- Email: [email protected]
Extracellular amino acid levels in hippocampus during pilocarpine-induced seizures
Epilepsy Research, 14 (1993) 139-148 0920-1211/93/$06.00 © 1993 Elsevier Science Publishers B.V. All rights reserved
139
EPIRES 00548
Extracellular amino acid levels in hippocampus during pilocarpine-induced seizures
Maria H. Millan, Astrid G. Chapman and Brian S. Meldrum Departnlent of Neurology, bzstitute of Psychiatry, London SE5 8AF, UK (Received 23 June 1992; revision received 23 October 1992; accepted 2 November 1992)
Key words: Pilocarpine; Aspartate; Glutamate; Glutamate uptake inhibition; Hippocampus; Microdialysis; (Epilepsy)
Extracellular levels of aspartate, glutamate and glutamine were monitored by microdialysis in the dorsal hippocampus of freely moving rats following the administration of a convulsant dose of pilocarpine (400 mg/kg, i.p.). Rats were either pretreated with the glutamate uptake inhibitor, I.trans.pyrrolidine-2,4-dicarboxylic acid (PDC, 1 mM in the perfusion medium, - 2 5 min), or received pilocarpine directly. All rats injected with pilocarpine (with or without PDC pretreatment) developed limbic seizures (latency 15.4+2.4 rain). Without PDC pretreatment there were no significant changes in extracellular levels of aspartate, glutamate and glutamine following pilocarpine administration until the onset of iimbic seizures when glutamine levels fell by 35%. Following PDC pretreatment there were ,arge and sustained increases in extracellular hippocampal aspartate (250%) and glutamate (55%) levels, but no significant change in the glutamine level. When pilocarpine was administered to this group of rats, there were f,,'ther selective, significant, transient increases in the extraceilular levels of aspartate (31%) and glutamate (18%) which preceded the onset of seizures. Aspartate and glutamate levels were not significantly increased (relative to PDC controls) during seizures. The conditions for pilocarpine-induced increases in aspartate and glutamate release were established in parallel groups of anaesthetised rats where pilocarpine was administered via a microdialysis probe in the dorsal hippocampus. Following the infusion of 10 mM pilocarpine there were large and rapid increases in the levels of aspartate (143%) and glutamate (179%), which were completely abolished by the absence of calcium in the perfusion medium, or by the presence of atropine (20 mM) or tetrodotoxin (i ~M).
Introduction In rats the systemic injection of a high dose of cholinergic agents, such as the muscarinic agonist pilocarpine, results in the development of motor limbic seizures 26'33. Pilocarpine-induced seizures can be prevented by focal injection of excitatory amino acid antagonists into several brain areas involved in the development and transmission of iimbic seizures 28. These findings suggest that activaCorrespondence to: A.G. Chapman, Department of Neurology, Institute of Psychiatry, De Crespigny Park, Denmark Hill, London SE5 8AF, UK.
tion of muscarinic receptors results in the enhancement of excitatory processes responsible for the development of limbic seizures. The hippocampus plays a major role in the development and maintenance of limbic seizures. It receives a major cholinergic innervation from the septum and contains numerous excitatory links using glutamate as a transmitter 27'3s. The hippocampal formation is thus a suitable area for studies on glutamate release during pilocarpine-induced limbic seizures. Several studies have used in vivo microdialysis in an effort to quantify a presumed increase in glutamate release during epileptic seizures in animal
140 models. Most of them failed to show such a phenomenon ~3"t4'35. It has been suggested that prompt reuptake of glutamate could markedly limit the presumed increase in extracellular concentration. Our previous s! udy on anaesthetised rats with on-line glutamate enzymatic detection using dihydrokainate to block glutamate reuptake did not confirm this hypothesis22. In this study a new, potent and selective glutamate uptake inhibitor, I.trans-pyrrolidine-2,4-dicarboxylic acid (PDC) i, was applied via the dialysis probe in freely moving rats that were exposed to pilocarpine-induced limbic seizures. This permitted assessment of excitatory amino acid release prior to and following seizure onset. Additionally, we have characterised the direct effects of pilocarpine on excitatory amino acid release, by using anaesthetised rats in which pilocarpine and other drugs were delivered via a dialysis probe in the dorsal hippocampus. Methods
Male Sprague-Dawley rats (250 g) were used for the experiments.
Freely moving rats The animals were anaesthetised with pentobarbital, 60 mg/kg, i.p., and placed in a stereotaxic frame (Kopf Instruments). Dialysis probes were chronically implanted into the right dorsal hippocampus (A + 3.8, V + 5.0 from intraaurai line and L + 2.8). 24 h after implantation of the probes, the animals were placed in a plexiglas rodent cage and the probes were connected to a counter-balanced, freely rotating swivel (Carnegie Medicin). The dialysis probes were flushed with Ringer solution (NaCI 125 mM, KC! 3.3 mM, MgSO4 2.4 raM, KH2PO4 !.25 mM, CaCI2 1.85 mM, pH 7.2) prior to perfusion (3 Fi/min) using a slow infusion pump (Carnegie Medicin, CMAI00). A 90-rain wash-out period was allowed before the onset of the experiments. There were three experimental groups of freely moving animals: Group l, pilocarpine-induced seizures without the glutamate uptake inhibitor, PDC; Group 2, pilocarpine-induced seizures, following pretreatment with PDC via the dialysis probe; Group 3, administration of PDC alone.
In group 1 (n = 6) five 5-min samples (15/al each) were collected for the determination of baseline levels of amino acids. Scopolamine methyl nitrate (1 mg/kg, s.c.) was injected 15 min prior to pilocarpine hydrochloride (400 mg/kg, i.p.) in order to prevent peripheral effects of pilocarpine. Follow ing pilocarpine administration, 15-/d samples were continuously collected during the latency period (latency to seizures ranged from 10 to 35 min; mean: 15.4 + 2.4), and for at least 30 min following the development of limbic seizures, or until the death of the animal (pilocarpine-induced seizures were fatal in 1/13 rats 15 rain after onset of seizures). In group 2 (n = 7) the collection of five 5-min baseline samples was followed by the collection of an additional five 5-min samples in the presence of the glutamate reuptake inhibitor PDC (delivered via probe as 1 mM in Ringer solution) before the injection of scopolamine and pilocarpine. PDC was present in the perfusate for the remainder of the experiment, otherwise conditions were identical to those of group 1 above. In group 3 (n =6) the collection of five 5-rain baseline samples was followed by the administration of PDC (delivered via the probe as 1 mM in Ringer solution) for the remainder of the experiment. 5-min samples were collected continuously for a 60-90-min time period designed to match the time course of the experiment in group 2. All the surviving animals were killed with an overdose of pentobarbital and the brains were taken for histological examination using frozen sections. This showed that the dialysis probe membrane vertically crossed CA~, dentate gyrus and CA4 hippocampal cell layers.
Anaesthetised animals Animals were anaesthetised with chloral hydrate (600 mg/kg, i.p.) and placed in a stereotaxic apparatus (Kopf Instruments) and a microdialysis probe was inserted in the dorsal hippocampus and perfused with Ringer solution (2 Fl/min perfusion rate). Microdialysis sampling began after a 90-min wash-out period. Three baseline samples were collected for the next 30 min (20 F1 each) before drugs were applied via the probe for 20 min (2 x 20-FI samples). This was followed by a 30-min recovery
141 period when the hippocampal probe was again perfused with Ringer solution (3 × 20-pl samples). The following experimental groups were studied: 1, 10 mM pilocarpine (20 min application, n = 6); 2, 10 mM pilocarpine + 20 mM atropine (20 min application, n = 4); 3, 10 mM pilocarpine + 1/~M tetrodotoxin (TTX) (20 min application, n = 4); 4, 10 mM pilocarpine (20 min application) in animals perfused with Ca 2÷-free medium containing 2 mM EGTA throughout the experiment (n=4). All drugs were dissolved in Ringer solution (pH adjusted to 7.2) and delivered via the hippocampal probe for 20 min.
Microdialysis and biochemical analysis Microdialysis probes were assembled as required (see Young et al.39). The permeable tip of the probes, made of cuprophane fibre, was 2.0-2.5 mm long. The in vitro recovery of amino acids was tested by inserting the tip of the probe in 10 #M solutions of amino acids. With a perfusion rate of 2 #l/min (at room temperature) the mean recovery values were 14.0 + 6% for aspartate and 15.8 _ 1.6% for glutamate (n = 12). Dialysate samples (15/zl for freely moving rats, 20 #1 for anaesthetised rats) were analysed by HPLC after ortho-phthaldialdehyde derivatisation of amino acids. The amino acid derivatives were separated on a Spherisorb ODS2 5# column (25 c m x 4.6 cm) using a 10-55% methanol gradient in 0.1 M sodium acetate, 2.5% tetrahydrofuran buffer, pH 5.75. The HPLC system consisted of a Spectraphysics SP 8800 ternary HPLC pump, a Kratos FS 950 Fluorimeter and a Spectraphysics Chromjet Integrator. Levels of the following amino acids were estimated: aspartate, glutamate, glutamine, gly.:ine, serine, alanine and taurine. GABA !evels c~t.,id not be reliably quantified with the ~iPLC conditions used. Statistical analysis was performed using Student's t-test.
Results
(1) Freely moving animals (A ) Behavioural results Intraperitoneal injection of pilocarpine resulted in the development of motor limbic seizures in all animals. Within 3 min post injection rats displayed gnawing followed by tremor of increasing intensity, ear twitching and head nodding. After 10-35 min (mean latency period 15.4 + 2.4 min) the animals displayed initially whole body clonus with loss of posture often with barrel rotation and elements of tonus, or full whole body tonus. The majority of animals (12/13) recovered from this initial seizure and after 5-10 min displayed further seizures with rearing, forelimb clonus and falling. This kind of behaviour was repeated with increased frequency and eventually became continuous (status epilepticus). Perfusion with PDC neither changed the animal's behaviour in the non-seizing state nor altered the severity of pilocarpine-induced seizures.
(B) i',iochemical results Only aspartate, glutamate and glutamine levels showed any significant changes in the hippocampal dialysate under the experimental conditions used in this study. The dialysate levels of other amino acids (extracellular hippocampal baseline levels given in parentheses): serine (2.6 + 0.4 #M), glycine (5.2 + 0.5 #M), alanine (5.1 + 0.7 #M) and taurine (3.9 + 0.5 gM) were also monitored, but are not discussed further because of their lack of significant response to either PDC, pilocarpine administration or seizures.
Group 1, pilocarpine-induced seizures without glutamate uptake inhibitor, PDC. Amino acid levels are expressed as the means of the pooled 5-min fractions from three time periods: (a) the 25-min baseline period, (b) after pilocarpine injection, but before onset of behavioural seizures (10-35 min) and (c) during seizures and status epilepticus (1545 min) (see Fig. 1). Concentration of aspartate in the dialysate during baseline recording was 0.20 + 0.08/zM, glutamate was 0.96 + 0.26/tM and glutamine was 12.9 -/- 1.3/zM (n = 6).
142
~a.
GLUTAMATE
AsPARTATE
¢Jr 0 . 3 Z
15 i_GLUTAMINE
1.2
0 (3
i..... i
tu
W >,
101
0.2~
0.81
There were no significant changes in the mean levels of aspartate and glutamate following the administration of pilocarpine or the onset of seizures (Fig. I). The level of glutamine was significantly decreased (by 35%, P <0.05) during seizures (Fig. 1).
I .J
~ 0.1~
1U
0.41
" 0
............
o!:. ..........................
.. .......
04 812 04 812 012 FRACTION NUMBER (5 rain or 15 pl each)
04
BASELINE (25 rain) !:!:! FOLLOWING PILOCARPINE,400 mglkg, I.p. (10.35 rain) FOLLOWING OffSET OF LIMBIC SEIZURE (15.45 rain)
Fig. I. Extracellular levels of aspartate, glutamate ~.ild glutamine in dorsal hippocampus of freely moving rats (a) during baseline conditions, (b) following pilocarpine administration (400 mg/kg, i.p.) and (c) following onset of limbic seizures. Cons¢cutive 5-rain samples ( 15 ld each) were pooled for each experimental phase, and the values are expressed as l~M in the hippocampal dialysate (mean + SEM). The asterisks denote values significantly different from the corresponding baseline control value (* P < 0.05).
:
d z
|
~
!
!
8 07si 10
~ 0.50'
o.2si _. =c
0~ 0 4 81216 0 4 81216 0 4 81216 FRACTION NUMBER (5 rain or 15 pl each) BASELINE (2S rain) Poe BASELINE (25 rain) i;i FOLLOWING PILOCARPINE,400 mg/k9, i.p. (10-35 rain) [ ] FOLLOWING ONSET OF LIMBIC SEIZURE (15-45 rain)
Fig. 2. Extracellular levels of aspartate, glutamate and glutamine in dorsal hippocampus of freely moving rats (a) during baseline conditions, (b) following perl~sion with i mM PDC (present in the perfusate lbr the remainder of the experiment), (c) following pilocarpine administration (400 mg/kg, i,p.) and (d) following onset of limbic seizures, Consecutive 5-min samples (15 141each) were pooled for each experimental phase, and the values are expressed as t~M in the hippocampal dialysate (mean + SEM). The asterisks denote values significantly different from the corresponding baseline control value (** P < 0.01; *** P <0.001).
Group 2, pilocarpine-induced seizures following pretreatment with PDC. Amino acid levels are expressed as the means of the pooled 5-min fractions from four time periods: (a) the 25-rain baseline period, (b) a 25-min period following PDC administration, (c) after pilocarpine injection, but before onset of behavioural seizures (1035 min) and (d) during seizures and status epilepticus ( 15-45 rain). Baseline concentrations of amino acids were similar to those of group I above except that glutamate levels were higher (aspartate, 0.21 + 0.02 /zM; glutamate, 2.16 + 0.33 #M; glutamine, 12.3 __. 1.0/aM; n = 7). Perfusion with ! m M PDC resulted in significant increases of extraceilular aspartate (by 233%, P <0.01) and glutamate (by 54%, P <0.01). Administration of pilocarpine resulted in further increases in aspartate and glutamate, but not in glutamine levels (Fig. 2), whereas following the onset of seizures (initially intermittent and subsequently sustained) the levels of aspartate, glutamate and glutamine apparently decreased compared with directly prior to seizures (not statistically significant). Mean values of aspartate and glutamate appeared increased relative to corresponding values after PDC (see Figs. 3 and 4) but these differences were not statistically significant. Group 3, administration of PDC alone. Amino acid levels are expressed as the means of the pooled 5-min fractions from four time periods, designed to correspond to the experimental time course in group 2 above: (a) the 25-min baseline period, (b-d) 25-rain + 15-min + 25-min periods following PDC administration (1 mM via dialysate probe). Baseline concentrations of amino acids were similar to those of group 2 above: aspartate, 0.22 + 0.02 #M, glutamate was 2.12 + 0.13/~M and glutamine was 12.8 + 0.7/~M (n = 6).
143
I F
=.
. . . .
i
~SP-ART~E i
:
'
!41
~0.75 o
rt ,.I
O. =C
15 ~
:, | | --
. . . .
!
.2_
10
0,50
[
)
~::,
0.25
:S 0 n.
GLUTAMINE
GLUTAMATE
!
31
rj uJ
~
:
~i~ ~i~i, 0
0 '" 0
12 16
,
4~ 8
12 16
. . . . . . . . .
0
4A 8
t . ,
•
• 1..
12 16
i; Po-6~
i ( 1 mM)._~
L(!mM) .
FRACTION NUMBER (5 mln or 15 pl each)
Fig. 3. Extracellular levels of aspartate, glutamate and glutamine in dorsal hippocampus of freely moving rats (a) during baseline conditions, (b--d) following perfusion with I mM PDC alone for 60:90 rain. Consecutive 5-rain samples (15 td each) were pooled for the baseline period and for three consecutive time periods designed to correspond to the experimental phases shown in Fig, 2. The values are expressed as pM in the hippocampal dialysate (mean + SEM). The asterisks denote values significantly different from the corresponding baseline control value (*** P <0.001).
Levels of aspartate and glutamate increased initially by 250% and 55% respectively (P <0.001),
ILl
+40
•k
~ : i + Pilocarpine
z -1
Seizures
ul
|
u Q
I
+20
a_
:Z 0 U,I
(9
ziz o
ASP
GLU
GLN
-20
Fig. 4. Calculated differences in aspartate, glutamate and glutamine levels between PDC-treated, freely moving rats with and without pilocarpine-induced seizures. The calculations are based on the data shown in Fig. 2 and Fig. 3. The cross-hatched bars represent the excess extracellular amino acid levels following pilocarpine administration, and the solid bars represent the excess extracellular amino acid levels following the onset of seizures. The values are expressed as per cent changes in the pilocarpine-treated group compared to corresponding values in the PDC only groups (* P <0.05).
closely resembling the PDC-induced changes in group 2. The PDC-enhanced extracellular glutamate level remained relatively stable throughout the experimental period, while the PDC-enhanced extracellular aspartate level showed a tendency towards a gradual decrease (statistically not significant) during consecutive phases of the experiment. Glutamine concentration in dialysate was not significantly changed by the addition of PDC (Fig. 3). Fig. 4 shows the pilocarpine-induced effect on extracellular aspartate, glutamate and glutamine levels in dorsal hippocampus of PDC-pretreated rats. Following pilocarpine administration the aspartate and glutamate levels were 31% and 18% higher (P < 0.05) than the corresponding values in rats receiving PDC alone (matching phases of groups 2 and 3 were compared). Glutamine values in PDC-pretreated rats were not affected by pilocarpine administration. (2) Anaesthetised animals The concentration of aspartate in the hippocampal dialysate during baseline recording was 0.44 + 0.06 pM, glutamate was 1.2 + 0.26/~M (n = 20), and glutamine was 10.9 + 1.0/~M. Perfusion with 10 mM pilocarpine resulted in significant increases in the dialysate levels of aspartate (143%) and glutamate (179%) in the first stimulated sample (Fig. 5). The aspartate and glutamate levels remained significantly elevated during the pilocarpine infusion and for the first 10 min of the recovery period. The aspartate and glutamate levels subsequently returned to basal levels. Dialysate levels of glutamine and other amino acids did not change significantly during pilocarpine perfusion. Perfusion with a calcium-free medium containing 2 mM EGTA did not significantly influence baseline levels of aspartate and glutamate, but prevented the pilocarpine-induced increase in the levels of glutamate and aspartate (Fig. 6, top). Instead, the level of glutamate showed a significant 25-35% reduction during and immediately fc,llowing the perfusion with pilocarpine in a calcium-free medium. There was a similar trend towards a reduction in the aspartate level under these conditions, but this failed to reach significance. Co-administra-
144 GLUTAMATE
I ASPARTATE
GLUTAMINE
soo i i ! i
4OO
| 600
i
.O
i
300!
IU (n
~
400 2C
U~
~200 .~
1©
0
,,,I,,,0,,,I
.....
,,I
Recovery Basal 10 mM Recovery Pllo Pllo i:ig. 5. Extracellular hippocampal levels of aspartate, glutamate and glutamine in anaesthetised rats before (solid bars), during (crosshatched bars) and after (solid bars) a 20-rain administration of pilocarpine (10 mM delivered via dialysate probe in dorsal hippocampus). Values are expressed as per cent of pooled pre-pilocarpine baseline samples (mean + SEM). The baseline dialysate levels were: aspartate, 0.44 + 0,06 l~M; glutamate, 1.20 + 0.26 l~M; glutamine, 10,9:1:1.0 I;M. Each bar represents a 10-rain (20-/d) sample, Asterisks denote values significantly different from corresponding pooled baseline values (* P <0.05; ** P <0.01; *** P <0.001). Basal 10 mM Recovery
Basal
10 mM
Pilo
GLUTAMATE
ASPARTATE ~-ff~
i ÷taw
;-~r- ........................
L--1
ii [
o[i~i:::it£:i~~ BASAL
PlkO
RtCOVERY
BASAl.
PlLO
RECOVERY
r ......................................................................
al
o
.,tl--~ BAIIAL
TTX
D.-91---II~
PlkO
RECOVERY
tion of I #M TTX with 10 mM pilocarpine also prevented the pilocarpine-induced increase in amino acid concentration. Glutamate levels dropped slightly (but significantly) below the baseline level after the end of perfusion with the drugs (Fig. 6, middle). Co-administration of 20 mM atropine together with pilocarpine blocked the enhanced release of aspartate and glutamate during perfusion with the drugs. However, 20 min after the end of pilocarpine administration a statistically significant increase of glutamate occurred (Fig. 6, bottom). Discu~ion
BASAL
A1TIOPlNE PlkO RECOVERY
IIASAL
ATROPINE PlLO RECOVERY
Fig. 6. Extracellular hippocampal levels of aspartate and glutamate in anaesthctised rats before (solid bars), during (crosshatched bars) and after (solid bars) a 20-rain administration of pilocarpine (10 mM via dialysate probe in dorsal hippocampus) delivered: (a) in a calcium-free Ringer solution containing 2 mM EGTA; (b) in normal Ringer solution containing i I~M TTX; (c) in normal Ringer solution containing 20 mM atropine. Values are expressed as per cent of pooled pre-pilocarpine baseline samples (mean + SEM). Each bar represents a 10-min (20-/d) sample. Asterisks denote values significantly different from corresponding pooled baseline values (* P <0.05; ** P <0.01; *** P <0.001).
This study shows that a systemic injection of a convulsant dose of pilocarpine tends to decrease the extracellular concentration of glutamate and aspartate in the hippocampus. In the presence of the glutamate uptake inhibitor, PDC, however, glutamate and aspartate concentrations rise significantly soon after the pilocarpine injection, prior to the development of seizures. This differs from our previous study with on-line enzymatic detection of glutamate 22 in which there was no change in extracellular glutamate prior to or during seizures in-
145 duced by systemic picrotoxin or focal bicuculline. In that study there continued to be no change in extracellular glutamate levels even in the presence of the glutamate uptake inhibitor, dihydrokainate, which produced a greater increase in baseline glutamate level than that found here ai'ter PDC. In order to interpret these differences and further 5ifferences with other microdialysis studies and also the effects of pilocarpine applied via the probe it is necessary to consider in some detail the mechanisms of action of pilocarpine and how these relate to seizure induction. The mechanisms by which pilocarpine induces seizures are not fully understood, but clearly involve changes in ionic conductances and second messenger systems. Four muscarinic receptor molecules have now been sequenced and cloned. M I and M3 induce hydrolysis of phosphoinositides and formation of inositol triphosphate and diacylglycerol, whereas M2 and M4 inhibit adenylate cyclase and decrease cAMP formation. M! and M2 receptors occur in the hippocampus with M! receptors being more dense in the CA~ subfield and dentate gyrus 31. Electrophysiological studies have defined at least four effects of muscarinic receptor activation on K ÷ conductances. These include blockade of the persistent K ÷ current known as Im, thereby producing a slow depolarisation 9 and blockade of the calcium-dependent slow K + current that is responsible for the after-hyperpolarisation that follows a burst discharge 5 and blockade of a transient outward K + current that controls excitability in hippocampal neurones 24. All these three effects occur in hippocampal neurones and tend to enhance excitability so that they could be contributing to epileptogenesis. They also, by decreasing the outward flux of K +, will tend to decrease its extracellular concentration. Uptake of glutamate is coupled to outward movement of K + down its concentration gl'adient; increasing the K ÷ gradient leads to a lower equilibrium extracellular concentration of glutamate and aspartate 25. Thus the lack of an increase in extracellular aspartate or glutamate concentration following pilocarpine and the appearance of an increase in the presence of PDC is consistent with enhancement of glutamate and aspartate uptake through pilocarpine's action on potassium conductances. Action at the
M2 receptor can increase a K + conductance that is responsible for a fast hyperpolarisation in thalamic neurones, which inhibits single spike activity but promotes burst discharges 2°. Some other effects of muscarinic activation are particularly linked to epileptogenesis. Acetylcholine acting on muscarinic receptors in the hippocampus produces a long-lasting facilitation of excitatory postsynaptic potentials that is dependent on enhanced responsiveness to NMDA (and not to quisqualate or kainate) 16'!7. This effect is not dependent on changes in K + conductance but is dependent on the second messenger function of inositol 1,4,5-triphosphate ~s. That this second messenger system is important in the convulsant effect of pilocarpine is suggested by the powerful proconvulsant effect of co-administration of lithium with pilocarpine. The presumed role of inhibition of inositol phosphatase activity by lithium and consequent depletion of the supply of myo-inositol is supported by experiments showing that the proconvulsant action of lithium in mice receiving pilocarpine is reversed by injection of myo-inositola2. The second messenger function of diacylglycerol is also involved; activation of protein kinase C leads to functional modification, by phosphorylation, of various receptors. In the case of the NMDA receptor (but not the AMPA or kainate receptor) protein kinase C activation potentiates responses to glutamate 4'8. Activation of protein kinase C also enhances glutamate uptake into glial cells2, providing another mechanism by which muscarinic receptor activation might decrease extracellular glutamate concentration. Although seizures induced by systemic kainate, bicuculline or picrotoxin do not produce significant increases in extracellular glutamate concentration in the hippocampus 13'22, seizures induced by the potent anticholinesterase, soman ~2'36, do. No change in glutamate levels occurred pre seizure but significant increases occurred during the first 20 min of soman-induced seizures in both CA~ (maximally + 135%) and CA3 (maximally -t-78%) 12. Measurement of [all]glutamate uptake into hippoeampal homogenates showed that uptake was enhanced 30-40 min after seizure onset. The soman study ~2 differs from ours in that there was a marked increase in extracellular taurine con-
146 centration, and a modest increase in glycine at 030 rain (when we saw no change) and a rise in glutamine at 0-10 rain (when we saw a fall). The apparent differences between the effects of pilocarpine and soman may relate to activation of nicotinic receptors, to a different balance between M! and M2 receptor activation, a different time course of receptor activation or may reflect only experimental differences such as probe placement. There is no correlation between the reported changes in extracellular glutamate concentration in the different seizure models and the occurrence of hippocampal pathology, it is likely that excitotoxic damage can occur without a general increase in extracellular glutamate concentration 21. The reduction we observe in glutamine levels during seizure activity requires explanation. Extracellular glutamine levels fall during depolarisation induced by high potassium or by ischaemia However, such depolarisation is associated with changes in other amino acid levels that were not observed here, and can therefore be excluded. A decreased rate of glutamine synthesis due to altered glial pH or ionic circumstances is more probable. The results obtained in anaesthetised animals show that pilocarpine delivered focally into the hippocampus via the microdialysis probe induces calcium-dependent and TTX-sensitive increases in the extracellular concentrations of glutamate and aspartate. The more dramatic nature of this change compared with the effects of systemic pilocarpine may simply reflect a greater local concentration of pilocarpine. Nevertheless it should aid interpretation of the immediate pharmacological interaction between the cholinergic and excitatory amino acid system within the hippocampus. Cholinergic afferents terminate on all pyramidal subfields and in the dentate gyrus 3'23. A presynaptic muscarinic receptor has been described that decreases the release of glutamate or aspartate 15'29"34"37. Direct perfusion of pilocarpine indicates that this cannot be having a predominant effect. Instead the predominant effect would appear to be the activating role of the cholinergic
system on hippocampal pyramidal neurones, which is described in many electrophysiological studies ~°'ll. There is a specific potentiation of NMDA receptor-mediated responses 16 and even a direct potentiation of glutamate toxicity in cultured neurones 19. As pyramidal neurones in CA! and CA3 are themselves glutamatergic their activation will lead to increased glutamate release. This interpretation is consistent with the observed calcium dependence and TTX sensitivity. We interpret our results as showing that a dose of pilocarpine sufficient to induce limbic seizures enhances uptake of glutamate and aspartate (into neurones and glia) and, when such uptake is blocked, increases the extracellular concentration of aspartate and glutamate prior to seizure onset, presumably by enhancing release. There is thus an interesting parallel between this study and three recent clinical reports of changes in extracellular aspartate and glutamate concentration associated with the onset of seizures in hippocampal or cortical foci 6'7'3°. Although these three studies employed different procedures (push-pull cannulae, acute microdialysis, and chronically implanted microdialysis probes), they all found that brief spontaneous or electrically induced seizures are associated with a very early small transient increase in the extracellular concentration of aspartate and, less consistently, of glutamate. This finding is similar to our finding of an increase in aspartate and glutamate preceding seizure onset and raises the possibility that the pathogenesis of spontaneous limbic seizures in man shares mechanisms (presumably involving the same ion channels and second messengers) with the pilocarpine-induced limbic seizures in the rat.
Acknowledgements We thank the Leverhulme Trust, the Medical Research Council and the Wellcome Trust for financial support.
147
References ! Bridges, R.J., Stanley, M.S., Anderson, M.W., Cotman, C.W. and Chamberlin, A.R., Conformationally defined neurotransmitter analogues. Selective inhibition of glutamate uptake by one pyrrolidine-2,4-dicarboxylate diastereomer, J. Med. Chem., 34 (1991) 717-725. 2 Casado, M., Zafra, F., Arag6n, C. and Gim6nez, C., Activation of high-affinity uptake of glutamate by phorbol esters in primary glial cell cultures, J. 3;eurochem., 57 (1991) 1185! 190. 3 Chandler, J.P. and Crutcher, K.A., The septohippocampal projection in the rat: an electron microscopic horseradish peroxidase study, Neuroscience, 10 (1983) 685-696. 4 Chen, L. and Huang, L.-Y.M., Protein kinase C reduces Mg 2+ block of NMDA-receptor channels as a mechanism of modulation, Nature, 356 (1992) 521--523. 5 Cole, A.E. and Nicoll, R.A., The pharmacology ofcholinergic excitatory responses in hippocampal pyramidal cells, Bra#a Res., 305 (1984) 283-290. 6 Do, K.Q., Klancnik, J., G/ihwiler, B., Perschak, H., Wieser, H.G. and Cuenod, M., Release of EAA: animal studies and epileptic foci studies in human. In: B.S. Meldrum, F. Moroni, R.P. Simon and J.H. Woods (Eds.), Excitatory Amino Acids, Raven Press, New York, NY, 1991, pp. 677-685. 7 During, M.J., Anderson, G.M., Roth, R.H. and Spencer, D.D., A human dialytrode: in vivo measurements of neuroactive substances in the human hippocampus with simultaneous depth EEG recordings, Soc. Neurosci. Abstr., 15 (1989) 340. 8 Gerber, G., Kangrga, I., Ryu, P.D., Larew, J.S.A. and Randic, M., Multiple effects of phorbol esters in the rat spinal dorsal horn, J. Neurosci., 9 (1989) 3606-3617. 9 Halliwell, J.V. and Adams, P.R., Voltage-clamp analysis of muscarinic excitation in hippocampal neurones, Brain Res., 250 (1982) 71-92. 10 Krnjevic, K. and Ropert, N., Electrophysiological and pharmacological characteristics of facilitation of hippocampai population spikes by stimulation of medial septum, Neuroscience, 7 (1982) 2165-2183. i i Krnjevic, K., Ropert, N. and Casullo, J., Septohippocampal disinhibition, Brain Res., 438 (1988) 182-192. 12 Lallement, G., Carpentier, P., Coilet, A., Pernot-Marino, !., Baubichon, D. and Blanchet, G., Effects of soman-induced seizures on different extracellular amino acid levels and on glutamate uptake in rat hippocampus, Brain Res., 563 (1991 ) 234-240. 13 Lehmann, A., Alterations in hippocampal extracellular amino acids and purine catabolites during limbic seizures induced by folate injections into the rabbit amygdala, Neuroscience, 22 (1987) 573-578. 14 Lehmann, A., Hagberg, H., Jacobson, I. and Hamberger, A., Effect of status epilepticus on extracellular amino acids in the hippocampus, Brain Res., 359 (1985) 147-151. 15 Marchi, M., Bocchieri, P., Garbarino, L. and Raiteri, M., Muscarinic inhibition of endogenous glutamate release from rat hippocampus synaptosomes, Neurosci. Lett., 96
(I 989) 229-234. 16 Markram, H. and Segal, M., Acetylcholine potentiates responses to N-methyl-o-aspartate in the rat hippocampus, Neurosci. Lett., ! 13 (1990) 62-65. 17 Markram, H. and Segal, M., Long-lasting facilitation of excitatory postsynaptic potentials in the rat hippocampus by acetylcholine, J. Physiol., 427 (1990) 381-393. 18 Markram, H. and Segal, M., The inositol 1,4,5-trisphosphate pathway mediates cholinergic potentiation of rat hippocampal neuronal responses to NMDA, J. Physiol., 447 (1992) 513-533. 19 Mattson, M.P., Acetylcholine potentiates glutamate-induced neurodegeneration in cultured hippocampal neurones, Brain Res., 497 (1989) 402-406. 20 McCormick, D.A. and Prince, D.A., Acetylcholine induces burst firing in thalamic reticular neurones by activating a potassium conductance, Nature, 319 (i 986) 402-405. 21 Meldrum, B.S., Excitotoxicity and epileptic brain damage, Epilepsy Res., 10 (1991) 55-61. 22 Millan, M.H., Obrenovitch, T.P., Sarna, G.S., Lok, S.Y., Symon, L. and Meldrum, B.S., Changes in rat brain extracellular glutamate concentration during seizures induced by systemic picrotoxin or focal bicuculline injection: An in vivo dialysis study with on-line enzymatic detection, Epilepsy Res., 9 (1991) 86-91. 23 Mosko, S., Lynch, G. and Cotman, C.W., The distribution of septal projections to the hippocampus of the rat, J. Comp. Neurol., 152 (1973) 163-174. 24 Nakajima, Y., Nakajima, S., Leonard, R.J. and Yamaguchi, K., Acetylcholine raises excitability by inhibiting the fast transient potassium current in cultured hippocampai neurons, Proc. Natl. Acad. Sci. USA, 83 (1986) 3022-3026. 25 Nicholis, D. and Attweil, D., The release and uptake of excitatory amino acids, Trends Pharmacoi. Sci., il (1990) 462--468. 26 Oiney, J.W., de Gubareff, T. and Labruyere, J., Seizure-related brain damage induced by cholinergic agents, Nature, 301 (1983) 520-522. 27 Ottersen, O.P. and Storm-Mathiesen, J., Excitatory amino acid pathways in the brain. In: R. Schwarz and Y. Ben-Ari (Eds.), Advance in Experimental Medicine and Biology, Plenum Press, New York, NY, 1986, pp. 263-284. 28 Patei, S., Chemically induced limbic seizures in rodents. In: B.S. Meldrum, J.A. Ferrendelli and H.G. Wieser (Eds.), Anatomy of Epileptogenesis. Current Problems in Epilepsy, Voi. 6, John Libbey, London, 1988, pp. 89-106. 29 Raiteri, M., Marchi, M., Costi, A. and Volpe, G., Endogenous aspartate release in the rat hippocampus is inhibited by M2 'cardiac' muscarinic receptors, Eur. J. Pharmacol., 177 (1990) 181-187. 30 Ronne-Engstr6m, E., Carlsson, H., Flink, R., Hillered, L., Spfinnare, B. and Ungerstedt, U., in vivo studies of the human epileptic focus using intracerebral microdialysis, J. Cereb. Blood Flow Metab., 11, Suppl. 2 ( ! 991 ) s656. 31 Spencer, D.G., Horvfith, E. and Traber, J., Direct autora-
148
32
33
34
35
diographic determination of M ! and M2 muscarinic acetylcholine receptor distribution in the rat brain: Relation to cholinergic nuclei and projections, Brain Res., 380 (1986) 59-68. Tricklebank, M.D., Singh, L., Jackson, A. and Oles, R.J., Evidence that a proconvulsant action of lithium is mediated by inhibition of myo-inositol phosphatase in mouse brain, Brain Res., 558 (1991) 145-148. Turski, W.A., Cavalheiro, E.A., Schwarz, M., Czuczwar, S.J., Kleinrok, Z. and Turski, L., Limbic seizures produced by pilocarpine in rats: bchavioural, electroencephalographic and neuropathological study, Behav. Brain Res., 9 (1983) 315-335. Valentine, R.J. and Dingledine, R., Presynaptic inhibitory effect of acetylcholine in the hippocampus, J. Neurosci., 7 (1981) 784-792. Vezzani, A., Ungerstedt, U., French, E.D. and Schwarcz, R., In rive brain dialysis of amino acids and simultaneous
36
37
38
39
EEG measurements following intrahippocampal quinolinic acid injection: Evidence for a dissociation between neurochemical cht, nges and seizures, J. Neurochem., 45 (1985) 335-344. Wade, J.V., Samson, F.E., Nelson, S.R. and Pazdernik, T.L., Changes in extracellular amino acids during somanand kainic acid-induced seizures, J. Neurochem., 49 (1987) 645-650. Williams, S. and Johnston, D., Muscarinic depression of synaptic transmission at the hippocampal mossy fiber synapse, J. Neurophysiol., 64 (1990) 1089-1097. Woolf, N.J., Eckenstein, F. and Butcher, L.L., Cholinergic systems in the rat brain: I. Projections to the limbic telencephalon, Brain Res. Bull., 13 (1984) 751-784. Young, A.M.J., Crowder, J.M. and Bradford, H.F., Potentiation by kainate of excitatory amino acid release in striatum: complementary in rive and in vitro experiments, J. Neurochem., 50 (1988) 337-345.
139
EPIRES 00548
Extracellular amino acid levels in hippocampus during pilocarpine-induced seizures
Maria H. Millan, Astrid G. Chapman and Brian S. Meldrum Departnlent of Neurology, bzstitute of Psychiatry, London SE5 8AF, UK (Received 23 June 1992; revision received 23 October 1992; accepted 2 November 1992)
Key words: Pilocarpine; Aspartate; Glutamate; Glutamate uptake inhibition; Hippocampus; Microdialysis; (Epilepsy)
Extracellular levels of aspartate, glutamate and glutamine were monitored by microdialysis in the dorsal hippocampus of freely moving rats following the administration of a convulsant dose of pilocarpine (400 mg/kg, i.p.). Rats were either pretreated with the glutamate uptake inhibitor, I.trans.pyrrolidine-2,4-dicarboxylic acid (PDC, 1 mM in the perfusion medium, - 2 5 min), or received pilocarpine directly. All rats injected with pilocarpine (with or without PDC pretreatment) developed limbic seizures (latency 15.4+2.4 rain). Without PDC pretreatment there were no significant changes in extracellular levels of aspartate, glutamate and glutamine following pilocarpine administration until the onset of iimbic seizures when glutamine levels fell by 35%. Following PDC pretreatment there were ,arge and sustained increases in extracellular hippocampal aspartate (250%) and glutamate (55%) levels, but no significant change in the glutamine level. When pilocarpine was administered to this group of rats, there were f,,'ther selective, significant, transient increases in the extraceilular levels of aspartate (31%) and glutamate (18%) which preceded the onset of seizures. Aspartate and glutamate levels were not significantly increased (relative to PDC controls) during seizures. The conditions for pilocarpine-induced increases in aspartate and glutamate release were established in parallel groups of anaesthetised rats where pilocarpine was administered via a microdialysis probe in the dorsal hippocampus. Following the infusion of 10 mM pilocarpine there were large and rapid increases in the levels of aspartate (143%) and glutamate (179%), which were completely abolished by the absence of calcium in the perfusion medium, or by the presence of atropine (20 mM) or tetrodotoxin (i ~M).
Introduction In rats the systemic injection of a high dose of cholinergic agents, such as the muscarinic agonist pilocarpine, results in the development of motor limbic seizures 26'33. Pilocarpine-induced seizures can be prevented by focal injection of excitatory amino acid antagonists into several brain areas involved in the development and transmission of iimbic seizures 28. These findings suggest that activaCorrespondence to: A.G. Chapman, Department of Neurology, Institute of Psychiatry, De Crespigny Park, Denmark Hill, London SE5 8AF, UK.
tion of muscarinic receptors results in the enhancement of excitatory processes responsible for the development of limbic seizures. The hippocampus plays a major role in the development and maintenance of limbic seizures. It receives a major cholinergic innervation from the septum and contains numerous excitatory links using glutamate as a transmitter 27'3s. The hippocampal formation is thus a suitable area for studies on glutamate release during pilocarpine-induced limbic seizures. Several studies have used in vivo microdialysis in an effort to quantify a presumed increase in glutamate release during epileptic seizures in animal
140 models. Most of them failed to show such a phenomenon ~3"t4'35. It has been suggested that prompt reuptake of glutamate could markedly limit the presumed increase in extracellular concentration. Our previous s! udy on anaesthetised rats with on-line glutamate enzymatic detection using dihydrokainate to block glutamate reuptake did not confirm this hypothesis22. In this study a new, potent and selective glutamate uptake inhibitor, I.trans-pyrrolidine-2,4-dicarboxylic acid (PDC) i, was applied via the dialysis probe in freely moving rats that were exposed to pilocarpine-induced limbic seizures. This permitted assessment of excitatory amino acid release prior to and following seizure onset. Additionally, we have characterised the direct effects of pilocarpine on excitatory amino acid release, by using anaesthetised rats in which pilocarpine and other drugs were delivered via a dialysis probe in the dorsal hippocampus. Methods
Male Sprague-Dawley rats (250 g) were used for the experiments.
Freely moving rats The animals were anaesthetised with pentobarbital, 60 mg/kg, i.p., and placed in a stereotaxic frame (Kopf Instruments). Dialysis probes were chronically implanted into the right dorsal hippocampus (A + 3.8, V + 5.0 from intraaurai line and L + 2.8). 24 h after implantation of the probes, the animals were placed in a plexiglas rodent cage and the probes were connected to a counter-balanced, freely rotating swivel (Carnegie Medicin). The dialysis probes were flushed with Ringer solution (NaCI 125 mM, KC! 3.3 mM, MgSO4 2.4 raM, KH2PO4 !.25 mM, CaCI2 1.85 mM, pH 7.2) prior to perfusion (3 Fi/min) using a slow infusion pump (Carnegie Medicin, CMAI00). A 90-rain wash-out period was allowed before the onset of the experiments. There were three experimental groups of freely moving animals: Group l, pilocarpine-induced seizures without the glutamate uptake inhibitor, PDC; Group 2, pilocarpine-induced seizures, following pretreatment with PDC via the dialysis probe; Group 3, administration of PDC alone.
In group 1 (n = 6) five 5-min samples (15/al each) were collected for the determination of baseline levels of amino acids. Scopolamine methyl nitrate (1 mg/kg, s.c.) was injected 15 min prior to pilocarpine hydrochloride (400 mg/kg, i.p.) in order to prevent peripheral effects of pilocarpine. Follow ing pilocarpine administration, 15-/d samples were continuously collected during the latency period (latency to seizures ranged from 10 to 35 min; mean: 15.4 + 2.4), and for at least 30 min following the development of limbic seizures, or until the death of the animal (pilocarpine-induced seizures were fatal in 1/13 rats 15 rain after onset of seizures). In group 2 (n = 7) the collection of five 5-min baseline samples was followed by the collection of an additional five 5-min samples in the presence of the glutamate reuptake inhibitor PDC (delivered via probe as 1 mM in Ringer solution) before the injection of scopolamine and pilocarpine. PDC was present in the perfusate for the remainder of the experiment, otherwise conditions were identical to those of group 1 above. In group 3 (n =6) the collection of five 5-rain baseline samples was followed by the administration of PDC (delivered via the probe as 1 mM in Ringer solution) for the remainder of the experiment. 5-min samples were collected continuously for a 60-90-min time period designed to match the time course of the experiment in group 2. All the surviving animals were killed with an overdose of pentobarbital and the brains were taken for histological examination using frozen sections. This showed that the dialysis probe membrane vertically crossed CA~, dentate gyrus and CA4 hippocampal cell layers.
Anaesthetised animals Animals were anaesthetised with chloral hydrate (600 mg/kg, i.p.) and placed in a stereotaxic apparatus (Kopf Instruments) and a microdialysis probe was inserted in the dorsal hippocampus and perfused with Ringer solution (2 Fl/min perfusion rate). Microdialysis sampling began after a 90-min wash-out period. Three baseline samples were collected for the next 30 min (20 F1 each) before drugs were applied via the probe for 20 min (2 x 20-FI samples). This was followed by a 30-min recovery
141 period when the hippocampal probe was again perfused with Ringer solution (3 × 20-pl samples). The following experimental groups were studied: 1, 10 mM pilocarpine (20 min application, n = 6); 2, 10 mM pilocarpine + 20 mM atropine (20 min application, n = 4); 3, 10 mM pilocarpine + 1/~M tetrodotoxin (TTX) (20 min application, n = 4); 4, 10 mM pilocarpine (20 min application) in animals perfused with Ca 2÷-free medium containing 2 mM EGTA throughout the experiment (n=4). All drugs were dissolved in Ringer solution (pH adjusted to 7.2) and delivered via the hippocampal probe for 20 min.
Microdialysis and biochemical analysis Microdialysis probes were assembled as required (see Young et al.39). The permeable tip of the probes, made of cuprophane fibre, was 2.0-2.5 mm long. The in vitro recovery of amino acids was tested by inserting the tip of the probe in 10 #M solutions of amino acids. With a perfusion rate of 2 #l/min (at room temperature) the mean recovery values were 14.0 + 6% for aspartate and 15.8 _ 1.6% for glutamate (n = 12). Dialysate samples (15/zl for freely moving rats, 20 #1 for anaesthetised rats) were analysed by HPLC after ortho-phthaldialdehyde derivatisation of amino acids. The amino acid derivatives were separated on a Spherisorb ODS2 5# column (25 c m x 4.6 cm) using a 10-55% methanol gradient in 0.1 M sodium acetate, 2.5% tetrahydrofuran buffer, pH 5.75. The HPLC system consisted of a Spectraphysics SP 8800 ternary HPLC pump, a Kratos FS 950 Fluorimeter and a Spectraphysics Chromjet Integrator. Levels of the following amino acids were estimated: aspartate, glutamate, glutamine, gly.:ine, serine, alanine and taurine. GABA !evels c~t.,id not be reliably quantified with the ~iPLC conditions used. Statistical analysis was performed using Student's t-test.
Results
(1) Freely moving animals (A ) Behavioural results Intraperitoneal injection of pilocarpine resulted in the development of motor limbic seizures in all animals. Within 3 min post injection rats displayed gnawing followed by tremor of increasing intensity, ear twitching and head nodding. After 10-35 min (mean latency period 15.4 + 2.4 min) the animals displayed initially whole body clonus with loss of posture often with barrel rotation and elements of tonus, or full whole body tonus. The majority of animals (12/13) recovered from this initial seizure and after 5-10 min displayed further seizures with rearing, forelimb clonus and falling. This kind of behaviour was repeated with increased frequency and eventually became continuous (status epilepticus). Perfusion with PDC neither changed the animal's behaviour in the non-seizing state nor altered the severity of pilocarpine-induced seizures.
(B) i',iochemical results Only aspartate, glutamate and glutamine levels showed any significant changes in the hippocampal dialysate under the experimental conditions used in this study. The dialysate levels of other amino acids (extracellular hippocampal baseline levels given in parentheses): serine (2.6 + 0.4 #M), glycine (5.2 + 0.5 #M), alanine (5.1 + 0.7 #M) and taurine (3.9 + 0.5 gM) were also monitored, but are not discussed further because of their lack of significant response to either PDC, pilocarpine administration or seizures.
Group 1, pilocarpine-induced seizures without glutamate uptake inhibitor, PDC. Amino acid levels are expressed as the means of the pooled 5-min fractions from three time periods: (a) the 25-min baseline period, (b) after pilocarpine injection, but before onset of behavioural seizures (10-35 min) and (c) during seizures and status epilepticus (1545 min) (see Fig. 1). Concentration of aspartate in the dialysate during baseline recording was 0.20 + 0.08/zM, glutamate was 0.96 + 0.26/tM and glutamine was 12.9 -/- 1.3/zM (n = 6).
142
~a.
GLUTAMATE
AsPARTATE
¢Jr 0 . 3 Z
15 i_GLUTAMINE
1.2
0 (3
i..... i
tu
W >,
101
0.2~
0.81
There were no significant changes in the mean levels of aspartate and glutamate following the administration of pilocarpine or the onset of seizures (Fig. I). The level of glutamine was significantly decreased (by 35%, P <0.05) during seizures (Fig. 1).
I .J
~ 0.1~
1U
0.41
" 0
............
o!:. ..........................
.. .......
04 812 04 812 012 FRACTION NUMBER (5 rain or 15 pl each)
04
BASELINE (25 rain) !:!:! FOLLOWING PILOCARPINE,400 mglkg, I.p. (10.35 rain) FOLLOWING OffSET OF LIMBIC SEIZURE (15.45 rain)
Fig. I. Extracellular levels of aspartate, glutamate ~.ild glutamine in dorsal hippocampus of freely moving rats (a) during baseline conditions, (b) following pilocarpine administration (400 mg/kg, i.p.) and (c) following onset of limbic seizures. Cons¢cutive 5-rain samples ( 15 ld each) were pooled for each experimental phase, and the values are expressed as l~M in the hippocampal dialysate (mean + SEM). The asterisks denote values significantly different from the corresponding baseline control value (* P < 0.05).
:
d z
|
~
!
!
8 07si 10
~ 0.50'
o.2si _. =c
0~ 0 4 81216 0 4 81216 0 4 81216 FRACTION NUMBER (5 rain or 15 pl each) BASELINE (2S rain) Poe BASELINE (25 rain) i;i FOLLOWING PILOCARPINE,400 mg/k9, i.p. (10-35 rain) [ ] FOLLOWING ONSET OF LIMBIC SEIZURE (15-45 rain)
Fig. 2. Extracellular levels of aspartate, glutamate and glutamine in dorsal hippocampus of freely moving rats (a) during baseline conditions, (b) following perl~sion with i mM PDC (present in the perfusate lbr the remainder of the experiment), (c) following pilocarpine administration (400 mg/kg, i,p.) and (d) following onset of limbic seizures, Consecutive 5-min samples (15 141each) were pooled for each experimental phase, and the values are expressed as t~M in the hippocampal dialysate (mean + SEM). The asterisks denote values significantly different from the corresponding baseline control value (** P < 0.01; *** P <0.001).
Group 2, pilocarpine-induced seizures following pretreatment with PDC. Amino acid levels are expressed as the means of the pooled 5-min fractions from four time periods: (a) the 25-rain baseline period, (b) a 25-min period following PDC administration, (c) after pilocarpine injection, but before onset of behavioural seizures (1035 min) and (d) during seizures and status epilepticus ( 15-45 rain). Baseline concentrations of amino acids were similar to those of group I above except that glutamate levels were higher (aspartate, 0.21 + 0.02 /zM; glutamate, 2.16 + 0.33 #M; glutamine, 12.3 __. 1.0/aM; n = 7). Perfusion with ! m M PDC resulted in significant increases of extraceilular aspartate (by 233%, P <0.01) and glutamate (by 54%, P <0.01). Administration of pilocarpine resulted in further increases in aspartate and glutamate, but not in glutamine levels (Fig. 2), whereas following the onset of seizures (initially intermittent and subsequently sustained) the levels of aspartate, glutamate and glutamine apparently decreased compared with directly prior to seizures (not statistically significant). Mean values of aspartate and glutamate appeared increased relative to corresponding values after PDC (see Figs. 3 and 4) but these differences were not statistically significant. Group 3, administration of PDC alone. Amino acid levels are expressed as the means of the pooled 5-min fractions from four time periods, designed to correspond to the experimental time course in group 2 above: (a) the 25-min baseline period, (b-d) 25-rain + 15-min + 25-min periods following PDC administration (1 mM via dialysate probe). Baseline concentrations of amino acids were similar to those of group 2 above: aspartate, 0.22 + 0.02 #M, glutamate was 2.12 + 0.13/~M and glutamine was 12.8 + 0.7/~M (n = 6).
143
I F
=.
. . . .
i
~SP-ART~E i
:
'
!41
~0.75 o
rt ,.I
O. =C
15 ~
:, | | --
. . . .
!
.2_
10
0,50
[
)
~::,
0.25
:S 0 n.
GLUTAMINE
GLUTAMATE
!
31
rj uJ
~
:
~i~ ~i~i, 0
0 '" 0
12 16
,
4~ 8
12 16
. . . . . . . . .
0
4A 8
t . ,
•
• 1..
12 16
i; Po-6~
i ( 1 mM)._~
L(!mM) .
FRACTION NUMBER (5 mln or 15 pl each)
Fig. 3. Extracellular levels of aspartate, glutamate and glutamine in dorsal hippocampus of freely moving rats (a) during baseline conditions, (b--d) following perfusion with I mM PDC alone for 60:90 rain. Consecutive 5-rain samples (15 td each) were pooled for the baseline period and for three consecutive time periods designed to correspond to the experimental phases shown in Fig, 2. The values are expressed as pM in the hippocampal dialysate (mean + SEM). The asterisks denote values significantly different from the corresponding baseline control value (*** P <0.001).
Levels of aspartate and glutamate increased initially by 250% and 55% respectively (P <0.001),
ILl
+40
•k
~ : i + Pilocarpine
z -1
Seizures
ul
|
u Q
I
+20
a_
:Z 0 U,I
(9
ziz o
ASP
GLU
GLN
-20
Fig. 4. Calculated differences in aspartate, glutamate and glutamine levels between PDC-treated, freely moving rats with and without pilocarpine-induced seizures. The calculations are based on the data shown in Fig. 2 and Fig. 3. The cross-hatched bars represent the excess extracellular amino acid levels following pilocarpine administration, and the solid bars represent the excess extracellular amino acid levels following the onset of seizures. The values are expressed as per cent changes in the pilocarpine-treated group compared to corresponding values in the PDC only groups (* P <0.05).
closely resembling the PDC-induced changes in group 2. The PDC-enhanced extracellular glutamate level remained relatively stable throughout the experimental period, while the PDC-enhanced extracellular aspartate level showed a tendency towards a gradual decrease (statistically not significant) during consecutive phases of the experiment. Glutamine concentration in dialysate was not significantly changed by the addition of PDC (Fig. 3). Fig. 4 shows the pilocarpine-induced effect on extracellular aspartate, glutamate and glutamine levels in dorsal hippocampus of PDC-pretreated rats. Following pilocarpine administration the aspartate and glutamate levels were 31% and 18% higher (P < 0.05) than the corresponding values in rats receiving PDC alone (matching phases of groups 2 and 3 were compared). Glutamine values in PDC-pretreated rats were not affected by pilocarpine administration. (2) Anaesthetised animals The concentration of aspartate in the hippocampal dialysate during baseline recording was 0.44 + 0.06 pM, glutamate was 1.2 + 0.26/~M (n = 20), and glutamine was 10.9 + 1.0/~M. Perfusion with 10 mM pilocarpine resulted in significant increases in the dialysate levels of aspartate (143%) and glutamate (179%) in the first stimulated sample (Fig. 5). The aspartate and glutamate levels remained significantly elevated during the pilocarpine infusion and for the first 10 min of the recovery period. The aspartate and glutamate levels subsequently returned to basal levels. Dialysate levels of glutamine and other amino acids did not change significantly during pilocarpine perfusion. Perfusion with a calcium-free medium containing 2 mM EGTA did not significantly influence baseline levels of aspartate and glutamate, but prevented the pilocarpine-induced increase in the levels of glutamate and aspartate (Fig. 6, top). Instead, the level of glutamate showed a significant 25-35% reduction during and immediately fc,llowing the perfusion with pilocarpine in a calcium-free medium. There was a similar trend towards a reduction in the aspartate level under these conditions, but this failed to reach significance. Co-administra-
144 GLUTAMATE
I ASPARTATE
GLUTAMINE
soo i i ! i
4OO
| 600
i
.O
i
300!
IU (n
~
400 2C
U~
~200 .~
1©
0
,,,I,,,0,,,I
.....
,,I
Recovery Basal 10 mM Recovery Pllo Pllo i:ig. 5. Extracellular hippocampal levels of aspartate, glutamate and glutamine in anaesthetised rats before (solid bars), during (crosshatched bars) and after (solid bars) a 20-rain administration of pilocarpine (10 mM delivered via dialysate probe in dorsal hippocampus). Values are expressed as per cent of pooled pre-pilocarpine baseline samples (mean + SEM). The baseline dialysate levels were: aspartate, 0.44 + 0,06 l~M; glutamate, 1.20 + 0.26 l~M; glutamine, 10,9:1:1.0 I;M. Each bar represents a 10-rain (20-/d) sample, Asterisks denote values significantly different from corresponding pooled baseline values (* P <0.05; ** P <0.01; *** P <0.001). Basal 10 mM Recovery
Basal
10 mM
Pilo
GLUTAMATE
ASPARTATE ~-ff~
i ÷taw
;-~r- ........................
L--1
ii [
o[i~i:::it£:i~~ BASAL
PlkO
RtCOVERY
BASAl.
PlLO
RECOVERY
r ......................................................................
al
o
.,tl--~ BAIIAL
TTX
D.-91---II~
PlkO
RECOVERY
tion of I #M TTX with 10 mM pilocarpine also prevented the pilocarpine-induced increase in amino acid concentration. Glutamate levels dropped slightly (but significantly) below the baseline level after the end of perfusion with the drugs (Fig. 6, middle). Co-administration of 20 mM atropine together with pilocarpine blocked the enhanced release of aspartate and glutamate during perfusion with the drugs. However, 20 min after the end of pilocarpine administration a statistically significant increase of glutamate occurred (Fig. 6, bottom). Discu~ion
BASAL
A1TIOPlNE PlkO RECOVERY
IIASAL
ATROPINE PlLO RECOVERY
Fig. 6. Extracellular hippocampal levels of aspartate and glutamate in anaesthctised rats before (solid bars), during (crosshatched bars) and after (solid bars) a 20-rain administration of pilocarpine (10 mM via dialysate probe in dorsal hippocampus) delivered: (a) in a calcium-free Ringer solution containing 2 mM EGTA; (b) in normal Ringer solution containing i I~M TTX; (c) in normal Ringer solution containing 20 mM atropine. Values are expressed as per cent of pooled pre-pilocarpine baseline samples (mean + SEM). Each bar represents a 10-min (20-/d) sample. Asterisks denote values significantly different from corresponding pooled baseline values (* P <0.05; ** P <0.01; *** P <0.001).
This study shows that a systemic injection of a convulsant dose of pilocarpine tends to decrease the extracellular concentration of glutamate and aspartate in the hippocampus. In the presence of the glutamate uptake inhibitor, PDC, however, glutamate and aspartate concentrations rise significantly soon after the pilocarpine injection, prior to the development of seizures. This differs from our previous study with on-line enzymatic detection of glutamate 22 in which there was no change in extracellular glutamate prior to or during seizures in-
145 duced by systemic picrotoxin or focal bicuculline. In that study there continued to be no change in extracellular glutamate levels even in the presence of the glutamate uptake inhibitor, dihydrokainate, which produced a greater increase in baseline glutamate level than that found here ai'ter PDC. In order to interpret these differences and further 5ifferences with other microdialysis studies and also the effects of pilocarpine applied via the probe it is necessary to consider in some detail the mechanisms of action of pilocarpine and how these relate to seizure induction. The mechanisms by which pilocarpine induces seizures are not fully understood, but clearly involve changes in ionic conductances and second messenger systems. Four muscarinic receptor molecules have now been sequenced and cloned. M I and M3 induce hydrolysis of phosphoinositides and formation of inositol triphosphate and diacylglycerol, whereas M2 and M4 inhibit adenylate cyclase and decrease cAMP formation. M! and M2 receptors occur in the hippocampus with M! receptors being more dense in the CA~ subfield and dentate gyrus 31. Electrophysiological studies have defined at least four effects of muscarinic receptor activation on K ÷ conductances. These include blockade of the persistent K ÷ current known as Im, thereby producing a slow depolarisation 9 and blockade of the calcium-dependent slow K + current that is responsible for the after-hyperpolarisation that follows a burst discharge 5 and blockade of a transient outward K + current that controls excitability in hippocampal neurones 24. All these three effects occur in hippocampal neurones and tend to enhance excitability so that they could be contributing to epileptogenesis. They also, by decreasing the outward flux of K +, will tend to decrease its extracellular concentration. Uptake of glutamate is coupled to outward movement of K + down its concentration gl'adient; increasing the K ÷ gradient leads to a lower equilibrium extracellular concentration of glutamate and aspartate 25. Thus the lack of an increase in extracellular aspartate or glutamate concentration following pilocarpine and the appearance of an increase in the presence of PDC is consistent with enhancement of glutamate and aspartate uptake through pilocarpine's action on potassium conductances. Action at the
M2 receptor can increase a K + conductance that is responsible for a fast hyperpolarisation in thalamic neurones, which inhibits single spike activity but promotes burst discharges 2°. Some other effects of muscarinic activation are particularly linked to epileptogenesis. Acetylcholine acting on muscarinic receptors in the hippocampus produces a long-lasting facilitation of excitatory postsynaptic potentials that is dependent on enhanced responsiveness to NMDA (and not to quisqualate or kainate) 16'!7. This effect is not dependent on changes in K + conductance but is dependent on the second messenger function of inositol 1,4,5-triphosphate ~s. That this second messenger system is important in the convulsant effect of pilocarpine is suggested by the powerful proconvulsant effect of co-administration of lithium with pilocarpine. The presumed role of inhibition of inositol phosphatase activity by lithium and consequent depletion of the supply of myo-inositol is supported by experiments showing that the proconvulsant action of lithium in mice receiving pilocarpine is reversed by injection of myo-inositola2. The second messenger function of diacylglycerol is also involved; activation of protein kinase C leads to functional modification, by phosphorylation, of various receptors. In the case of the NMDA receptor (but not the AMPA or kainate receptor) protein kinase C activation potentiates responses to glutamate 4'8. Activation of protein kinase C also enhances glutamate uptake into glial cells2, providing another mechanism by which muscarinic receptor activation might decrease extracellular glutamate concentration. Although seizures induced by systemic kainate, bicuculline or picrotoxin do not produce significant increases in extracellular glutamate concentration in the hippocampus 13'22, seizures induced by the potent anticholinesterase, soman ~2'36, do. No change in glutamate levels occurred pre seizure but significant increases occurred during the first 20 min of soman-induced seizures in both CA~ (maximally + 135%) and CA3 (maximally -t-78%) 12. Measurement of [all]glutamate uptake into hippoeampal homogenates showed that uptake was enhanced 30-40 min after seizure onset. The soman study ~2 differs from ours in that there was a marked increase in extracellular taurine con-
146 centration, and a modest increase in glycine at 030 rain (when we saw no change) and a rise in glutamine at 0-10 rain (when we saw a fall). The apparent differences between the effects of pilocarpine and soman may relate to activation of nicotinic receptors, to a different balance between M! and M2 receptor activation, a different time course of receptor activation or may reflect only experimental differences such as probe placement. There is no correlation between the reported changes in extracellular glutamate concentration in the different seizure models and the occurrence of hippocampal pathology, it is likely that excitotoxic damage can occur without a general increase in extracellular glutamate concentration 21. The reduction we observe in glutamine levels during seizure activity requires explanation. Extracellular glutamine levels fall during depolarisation induced by high potassium or by ischaemia However, such depolarisation is associated with changes in other amino acid levels that were not observed here, and can therefore be excluded. A decreased rate of glutamine synthesis due to altered glial pH or ionic circumstances is more probable. The results obtained in anaesthetised animals show that pilocarpine delivered focally into the hippocampus via the microdialysis probe induces calcium-dependent and TTX-sensitive increases in the extracellular concentrations of glutamate and aspartate. The more dramatic nature of this change compared with the effects of systemic pilocarpine may simply reflect a greater local concentration of pilocarpine. Nevertheless it should aid interpretation of the immediate pharmacological interaction between the cholinergic and excitatory amino acid system within the hippocampus. Cholinergic afferents terminate on all pyramidal subfields and in the dentate gyrus 3'23. A presynaptic muscarinic receptor has been described that decreases the release of glutamate or aspartate 15'29"34"37. Direct perfusion of pilocarpine indicates that this cannot be having a predominant effect. Instead the predominant effect would appear to be the activating role of the cholinergic
system on hippocampal pyramidal neurones, which is described in many electrophysiological studies ~°'ll. There is a specific potentiation of NMDA receptor-mediated responses 16 and even a direct potentiation of glutamate toxicity in cultured neurones 19. As pyramidal neurones in CA! and CA3 are themselves glutamatergic their activation will lead to increased glutamate release. This interpretation is consistent with the observed calcium dependence and TTX sensitivity. We interpret our results as showing that a dose of pilocarpine sufficient to induce limbic seizures enhances uptake of glutamate and aspartate (into neurones and glia) and, when such uptake is blocked, increases the extracellular concentration of aspartate and glutamate prior to seizure onset, presumably by enhancing release. There is thus an interesting parallel between this study and three recent clinical reports of changes in extracellular aspartate and glutamate concentration associated with the onset of seizures in hippocampal or cortical foci 6'7'3°. Although these three studies employed different procedures (push-pull cannulae, acute microdialysis, and chronically implanted microdialysis probes), they all found that brief spontaneous or electrically induced seizures are associated with a very early small transient increase in the extracellular concentration of aspartate and, less consistently, of glutamate. This finding is similar to our finding of an increase in aspartate and glutamate preceding seizure onset and raises the possibility that the pathogenesis of spontaneous limbic seizures in man shares mechanisms (presumably involving the same ion channels and second messengers) with the pilocarpine-induced limbic seizures in the rat.
Acknowledgements We thank the Leverhulme Trust, the Medical Research Council and the Wellcome Trust for financial support.
147
References ! Bridges, R.J., Stanley, M.S., Anderson, M.W., Cotman, C.W. and Chamberlin, A.R., Conformationally defined neurotransmitter analogues. Selective inhibition of glutamate uptake by one pyrrolidine-2,4-dicarboxylate diastereomer, J. Med. Chem., 34 (1991) 717-725. 2 Casado, M., Zafra, F., Arag6n, C. and Gim6nez, C., Activation of high-affinity uptake of glutamate by phorbol esters in primary glial cell cultures, J. 3;eurochem., 57 (1991) 1185! 190. 3 Chandler, J.P. and Crutcher, K.A., The septohippocampal projection in the rat: an electron microscopic horseradish peroxidase study, Neuroscience, 10 (1983) 685-696. 4 Chen, L. and Huang, L.-Y.M., Protein kinase C reduces Mg 2+ block of NMDA-receptor channels as a mechanism of modulation, Nature, 356 (1992) 521--523. 5 Cole, A.E. and Nicoll, R.A., The pharmacology ofcholinergic excitatory responses in hippocampal pyramidal cells, Bra#a Res., 305 (1984) 283-290. 6 Do, K.Q., Klancnik, J., G/ihwiler, B., Perschak, H., Wieser, H.G. and Cuenod, M., Release of EAA: animal studies and epileptic foci studies in human. In: B.S. Meldrum, F. Moroni, R.P. Simon and J.H. Woods (Eds.), Excitatory Amino Acids, Raven Press, New York, NY, 1991, pp. 677-685. 7 During, M.J., Anderson, G.M., Roth, R.H. and Spencer, D.D., A human dialytrode: in vivo measurements of neuroactive substances in the human hippocampus with simultaneous depth EEG recordings, Soc. Neurosci. Abstr., 15 (1989) 340. 8 Gerber, G., Kangrga, I., Ryu, P.D., Larew, J.S.A. and Randic, M., Multiple effects of phorbol esters in the rat spinal dorsal horn, J. Neurosci., 9 (1989) 3606-3617. 9 Halliwell, J.V. and Adams, P.R., Voltage-clamp analysis of muscarinic excitation in hippocampal neurones, Brain Res., 250 (1982) 71-92. 10 Krnjevic, K. and Ropert, N., Electrophysiological and pharmacological characteristics of facilitation of hippocampai population spikes by stimulation of medial septum, Neuroscience, 7 (1982) 2165-2183. i i Krnjevic, K., Ropert, N. and Casullo, J., Septohippocampal disinhibition, Brain Res., 438 (1988) 182-192. 12 Lallement, G., Carpentier, P., Coilet, A., Pernot-Marino, !., Baubichon, D. and Blanchet, G., Effects of soman-induced seizures on different extracellular amino acid levels and on glutamate uptake in rat hippocampus, Brain Res., 563 (1991 ) 234-240. 13 Lehmann, A., Alterations in hippocampal extracellular amino acids and purine catabolites during limbic seizures induced by folate injections into the rabbit amygdala, Neuroscience, 22 (1987) 573-578. 14 Lehmann, A., Hagberg, H., Jacobson, I. and Hamberger, A., Effect of status epilepticus on extracellular amino acids in the hippocampus, Brain Res., 359 (1985) 147-151. 15 Marchi, M., Bocchieri, P., Garbarino, L. and Raiteri, M., Muscarinic inhibition of endogenous glutamate release from rat hippocampus synaptosomes, Neurosci. Lett., 96
(I 989) 229-234. 16 Markram, H. and Segal, M., Acetylcholine potentiates responses to N-methyl-o-aspartate in the rat hippocampus, Neurosci. Lett., ! 13 (1990) 62-65. 17 Markram, H. and Segal, M., Long-lasting facilitation of excitatory postsynaptic potentials in the rat hippocampus by acetylcholine, J. Physiol., 427 (1990) 381-393. 18 Markram, H. and Segal, M., The inositol 1,4,5-trisphosphate pathway mediates cholinergic potentiation of rat hippocampal neuronal responses to NMDA, J. Physiol., 447 (1992) 513-533. 19 Mattson, M.P., Acetylcholine potentiates glutamate-induced neurodegeneration in cultured hippocampal neurones, Brain Res., 497 (1989) 402-406. 20 McCormick, D.A. and Prince, D.A., Acetylcholine induces burst firing in thalamic reticular neurones by activating a potassium conductance, Nature, 319 (i 986) 402-405. 21 Meldrum, B.S., Excitotoxicity and epileptic brain damage, Epilepsy Res., 10 (1991) 55-61. 22 Millan, M.H., Obrenovitch, T.P., Sarna, G.S., Lok, S.Y., Symon, L. and Meldrum, B.S., Changes in rat brain extracellular glutamate concentration during seizures induced by systemic picrotoxin or focal bicuculline injection: An in vivo dialysis study with on-line enzymatic detection, Epilepsy Res., 9 (1991) 86-91. 23 Mosko, S., Lynch, G. and Cotman, C.W., The distribution of septal projections to the hippocampus of the rat, J. Comp. Neurol., 152 (1973) 163-174. 24 Nakajima, Y., Nakajima, S., Leonard, R.J. and Yamaguchi, K., Acetylcholine raises excitability by inhibiting the fast transient potassium current in cultured hippocampai neurons, Proc. Natl. Acad. Sci. USA, 83 (1986) 3022-3026. 25 Nicholis, D. and Attweil, D., The release and uptake of excitatory amino acids, Trends Pharmacoi. Sci., il (1990) 462--468. 26 Oiney, J.W., de Gubareff, T. and Labruyere, J., Seizure-related brain damage induced by cholinergic agents, Nature, 301 (1983) 520-522. 27 Ottersen, O.P. and Storm-Mathiesen, J., Excitatory amino acid pathways in the brain. In: R. Schwarz and Y. Ben-Ari (Eds.), Advance in Experimental Medicine and Biology, Plenum Press, New York, NY, 1986, pp. 263-284. 28 Patei, S., Chemically induced limbic seizures in rodents. In: B.S. Meldrum, J.A. Ferrendelli and H.G. Wieser (Eds.), Anatomy of Epileptogenesis. Current Problems in Epilepsy, Voi. 6, John Libbey, London, 1988, pp. 89-106. 29 Raiteri, M., Marchi, M., Costi, A. and Volpe, G., Endogenous aspartate release in the rat hippocampus is inhibited by M2 'cardiac' muscarinic receptors, Eur. J. Pharmacol., 177 (1990) 181-187. 30 Ronne-Engstr6m, E., Carlsson, H., Flink, R., Hillered, L., Spfinnare, B. and Ungerstedt, U., in vivo studies of the human epileptic focus using intracerebral microdialysis, J. Cereb. Blood Flow Metab., 11, Suppl. 2 ( ! 991 ) s656. 31 Spencer, D.G., Horvfith, E. and Traber, J., Direct autora-
148
32
33
34
35
diographic determination of M ! and M2 muscarinic acetylcholine receptor distribution in the rat brain: Relation to cholinergic nuclei and projections, Brain Res., 380 (1986) 59-68. Tricklebank, M.D., Singh, L., Jackson, A. and Oles, R.J., Evidence that a proconvulsant action of lithium is mediated by inhibition of myo-inositol phosphatase in mouse brain, Brain Res., 558 (1991) 145-148. Turski, W.A., Cavalheiro, E.A., Schwarz, M., Czuczwar, S.J., Kleinrok, Z. and Turski, L., Limbic seizures produced by pilocarpine in rats: bchavioural, electroencephalographic and neuropathological study, Behav. Brain Res., 9 (1983) 315-335. Valentine, R.J. and Dingledine, R., Presynaptic inhibitory effect of acetylcholine in the hippocampus, J. Neurosci., 7 (1981) 784-792. Vezzani, A., Ungerstedt, U., French, E.D. and Schwarcz, R., In rive brain dialysis of amino acids and simultaneous
36
37
38
39
EEG measurements following intrahippocampal quinolinic acid injection: Evidence for a dissociation between neurochemical cht, nges and seizures, J. Neurochem., 45 (1985) 335-344. Wade, J.V., Samson, F.E., Nelson, S.R. and Pazdernik, T.L., Changes in extracellular amino acids during somanand kainic acid-induced seizures, J. Neurochem., 49 (1987) 645-650. Williams, S. and Johnston, D., Muscarinic depression of synaptic transmission at the hippocampal mossy fiber synapse, J. Neurophysiol., 64 (1990) 1089-1097. Woolf, N.J., Eckenstein, F. and Butcher, L.L., Cholinergic systems in the rat brain: I. Projections to the limbic telencephalon, Brain Res. Bull., 13 (1984) 751-784. Young, A.M.J., Crowder, J.M. and Bradford, H.F., Potentiation by kainate of excitatory amino acid release in striatum: complementary in rive and in vitro experiments, J. Neurochem., 50 (1988) 337-345.