Both these authors contributed equally to this work.
NMDA receptor-mediated excitotoxic neuronal apoptosis in vitro and in vivo occurs in an ER stress and PUMA independent manner
Version of Record online: 13 DEC 2007
© 2008 The Authors
Journal of Neurochemistry
Volume 105, Issue 3, pages 891–903, May 2008
How to Cite
Concannon, C. G., Ward, M. W., Bonner, H. P., Kuroki, K., Tuffy, L. P., Bonner, C. T., Woods, I., Engel, T., Henshall, D. C. and Prehn, J. H. M. (2008), NMDA receptor-mediated excitotoxic neuronal apoptosis in vitro and in vivo occurs in an ER stress and PUMA independent manner. Journal of Neurochemistry, 105: 891–903. doi: 10.1111/j.1471-4159.2007.05187.x
- Issue online: 13 DEC 2007
- Version of Record online: 13 DEC 2007
- Received October 4, 2007; revised manuscript received December 5, 2007; accepted December 7, 2007.
- BH3-only protein;
- endoplasmic reticulum stress;
- Top of page
- Materials and methods
- Supporting Information
Disruption of endoplasmic reticulum (ER) Ca2+ homeostasis and ER dysfunction have been suggested to contribute to excitotoxic and ischaemic neuronal injury. Previously, we have characterized the neural transcriptome following ER stress and identified the BH3-only protein, p53 up-regulated mediator of apoptosis (PUMA), as a central mediator of ER stress toxicity. In this study, we investigated the effects of excitotoxic injury on ER Ca2+ levels and induction of ER stress responses in models of glutamate- and NMDA-induced excitotoxic apoptosis. While exposure to the ER stressor tunicamycin induced an ER stress response in cerebellar granule neurons, transcriptional activation of targets of the ER stress response, including PUMA, were absent following glutamate-induced apoptosis. Confocal imaging revealed no long-term changes in the ER Ca2+ level in response to glutamate. Murine cortical neurons and organotypic hippocampal slice cultures from PUMA+/+ and PUMA−/− animals provided no evidence of ER stress and did not differ in their sensitivity to NMDA. Finally, NMDA-induced excitotoxic apoptosis in vivo was not associated with ER stress, nor did deficiency in PUMA alleviate the injury induced. Our data suggest that NMDA receptor-mediated excitotoxic apoptosis occurs in vitro and in vivo in an ER stress and PUMA independent manner.
activating transcription factor
CIEBP homology protein
glucose regulated protein
inositol-requiring ER-to-nucleus signal kinase 1
minimal essential medium
oxygen and glucose deprivation
protein kinase-like ER kinase
p53 up-regulated mediator of apoptosis
sarco/endoplasmic reticulum Ca2+-ATPase
terminal deoxynucleotidyl transferase-mediated dUTP nick end labelling
unfolded protein response
X-box binding protein 1
Glutamate receptor excitation is known to play an important role during neuronal injury associated with ischaemic stroke, head trauma and prolonged seizures (Schanne et al. 1979; Choi 1994). The activation of NMDA receptors, and the subsequent Ca2+ influx, have been identified as critical in mediating this form of injury (Choi et al. 1987; Tymianski et al. 1993a) with much research focused on the sequestration of Ca2+ within the mitochondrial matrix following excitation. The endoplasmic reticulum (ER) is a subcellular compartment which also exhibits a high Ca2+ sequestering activity and plays a functional role in cellular Ca2+ regulation, however, the role of the ER during excitotoxic neuronal injury is currently poorly understood.
Within cells the ER is the major site for protein folding and maturation. Disruption of homeostasis within the ER such as depletion of ER Ca2+ can interfere with protein folding and transport, and can lead to the activation of a highly conserved response termed ER stress. This disturbance in ER homeostasis is sensed by three ER-localized proteins: activating transcription factor 6 (ATF6), RNA-dependent protein kinase-like ER kinase (PERK) and inositol-requiring ER-to-nucleus signal kinase 1 (IRE1) which under normal physiological condition are kept in check bound to the molecular chaperone, glucose regulated protein 78 (Grp78) (Zhang and Kaufman 2004). Upon accumulation of misfolded proteins within the ER, Grp78 dissociates from each of the sensors leading to activation of their associated signalling pathways, culminating in the attenuation of general protein synthesis and increased transcription of ER resident chaperones and protein foldases. However, under conditions of prolonged or severe stress, apoptosis ensues.
We have previously characterized the gene expression profile during ER stress-induced apoptosis of neural cells treated with the inhibitor of N-glycosylation, tunicamycin (Reimertz et al. 2003). p53 up-regulated mediator of apoptosis (PUMA), a member of the Bcl-2 homology 3 (BH3)-only family of proteins was the most prominent pro-apoptotic gene induced. BH3-only proteins comprise a pro-apoptotic subset of the Bcl-2 family which function to couple cellular stress events to the release of pro-apoptotic proteins from mitochondrial in a Bax-dependent manner (Labi et al. 2006). Expression of PUMA has been demonstrated to be both necessary and sufficient for inducing ER stress-mediated apoptosis in a number of cell models (Reimertz et al. 2003; Luo et al. 2005; Li et al. 2006; Nickson et al. 2007). Recently, increased X-box binding protein 1 (XBP1) splicing and CIEBP homology protein (CHOP) mRNA levels have been observed during nitric oxide- and excitotoxicity-induced inactivation of protein-disulphide isomerases (Uehara et al. 2006), suggesting that excitotoxic injury may activate an ER stress response. Moreover, NMDA-mediated excitotoxic apoptosis has been previously demonstrated to occur in a Bax-dependent manner (Xiang et al. 1998). In the present study, we therefore sought to clarify whether an ER stress response is induced following NMDA-receptor mediated excitotoxic apoptosis in vitro or in vivo, and explored the potential role of PUMA in mediating such injury.
Materials and methods
- Top of page
- Materials and methods
- Supporting Information
Foetal calf serum, horse serum, B27 supplement, neurobasal medium and minimal essential medium (MEM) were from Invitrogen (Paisley, Strathclyde, UK). Tunicamycin was purchased from Alexis (Axxora, Nottingham, UK). All other chemicals came in analytical grade purity from Sigma-Aldrich (Tallaght, Dublin, Ireland).
Preparation of primary cerebellar granule neurons
Cerebellar granule neurons were prepared as described previously (Ward et al. 2000) with minor modifications. Cerebella from 7-day-old Wistar rats of both sexes were dissected and pooled. The tissue was placed in 20 mL of filter sterilized phosphate-buffered saline (PBS) supplemented with 0.25 mg/mL trypsin and incubated at 37°C for 20 min. Trypsinization was terminated by addition of an equal volume of PBS supplemented with 0.05 mg/mL soyabean trypsin inhibitor, 3 mmol/L MgSO4 and 30 U/mL Dnase I. The neurons were then triturated and the resulting neurons were resuspended in supplemented culture medium. Cells were then plated on poly-l-lysine coated glass Willco dishes (Willco BV, Amsterdam, The Netherlands) or six-well plates at 1 × 106 cells per mL and maintained at 37°C in a humidified atmosphere of 5% CO2 : 95% air. All experiments were performed on days 7–9 in vitro.
Preparation of primary cortical neurons
Cortical neurons were prepared as described previously (Kushnareva et al. 2005). Briefly days 17–19 (E17–19) were isolated from inbred PUMA+/+ and PUMA−/− C57BL6/J mouse embryos by hysterectomy of the uterus, using an abdominal injection of 0.2 μL of pentobarbital as lethal anaesthesia. The embryos were transferred to a dissection medium on ice (PBS with 0.25% glucose and 0.3% bovine serum albumin). The cerebral cortices from each of the embryos were isolated, the surrounding meninges were removed and the tissue was pooled in dissection media on ice. The tissue was incubated with trypsin (0.125 mg/mL) at 37°C for 15 min. Soyabean trypsin inhibitor was used to stop the reaction (0.5 mg/mL). The neurons were dissociated from tissue in 5 mL of the trypsin inhibitor by gentle pipetting. The neurons were resuspended in plating media MEM containing 5% foetal calf serum, 5% horse serum, 100 U/mL penicillin/streptomycin, 0.5 mmol/L l-glutamine and 6% glucose) and plated at 7 × 105 cells/mL on poly-lysine coated plates. Cells were incubated at 37°C, 5% CO2 in media (new basal medium (NBM)-embryonic containing 180 μmol/L glutamate, 100 U/mL of penicillin/streptomycin, 1% B27 and 0.5 mmol/L l-glutamine) was exchanged every 3–4 days. Experiments were carried out on day 11. The generation of PUMA knockout mice was described previously (Villunger et al. 2003).
Lactate dehydrogenase release and Hoechst staining
Cerebellar granule neurons (1 × 106 cells per well) were plated on poly-l-lysine coated 24-well plates. In each case neuronal injury was assessed at each of the time points by quantitative measurement of lactate dehydrogenase (LDH) release using a cytotoxicity Detection Kit (Roche Diagnostics Ltd, Bell lane, Lewes, UK). Total cell death was obtained by incubating cultures in 0.8% Triton-X 100 for 45 min. Levels of cell death in cultures were normalized as a percentage of total cell death as per manufacturers instructions. All conditions were carried out in triplicate in each experiment and each experiment was carried out three times. For analysis of the nuclear structure of cortical neurons, cell were stained live with Hoechst 33258 (Sigma, Dublin, Ireland) at a final concentration of 1 μg/mL for 10 min, fixed and nuclear morphology was observed using an Eclipse TE 300 inverted microscope (Nikon, Düsseldorf, Germany) and a 63× oil objective.
Organotypic hippocampal slice cultures
Organotypic hippocampal slice cultures were prepared and grown according to the modified procedure described previously (Kristensen et al. 2001). Briefly 10-day-old mice were decapitated; brain was removed and gently submerged in dissection medium containing Hank’s balanced salt solution (Gibco, Biosciences, Dublin, Ireland), 20 mmol/L HEPES, 100 U/mL penicillin and 100 μg/mL streptomycin and 0.65% glucose. After separation of both hemispheres, each hippocampus was carefully removed. Both hippocampi were placed on the Teflon plate of a Mc Illwain tissue Chopper (Mickle Laboratory, Guldfort, UK), and cut into 450-μm thick sections. The slices were transferred into fresh dissection medium, and selected for clear hippocampal morphology [intact CA regions and dentate gyrus (DG)]. The slices were then transferred onto the porous (0.4 μm) membrane of the millicell inserts (Millipore, Cork, Ireland). Per animal up to six sections per hemisphere could be obtained and three were placed on each membrane. The inserts were placed in 24-well tissue culture plate with 250 μL of culture medium consisting of 50% MEM (Sigma, Dublin, Ireland), 25% horse serum (Biosciences), 4 mmol/L l-glutamine, 6 mg/mL d-glucose, 2% B27, 50 U/mL Penicillin G and 50 μg/mL streptomycin/mL (Sigma, Dublin, Ireland) The slices were placed in a humidified chamber with 5% CO2 at 35°C. Cultures were maintained for 10 days in vitro with media changes every other day.
Stimulus of organotypic hippocampal slices, quantification of injury and OGD experiments
Slices were pre-stained with propidium iodide (PI) at 5 μg/mL prior to initiation of experiments. Healthy slices from PUMA+/+ and PUMA−/− animals were treated with 0.05 mmol/L NMDA for 30 min in experimental buffer (120 mmol/L NaCl, 3.5 mmol/L KCl, 0.4 mmol/L KH2PO4, 5 mmol/L NaHCO3, 20 mmol/L HEPES and 1.2 mmol/L Na2SO4) supplemented with glucose (15 mmol/L) and CaCl2 (1.2 mmol/L) for 30 min at 35°C. Control slices contained were treated with experimental buffer without the NMDA. In alternative experiments healthy slices from PUMA+/+ and PUMA−/− were exposed to tunicamycin (30 μmol/L) for 24 h (vehicle addition was used as control). Observation of cell death was performed by PI staining (5 μg/mL) for 15 min prior to image acquisition. Images were acquired with an Eclipse TE 300 inverted microscope (Nikon) and a 4× dry objective with Wasabi Software (Hamamatsu Photonics, Germany). The CA1 and DG region were selected from PUMA+/+ and PUMA−/− hippocampal slices following NMDA receptor over-activation and the addition of tunicamycin respectively. The regional grey-level intensity was measured using the Wasabi interactive drawing tool, background grey-level was corrected for and the average intensity plotted for nine slices (triplicates from three separate cultures) for each of the conditions. Cell death induced via each condition was expressed as a percentage of total cell death within the slice region induced via treatment with 1 mmol/L NMDA for 24 h which induced total cell death in the CA1 or via treatment with 30 μmol/L tunicamycin for 72 h which induced total cell death in the DG. For the oxygen and glucose deprivation (OGD) experiments OGD solution containing 124 mmol/L NaCl, 2.5 mmol/L KCl, 1.25 mmol/L KH2PO4, 26 mmol/L NaHCO3, 1.5 mmol/L MgSO4 and 2.5 mmol/L CaCl2 (pH 7.4) was bubbled with oxygen-free nitrogen for 1 h prior to being placed in a COY hypoxic chamber (COY Lab Products Inc., Grass Lake, MI, USA) saturated with 94% N2, 1% O2 and 5% CO2 at 35°C for a further hour. The hippocampal slices were taken from the incubator and placed in the hypoxic chamber, the media was rapidly replaced with O2 depleted Roswell Park Memorial Institute (RPMI) containing zero glucose and slices were maintained in the hypoxic chamber for 30 min. The slices were removed from the chamber and the experimental media replaced with normal culture media and returned to the incubator under normoxic conditions until neuronal injury was evaluated 24 h later with PI.
Cerebellar granule neurons on Willco dishes were loaded with Fluo-4AM (3 μmol/L) for 30 min at 37°C (in the dark), in experimental buffer (120 mmol/L NaCl, 3.5 mmol/L KCl, 0.4 mmol/L KH2PO4, 20 mmol/L HEPES, 5 mmol/L NaHCO3, 1.2 mmol/L Na2SO4, 1.2 mmol/L CaCl2, 1.2 mmol/L MgCl2 and 15 mmol/L glucose, pH 7.4). The cells were washed in fresh medium after loading before being mounted in a thermostatically regulated chamber (37°C) and placed on the stage of a LSM 510 Meta Zeiss (Jena, Germany) confocal microscope. An objective heater was also utilized enabling the maintenance of a steady temperature of 37°C ± 0.26°C within the extracellular media. The neurons (1 × 106) were bathed in 2 mL of experimental buffer and a thin layer of mineral oil was added to prevent evaporation. No MgCl2 was present in buffers for experiments that involve the addition of glutamate. For glutamate-induced apoptosis, neurons were exposed to glutamate and glycine (100 and 10 μmol/L) for 5 min with MK801 (10 μmol/L) added to block glutamate receptor activation. Fluo-4AM was excited at 488 nm with an Argon laser (1%) and the emission collected through a 505–550 nm barrier filter. Images were collected at 15 s intervals during glutamate excitation and every 5 min during the rest of the experiment and the resulting fluorescent images were processed using MetaMorph Software (Universal Imaging Co., West Chester, PA, USA). To monitor intracellular Ca2+ levels while inhibiting ER Ca2+ uptake with thapsigargin, cerebellar granule neurons on Willco dishes were loaded with Fluo-4AM as above and placed on a LSM 510 Meta Zeiss confocal microscope and images were taken every 2 min. Thapsigargin (10 μmol/L) was added to control neurons (non-stimulated) after 0.5 h and peak Ca2+ responses measured. In addition thapsigargin (10 μmol/L) was added to glutamate-stimulated neurons at 0.5 and 5 h post-excitation and the peak Ca2+ responses measured. Images were again processed using MetaMorph Software and the data presented as a percentage of the baseline response.
Total RNA was extracted using the RNeasy mini Kit (Qiagen, Hilden, Germany). First strand cDNA synthesis was performed using 2 μg of total RNA as template and reverse transcribed using Superscript II (Invitrogen) primed with 50 pmol of random hexamers. Quantitative real-time PCR was performed using the LightCycler 2.0 (Roche Diagnostics, Basel, Switzerland) and the QuantiTech SYBR Green PCR kit (Qiagen) as per manufacturer’s protocol. Specific primers for each gene analysed were designed using Primer3 Software (http://frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi). Sense and antisense primers were: TCTCAGTGCAATGGCTTCC and CAATGCATTCTCCACACCAG for Grp78; TCACAAGCACCTCCCAAAGC and AGCAAGCTGTGCCACTTTCC for CHOP; GGTCTGCTGAGTCCGCAGCAGG and AAAGGGAGGCTGGTAAGGAA for XBP1s; ATGGACTCAGCATCGGAAGG and TGGCTCATTTGCTCTTCACG for PUMA; TCGTTCCTCCAGTCCGAGAG and TGAGTGGGGGTCGGTGTAGT for c-Jun; and AGCCATCCAGGCTGTGTTGT and CAGCTGTGGTGGTGAAGCTG for β-Actin. Each primer pair was tested with a logarithmic dilution of a cDNA mix to generate a linear standard curve, which was used to calculate the primer pair efficiency. The PCR reactions were performed in 20 μL volumes with following parameters: 95°C for 15 min followed by 40 cycles of 94°C for 20 s, 59°C for 20 s and 72°C for 20 s. The generation of specific PCR products was confirmed by melting curve analysis and gel electrophoresis. The data was analysed using the Lightcycler Software 4.0 with all samples normalized to β-actin.
Sodium dodecyl sulphate–polyacrylamide gel electrophoresis and western blotting
Preparation of cell lysates and western blotting was carried out as described (Reimertz et al. 2003). The resulting blots were probed with a mouse monoclonal anti-KDEL antibody (StressGen, Victoria, BC, Canada) diluted 1 : 1000, a rabbit polyclonal anti-CHOP antibody (Santa Cruz Biotechnology, Santa Cruz, CA, USA) diluted 1 : 250, a rabbit polyclonal anti-XBP1 antibody (Santa Cruz Biotechnology), a mouse monoclonal anti-Ubiquitin antibody (Affiniti, Victoria, BC, Canada) diluted 1 : 1000, a rabbit polyclonal anti-protein disulphide isomerase antibody (StressGen) diluted 1 : 1000, or a mouse monoclonal anti-β-Actin antibody (clone DM 1A; Sigma), diluted 1 : 5000. Horseradish peroxidase-conjugated secondary antibodies diluted 1 : 10 000 (Pierce, Northumberland, UK) were detected using SuperSignal West Pico Chemiluminescent Substrate (Pierce) and imaged using a FujiFilm LAS-3000 imaging system (Fuji, Sheffield, UK).
NMDA-induced hippocampal lesions
In vivo excitoxicity was modelled using bilateral intra-hippocampal NMDA injections to trigger death of CA1 pyramidal neurons. Briefly, adult (20–25 g) male C57BL/6 mice were anaesthetized with isoflurane and placed in a stereotaxic frame (Stoelting, Wood Dale, IL, USA) equipped with a neonatal rat/mouse adaptor. Mice were maintained normothermic by means of a feedback controlled homeothermic blanket. Following a midline incision, two craniectomies were performed. Coordinates from Bregma: AP = −1.9 mm posterior, ±1.5 mm lateral. NMDA (1 μg) or vehicle was injected at a depth of 1.3 mm below the dura in 0.1 μL volume. Mice were killed 2, 8, 24 or 48 h later, perfused with saline to remove intravascular blood components and brains fresh-frozen and processed for immunohistochemical analysis. Ethical approval was provided by the Royal College of Surgeons in Ireland for all animal work carried out within this study.
Immunohistochemistry and TUNEL staining
Immunohistochemistry was performed as previously described (Yamamoto et al. 2006). Mouse brain sections (12 μm) were air dried (15 min), post-fixed in 4% formalin (30 min), washed in PBS and then processed for immunohistochemistry and detection of DNA fragmentation. Following blocking in 5% goat serum, sections were incubated overnight at 4°C with an anti-KDEL antibody (mouse monoclonal; Stressgen Biotechnologies, Victoria, Canada). Sections were washed in PBS and then incubated for 2 h at 20–25°C in a 1 : 1000 dilution of Rhodamine goat anti-mouse secondary antibody (Jackson Immunoresearch, Plymouth, PA, USA). Detection of DNA fragmentation was performed using a fluorescein-based terminal deoxynucleotidyl transferase-mediated dUTP nick end labelling (TUNEL) technique (Roche Molecular Biochemicals, Indianapolis, IN, USA) as per manufacturer’s instructions. After additional washes, sections were counterstained with 4′, 6 diamidino-2-phenylindole (Vector Laboratories, Burlingame, CA, USA) for assessment of nuclear morphology. Images were visualized using a Hamamatsu Orca 285 camera attached to a Nikon 2000s epifluorescence microscope (Micro-optica, Dublin, Ireland) under Ex/Em wavelengths of 330–380/420 nm (blue), 472/520 nm (green) and 540–580/600–660 nm (red).
Data are given as mean ± SEM. For statistical comparison, one-way anova followed by Tukey’s test were employed. Values of p < 0.05 were considered to be statistically significant.
- Top of page
- Materials and methods
- Supporting Information
Transient activation of glutamate receptors induces a time-dependent increase in excitotoxic apoptosis of cultured rat cerebellar granule cells in the absence of long-term ER Ca2+ level changes
Excitotoxic neuronal injury has been shown to be highly dependent on excessive Ca2+ uptake (Tymianski et al. 1993b; Sattler et al. 1998; Ward et al. 2005). We have previously established a model in which a 5 min transient activation of glutamate receptors results in a delayed, apoptotic neuronal injury in cultured rat cerebellar granule neurons (Ward et al. 2006). In agreement with previous studies (Ward et al. 2006), cerebellar granule neurons loaded with the Ca2+ sensitive probe Fluo-4AM and exposed to glutamate went through a delayed collapse of Ca2+ homeostasis 8–16 h downstream of glutamate excitation (Fig. 1a). In order to quantify the onset and development of injury within our model we measured LDH release over a 24 h period post-glutamate excitation (Fig. 1b). The LDH release showed a similar trend to the Ca2+ data in that there was no significant increase in cell death prior to the 4 h time point with most of the injury occurring between 8 and 24 h. In order to investigate whether transient glutamate receptor activation affected Ca2+ levels within the ER, we treated cells with thapsigargin, an inhibitor of Sarco/Endoplasmic Reticulum Ca2+-ATPase pumps, which mobilizes Ca2+ from ER stores to the cytosol. Control cerebellar granule neurons and glutamate excited (5 min) neurons were loaded with Fluo-4AM at 0.5 and 5 h post-glutamate excitation, time points which preceded the onset of injury (Fig. 1a and b). The neurons were subsequently exposed to 10 μmol/L thapsigargin and changes in cytosolic Ca2+ levels were monitored. Addition of thapsigargin to the neurons resulted in a rapid increase in cytosolic Ca2+ levels. Quantification of this thapsigargin sensitive Ca2+ response between glutamate excited and control neurons revealed no significant difference in the level of peak Ca2+ responses between the two groups (Fig. 1c). Taken together, this data suggests that although transient glutamate receptor activation acutely increases cytosolic Ca2+ levels uptake of Ca2+ into ER is not suppressed post-insult.
Glutamate receptor activation does not induce changes in the gene expression of UPR associated genes
To determine if activation of glutamate receptors was associated with activation of the three main signalling pathways associated with accumulation of unfolded proteins within the ER, we exposed cerebellar granular cells to glutamate for 5 min and extracted RNA at various time periods. In parallel, neurons were treated with 3 μmol/L tunicamycin, an inhibitor of N-glycosylation, and a classical mediator of the ER stress response. Activation of ATF6 results in increased transcription of the ER resident chaperones Grp78 (Yoshida et al. 1998). As illustrated in Fig. 2a, glutamate treatment was not associated with a significant modulation of Grp78 mRNA expression. In contrast, tunicamycin treatment resulted in a robust and prolonged increase in Grp78 expression (Fig. 2a). Signalling through IRE1α is associated with increased splicing of the transcription factor, (XBP1 (Yoshida et al. 2001). In order to detect XBP1 splicing we selected primers which detect only the spliced form of XBP1 as described by Hirota et al. (2006). Real-time qPCR analysis revealed a highly potent and rapid increase in splicing of XBP1 following tunicamycin treatment which was sustained over the time period investigated, an event which was only modestly affected in glutamate-treated samples (Fig. 2b). Similarly, PERK-dependent signalling is associated with increases in CHOP mRNA expression via increased translation of ATF4 (Harding et al. 2000). In line with the previous observations, increased CHOP mRNA expression was only evident in tunicamycin-treated samples and not in the glutamate model (Fig. 2c). ER stress is also associated with the transcriptional activation of ATF4 (Reimertz et al. 2003) and ATF6 (Namba et al. 2007), however, despite being induced by tunicamycin treatment, glutamate treatment failed to result in their increased expression (data not shown). To confirm the results of the gene expression studies, we performed western blot analysis of lysates from glutamate- and tunicamycin-treated samples. As demonstrated in Fig. 2d, western blot analysis using an anti-KDEL antibody, which detects Grp94 and Grp78 expression, revealed only the tunicamycin-treated samples exhibited increased levels of the ER chaperones, Grp94 and Grp78. Similar results were found by western blotting for CHOP and protein disulphide isomerase expression (Fig. 2d). Furthermore, western blot analysis revealed the presence of the spliced form of XBP1 following tunicamycin treatment, an observation which was absent in the glutamate-treated samples.
Proteasome function is not affected by glutamate receptor activation
Given that disturbances in protein folding may also modulate the rate of protein degradation and the activity of the proteasome, the major site of protein degradation within a cell, we investigated whether glutamate-induced excitotoxic apoptosis was associated with the modulation of proteasome function. Proteins designated for proteasomal degradation are tagged with poly-ubiquitin chains resulting in their targeting to the proteasome complex where they are degraded. As demonstrated in Fig. 3a, treatment with the proteasomal inhibitor, epoxomicin (Epoxo), resulted in the accumulation of mono- and poly-ubiquitinylated proteins, an observation which was absence in glutamate-treated samples. We have previously profiled the transcriptional changes which occur following inhibition of the proteasome in neural cells (Concannon et al. 2007). Amongst the most prominently induced genes is the transcription factor c-Jun. Real-time qPCR analysis revealed a potent time-dependent increase in c-Jun transcription following inhibition of the proteasome with epoxomicin (Fig. 3c). This potent increase in c-Jun transcription was absent following glutamate treatment (Fig. 3b). Taken together, these data suggest that glutamate-induced excitotoxic apoptosis does not result in modulation of proteasome function and associated stress events.
NMDA receptor-mediated excitotoxic apoptosis of cultured mouse cortical neurons does not require PUMA
Endoplasmic reticulum stress-induced apoptosis is mediated, at least in part, by induction of the BH3-only protein PUMA (Reimertz et al. 2003), with increased transcription of PUMA mRNA preceding the onset of apoptosis. Real-time qPCR analysis revealed that the expression of levels of PUMA mRNA was not significantly altered in glutamate-treated cerebellar granule neurons (Fig. 4a). In contrast, tunicamycin treatment resulted in time-dependent increase in the expression of PUMA mRNA (Fig. 4b). Furthermore, this increase in PUMA mRNA within tunicamycin-treated samples preceded injury onset (Fig. 4c), supporting a requirement for PUMA in ER stress-mediated injury.
We next investigated the requirement for PUMA expression in an NMDA-mediated model of excitotoxic apoptosis in mouse cortical neurons. Primary cortical neurons were derived from both PUMA+/+ and PUMA−/− mice on the same day and maintained in culture for 11 days prior to NMDA treatment. Neurons were stimulated with 300 μmol/L NMDA for 15 min and allowed to recover for 24 h and cell viability in the cultures assessed by the ability to exclude the vital dye, PI. This treatment regime induced morphological changes indicative of apoptotic cell death including condensation and fragmentation of nuclei as assessed by Hoechst staining (Fig. 5a). Subsequent quantification of the number of PI-positive cells in PUMA+/+ and PUMA−/− cultures revealed that the loss of PUMA expression had no effect on the levels of cell death resulting from excitation with NMDA (Fig. 5b). In contrast, loss of PUMA expression significantly attenuated the levels of cell death following tunicamycin treatment (Fig. 5c). In agreement with these results, western blot analysis revealed an absence of increased expression of the ER stress markers, Grp78 or CHOP, in NMDA-treated cortical cultures (Fig. 5d), further demonstrating the absence of an ER stress response in NMDA receptor dependent glutamate toxicity.
Loss of PUMA expression does not protect against NMDA-induced excitotoxic apoptosis in an organotypic hippocampal slice culture model
In order to explore whether ER stress may be involved in mediating excitotoxic cell death in selectively vulnerable brain regions, we employed a mouse model of organotypic hippocampal slice culture. A transient activation of glutamate receptors with 0.05 mmol/L NMDA for 30 min resulted in cell death and a progression of apoptosis over time within the CA1 region of the organotypic hippocampal slice cultures. 4′, 6 Diamidino-2-phenylindole staining of these neurons within this CA1 region revealed the presence of condensed nuclei indicative of apoptotic cell death (Fig. 6a). Subsequent quantitative analysis of the levels of injury induced in slices derived from either PUMA+/+ and PUMA−/− slices mice revealed similar levels of cell death in PUMA+/+ and PUMA−/− slices suggesting that loss of PUMA expression did not protect from NMDA-mediated excitotoxic apoptosis (Fig. 6a and b). In contrast, loss of PUMA expression offered significant protection from ER stress mediated via tunicamycin treatment. As demonstrated in Fig. 6d, addition of 30 μmol/L tunicamycin for 24 h resulted in increased PI staining and damage with the DG region of the hippocampus. However, in mice deficient for PUMA expression, the levels of injury were dramatically impaired supporting a central role of PUMA in mediating ER stress-dependent injury in this model (Fig. 6c and d). It has been previously demonstrated that mice injected intraperitoneally with tunicamycin have marked neuronal damage within the DG of the hippocampus (Chae et al. 2004). Consistent with the absence of a role for PUMA in mediating NMDA-induced excitotoxic apoptosis, western blot analysis revealed increased levels of Grp78 only in tunicamycin-treated slices and not following NMDA treatment, demonstrating an absence of an ER stress response (Fig. 6e). Furthermore, in order to investigate whether conditions of OGD, which mimic the effects of ischaemia, could potentially induce an ER stress response we incubated hippocampal slice cultures in OGD conditions for 30 min and allowed them to subsequently recover under normal culture conditions. As demonstrated in Fig. 6f, 30 min OGD was sufficient to induce CA1 damage (Fig. 6f). Interestingly, a time-dependent increase in the expression of both Grp78 and CHOP was evident in this model (Fig. 6g).
NMDA-induced lesions in the mouse hippocampus occur independent of ER stress and PUMA expression
In order to examine whether excitotoxicity-induced apoptosis in vivo may be mediated in a PUMA-dependent manner, excitotoxic hippocampal lesions were induced in wild-type mice by bilateral intra-hippocampal NMDA injections. Injury in this model is characterized by widespread degeneration of CA1 neurons, which became TUNEL-positive at 8 h post-injection (Fig. 7a). Immunohistochemical analysis of levels of KDEL revealed modest immunoreactivity in CA1 neurons in control mice. Following NMDA injection KDEL immunostaining rapidly declined as early as 2 h post-treatment, supporting an absence of ER stress in this model. A similar immunostaining pattern was observed for calnexin, an additional ER stress marker (Supplemental Fig. S1). We next compared neuronal death induced by intra-hippocampal NMDA between PUMA+/+, PUMA+/− and PUMA−/− mice. Importantly, counting of neuronal within the hippocampus revealed similar counts in PUMA+/+, PUMA+/− and PUMA−/− mice (Supplemental Fig. S2). PUMA+/+ mice exhibited near-complete death of CA1 neurons when examined 48 h following NMDA injection. Quantification of the levels of apoptosis induced in this model revealed no significant differences between PUMA+/+, PUMA+/− and PUMA−/− mice (Fig. 7b), confirming that NMDA-induced excitotoxic apoptosis is mediated in an ER stress and PUMA independent manner.
- Top of page
- Materials and methods
- Supporting Information
Activation of glutamate receptors has been inextricably linked with the neuronal injury associated with acute neurological insults such as cerebral ischaemia, trauma, and seizures. Several studies have suggested the hypothesis that ER dysfunction and ER stress are responsible for transducing excitotoxic and ischaemic injury (Paschen 2003; Paschen et al. 2003; Tarabal et al. 2005; Uehara et al. 2006). In this study, we have demonstrated that the activation of NMDA receptors per se is not associated with significant modulation of ER stress signalling pathways. Moreover, loss of expression of the BH3-only protein, PUMA, a previously implicated mediator of ER stress-induced cell death had no effect in modulation the injury associated with excitotoxicity.
The classical ER stress response involves modulation of three distinct signalling pathways: ATF6, PERK and IRE1α pathways. Analysis of the expression levels of target genes of each of these pathways failed to detect significant changes in their expression levels in several models of delayed excitotoxic injury. The activation of a typical ER stress response is associated with a rapid and sustained increase in splicing of XBP1 mRNA, resulting in an alteration in the open reading frame of the XBP1 protein leading to the translation of a highly active transcription factor. In our studies glutamate-induced injury resulted in a very moderate increase in the splicing of XBP1 mRNA as assessed by qPCR. However, subsequent western blot analysis failed to demonstrate any increase in the expression of spliced XBP1 on a protein level. In contrast, tunicamycin-induced injury was associated with a rapid and robust increase in XBP1 splicing at both mRNA and protein level. Furthermore, induction of the classical XBP1 target gene, grp78, was unaffected, both at the mRNA and protein level. A recent study by Uehara et al. (2006) demonstrated increased splicing of XBP1 mRNA following NMDA-mediated excitotoxic injury, however, no increases in the translation of this potent transcription factor or induction of its target genes were demonstrated suggesting that although XBP1 splicing may be moderately affected it is not sufficient to allow for increases on a protein level. Finally, we demonstrate that the transcription rates of the BH3-only protein, PUMA, which has previously been demonstrated to be central to mediating ER stress-mediated cell death (Reimertz et al. 2003; Luo et al. 2005; Li et al. 2006; Nickson et al. 2007), remained unaltered during NMDA receptor-mediated excitotoxic apoptosis. Indeed, utilizing primary cortical neurons, organotypic hippocampal slices and in vivo hippocampal NMDA injection, we were able to demonstrate that excitotoxic injury does not require PUMA expression.
A previous study has demonstrated acute increases in ER Ca2+ levels following NMDA receptor activation (Wang et al. 2006). Our Fluo-4 experiments revealed that the influx of Ca2+ associated with transient NMDA receptor activation does not significantly alter the overall rate of Ca2+ uptake into the ER in the long-term. In line with this observation, previous studies have illustrated that mitochondria appear to be the major subcellular site for accumulation of Ca2+ following its influx into neurons after excitation of ionotropic receptors (Budd and Nicholls 1996; Stout et al. 1998; Ward et al. 2005). Of note, several studies have demonstrated a decrease in ER Ca2+ stores post-ischaemia (Kohno et al. 1997; Xing et al. 2004) which may also potentially account for the increased expression of ER stress markers following ischaemia.
Several studies have demonstrated that ischaemic injury in neurons is associated with activation of several pathways of the unfolded protein response (UPR) including phosphorylation of PERK (Kumar et al. 2001) and activation of IRE1α-dependent signalling (Paschen et al. 2003). The accumulation of protein aggregates containing ubiquitinylated proteins in the cytosol of neurons is also an early evident following cerebral ischaemia (Hu et al. 2000, 2001; Asai et al. 2002). Defects in proteasome-mediated protein degradation trigger certain similar cellular stress responses as observed during ER stress, such as attenuation of protein synthesis and induction of transcription factors of the ATF and C/enhancer binding protein (EBP) families (Jiang and Wek 2005), although the global pattern of gene expression pattern is quite distinct from that induced during ER stress (Reimertz et al. 2003; Concannon et al. 2007). In our in vitro and in vivo models we were also unable to detect increased levels of ubiquitinylated proteins or gene expression changes indicative of proteasomal dysfunction in response to excitotoxic injury. However, OGD in organotypic hippocampal slices was sufficient to induce increased expression of both Grp78 and CHOP. Our results suggest that application of exogenous glutamate/NMDA both in vitro and in vivo does not appear to contribute to ER stress responses or defects in protein degradation. Interestingly, hypoxia alone has been shown to increase the expression of ER stress response genes (Blais et al. 2004; Bi et al. 2005), suggesting that the activation of ER stress during ischaemia may be the result of the events linked to the absence of sufficient levels of oxygen and/or glucose. Ιndeed, hypoxia within tumour cells has recently been described to activate both the IRE1α and PERK signalling pathways of the UPR (Feldman et al. 2005). We and others have previously demonstrated a transcriptional activation of CHOP expression in neural cells exposed to hypoglycaemic conditions (Carlson et al. 1993; Ikesugi et al. 2006; Kogel et al. 2006). Furthermore, Hori et al. (1996) have demonstrated that hypoxia and subsequent reoxygenation results in a depletion of intracellular glucose levels and the subsequent induction of Grp78. Indeed, Grp78 and Grp94 were first identified on the basis of their induction in cells following glucose starvation (Pouyssegur et al. 1977). Similarly, the generation of reactive oxygen intermediates during reperfusion following ischaemia may also be detrimental to ER homeostasis and may contribute to ischaemia-induced ER stress or defect in protein degradation.
In conclusion, our study demonstrates that NMDA-mediated excitotoxic apoptosis does not appear to significantly alter ER stress responses and does not require PUMA expression. They also suggest that the contribution of NMDA-mediated pathways to the activation of the ER stress response seen during OGD or ischaemia may not be as significant as previously thought. This study has important implications for the reassessment of the design of therapies aimed at the ER stress response for the treatment of injuries associated with excitotoxic processes.
- Top of page
- Materials and methods
- Supporting Information
The authors wish to thank Drs A. Strasser and A. Villunger for gifts of knockout mice. This study was supported by Grants from Science Foundation Ireland (03/RP/B344) and the Health Research Board (RP/2005/206) to JHMP. DCH is supported by the Wellcome Trust, and LPT is a recipient of an IRCSET Post-graduate Studentship (325/2005).
- Top of page
- Materials and methods
- Supporting Information
- 2002) Selective proteasomal dysfunction in the hippocampal CA1 region after transient forebrain ischemia. J. Cereb. Blood Flow Metab. 22, 705–710. , , , , , , and (
- 2005) ER stress-regulated translation increases tolerance to extreme hypoxia and promotes tumor growth. EMBO J. 24, 3470–3481. , , et al. (
- 2004) Activating transcription factor 4 is translationally regulated by hypoxic stress. Mol. Cell. Biol. 24, 7469–7482. , , , , , , and (
- 1996) Mitochondria, calcium regulation, and acute glutamate excitotoxicity in cultured cerebellar granule cells. J. Neurochem. 67, 2282–2291. and (
- 1993) Regulation of the C/EBP-related gene gadd153 by glucose deprivation. Mol. Cell. Biol. 13, 4736–4744. , , , and (
- 2004) BI-1 regulates an apoptosis pathway linked to endoplasmic reticulum stress. Mol. Cell 15, 355–366. , , et al. (
- 1994) Glutamate receptors and the induction of excitotoxic neuronal death. Prog. Brain Res. 100, 47–51. (
- 1987) Glutamate neurotoxicity in cortical cell culture. J. Neurosci. 7, 357–368. , and (
- 2007) Apoptosis induced by proteasome inhibition in cancer cells: predominant role of the p53/PUMA pathway. Oncogene 26, 1681–1692. , , et al. (
- 2005) The unfolded protein response: a novel component of the hypoxic stress response in tumors. Mol. Cancer Res. 3, 597–605. , and (
- 2000) Regulated translation initiation controls stress-induced gene expression in mammalian cells. Mol. Cell 6, 1099–1108. , , , , , and (
- 2006) Quantitative measurement of spliced XBP1 mRNA as an indicator of endoplasmic reticulum stress. J. Toxicol. Sci. 31, 149–156. , , and (
- 1996) Exposure of astrocytes to hypoxia/reoxygenation enhances expression of glucose-regulated protein 78 facilitating astrocyte release of the neuroprotective cytokine interleukin 6. J. Neurochem. 66, 973–979. , , , , , , , , and (
- 2000) Protein aggregation after transient cerebral ischemia. J. Neurosci. 20, 3191–3199. , , and (
- 2001) Protein aggregation after focal brain ischemia and reperfusion. J. Cereb. Blood Flow Metab. 21, 865–875. , , , , , , and (
- 2006) Induction of endoplasmic reticulum stress in retinal pericytes by glucose deprivation. Curr. Eye Res. 31, 947–953. , , , , , and (
- 2005) Phosphorylation of the α-subunit of the eukaryotic initiation factor-2 (eIF2α) reduces protein synthesis and enhances apoptosis in response to proteasome inhibition. J. Biol. Chem. 280, 14189–14202. and (
- 2006) Induction of transcription factor CEBP homology protein mediates hypoglycaemia-induced necrotic cell death in human neuroblastoma cells. J. Neurochem. 99, 952–964. , , et al. (
- 1997) Neuroprotective nitric oxide synthase inhibitor reduces intracellular calcium accumulation following transient global ischemia in the gerbil. Neurosci. Lett. 224, 17–20. , , , , and (
- 2001) Comparison of excitotoxic profiles of ATPA, AMPA, KA and NMDA in organotypic hippocampal slice cultures. Brain Res. 917, 21–44. , and (
- 2001) Brain ischemia and reperfusion activates the eukaryotic initiation factor 2alpha kinase, PERK. J. Neurochem. 77, 1418–1421. , , et al. (
- 2005) Excitotoxic injury to mitochondria isolated from cultured neurons. J. Biol. Chem. 280, 28894–28902. , , , and (
- 2006) BH3-only proteins in cell death initiation, malignant disease and anticancer therapy. Cell Death Differ. 13, 1325–1338. , , and (
- 2006) Endoplasmic reticulum stress-induced apoptosis: multiple pathways and activation of p53-up-regulated modulator of apoptosis (PUMA) and Noxa by p53. J. Biol. Chem. 281, 7260–7270. , and (
- 2005) Transcriptional upregulation of PUMA modulates endoplasmic reticulum calcium pool depletion-induced apoptosis via Bax activation. Cell Death Differ. 12, 1310–1318. , , and (
- 2007) Transcriptional activation of ATF6 by endoplasmic reticulum stressors. Biochem. Biophys. Res. Commun. 355, 543. , , , and (
- 2007) PUMA is critical for neonatal cardiomyocyte apoptosis induced by endoplasmic reticulum stress. Cardiovasc. Res. 73, 48. , and (
- 2003) Endoplasmic reticulum: a primary target in various acute disorders and degenerative diseases of the brain. Cell Calcium 34, 365. (
- 2003) Transient cerebral ischemia activates processing of xbp1 messenger RNA indicative of endoplasmic reticulum stress. J. Cereb. Blood Flow Metab. 23, 449–461. , , and (
- 1977) Induction of two transformation-sensitive membrane polypeptides in normal fibroblasts by a block in glycoprotein synthesis or glucose deprivation. Cell 11, 941–947. , and (
- 2003) Gene expression during ER stress-induced apoptosis in neurons: induction of the BH3-only protein Bbc3/PUMA and activation of the mitochondrial apoptosis pathway. J. Cell Biol. 162, 587–597. , , , and (
- 1998) Distinct influx pathways, not calcium load, determine neuronal vulnerability to calcium neurotoxicity. J. Neurochem. 71, 2349–2364. , , and (
- 1979) Calcium dependence of toxic cell death: a final common pathway. Science 206, 700–702. , , and (
- 1998) Glutamate-induced neuron death requires mitochondrial calcium uptake. Nat. Neurosci. 1, 366–373. , , , and (
- 2005) Protein retention in the endoplasmic reticulum, blockade of programmed cell death and autophagy selectively occur in spinal cord motoneurons after glutamate receptor-mediated injury. Mol. Cell. Neurosci. 29, 283–298. , , , and (
- 1993a) Source specificity of early calcium neurotoxicity in cultured embryonic spinal neurons. J. Neurosci. 13, 2085–2104. , , and (
- 1993b) Cell-permeant Ca2+ chelators reduce early excitotoxic and ischemic neuronal injury in vitro and in vivo. Neuron 11, 221–235. , , , , , and (
- 2006) S-Nitrosylated protein-disulphide isomerase links protein misfolding to neurodegeneration. Nature 441, 513. , , , , , , , and (
- 2003) p53- and drug-induced apoptotic responses mediated by BH3-only proteins puma and noxa. Science 302, 1036–1038. , , , , , , and (
- 2006) Ester derivatives of tournefolic acid B attenuate N-methyl-D-aspartate-mediated excitotoxicity in rat cortical neurons. Mol. Pharmacol. 69, 950–959. , , , and (
- 2000) Mitochondrial membrane potential and glutamate excitotoxicity in cultured cerebellar granule cells. J. Neurosci. 20, 7208–7219. , , and (
- 2005) Cellular and subcellular calcium accumulation during glutamate-induced injury in cerebellar granule neurons. J. Neurochem. 92, 1081–1090. , , and (
- 2006) Real time single cell analysis of Bid cleavage and Bid translocation during caspase-dependent and neuronal caspase-independent apoptosis. J. Biol. Chem. 281, 5837–5844. , , , , and (
- 1998) Bax involvement in p53-mediated neuronal cell death. J. Neurosci. 18, 1363–1373. , , , , and (
- 2004) Caffeine releasable stores of Ca2+ show depletion prior to the final steps in delayed CA1 neuronal death. J. Neurophysiol. 92, 2960–2967. , , and (
- 2006) Endoplasmic reticulum stress and apoptosis signaling in human temporal lobe epilepsy. J. Neuropathol. Exp. Neurol. 65, 217–225. , , , , , , and (
- 1998) Identification of the cis-acting endoplasmic reticulum stress response element responsible for transcriptional induction of mammalian glucose-regulated proteins. Involvement of basic leucine zipper transcription factors. J. Biol. Chem. 273, 33741–33749. , , , and (
- 2001) XBP1 mRNA is induced by ATF6 and spliced by IRE1 in response to ER stress to produce a highly active transcription factor. Cell 107, 881. , , , and (
- 2004) Signaling the unfolded protein response from the endoplasmic reticulum. J. Biol. Chem. 279, 25935–25938. and (
- Top of page
- Materials and methods
- Supporting Information
Fig. S1C57BL/6 mice.
Fig. S2NeuN positive cells.
Please note: Blackwell Publishing are not responsible for the content or functionality of any supplementary materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.
|JNC_5187_sm_Figure S1.tif||5957K||Supporting info item|
|JNC_5187_sm_Figure S2.tif||2924K||Supporting info item|
|JNC_figure+s1.tif||5957K||Supporting info item|
|JNC_figure+s2.tif||2924K||Supporting 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.