Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Mar 9.
Published in final edited form as: Brain. 2005 Aug 25;128(Pt 10):2408–2420. doi: 10.1093/brain/awh619

The electrical response of cerebellar Purkinje neurons to simulated ischaemia

Martine Hamann 1,2,*, David J Rossi 1,3,*, Claudia Mohr 3, Adriana L Andrade 3, David Attwell 1
PMCID: PMC8906496  NIHMSID: NIHMS1769209  PMID: 16123143

Abstract

Despite lacking N-methyl-d-aspartate receptors, cerebellar Purkinje cells are highly vulnerable to ischaemic insults, which lead them to die necrotically in an α-amino-3-hydroxy-5-methyl-isoxazole-4-propionic acid (AMPA) receptor-dependent manner. To investigate the electrical events leading to this cell death, we whole-cell clamped Purkinje cells in cerebellar slices. Simulated ischaemia evoked an initial hyperpolarization of Purkinje cells by 8.5 mV, followed by a regenerative ‘anoxic depolarization’ (AD) to −14 mV. The AD was prevented by glutamate receptor blockers. In voltage-clamp mode, we used the cells’ glutamate receptors to sense the rise of extracellular glutamate concentration induced by ischaemia, with GABAA and GABAB receptors blocked and Cs+ as the main pipette cation. Ischaemia induced a small (<500 pA) slowly developing inward current in Purkinje cells, followed by a sudden large inward ‘AD current’ (~6 nA) which was largely prevented by blocking AMPA receptors. Removing extracellular calcium reduced the large glutamate-mediated current by ~70% at early times (after 10 min ischaemia), but had no effect at later times (15 min). Blocking the operation of glutamate transporters, by preloading cells with the slowly transported glutamate analogue PDC (l-trans-pyrrolidine-2,4-dicarboxylate), reduced the current by ~88% at early and 83% at later times. In Purkinje cells in slices from mice lacking the glial glutamate transporters GLAST or GLT-1, the ischaemia-evoked AD current was indistinguishable from that in wild-type slices. These data suggest that, in cerebellar ischaemia, the dominant cause of the electrophysiological dysfunction of Purkinje cells is an activation of Purkinje cell AMPA receptors. The glutamate activating these receptors is released both by exocytosis (at early times) and by reversal of a glutamate transporter, apparently in neurons.

Keywords: glutamate transporter, exocytosis, ischaemia, anoxia, cerebellum

Introduction

The cerebellum and hippocampus are particularly vulnerable to brain anoxia or ischaemia, showing loss of Purkinje cells and pyramidal cells, respectively (Pulsinelli, 1985; Cervos-Navarro and Diemer, 1991). There have been numerous studies of the events evoked by hippocampal ischaemia, in which the cessation of ATP production inhibits the Na/K pump, generating a rise of extracellular potassium concentration (Hansen, 1985). This leads to a release of glutamate by reversed uptake which activates N-methyl-d-aspartate (NMDA) and α-amino-3-hydroxy-5-methyl-isoxazole-4-propionic acid (AMPA) receptors (Rossi et al., 2000), triggering an excessive cation influx and [Ca2+]i rise which lead to pyramidal cells dying (Choi, 1987; Deshpande et al., 1987; Andine et al., 1988; Abdel-Hamid and Tymianski, 1997; Zhang and Lipton, 1999).

By contrast, little is known about the electrical response of cerebellar neurons to ischaemia. Purkinje cells die necrotically like hippocampal pyramidal cells, despite lacking (after ~8 days post-natally) the NMDA receptors which usually mediate a large glutamate-evoked Ca2+ influx into neurons (Yue et al., 1997; Martin et al., 2000). Glutamate, acting at Purkinje cell AMPA receptors, must be involved because AMPA receptor blockers prevent ischaemia-evoked Purkinje cell death in vivo (Balchen and Diemer, 1992). Since Purkinje cell AMPA receptors have a low Ca2+ permeability (Tempia et al., 1996), the [Ca2+]i rise occurring in ischaemic Purkinje cells (Mitani et al., 1995) may be generated by voltage-gated Ca2+ channels in response to the depolarization produced by incompletely desensitizing AMPA receptors (Brorson et al., 1995).

In this study we provide the first detailed description of the electrophysiological response of cerebellar Purkinje neurons to ischaemia. We show that glutamate release is a key determinant of the cells’ response to ischaemia and, by using the cells’ glutamate receptors to sense the rise in extracellular glutamate concentration, we assess the source of the glutamate release which triggers Purkinje cell death.

Methods

Brain slices and extracellular solution

Sprague–Dawley rats or transgenic mice (see below), 12–19 days old, were killed by cervical dislocation in accordance with UK animal experimentation regulations. Patch-clamp recordings from visually identified Purkinje cells in thin (160–220 μm) parasagittal cerebellar slices were performed as previously described (Hamann et al., 2002a). Slices were from animals of either sex. Recordings were made at 32 ± 2°C. Normal extracellular solution contained (mM): NaCl 126, NaHCO3 24, NaH2PO4 1, KCl 2.5, CaCl2 2.5, MgCl2 2, d-glucose 10 (gassed with 95% O2/5% CO2), pH 7.4. For experiments with no extracellular Ca2+, CaCl2 was replaced by 2.5 mM MgCl2 and 2 mM EGTA to chelate trace calcium. Kynurenic acid (1 mM) was included in the dissection and incubation solution (to block glutamate receptors, to reduce potential excitotoxic damage) but was omitted from the superfusion solution. To simulate the energy deprivation that occurs in ischaemia, glucose was replaced by 7 mM sucrose, 95% O2/5% CO2 was replaced with 95% N2/5% CO2 and 2 mM sodium iodoacetate and 1 mM sodium cyanide were added to block glycolysis and oxidative phosphorylation (Reiner et al., 1990).

Preloading slices with PDC

To block reversed transport of glutamate, slices were preloaded with the slowly transported glutamate analogue l-trans-pyrrolidine-2,4-dicarboxylic acid (PDC) (Longuemare and Swanson, 1995; Roettger and Lipton, 1996; Rossi et al., 2000). Slices were soaked in PDC (1 mM) containing solution for 1 h, in the presence of 1 mM kynurenate to block glutamate receptors and prevent neuronal death. Control slices were soaked in 1 mM kynurenate alone. Roettger and Lipton (1996) have shown that this procedure accumulates PDC within the cells of hippocampal slices without leading to a loss of glutamate. Rossi et al. (2000) showed that it reduced the activity of glutamate transporters in hippocampal slices without affecting glutamate receptors (as we also show below for cerebellar slices).

Patch-clamp recording from cerebellar slices

For recording, slices were placed under flowing solution on the stage of an upright microscope and viewed with a ×40 or ×60 water immersion objective with differential interference contrast and infrared optics. Whole-cell voltage-clamp recordings were made using an Axopatch 200B amplifier (Axon Instruments, USA) from the somata of visually identified Purkinje neurons. Patch pipettes were constructed from thick-walled borosilicate glass capillaries. For current-clamp recordings of membrane potential and for voltage-clamp experiments with ECl set to −65 mV, the solution contained (mM) K-gluconate 120, KCl 7.7, NaCl 4, HEPES 10, BAPTA 10, MgATP 4 and Na2GTP 0.5, pH adjusted to 7.2 with KOH. For voltage-clamp recordings of membrane current with ECl set to 0 mV they were filled with (mM) CsCl 130, NaCl 4, HEPES 10, BAPTA 10, MgATP 4, Na2GTP 0.5 and QX-314 10 (to block voltage-gated sodium currents and GABAB receptors), pH set to 7.2 with CsOH. Corrections for electrode junction potentials were made.

Series resistance voltage errors

In whole-cell mode, after series resistance compensation by ~75%, the residual series resistance was ~1 MΩ. The large size of the current changes evoked by ischaemia (up to 10 nA) at the peak of the anoxic depolarization (AD) means that, even after compensation, significant series resistance voltage errors (up to 10 mV) will inevitably occur (a detailed analysis of the effect of these errors for hippocampal pyramidal cells is presented in Hamann et al., 2002b). Data are presented in this paper without correction for this, because the series resistance was similar in different experimental conditions, and correcting for series resistance voltage errors would not alter the conclusions reached.

Transgenic mice

The generation of mice lacking GLT-1 or GLAST, by disrupting the parts of the genes encoding the third transmembrane region of the GLT-1 protein, or the 4th transmembrane region of GLAST, and their genotyping, have been described in detail previously (Tanaka et al., 1997b; Watase et al., 1998). Knock-out (−/−) and wild-type (+/+) mice were produced by breeding heterozygote (+/−) mice which had been backcrossed 6–9 times with C57 Black/6 mice. Experiments were on 2-week-old animals killed by cervical dislocation, and were carried out on pairs of +/+ and −/− animals from the same litter to reduce any variability that may occur between litters. Cerebellar slicing and electrophysiological recording in mice were as for rats.

Previous work has shown that Purkinje cells in mice lacking GLAST, the main glutamate transporter in the cerebellar molecular layer, showed anatomical and physiological properties broadly similar to those in wild-type littermates, apart from small changes of excitatory post-synaptic current (EPSC) properties (Marcaggi et al., 2003) and the persistence of multiple climbing fibre innervation to older ages in the knock-out (Watase et al., 1998). The overall anatomy of the cerebellum, including its size and foliation, Purkinje cell dendrite arborization, the area of the granule cell layer and the density of granule cells, the structure of parallel and climbing fibre synapses, and the parallel fibre synapse density, have been shown by a combination of light and electron microscopy not to differ significantly in the wild-type and GLAST knock-out mice and, like in wild-type mice, the parallel fibre fast EPSC is mediated solely by AMPA receptors with no NMDA receptor component (Watase et al., 1998). Western blotting of cerebellum from wild-type and knock-out mice has indicated that knocking out GLAST does not alter the total expression of the other cerebellar glutamate transporters GLT-1, EAAC1 and EAAT4 (Watase et al., 1998). In mice lacking GLT-1, there is no appreciable change in mRNA level for GLAST, EAAT4 or EAAC1 (Tanaka et al., 1997b), and no obvious change in cerebellar anatomy or Purkinje cell morphology at the light microscopic level (our observations).

Statistics

The effects of drugs (or transporter knock-out) on the ischaemic responses were assessed by comparing data from interleaved slices studied in the absence or presence of each drug (or transporter). Data are presented as mean ± standard error of the mean and significance of changes was assessed with a two-tailed Student’s t-test or χ2-test as appropriate.

Results

The voltage response of Purkinje cells to ischaemia

Purkinje cells whole-cell clamped with electrodes containing a mock-physiological solution (K+ as the main cation and ECl set to −65 mV) had an input resistance at −70 mV of 135 ± 33 MΩ (7 cells). When not voltage-clamped, they tended to fire action potentials spontaneously; the voltage response of cells to solution mimicking ischaemia was studied after spontaneous spiking in control solution had been terminated by the injection of hyperpolarizing current to bring the resting potential to ~−75 to −80 mV (mean value −78.3 ± 0.7 mV in four cells).

At the onset of simulated ischaemia, Purkinje cells initially hyperpolarized to ~−87 mV (mean hyperpolarization after 1–2.5 min in ischaemic solution was 8.5 ± 1.3 mV in four cells), but after ~8 min (504 ± 8 s) this was followed, after a short burst of action potentials, by a rapid prolonged depolarization to −14.3 ± 1.6 mV (Fig. 1A, mean data are shown in Fig. 1C). The initial hyperpolarization and subsequent prolonged depolarization are analogous to the voltage changes seen in ischaemic hippocampus, where the hyperpolarization may reflect an activation of Ca2+- and ATP-gated K+ channels when [ATP] falls following metabolic inhibition (Yamamoto et al., 1997; Nowicky and Duchen, 1998) and the sudden and prolonged depolarization is termed the AD (Hansen, 1985).

Fig. 1.

Fig. 1

Open in a new tab

Simulated ischaemia evokes a small hyperpolarization followed by a large depolarization in current-clamped cerebellar Purkinje cells. (A) Voltage response of a Purkinje cell to ischaemia solution. Before the large AD there is a burst of action potentials (arrow). Resting potential before ischaemia was −78 mV. (B) In ionotropic glutamate receptor blockers (NBQX and D-AP5), the AD is blocked, leaving the ischaemic hyperpolarization. Resting potential before ischaemia was −75 mV. (C) Mean data from experiments like A and B (four control cells, filled circles; three cells in blockers, open circles). (D) Specimen response to ischaemia of limited duration of a Purkinje cell which recovered its resting potential after the ischaemia. (E) Specimen response to ischaemia of limited duration of a Purkinje cell which failed to recover its resting potential after the ischaemia. (F) Number of cells recovering a resting potential more negative than −50 mV or failing to recover more negative than −30 mV, after termination of ischaemia 1 or 2.5 min after the AD. (G) Current response to ischaemia at −40 mV shows an outward current followed by a large inward current which declines to a smaller plateau. (H) Applying ionotropic glutamate receptor blockers greatly reduces the inward current. All records from cells studied with K-gluconate based internal solution.

To test the involvement of glutamate release in this series of events, we repeated this experiment in the presence throughout of the glutamate receptor blockers NBQX (25 μM) and D-AP5 (50 μM). The NMDA receptor blocker D-AP5 was included to guard against the possibility that the Purkinje cell still expressed some NMDA receptors (although these are normally absent after ~8 days post-natally: Llano et al., 1991; Rosenmund et al., 1992; Hausser and Roth, 1997). In the presence of these blockers the Purkinje cell showed only the hyperpolarization to ~−87 mV produced by ischaemia (12.3 ± 1.5 mV, from a resting potential of −74.7 ± 0.9 mV in three cells: Fig. 1B and C) with a subsequent very slow depolarization but no AD. The absence of the AD when ionotropic glutamate receptor blockers are present, in recordings lasting 20 min in ischaemia, demonstrates that the AD is produced by glutamate release, and that current generated by metabotropic glutamate receptors (Crepel et al., 1991; Vranesic et al., 1991; Glaum et al., 1992; Linden et al., 1994) is not involved in generating the AD. The possible mechanism of the ischaemia-evoked hyperpolarization remaining when glutamate receptors are blocked is considered in the Discussion.

Reversibility of the anoxic depolarization

To determine for how long glutamate needs to depolarize Purkinje cells, in order to produce irreversible damage, we carried out experiments in which the superfusion solution was switched back from ischaemia solution to normal solution either ~1 min (67 ± 6 s in 14 cells) or ~2.5 min (147 ± 10 s in 16 cells) after the AD (the timing of which was defined as the time of maximum rate of change of the potential), and observed whether the membrane potential was able to recover from the AD. Some cells recovered towards the normal resting potential, although they tended to show some instability of the resting potential even after this recovery (Fig. 1D), while others failed to show a significant recovery (Fig. 1E).

When cells fail to recover it is sometimes difficult to be certain that the lack of recovery does not result from movement of the cell relative to the electrode when the slice swells after the AD, so we rejected all cells in which the apparent potential fell to 0 mV and the capacity transient produced by a voltage step indicated complete loss of the cell. Having rejected these lost cells, to quantify the degree of irreversible depolarization we defined ‘recovered’ cells as those which hyperpolarized to at least −50 mV after ischaemia, while ‘no-recovery’ cells were those that failed to hyperpolarize beyond −30 mV (there were no cells which adopted potentials between −30 and −50 mV). Membrane potential recovery was greater when switching back to normal solution 1 min after the AD (10 out of 14 cells recovered) than when switching 2.5 min after the AD (4 out of 16 cells recovered), and this difference was significant (P = 0.03 by a χ2-test with Yates correction; Fig. 1F).

These data suggest that when glutamate is released and produces the AD it evokes irreversible damage to the cell in the first few minutes after the AD.

The current response of Purkinje cells to ischaemia

To study the mechanism of glutamate release, we voltage-clamped Purkinje cells, and used their glutamate receptors to sense the rise of glutamate occurring. Figure 1G shows voltage-clamp data obtained with the same K+-containing internal solution as in the voltage recording experiments of Fig. 1A–F. On applying ischaemic solution the current at −40 mV initially became more outward, reflecting the initial hyperpolarization seen in Fig. 1A, but then a large inward current suddenly developed, reflecting the AD in Fig. 1A. Applying NBQX and D-AP5 during the plateau of this current led to an almost complete block of the current (81 ± 3% in five cells, calculated assuming a current baseline at the level of the outward current before the AD), demonstrating that the depolarizing current is generated largely by glutamate release activating ionotropic receptors (Fig. 1H).

Having established that glutamate is the main agent producing the AD, subsequent experiments to define the receptor types generating the AD and to investigate the mechanism of glutamate release were performed in the presence of bicuculline (40 μM) plus picrotoxin (100 μM), to block GABAA receptor-mediated currents and thus help to isolate glutamate-mediated currents. In addition, Cs+ was used as the main intracellular cation and QX-314 was included in the patch pipette (see Methods) to improve voltage uniformity in the Purkinje cell dendritic tree and to block GABAB receptors. Figure 2A shows a specimen current response to ischaemia solution recorded at −33 mV. When Cs+ replaced K+ as the main internal cation and GABAA receptors were blocked, the initial ischaemia-evoked outward current was absent and was replaced by a slowly developing inward current which may partly reflect K+ entry through K+ channels activated in ischaemia. After ~8 min, 55 of the 56 cells studied showed an AD current that rapidly reached a peak and then decayed slowly (Fig. 2A), and sometimes increased again after that (e.g. Figs 3C and 4C). The time to the AD did not differ significantly (P = 0.44, Fig. 2B) when voltage-clamping with Cs+ in the pipette as in Fig. 2A, or recording in physiological conditions with K+ in the pipette and the membrane potential unclamped as in Fig. 1A. When ischaemia solution containing NBQX and D-AP5 was applied, no AD current was seen in 5 out of 7 cells (Fig. 2C, cf. Fig. 1B) while the remaining 2 out of 7 cells showed a smaller and more transient AD current than normal: overall the peak inward current generated was only 7.5% of that seen in 4 interleaved cells and 5% of that seen in all 56 cells studied (Fig. 2D).

Fig. 2.

Fig. 2

Open in a new tab

With K+ omitted from the pipette solution and GABAA receptors blocked, the ischaemia-evoked current is purely inward. (A) Current response of a Purkinje cell at −33 mV to ischaemia solution, recorded with a CsCl-based internal solution and bicuculline (40 μM) plus picrotoxin (100 μM) in the external solution. (B) Mean time to the AD voltage or current for 4 cells recorded in current clamp (I clamp) with a K-gluconate internal as in Fig. 1A and 55 cells voltage-clamped (V clamp) with a CsCl internal solution and GABAA receptors blocked as in A. (C) Response to ischaemia as in A but with ionotropic glutamate receptors blocked with NBQX and D-AP5. (D) Comparison of the peak inward current produced by 10 min ischaemia solution in the absence (56 cells, con) and the presence (7 cells) of NBQX and D-AP5.

Fig. 3.

Fig. 3

Open in a new tab

AMPA receptors generate the AD and action potentials do not contribute significantly to glutamate release in cerebellar ischaemia. (A) Current response to ischaemia, and lack of suppression of the post-AD current by D-AP5 (50 μM). (B) Superimposing NBQX (25 μM) on AP5 suppresses most of the post-AD current. (C) NBQX alone suppresses most of the post-AD current. (D) The AMPA receptor blocker GYKI 53655 (20 μM) blocks most of the post-AD current. (E) Mean current suppressed after the AD by D-AP5, NBQX and GYKI 53655, in 12, 8 and 7 cells, respectively. (F) Fraction of the current blocked by GYKI 53655 plus NBQX that was blocked by GYKI 53655 alone, in six cells. (G) Current response to ischaemia in the presence of TTX. (H) TTX did not significantly affect (in four cells in TTX interleaved with four control cells) the time to the AD, the amplitude of the current change at the time of the AD (IAD), or the glutamate-mediated current after the AD, Iglu, defined as the current suppressed by 25 μM NBQX and 50 μM D-AP5. All data recorded at −33 mV with a CsCl-based internal solution with GABAA receptors blocked.

Fig. 4.

Fig. 4

Open in a new tab

Removal of external calcium reduces early but not late glutamate release evoked by ischaemia. (A) Specimen response to ischaemia in normal solution at −33 mV, with 25 μM NBQX and 50 μM D-AP5 (blockers) applied to measure the glutamate-mediated current 10 min after the start of ischaemia. (B) Response to ischaemia in solution lacking calcium (and containing EGTA). (C) As in A but with blockers applied to measure Iglu after 15 min. (D) As in C but in zero-calcium solution. (E–H) Comparison of time to the AD (E), AD current (F), and glutamate-mediated current (Iglu) after 10 min (G) or 15 min (H) ischaemia, in 36 cells in normal solution (of which 18 and 10 cells provided Iglu data at 10 and 15 min) and 21 interleaved cells in calcium-free solution (of which 9 and 8 cells provided Iglu data at 10 and 15 min). (I) Effect of zero-calcium solution on the response to 1 μM AMPA (in 50 μM D-AP5, 40 μM bicuculline and 1 μM TTX) at −33 mV. (J) Average data from experiments as in I on three cells.

The AD is generated largely by AMPA receptors

Purkinje cells lack NMDA receptors after ~8 days post-natally (Llano et al., 1991; Rosenmund et al., 1992; Hausser and Roth, 1997), but NMDA receptors on granule cells might contribute to controlling the release of glutamate onto Purkinje cells in ischaemia. However, applying D-AP5 to block the effects of glutamate on NMDA receptors after the AD had no effect on the ischaemia-evoked AD current (Fig. 3A, B and E). Superfusing the AMPA/kainate receptor blocker NBQX, either on top of D-AP5 (Fig. 3B) or alone (Fig. 3C) resulted in most of the post-AD inward current being blocked.

To distinguish the possible contributions of AMPA- and kainate-receptors to the maintained depolarization produced by ischaemia, we studied the effect of the AMPA receptor blocker GYKI 53655 [20 μM, which blocks AMPA receptors by >90% but blocks kainate receptors by <5% (Wilding and Huettner, 1995; Bleakman et al., 1997)]. GYKI 53655 blocked most of the ischaemia-evoked inward current after the AD (Fig. 3D and E) and superimposing NBQX produced only a small further block, some of which may be block of the small fraction of AMPA receptors remaining unblocked by the GYKI 53655. On average GYKI 53655 blocked ~80% of the total current blockable by GYKI 53655 and NBQX together (Fig. 3F). Thus, the great majority of the ischaemia-evoked, glutamate-mediated current in Purkinje cells is mediated by AMPA receptors which do not completely desensitize in the maintained presence of glutamate.

Glutamate release is action potential independent

The occurrence of ischaemia-evoked action potentials in cells that are not voltage-clamped (Fig. 1A) suggested that the AD and glutamate release may be triggered by action potentials. To test this we applied ischaemic solution containing 1 μM TTX to block voltage-gated Na+ channels (Fig. 3G). TTX had no effect on the time to the AD, the magnitude of the AD current, or the current suppressed by ionotropic glutamate receptor blockers after the AD (Iglu) (Fig. 3H).

Glutamate release is initially calcium-dependent

When ischaemia solution lacking calcium (and containing 2 mM EGTA to chelate trace Ca2+: see Methods) was applied (Fig. 4B and D), although an AD current was present at the normal latency (Fig. 4E), it was smaller in amplitude (P < 0.01) than in interleaved slices in the control solution (Fig. 4A, C and F). Applying glutamate receptor blockers 10 min after the start of ischaemia suppressed a glutamate-mediated current that was only 30% of the amplitude seen in normal calcium-containing solution (Fig. 4A, B and G, P = 0.016 compared with control solution). However, by 15 min after the start of ischaemia, the ischaemia-evoked inward current in zero-calcium solution had become much larger (Fig. 4D), and the glutamate-mediated current was not significantly different from that seen after 15 min in normal solution (Fig. 4C and H, P = 0.36). The response of Purkinje cells to superfused AMPA (1 μM, in 40 μM bicuculline, 50 μM D-AP5 and 1 μM TTX) was not affected by calcium removal (Fig. 4I and J). Consequently, the suppression of ischaemia-induced currents produced by removing calcium (Fig. 4A–D) reflects reduced glutamate release. These results indicate that Ca2+-dependent exocytosis generates glutamate release for at least the first few minutes after the AD, and that 15 min after the start of ischaemia the Ca2+-dependent release has stopped.

The use of zero-calcium solution to block exocytotic release of glutamate assumes that exocytosis is indeed Ca2+-dependent. This is a reasonable assumption (which is supported by the reduction of glutamate release seen in Fig. 4); although it has been suggested that early ischaemia-evoked spontaneous exocytosis of transmitter is Ca2+-independent (Katchman and Hershkowitz, 1993; Fleidervish et al., 2001), these studies did not use EGTA to chelate trace Ca2+, and subsequent work has suggested that this protocol did not lower [Ca2+]o sufficiently to block Ca2+-dependent exocytosis (Allen and Attwell, 2004).

Glutamate release is partly by reversed uptake

The run-down of ion gradients occurring during hippocampal ischaemia has been shown to lead to release of glutamate by reversal of glutamate transporters (Madl and Burgesser 1993; Roettger and Lipton, 1996; Phillis et al., 2000; Rossi et al., 2000). A similar run-down of gradients occurs in ischaemic cerebellum, with [K+]o rising to 40 mM (Kraig et al., 1983). To test whether this [K+]o rise (and the associated depolarization and fall of [Na+]o) leads to significant glutamate release by transporter reversal, we used the approach of Longuemare and Swanson (1995), Roettger and Lipton (1996) and Rossi et al. (2000) to block transporter function. We preloaded slices with the slowly transported glutamate analogue PDC (see Methods), and then washed extracellular PDC out of the slice. The aim was to accumulate PDC intracellularly, so that if the run-down of ion gradients in ischaemia reverses the operation of glutamate transporters, the slowly transported PDC will bind preferentially, occupying the transporter and preventing glutamate release.

To assess whether this procedure succeeded in blocking transporter function, we compared the response of Purkinje cells in non-ischaemic slices to superfused glutamate (100 μM), the extracellular concentration of which is normally reduced by uptake, and to the non-transported glutamate analogue AMPA (1 μM). If PDC preloading blocks transporters then superfused glutamate should penetrate further into the slice and generate a larger response, relative to the response produced by AMPA which should be unaffected by block of uptake. Figure 5A and B show responses to glutamate and AMPA in a control slice and a slice preloaded with PDC. While the responses to AMPA were similar in control conditions and after PDC preloading (Fig. 5C), the ratio of (response to glutamate)/(response to AMPA) was significantly greater (P = 0.04) in six slices preloaded with PDC than in five control slices (Fig. 5C), confirming that the activity of transporters was reduced by the preloading procedure.

Fig. 5.

Fig. 5

Open in a new tab

Preloading with the glutamate transport blocker PDC reduces the fraction of cells showing an AD and greatly reduces glutamate release after the AD. (A–C) Effects of PDC-preloading on the current response of Purkinje cells (at −33 mV) to glutamate (100 μM) and AMPA (1 μM). (A) Specimen response of a cell to AMPA and glutamate in control solution. (B) Specimen response to AMPA and glutamate after PDC preloading: the AMPA response is similar to that in control solution, but the glutamate response is increased. (C) Mean data from experiments as in A and B, on five control cells and six cells after PDC preloading. (D) Specimen responses of two different Purkinje cells (at −33 mV) to ischaemia after PDC preloading, with 25 μM NBQX and 50 μM D-AP5 (blockers) applied after 10 or 15 min. No AD and little glutamate-mediated current are seen in these cells. (E) Comparison of the response to ischaemia in 36 control cells (of which 18 and 10 cells provided Iglu data at 10 and 15 min) and 12 interleaved cells after PDC preloading (of which 5 and 5 cells provided Iglu data at 10 and 15 min).

Figure 5D shows the response to ischaemia of two PDC-preloaded slices, with NBQX and D-AP5 applied at ~10 and ~15 min after the AD (cf. the control data in Fig. 4A and C). The PDC-preloaded slices show no AD current and a much smaller glutamate-mediated current. Out of 12 PDC preloaded slices studied, 7 showed complete abolition of the AD current (while only 1 out of 35 interleaved control slices failed to show an AD current; significantly different, P = 7 × 10−5 by χ2-test; Fig. 4E). The remainder did show an AD current (presumably the PDC preloading was less successful in blocking reversed uptake, possibly because some PDC was lost from the cells during the time needed to locate and record from a cell), but the subsequent glutamate-mediated current was reduced by 60–80% in magnitude. Averaging over all the slices, PDC-preloading reduced the glutamate-mediated current measured 10 and 15 min after the start of ischaemia by 88 and 83%, respectively (Fig. 4E).

These data, together with those recorded in 0 Ca2+ solution, suggest that in the first 10 min of ischaemia glutamate is released both by exocytosis and by the reversed operation of glutamate transporters, but that by 15 min in ischaemia exocytotic glutamate release has stopped and the remaining release is almost entirely by reversed uptake.

GLAST and GLT-1 glutamate transporters do not release glutamate in ischaemia

Glutamate released by reversed uptake could in principle be released by GLAST and GLT-1, which are located in Bergmann glia and molecular layer astrocytes, or by the neuronal transporters EAAC1 and EAAT4, which are located in Purkinje cells (Chaudhry et al., 1995; Yamada et al., 1996; Furuta et al., 1997a, b; Nagao et al., 1997). In the cerebellar cortex the density of glial transporters [21 000/μm3 for GLAST + GLT-1; Lehre and Danbolt (1998)] is higher than that of neuronal transporters [21000/μm3 for EAAT4, and EAAC1 is present at a lower level: Dehnes et al. (1998)], but glial cells have a lower intracellular concentration of glutamate because of conversion to glutamine by glial glutamine synthetase, and so glial transporters may be less prone to reverse during the change of ionic gradients occurring in ischaemia. To investigate the possible role of glial transporters in releasing glutamate we used transgenic mice lacking either the GLAST or GLT-1 transporters, neither of which shows any major upregulation of other transporters (Tanaka et al., 1997b; Watase et al., 1998).

The response to ischaemia of Purkinje cells in wild-type littermates of the GLT-1 and GLAST knock-out mice was similar to that in rats (Fig. 6A and D), except that the AD current occurred earlier than in rat Purkinje cells [after 235 ± 15 s (n = 16) in wild-type controls for GLT-1 knock-out mice, and 209 ± 14 s (n = 14) in wild-type controls for GLAST knock-out mice, compared with 454 ± 17 s in 55 rat Purkinje cells]. The subsequently maintained inward current was blocked by 25 μM NBQX and 50 μM D-AP5. As in rat Purkinje cells, PDC preloading greatly reduced the ischaemia-evoked glutamate-mediated current (measured after 10 min ischaemia), to 5.3% of its control value in wild-type littermates of GLT-1 KO mice (4 cells for PDC preloading, Iglu = 110 ± 41 pA, compared with 18 cells without preloading, Iglu = 2061 ± 242 pA, P = 0.0013; data not shown) and to 10.6% of its control value in littermates of GLAST KO mice (3 cells for PDC preloading, Iglu = 248 ± 100 pA, compared with 14 cells without preloading, Iglu = 2335 ± 392 pA, P = 0.03; data not shown). Thus, as for rat Purkinje cells, the ischaemia-evoked glutamate-mediated current is generated to a large extent by the reversal of glutamate transporters.

Fig. 6.

Fig. 6

Open in a new tab

Knocking out the glutamate transporters GLT-1 or GLAST has little effect on the response to ischaemia in mice. (A) and (B) Specimen response to ischaemia at −33 mV in wild-type (A) and GLT-1 knock-out (KO) mouse (B) Purkinje cells. (C) Mean AD current, and glutamate-mediated current from experiments as in A and B on 18 control and 12 KO cells. (D) and (E) Specimen response to ischaemia at −33 mV in wild-type (D) and GLAST knock-out mouse (E) Purkinje cells. (F) Mean AD current, and glutamate-mediated current from experiments as in D and E on 14 wild-type (WT) and 8 KO cells.

The response to ischaemia of GLT-1 knock-out and GLAST knock-out mice was indistinguishable from that in wild-type littermates (Fig. 6B and E), with no differences in the time to the AD current, the amplitude of the AD current or the magnitude of the glutamate-mediated current suppressed by NBQX and D-AP5 after the AD (Fig. 6C and F).

These data suggest that, as in the hippocampus (Hamann et al., 2002b), glial glutamate transporters do not contribute significantly to releasing glutamate early in severe cerebellar ischaemia, and thus that the release blockable by PDC preloading is via reversed operation of neuronal glutamate transporters, presumably EAAT4 and/or EAAC1.

Discussion

We have characterized for the first time the early electrical responses of cerebellar Purkinje cells to ischaemia, with the aim of understanding the mechanism of the ischaemia-evoked rise of extracellular glutamate concentration which leads to Purkinje cell death and motor dysfunction (Balchen and Diemer, 1992).

Ischaemia initially hyperpolarizes Purkinje cells

When the cerebellum is made ischaemic, the loss of the ATP supply leads to the Na+/K+ pump being inhibited and, as a result of the lack of pumping back into the cell of K+ which leaks out because the membrane potential is more positive than EK (Rossi et al., 2000), there is an initial slow rise of extracellular potassium concentration, [K+]o, to ~12 mM, followed by a more rapid rise to ~40 mM (Kraig et al., 1983). We find that Purkinje cells initially hyperpolarize when exposed to ischaemic solution. Whole-cell voltage-clamping the cells using a K-gluconate based pipette solution revealed an outward current corresponding to the initial hyperpolarization (Fig. 1G). This was absent when using a CsCl-based internal solution and GABAA receptors were blocked (Fig. 2). The hyperpolarization cannot be produced by ischaemia-evoked GABA release activating GABAA receptors (Allen et al., 2004) because the initial resting potential (−78 mV) was more negative than ECl (−65 mV). It may, therefore, reflect the activation of a Ca2+- or ATP-gated K+ current in Purkinje cells, as seen previously in hippocampal neurons (Nowicky and Duchen, 1998), which could contribute to the initial rise of [K+]o to ~12 mM (Kraig et al., 1983).

The AD of Purkinje cells is generated by glutamate and rapidly causes irrevocable malfunction

After the initial hyperpolarizing phase of the response to ischaemia, Purkinje cells exhibit a large and rapid AD which probably correlates with the rise of [K+]o to ~40 mM (Kraig et al., 1983). The Purkinje cell AD was blocked by glutamate receptor blockers, and so reflects ischaemia-evoked glutamate release. By using blockers of NMDA, AMPA and AMPA-plus-kainate receptors, we have shown that the great majority of the glutamate-mediated current is generated by AMPA receptors (Fig. 3A–F). The rapid speed of the AD suggests that it is a regenerative event produced by a positive feedback loop like that in the hippocampus, where ionic gradient run-down releases K+ which depolarizes cells and thus releases glutamate, which in turn depolarizes cells further and releases more K+. Recovery of the resting membrane potential after termination of simulated ischemia was prevented if the ischaemia was maintained for even just a few minutes after the AD occurred (Fig. 1D–F).

The complete abolition of the AD that we see with glutamate receptors blocked is similar to the effect of receptor blockers in abolishing the AD in hippocampal pyramidal cells (Rossi et al., 2000). The different types of glutamate receptors expressed in different brain areas will determine which receptor blockers are most effective at preventing the AD. In hippocampal pyramidal cells, where both NMDA and (to a lesser extent) AMPA receptors generate the ischaemia-evoked inward current (Rossi et al., 2000), it is essential to have both receptor types strongly blocked to prevent the AD; blocking only one type delays, but does not prevent, the AD (Tanaka et al., 1997a). Interestingly, some authors report an AD occurring even with blockers of both receptors present (Lauritzen and Hansen, 1992; Muller and Somjen, 2000) but in some cases at least (Jarvis et al., 2001) the blocker concentrations used were lower than those Rossi et al. (2000) found necessary to block the AD in the hippocampus: the positive feedback mechanisms involved in generating the AD require that a strong block of all glutamate receptors is achieved to prevent the AD. In the neocortex, blocking NMDA receptors alone is sufficient to block spreading depression (which has some similarities to the AD) and AMPA receptor block is ineffective (Nellgard and Wieloch, 1992), probably because of the fairly complete desensitization of AMPA receptors in the face of a maintained glutamate concentration rise. In contrast, in cerebellar Purkinje cells, which lack NMDA receptors but have incompletely desensitizing AMPA receptors (Brorson et al., 1995), it is essential to block AMPA receptors to prevent the AD (Fig. 3).

Incompletely desensitizing AMPA receptors generate the AD

Rossi et al. (2000) used a simple model of the ischaemic hippocampus to predict the rise of [K+]o, membrane depolarization and rise of extracellular glutamate concentration produced by ischaemia, based on the idea that cutting off the ATP supply to the Na+/K+ pump leads to run-down of the transmembrane ionic gradients, reversal of glutamate transporters and activation of glutamate receptors. In that simulation, activation of NMDA receptors by glutamate played a key role in producing a sustained depolarization of cells, because hippocampal AMPA receptors desensitize almost completely (Spruston et al., 1995). Cerebellar Purkinje cells, by contrast, lack NMDA receptors after postnatal day 8, so the fact that ischaemia-evoked glutamate release generates a sustained depolarizing current (as in Figs 1–3) depends on Purkinje cell AMPA receptors desensitizing incompletely, as suggested by Brorson et al. (1995).

Glutamate is released both by exocytosis and by reversal of glutamate transporters

In hippocampal slices, blocking glutamate release by reversed uptake can completely abolish the AD and subsequent glutamate release, whereas removing extracellular Ca2+ has little effect (Rossi et al., 2000). However, in cerebellar slices reversed uptake is not the only significant means of glutamate release. Blocking Ca2+-dependent exocytosis by removing extracellular Ca2+ reduces glutamate release and thus reduces the glutamate-mediated membrane current by ~70% early after the AD (10 min after the start of ischaemia; Fig. 4). Later than 15 min after the start of ischaemia, however, removing extracellular Ca2+ has no effect on the ischaemia-evoked glutamate release. This situation is similar to that in striatum, where microdialysis experiments in vivo have shown that in ischaemia there is a brief period of Ca2+-dependent glutamate release, after which glutamate release is Ca2+-independent (Wahl et al., 1994).

Even with exocytosis blocked there is still substantial glutamate release in cerebellar ischaemia (Fig. 4). Blocking reversed uptake by preloading with PDC produces a larger reduction of early (10 min) ischaemic glutamate release than does blocking exocytosis (Fig. 5). Further, blocking reversed uptake produces a similar reduction of late glutamate release (15 min after the start of ischaemia; Fig. 4), but late glutamate release is unaffected by removing calcium (Fig. 5). Thus, the major fraction of ischaemic glutamate release is by reversed uptake. Interestingly, after 10 min ischaemia, removal of calcium blocks ~70% of the glutamate-mediated current, whereas blocking reversed uptake blocks ~88%, implying (since the sum of these blocks is >100%) that these two release mechanisms do not have independent effects on the glutamate-mediated current. This could result simply from non-linearity in the conversion of extracellular glutamate concentration to AMPA receptor-mediated current. Alternatively, it could result from a synergy between the effects of release by Ca2+-dependent exocytosis and reversed uptake, as follows. Glutamate released by exocytosis will activate AMPA receptors and promote uptake reversal by raising [Na+]i and depolarizing cells, and in addition the resulting ATP consumption on ion pumping (Attwell and Laughlin, 2001) will accelerate the run-down of ion gradients that reverses transporters. Thus, blocking exocytosis will reduce reversed uptake as well. Conversely, glutamate released by reversed uptake will depolarize cells and release more glutamate by exocytosis, so that blocking reversed uptake will reduce exocytotic glutamate release as well.

Glial glutamate transporters do not contribute to ischaemic glutamate release

Purkinje cells in knock-out mice lacking the glial transporters GLT-1 or GLAST showed electrophysiological responses to ischaemia that were indistinguishable from those in wild-type littermates, with a large AD current and large glutamate-mediated membrane current. This suggests that the ischaemia-evoked transporter-mediated release of glutamate occurs by reversal of pre- or post-synaptic neuronal glutamate transporters. Neuronal transporters are thermodynamically more likely to reverse in the altered ionic conditions of ischaemia than are glial-transporters, because the glutamate concentration is initially lower in glia, as a result of glutamate being converted to glutamine by the glial specific enzyme glutamine synthetase. Welsh et al. (2002) found that lesioning the climbing fibres reduced ischaemia-evoked Purkinje cell death in vivo, but it is unclear whether this reflects an abolition of glutamate release by reversed uptake in pre-synaptic terminals during ischaemia, or an abolition of exocytotic release after the ischaemic episode, since glutamate release after an ischaemic episode contributes to cell death in the hippocampus (Hammond et al., 1994; Szatkowski, 1994). Interestingly, Welsh et al. (2002) also found that Purkinje cell death was inversely correlated with the expression of postsynaptic EAAT4 and the glycolytic enzyme aldolase, occurring preferentially in areas where EAAT4 and aldolase were at a low level. Although the increased death in areas of low EAAT4/aldolase might simply reflect less ATP production via aldolase, EAAT4 could also have a protective effect by lowering the glutamate concentration produced by exocytotic release of glutamate from the climbing fibres after a period of ischaemia (Otis et al., 1997; Auger and Attwell, 2000; Welsh et al., 2002).

If neuronal glutamate transporters reverse and release glutamate in ischaemia, then why do glial transporters not take up glutamate and reduce the rise of extracellular glutamate concentration produced? For GLAST and GLT-1 to take up a substantial amount of glutamate, there may need to be continuous conversion of glutamate to glutamine within astrocytes. This process requires ATP, which is absent during ischaemia, so the intracellular concentration of glutamate in astrocytes rises at the start of ischaemia (Storm-Mathisen et al., 1992) when only a small amount of glutamate has been taken up. This will bring the driving force on the transporter closer to equilibrium and reduce further uptake of glutamate. A similar situation may occur in the hippocampus, where block or knock-out of GLT-1 transporters does not affect the time to the AD or the glutamate-mediated current after the AD (Rossi et al., 2000; Hamann et al., 2002b), but knock-out of the neuronal glutamate transporter EAAC1 prolongs the time to the AD 3-fold (Gebhardt et al., 2002b).

In summary, our data suggest that the glutamate release which triggers Purkinje cell death and motor dysfunction in ischaemia occurs both by exocytosis and by the reversal of neuronal glutamate transporters, and that this depolarizes Purkinje cells by acting on incompletely desensitizing AMPA receptors.

Acknowledgements

We thank Kohichi Tanaka for his generous provision of the knock-out mice, and Hélène Marie and Stephen McGuiness for mouse genotyping. This study was supported by the Wellcome Trust, the European Union, a Wolfson-Royal Society Award (D.A.), and a Burroughs-Wellcome Fellowship (D.J.R.).

Abbreviations:

AD

anoxic depolarization

NMDA

N-methyl-d-aspartate

AMPA

α-amino-3-hydroxy-5-methyl-isoxazole-4-propionic acid

EPSC

excitatory postsynaptic current

PDC

l-trans-pyrrolidine-2,4-dicarboxylic acid

References

  1. Abdel-Hamid KM, Tymianski M. Mechanisms and effects of intracellular calcium buffering on neuronal survival in organotypic hippocampal cultures exposed to anoxia/aglycemia or to excitotoxins. J Neurosci 1997; 17: 3538–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Allen NJ, Attwell D. The effect of simulated ischaemia on spontaneous GABA release in area CA1 of the juvenile rat hippocampus. J Physiol 2004; 561: 485–98. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Allen NJ, Rossi DJ, Attwell D. Sequential release of GABA by exocytosis and reversed uptake leads to neuronal swelling in simulated ischemia of hippocampal slices. J Neurosci 2004; 24: 3837–49. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Andine P, Jacobson I, Hagberg H. Calcium uptake evoked by electrical stimulation is enhanced postischemically and precedes delayed neuronal death in CA1 of rat hippocampus: involvement of N-methyl-d-aspartate receptors. J Cereb Blood Flow Metab 1988; 8: 799–807. [DOI] [PubMed] [Google Scholar]
  5. Attwell D, Laughlin SB. An energy budget for signalling in the grey matter of the brain. J Cereb Blood Flow Metab 2001; 21: 1133–45. [DOI] [PubMed] [Google Scholar]
  6. Auger C, Attwell D. Fast removal of synaptic glutamate by postsynaptic transporters. Neuron 2000; 28: 547–58. [DOI] [PubMed] [Google Scholar]
  7. Balchen T, Diemer NH. The AMPA antagonist, NBQX, protects against ischemia-induced loss of cerebellar Purkinje cells. Neuroreport 1992; 3: 785–8. [DOI] [PubMed] [Google Scholar]
  8. Bleakman D, Ballyk BA, Schoepp DD, Palmer AJ, Bath CP, Sharpe EF, Wolley ML, Bufton HR, Kamboj RK, Tarnawa I, Lodge D. Activity of 2,3-benzodiazepines at native rat and recombinant human glutamate receptors in vitro: stereospecificity and selectivity profiles. Neuropharmacology 1997; 35: 1689–702. [DOI] [PubMed] [Google Scholar]
  9. Brorson JR, Mansolillo PA, Gibbons SJ, Miller RJ. AMPA receptor desensitization predicts the selective vulnerability of cerebellar Purkinje cells to excitotoxicity. J Neurosci 1995; 15: 4515–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Cervos-Navarro J, Diemer NH. Selective vulnerability in brain hypoxia. Crit Rev Neurobiol 1991; 6: 149–82. [PubMed] [Google Scholar]
  11. Chaudhry FA, Lehre KP, van Lookeren Camagne M, Ottersen OP, Danbolt NC, Storm-Mathisen J. Glutamate transporters in glial plasma membranes: highly differentiated localizations revealed by quantitative ultrastructural immunocytochemistry. Neuron 1995; 15: 711–20. [DOI] [PubMed] [Google Scholar]
  12. Choi DW. Ionic dependence of glutamate neurotoxicity. J Neurosci 1987; 7: 369–79. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Crepel F, Daniel H, Hemart N, Jaillard D. Effects of ACPD and AP3 on parallel-fibre-mediated EPSPs of Purkinje cells in cerebellar slices in vitro. Exp Brain Res 1991; 86: 402–6. [DOI] [PubMed] [Google Scholar]
  14. Dehnes Y, Chaudhry FA, Ullensvang K, Lehre KP, Storm-Mathisen J, Danbolt NC. The glutamate transporter EAAT4 in rat cerebellar Purkinje cells: a glutamate-gated chloride channel concentrated near the synapse in parts of the dendritic membrane facing astroglia. J Neurosci 1998; 18: 3606–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Deshpande JK, Siesjo BK, Wieloch T. Calcium accumulation and neuronal damage in the rat hippocampus following cerebral ischemia. J Cereb Blood Flow Metab 1987; 7: 89–95. [DOI] [PubMed] [Google Scholar]
  16. Fleidervish IA, Gebhardt C, Astman N, Gutnick MJ, Heinemann U. Enhanced spontaneous transmitter release is the earliest consequence of neocortical hypoxia that can explain the disruption of normal circuit function. J Neurosci 2001; 21: 4600–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Furuta A, Rothstein JD, Martin LJ. Glutamate transporter protein subtypes are expressed differentially during rat CNS development. J Neurosci 1997a; 17: 8363–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Furuta A, Martin LJ, Lin CL, Dykes-Hoberg M, Rothstein JD. Cellular and synaptic localization of the neuronal glutamate transporters excitatory amino acid transporter 3 and 4. Neuroscience 1997b; 81: 1031–42. [DOI] [PubMed] [Google Scholar]
  19. Gebhardt C, Korner R, Heinemann U. Delayed anoxic depolarizations in hippocampal neurons of mice lacking the excitatory amino acid carrier 1. J Cereb Blood Flow Metab 2002; 22: 569–75. [DOI] [PubMed] [Google Scholar]
  20. Glaum SR, Slater NT, Rossi DJ, Miller RJ. Role of metabotropic glutamate (ACPD) receptors at the parallel fiber-Purkinje cell synapse. J Neurophysiol 1992; 68: 1453–62. [DOI] [PubMed] [Google Scholar]
  21. Hamann M, Rossi DJ, Attwell D. Tonic and spillover inhibition of granule cells control information flow through cerebellar cortex. Neuron 2002a; 33: 625–33. [DOI] [PubMed] [Google Scholar]
  22. Hamann M, Rossi DJ, Marie H, Attwell D. Knocking out the glial glutamate transporter GLT-1 reduces glutamate uptake but does not affect hippocampal glutamate dynamics in early simulated ischaemia. Eur J Neurosci 2002b; 15: 308–14. [DOI] [PubMed] [Google Scholar]
  23. Hammond C, Crepel V, Gozlan H, Ben-Ari Y. Anoxic LTP sheds light on the multiple facets of NMDA receptors. Trends Neurosci 1994; 17: 497–503. [DOI] [PubMed] [Google Scholar]
  24. Hansen AJ. Effect of anoxia on ion distribution in the brain. Physiol Rev 1985; 65: 101–48. [DOI] [PubMed] [Google Scholar]
  25. Hausser M, Roth A. Dendritic and somatic glutamate receptor channels in rat cerebellar Purkinje cells. J Physiol 1997; 501: 77–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Jarvis CR, Anderson TR, Andrew RD. Anoxic depolarization mediates acute damage independent of glutamate in neocortical brain slices. Cereb Cortex 2001; 11: 249–59. [DOI] [PubMed] [Google Scholar]
  27. Kraig RP, Ferreira-Filho CR, Nicholson C. Alkaline and acid transients in cerebellar microenvironment. J Neurophysiol 1983; 49: 831–50. [DOI] [PubMed] [Google Scholar]
  28. Katchman AN, Hershkowitz N. Early anoxia-induced vesicular glutamate release results from mobilization of calcium from intracellular stores. J Neurophysiol 1993; 70: 1–7. [DOI] [PubMed] [Google Scholar]
  29. Lauritzen M, Hansen AJ. The effect of glutamate receptor blockade on anoxic depolarization and cortical spreading depression. J Cereb Blood Flow Metab 1992; 12: 223–29. [DOI] [PubMed] [Google Scholar]
  30. Lehre KP, Danbolt NC. The number of glutamate transporter subtype molecules at glutamatergic synapses: chemical and stereological quantification in young adult rat brain. J Neurosci 1998; 18: 8751–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Linden DJ, Smeyne M, Connor JA. Trans-ACPD, a metabotropic receptor agonist, produces calcium mobilization and an inward current in cultured cerebellar Purkinje neurons. J Neurophysiol 1994; 71: 1992–98. [DOI] [PubMed] [Google Scholar]
  32. Llano I, Marty A, Armstrong CM, Konnerth A. Synaptic- and agonist-induced excitatory currents of Purkinje cells in rat cerebellar slices. J Physiol 1991; 434: 183–213. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Longuemare MC, Swanson RA. Excitatory amino acid release from astrocytes during energy failure by reversal of sodium-dependent uptake. J Neurosci Res 1995; 40: 379–86. [DOI] [PubMed] [Google Scholar]
  34. Madl JE, Burgesser K. Adenosine triphosphate depletion reverses sodium-dependent, neuronal uptake of glutamate in rat hippocampal slices. J Neurosci 1993; 13: 4429–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Marcaggi P, Billups D, Attwell D. The role of glial glutamate transporters in maintaining the independent operation of juvenile mouse cerebellar parallel fibre synapses. J Physiol 2003; 552: 89–107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Martin LJ, Sieber FE, Traystman RJ. Apoptosis and necrosis occur in separate neuronal populations in hippocampus and cerebellum after ischemia and are associated with differential alterations in metabotropic glutamate receptor signaling pathways. J Cereb Blood Flow Metab 2000; 20: 153–67. [DOI] [PubMed] [Google Scholar]
  37. Mitani A, Yanase H, Namba S, Shudo M, Kataoka A. In vitro ischemia-induced intracellular Ca2+ elevation in cerebellar slices: a comparative study with the values found in hippocampal slices. Acta Neuropathol (Berl) 1995; 89: 2–7. [DOI] [PubMed] [Google Scholar]
  38. Muller M, Somjen GG. Na+ dependence and the role of glutamate receptors and Na+ channels in ion fluxes during hypoxia of rat hippocampal slices. J Neurophysiol 2000; 84: 1869–80. [DOI] [PubMed] [Google Scholar]
  39. Nagao S, Kwak S, Kanazawa I. EAAT4, a glutamate transporter with properties of a chloride channel, is predominantly localized in Purkinje cell dendrites, and forms parasagittal compartments in rat cerebellum. Neuroscience 1997; 78: 929–33. [DOI] [PubMed] [Google Scholar]
  40. Nellgard B, Wieloch T. NMDA-receptor blockers but not NBQX, an AMPA-receptor antagonist, inhibit spreading depression in the rat brain. Acta Physiol Scand 1992; 146: 497–503. [DOI] [PubMed] [Google Scholar]
  41. Nowicky AV, Duchen MR. Changes in [Ca2+]i and membrane currents during impaired mitochondrial metabolism in dissociated rat hippocampal neurons. J Physiol 1998; 507: 131–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Otis TS, Kavanaugh MP, Jahr CE. Postsynaptic glutamate transport at the climbing fiber-Purkinje cell synapse. Science 1997; 277: 1515–18. [DOI] [PubMed] [Google Scholar]
  43. Phillis JW, Ren J, O’Regan MH. Transporter reversal as a mechanism of glutamate release from the ischemic rat cerebral cortex: studies with DL-threo-beta-benzyloxyaspartate. Brain Res 2000; 880: 105–12. [DOI] [PubMed] [Google Scholar]
  44. Pulsinelli WA. Selective neuronal vulnerability: morphological and molecular characteristics. Prog Brain Res 1985; 63: 29–37. [DOI] [PubMed] [Google Scholar]
  45. Reiner PB, Laycock AG, Doll CJ. A pharmacological model of ischemia in the hippocampal slice. Neurosci Lett 1990; 119: 175–8. [DOI] [PubMed] [Google Scholar]
  46. Roettger V, Lipton P. Mechanism of glutamate release from rat hippocampal slices during in vitro ischemia. Neuroscience 1996; 75: 677–85. [DOI] [PubMed] [Google Scholar]
  47. Rosenmund C, Legendre P, Westbrook GL. Expression of NMDA channels on cerebellar Purkinje cells acutely dissociated from newborn rats. J Neurophysiol 1992; 68: 1901–5. [DOI] [PubMed] [Google Scholar]
  48. Rossi D, Oshima T, Attwell D. Glutamate release in severe brain ischaemia is mainly by reversed uptake. Nature 2000; 403: 316–21. [DOI] [PubMed] [Google Scholar]
  49. Spruston N, Jonas P, Sakmann B. Dendritic glutamate receptor channels in rat hippocampal CA3 and CA1 pyramidal neurons. J Physiol 1995; 482: 325–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Storm-Mathisen J, Storm-Mathisen J, Danbolt NC, Rothe F, Torp R, Zhang N, et al. Ultrastructural immunocytochemical observations on the localization, metabolism and transport of glutamate in normal and ischemic brain tissue. Prog Brain Res 1992; 94: 225–41. [DOI] [PubMed] [Google Scholar]
  51. Szatkowski M, Attwell D. Triggering and execution of neuronal death in brain ischaemia: two phases of glutamate release by different mechanisms. Trends Neurosci 1994; 17: 359–65. [DOI] [PubMed] [Google Scholar]
  52. Tanaka E, Yamamoto S, Kudo Y, Mihara S, Higashi H. Mechanisms underlying the rapid depolarization produced by deprivation of oxygen and glucose in rat hippocampal CA1 neurons in vitro. J Neurophysiol 1997a; 78: 891–902. [DOI] [PubMed] [Google Scholar]
  53. Tanaka K, Watase K, Manabe T, Yamada K, Watanabe M, Takahashi K, et al. Epilepsy and exacerbation of brain injury in mice lacking the glutamate transporter GLT-1. Science 1997b; 276: 1699–702. [DOI] [PubMed] [Google Scholar]
  54. Tempia F, Kano M, Schneggenburger R, Schirra C, Garaschuk O, Plant T, et al. Fractional calcium current through neuronal AMPA-receptor channels with a low calcium permeability. J Neurosci 1996; 16: 456–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Vranesic I, Batchelor A, Gahwiler BH, Garthwaite J, Staub C, Knopfel T. Trans-ACPD-induced Ca2+ signals in cerebellar Purkinje cells. Neuroreport 1991; 2: 759–62. [DOI] [PubMed] [Google Scholar]
  56. Wahl F, Obrenovitch TP, Hardy AM, Plotkine M, Boulu R, Symon L. Extracellular glutamate during focal cerebral ischaemia in rats: time course and calcium-dependency. J Neurochem 1994; 63: 1003–11. [DOI] [PubMed] [Google Scholar]
  57. Watase K, Hashimoto K, Kano M, Yamada K, Watanabe M, Inoue Y, et al. Motor discoordination and increased susceptibility to cerebellar injury in GLAST mutant mice. Eur J Neurosci 1998; 10: 976–88. [DOI] [PubMed] [Google Scholar]
  58. Welsh JP, Yuen G, Placantonakis DG, Vu TQ, Haiss F, O’Hearn E, et al. Why do Purkinje cells die so easily after global brain ischemia? Aldolase C, EAAT4, and the cerebellar contribution to posthypoxic myoclonus. Adv Neurol 2002; 89: 331–59. [PubMed] [Google Scholar]
  59. Wilding TJ, Huettner JE. Differential antagonism of alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid-preferring and kainite-preferring receptors by 2,3-benzodiazepines. Mol Pharmacol 1995; 47: 582–7. [PubMed] [Google Scholar]
  60. Yamada K, Watanabe M, Shibata T, Tanaka K, Wada K, Inoue Y. EAAT4 is a post-synaptic glutamate transporter at Purkinje cell synapses. Neuroreport 1996; 7: 2013–17. [DOI] [PubMed] [Google Scholar]
  61. Yamamoto S, Tanaka E, Shoji Y, Kudo Y, Inokuchi H, Higashi H. Factors that reverse the persistent depolarization produced by deprivation of oxygen and glucose in rat hippocampal CA1 neurons in vitro. J Neurophysiol 1997; 78: 903–11. [DOI] [PubMed] [Google Scholar]
  62. Yue X, Mehmet H, Penrice J, Cooper C, Cady E, Wyatt JS, et al. Apoptosis and necrosis in the newborn piglet brain following transient cerebral hypoxia-ischaemia. Neuropathol Appl Neurobiol 1997; 23: 16–25. [PubMed] [Google Scholar]
  63. Zhang Y, Lipton P. Cytosolic Ca2+ changes during in vitro ischemia in rat hippocampal slices: major roles for glutamate and Na+-dependent Ca2+ release from mitochondria. J Neurosci 1999; 19: 3307–15. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES