
  • Beattie E. C., Stellwagen D., Morishita W., Bresnahan J. C., Ha B. K., Von Zastrow M., Beattie M. S. and Malenka R. C. (2002) Control of synaptic strength by glial TNFalpha. Science 295, 22822285.
  • Bernardino L., Xapelli S., Silva A. P., Jakobsen B., Poulsen F. R., Oliveira C. R., Vezzani A., Malva J. O. and Zimmer J. (2005) Modulator effects of interleukin-1beta and tumor necrosis factor-alpha on AMPA-induced excitotoxicity in mouse organotypic hippocampal slice cultures. J. Neurosci. 25, 67346744.
  • Bigini P., Repici M., Cantarella G. et al. (2008) Recombinant human TNF-binding protein-1 (rhTBP-1) treatment delays both symptoms progression and motor neuron loss in the wobbler mouse. Neurobiol. Dis. 29, 465476.
  • Bruce A. J., Boling W., Kindy M. S., Peschon J., Kraemer P. J., Carpenter M. K., Holtsberg F. W. and Mattson M. P. (1996) Altered neuronal and microglial responses to excitotoxic and ischemic brain injury in mice lacking TNF receptors. Nat. Med. 2, 788794.
  • Camu W. and Henderson C. E. (1992) Purification of embryonic rat motoneurons by panning on a monoclonal antibody to the low-affinity NGF receptor. J. Neurosci. Methods 44, 5970.
  • Canton T., Bohme G. A., Boireau A. et al. (2001) RPR 119990, a novel alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid antagonist: synthesis, pharmacological properties, and activity in an animal model of amyotrophic lateral sclerosis. J. Pharmacol. Exp. Ther. 299, 314322.
  • Carlson N. G., Bacchi A., Rogers S. W. and Gahring L. C. (1998) Nicotine blocks TNF-alpha-mediated neuroprotection to NMDA by an alpha-bungarotoxin-sensitive pathway. J. Neurobiol. 35, 2936.
  • Carlson N. G., Wieggel W. A., Chen J., Bacchi A., Rogers S. W. and Gahring L. C. (1999) Inflammatory cytokines IL-1 alpha, IL-1 beta, IL-6, and TNF-alpha impart neuroprotection to an excitotoxin through distinct pathways. J. Immunol. 163, 39633968.
  • Carriedo S. G., Yin H. Z. and Weiss J. H. (1996) Motor neurons are selectively vulnerable to AMPA/kainate receptor-mediated injury in vitro. J. Neurosci. 16, 40694079.
  • Carriedo S. G., Sensi S. L., Yin H. Z. and Weiss J. H. (2000) AMPA exposures induce mitochondrial Ca(2+) overload and ROS generation in spinal motor neurons in vitro. J. Neurosci. 20, 240250.
  • Chadwick W., Magnus T., Martin B., Keselman A., Mattson M. P. and Maudsley S. (2008) Targeting TNF-α receptors for neurotherapeutics. Trends Neurosci. 31, 504511.
  • Chao C. C. and Hu S. (1994) Tumor necrosis factor-alpha potentiates glutamate neurotoxicity in human fetal brain cell cultures. Dev. Neurosci. 16, 172179.
  • Cheng B., Christakos S. and Mattson M. P. (1994) Tumor necrosis factors protect neurons against metabolic-excitotoxic insults and promote maintenance of calcium homeostasis. Neuron 12, 139153.
  • Couratier P., Hugon J., Sindou P., Vallat J. M. and Dumas M. (1993) Cell culture evidence for neuronal degeneration in amyotrophic lateral sclerosis being linked to glutamate AMPA/kainate receptors. Lancet 341, 265268.
  • De A., Krueger J. M. and Simasko S. M. (2003) Tumor necrosis factor alpha increases cytosolic calcium responses to AMPA and KCl in primary cultures of rat hippocampal neurons. Brain Res. 981, 133142.
  • Dolga A. M., Granic I., Blank T., Knaus H. G., Spiess J., Luiten P. G., Eisel U. L. and Nijholt I. M. (2008) TNF-alpha mediates neuroprotection against glutamate-induced excitotoxicity via NF-kappaB-dependent up-regulation of K2.2 channels. J. Neurochem. 107, 11581167.
  • Duong F. H., Warter J. M., Poindron P. and Passilly P. (1999) Effect of the nonpeptide neurotrophic compound SR 57746A on the phenotypic survival of purified mouse motoneurons. Br. J. Pharmacol. 128, 13851392.
  • Elliott J. L. (2001) Cytokine upregulation in a murine model of familial amyotrophic lateral sclerosis. Mol. Brain Res. 95, 172178.
  • Ferguson A. R., Christensen R. N., Gensel J. C., Miller B. A., Sun F., Beattie E. C., Bresnahan J. C. and Beattie M. S. (2008) Cell death after spinal cord injury is exacerbated by rapid TNF alpha-induced trafficking of GluR2-lacking AMPARs to the plasma membrane. J. Neurosci. 28, 1139111400.
  • Fiers W., Beyaert R., Boone E. et al. (1995) TNF-induced intracellular signaling leading to gene induction or to cytotoxicity by necrosis or by apoptosis. J. Inflamm. 47, 6775.
  • Fontaine V., Mohand-Said S., Hanoteau N., Fuchs C., Pfizenmaier K. and Eisel U. (2002) Neurodegenerative and neuroprotective effects of tumor necrosis factor (TNF) in retinal ischemia: opposite roles of TNF receptor 1 and TNF receptor 2. J. Neurosci. 22, RC216.
  • Ghezzi P. and Mennini T. (2001) Tumor necrosis factor and motoneuronal degeneration: an open problem. Neuroimmunomodulation 9, 178182.
  • Gowing G., Dequen F., Soucy G. and Julien J. P. (2006) Absence of tumor necrosis factor-alpha does not affect motor neuron disease caused by superoxide dismutase 1 mutations. J. Neurosci. 26, 1139711402.
  • Gustin J. A., Ozes O. N., Akca H. et al. (2004) Cell type-specific expression of the IkappaB kinases determines the significance of phosphatidylinositol 3-kinase/Akt signaling to NF-kappa B activation. J. Biol. Chem. 279, 16151620.
  • He B. P., Wen W. and Strong M. J. (2002) Activated microglia (BV-2) facilitation of TNF-alpha-mediated motor neuron death in vitro. J. Neuroimmunol. 128, 3138.
  • Hensley K., Floyd R. A., Gordon B., Mou S., Pye Q. N., Stewart C., West M. and Williamson K. (2002) Temporal patterns of cytokine and apoptosis-related gene expression in spinal cords of the G93A-SOD1 mouse model of amyotrophic lateral sclerosis. J. Neurochem. 82, 365374.
  • Hollmann M., Hartley M. and Heinemann S. (1991) Ca2+ permeability of KA-AMPA-gated glutamate receptor channels depends on subunit composition. Science 252, 851853.
  • Jara J. H., Singh B. B., Floden A. M. and Combs C. K. (2007) Tumor necrosis factor alpha stimulates NMDA receptor activity in mouse cortical neurons resulting in ERK-dependent death. J. Neurochem. 100, 14071420.
  • Leonoudakis D., Zhao P. and Beattie E. C. (2008) Rapid tumor necrosis factor alpha-induced exocytosis of glutamate receptor 2-lacking AMPA receptors to extrasynaptic plasma membrane potentiates excitotoxicity. J. Neurosci. 28, 21192130.
  • Lobsiger C. S. and Cleveland D. W. (2007) Glial cells as intrinsic components of non-cell-autonomous neurodegenerative disease. Nat. Neurosci. 10, 13551360.
  • Loustau V., Foltz V., Poulain C., Rozenberg S. and Bruneteau G. (2009) Diagnosis of amyotrophic lateral sclerosis in a patient treated with TNFalpha blockers for ankylosing spondylitis: fortuitus association or new side effect of TNFalpha blockers? Joint Bone Spine 76, 213214.
  • Lu W., Man H., Ju W., Trimble W. S., MacDonald J. F. and Wang Y. T. (2001) Activation of synaptic NMDA receptors induces membrane insertion of new AMPA receptors and LTP in cultured hippocampal neurons. Neuron 29, 243254.
  • Liu S. Q. and Cull-Candy S. G. (2000) Synaptic activity at calcium permeable AMPA receptors induces a change in receptor subtype. Nature 405, 454458.
  • Man H. Y., Wang Q., Lu W. Y. et al. (2003) Activation of PI3-kinase is required for AMPA receptor insertion during LTP of mEPSCs in cultured hippocampal neurons. Neuron 38, 611624.
  • Marchetti L., Klein M., Schlett K., Pfizenmaier K. and Eisel U. L. (2004) Tumor necrosis factor (TNF)-mediated neuroprotection against glutamate-induced excitotoxicity is enhanced by N-methyl-d-aspartate receptor activation. Essential role of a TNF receptor 2-mediated phosphatidylinositol 3-kinase-dependent NF-kappa B pathway. J. Biol. Chem. 279, 3286932881.
  • McGeer P. L. and McGeer E. G. (2002) Inflammatory processes in amyotrophic lateral sclerosis. Muscle Nerve 26, 459470.
  • Newpher T. M. and Ehlers M. D. (2008) Glutamate receptor dynamics in dendritic microdomains. Neuron 58, 472497.
  • Nguyen M. D., Julien J. P. and Rivest S. (2001) Induction of proinflammatory molecules in mice with amyotrophic lateral sclerosis: no requirement for proapoptotic interleukin-1beta in neurodegeneration. Ann. Neurol. 50, 630639.
  • Ogoshi F., Yin H. Z., Kuppumbatti Y., Song B., Amindari S. and Weiss J. H. (2005) Tumor necrosis-factor-alpha (TNF-alpha) induces rapid insertion of Ca2+ permeable alpha-amino-3-hydroxyl-5-methyl-4-isoxazole-propionate (AMPA)/kainate (Ca-A/K) channels in a subset of hippocampal pyramidal neurons. Exp. Neurol. 193, 384393.
  • Peacey E., Miller C. C., Dunlop J. and Rattray M. (2009) The four major N- and C-terminal splice variants of the excitatory amino acid transporter GLT-1 form cell surface homomeric and heteromeric assemblies. Mol. Pharmacol. 75, 10621073.
  • Perkinton M. S., Sihra T. S. and Williams R. J. (1999) Ca(2+)-permeable AMPA receptors induce phosphorylation of cAMP response element-binding protein through a phosphatidylinositol 3-kinase-dependent stimulation of the mitogen-activated protein kinase signaling cascade in neurons. J. Neurosci. 19, 58615874.
  • Pezet S., Marchand F., D’Mello R., Grist J., Clark A. K., Malcangio M., Dickenson A. H., Williams R. J. and McMahon S. B. (2008) Phosphatidylinositol 3-kinase is a key mediator of central sensitization in painful inflammatory conditions. J. Neurosci. 28, 42614270.
  • Pickering M., Cumiskey D. and O’Connor J. J. (2005) Actions of TNF-alpha on glutamatergic synaptic transmission in the central nervous system. Exp. Physiol. 90, 663670.
  • Raivich G., Liu Z. Q., Kloss C. U., Labow M., Bluethmann H. and Bohatschek M. (2002) Cytotoxic potential of proinflammatory cytokines: combined deletion of TNF receptors TNFR1 and TNFR2 prevents motoneuron cell death after facial axotomy in adult mouse. Exp. Neurol. 178, 186193.
  • Raoul C., Buhler E., Sadeghi C., Jacquier A., Aebischer P., Pettmann B., Henderson C. E. and Haase G. (2006) Chronic activation in presymptomatic amyotrophic lateral sclerosis (ALS) mice of a feedback loop involving Fas, Daxx, and FasL. Proc. Natl Acad. Sci. USA 103, 60076012.
  • Rattray M. and Bendotti C. (2006) Does excitotoxic cell death of motor neurons in ALS arise from glutamate transporter and glutamate receptor abnormalities? Exp. Neurol. 201, 1523.
  • Robertson J., Beaulieu J. M., Doroudchi M. M., Durham H. D., Julien J. P. and Mushynski W. E. (2001) Apoptotic death of neurons exhibiting peripherin aggregates is mediated by the proinflammatory cytokine tumor necrosis factor-alpha. J. Cell Biol. 155, 217226.
  • Roy J., Minotti S., Dong L., Figlewicz D. A. and Durham H. D. (1998) Glutamate potentiates the toxicity of mutant Cu/Zn-superoxide dismutase in motor neurons by postsynaptic calcium-dependent mechanisms. J. Neurosci. 18, 96739684.
  • Sanna P. P., Cammalleri M., Berton F., Simpson C., Lutjens R., Bloom F. E. and Francesconi W. (2002) Phosphatidylinositol 3-kinase is required for the expression but not for the induction or the maintenance of long-term potentiation in the hippocampal CA1 region. J. Neurosci. 22, 33593365.
  • Schroeter H., Bahia P., Spencer J. P., Sheppard O., Rattray M., Cadenas E., Rice-Evans C. and Williams R. J. (2007) (-)Epicatechin stimulates ERK-dependent cyclic AMP response element activity and upregulates GluR2 in cortical neurons. J. Neurochem. 101, 15961606.
  • Sedel F., Béchade C., Vyas S. and Triller A. (2004) Macrophage-derived tumor necrosis factor alpha, an early developmental signal for motoneuron death. J. Neurosci. 24, 22362246.
  • Stellwagen D. and Malenka R. C. (2006) Synaptic scaling mediated by glial TNF-alpha. Nature 440, 10541059.
  • Stellwagen D., Beattie E. C., Seo J. Y. and Malenka R. C. (2005) Differential regulation of AMPA receptor and GABA receptor trafficking by tumor necrosis factor-alpha. J. Neurosci. 25, 32193228.
  • Stommel E. W., Van Hoff R. M., Graber D. J., Bercury K. K., Langford G. M. and Harris B. T. (2007) Tumor necrosis factor-alpha induces changes in mitochondrial cellular distribution in motor neurons. Neuroscience 146, 10131019.
  • Tartaglia L. A., Rothe M., Hu Y. F. and Goeddel D. V. (1993) Tumor necrosis factor’s cytotoxic activity is signaled by the p55 TNF receptor. Cell 73, 213216.
  • Tateno M., Sadakata H., Tanaka M. et al. (2004) Calcium permeable AMPA receptors promote misfolding of mutant SOD1 protein and development of amyotrophic lateral sclerosis in a transgenic mouse model. Hum. Mol. Genet. 13, 21832196.
  • Terrado J., Monnier D., Perrelet D. et al. (2000) Soluble TNF receptors partially protect injured motoneurons in the postnatal CNS. Eur. J. Neurosci. 12, 34433447.
  • Tortarolo M., Crossthwaite A. J., Conforti L., Spencer J. P., Williams R. J., Bendotti C. and Rattray M. (2004) Expression of SOD1 G93A or wild-type SOD1 in primary cultures of astrocytes down-regulates the glutamate transporter GLT-1: lack of involvement of oxidative stress. J. Neurochem. 88, 481493.
  • Van Damme P., Van Den Bosch L., Van Houtte E., Callewaert G. and Robberecht W. (2002) GluR2-dependent properties of AMPA receptors determine the selective vulnerability of motor neurons to excitotoxicity. J. Neurophysiol. 88, 12791287.
  • Van Damme P., Leyssen M., Callewaert G., Robberecht W. and Van Den Bosch L. (2003) The AMPA receptor antagonist NBQX prolongs survival in a transgenic mouse model of amyotrophic lateral sclerosis. Neurosci. Lett. 343, 8184.
  • Vandenberghe W., Robberecht W. and Brorson J. R. (2000) AMPA receptor calcium permeability, GluR2 expression, and selective motoneuron vulnerability. J. Neurosci. 20, 123132.
  • Veglianese P., Lo Coco D., Bao Cutrona M. et al. (2006) Activation of the p38MAPK cascade is associated with upregulation of TNF alpha receptors in the spinal motor neurons of mouse models of familial ALS. Mol. Cell. Neurosci. 31, 218231.
  • Wen W., Sanelli T., Ge W., Strong W. and Strong M. J. (2006) Activated microglial supernatant induced motor neuron cytotoxicity is associated with upregulation of the TNFR1 receptor. Neurosci. Res. 55, 8795.
  • Williams R. J. and Glowinski J. (1996) Cyclothiazide unmasks an AMPA-evoked release of arachidonic acid from cultured striatal neurones. J. Neurochem. 67, 15511558.
  • Yang L., Lindholm K., Konishi Y., Li R. and Shen Y. (2002) Target depletion of distinct tumor necrosis factor receptor subtypes reveals hippocampal neuron death and survival through different signal transduction pathways. J. Neurosci. 22, 30253032.
  • Yoshihara T., Ishigaki S., Yamamoto M., Liang Y., Niwa J., Takeuchi H., Doyu M. and Sobue G. (2002) Differential expression of inflammation- and apoptosis-related genes in spinal cords of a mutant SOD1 transgenic mouse model of familial amyotrophic lateral sclerosis. J. Neurochem. 80, 158167.
  • Yu Z., Cheng G., Wen X., Wu G. D., Lee W. T. and Pleasure D. (2002) Tumor necrosis factor alpha increases neuronal vulnerability to excitotoxic necrosis by inducing expression of the AMPA-glutamate receptor subunit GluR1 via an acid sphingomyelinase- and NF-kappaB-dependent mechanism. Neurobiol. Dis. 11, 199213.
  • Yuasa T., Ohno S., Kehrl J. H. and Kyriakis J. M. (1998) Tumor necrosis factor signaling to stress-activated protein kinase (SAPK)/Jun NH2-terminal kinase (JNK) and p38. Germinal center kinase couples TRAF2 to mitogen-activated protein kinase/ERK kinase kinase 1 and SAPK while receptor interacting protein associates with a mitogen-activated protein kinase kinase kinase upstream of MKK6 and p38. J. Biol. Chem. 273, 2268122692.

Supporting Information

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

Figure S1 Fura-2 AM microfluorimetry demonstrating AMPA and KCl-evoked intracellular Ca2+ rises in primary mouse cortical neurones. (A,B,C) Pseudocoloured images illustrating [Ca 2+]i at baseline (A) and [Ca 2+]i responses evoked by 50 μM AMPA (B) and 50 mM KCl (C). (D) Changes in [Ca 2+]i (340 nm/380 nm ratio) of 44 individual neurones in response to sequential application of AMPA (50 μM) and KCl (50 mM). (E) Average 340 nm/380 nm ratio trace of neurones monitored in (D). Compounds were administered by superfusion in Na+ containing buffer.

Figure S2 Motor neurones pretreated with TNFα demonstrate decreased susceptibility to excitotoxicity. Neurones were left untreated (control) or incubated with TNFα (10 ng/ml) for 15 min prior to administration of kainate at 300 μM, or vehicle, for 16 h. Cells were fixed and labelled with immunofluorescence indirectly using SMI-32 antibody. The graph represents SMI-32 positive cell counts normalised to untreated controls. Error bars indicate SEM. *p<0.05 compared to untreated group; Student’s unpaired t-test; n=6 separate experiments. Representative immunofluorescent SMI-32 positive cells (green) are shown, Hoechst 33342-stained nuclei appear blue. Motor neurones with an intact cell body and at least one undamaged cell process were counted. Scale bar, 50 μm. Results show that TNFα is not toxic to motor neurones under these conditions, and that TNFα affords a small but significant protection against kainate toxicity.

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer-reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

FilenameFormatSizeDescription
JNC_6634_sm_Figure1-2.pdf191KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.