Neuroscience Letters, 138 (1992) 141 144 © 1992 Elsevier Scientific Publishers Ireland Ltd. All rights reserved 0304-3940/92/$ 05.00
141
NSL 08544
The immediate consequences of middle cerebral artery occlusion on G A B A synthesis in mouse cortex and cerebellum A.R. Green, A.J. Cross, M.F. Snape and R.J. De Souza Astra Neuroscience Research Unit, London (UK) (Received 14 October 1991; Revised version received 12 January 1992; Accepted 21 January 1992)
Key words:
7-Aminobutyric acid synthesis; Ischaemia; Middle cerebral artery occlusion
The effect on y-aminobutyric acid (GABA) synthesis of focal ischaemia in the right cortex of the mouse was investigated by performing a right middle cerebral artery (MCA) occlusion. Synthesis of GABA was determined by measurement of the rate of GABA accumulation in tissue following injection of amino oxyacetic acid (AOAA; 30 mg/kg, i.p.). Five min following the MCA occlusion, the rate of GABA synthesis in the right (ischaemic) cortex was decreased by approximately 70% compared to either the left cortex or the right cortex of untreated controls. The basal GABA concentration was however unaffected. Four hours after the occlusion the rate of GABA synthesis was similar in the right and left cortex. The rate of GABA accumulation in the cerebellum was unchanged at both times after the right MCA occlusion compared with untreated control mice. The data suggest that there is a rapid but short lasting decrease in GABA synthesis following an ischaemic insult and it is suggested that this might be associated with the EEG spiking activity that occurs at this time.
There is now substantial evidence that competitive and non-competitive NMDA antagonists are neuroprotective in various models of stroke and cerebral ischaemic damage. Dizocilpine (MK-801), for example, prevents the hippocampal damage that normally occurs after a brief occlusion of the cerebral arteries of the gerbil [9, 10]. Recently, we have demonstrated that chlormethiazole is also neuroprotective in this model, even when the drug has been given 3 h after the period of occlusion [4]. There is, however, no evidence for chlormethiazole binding to any site on the N-methyl-o-aspartate (NMDA) receptor complex [5]. What has been established is that chlormethiazole can potentiate the action of GABA at the GABAA receptor and modulate channel opening [7, 15, 19, 20, 25]. This raises the possibility that chlormethiazole may act by enhancing GABA function and that this enhancement results in decreased NMDA function, an hypothesis which is given credence by the observation that chlormethiazole is an effective inhibitor of the convulsive behaviour induced by injection of N-methyl-DLaspartate (NMDLA) [5]. Following a global ischaemic episode 'spiking' occurs in the EEG, suggesting convulsive-like changes in the Correspondence: A.R. Green, Astra Neuroscience Research Unit, 1 Wakefield Street, London WCIN IPJ, UK.
brain [3]. Since previous studies have found that GABA synthesis in the brain is markedly inhibited after a convulsion elicited either chemically or electrically [13], it seemed possible that cerebral ischaemia might also result in a decrease in GABA synthesis. The current study has investigated this possibility. The model used to study the consequences of ischaemia on GABA synthesis was that of middle cerebral artery (MCA) occlusion in the mouse. This focal model was chosen both for its relative simplicity and also because it induces ischaemic damage on one side of the cortex whilst sparing the other. The study could therefore be conducted by investigating the intact and damaged sides of the cortex in the same animal. In addition, the consequences of the unilateral right MCA occlusion on GABA synthesis in the cerebellum, a region remote from the lesion, could also be examined. Male C57 B 1 / 6 0 l a mice (20-25 g) obtained from Harlan Olac (Bicester, UK) were used throughout the study. They were housed in groups of 6 in a 12 h light/12 h dark cycle (lights on 07,00 h) with food and water freely available. Middle cerebral artery occlusion was performed by anaesthetising the animals with 4% halothane in O2/N20 (ratio 2:1). The distal portion of the right middle cerebral artery was permanently occluded as follows. The animals underwent a right subtemporal craniotomy after partial
142 TABLE I BINDING OF PK 11195 IN MOUSE RIGHT CEREBRAL CORTEX 7 DAYS AFTER A MIDDLE CEREBRAL ARTERY OCCLUSION Results expressed as mean _+S.E.M.
Sham Ischaemic
PK 11195 Binding (fmol/mg protein)
n
% increase in binding
164_+15 379_+58
8 8
231 _+19
removal o f the temporalis muscle. The exposed M C A was electrocauterised and severed. The w o u n d was then sutured and anaesthesia discontinued. The animals were conscious and displaying l o c o m o t o r activity within another 2 min. The procedure used resulted in the formation o f an ischaemic infarct extending across the posterior frontal cortex and anterior parietal cortex, in the territory o f the middle, cerebral artery. Infarcts when visualised using ex vivo tetrazolium salt histochemistry were approximately 3 m m in diameter. The extent o f variability in size o f the infarct produced was assessed by the use o f P K 11 195 binding [8] following dissection o f the left and right cerebral cortex. Results obtained using this assay were similar to those previously reported [1] and indicate the reproducibility o f the operative procedure (Table I). G A B A synthesis was determined by measurement o f the rate o f accumulation o f G A B A following administration o f the G A B A transaminase inhibitor amino-oxyacetic acid ( A O A A ) , a m e t h o d originally described by Walters et al. [19] and recently validated [13]. Accumula-
tion o f G A B A after A O A A has been shown to be linear for at least 60 min [13]. A single measure o f G A B A concentration 40 min after A O A A is therefore justified, and has been used previously [7, 13] to calculate the rate o f G A B A synthesis. Results using this method [17] have been essentially identical with those obtained by other methods including mass f r a g m e n t o g r a p h y [16]. Animals were injected with A O A A (30 mg/kg, i.p.) either 5 rain or 4 h after the completion o f the surgical procedure. F o r t y rain after A O A A injection the mice were killed by exposure o f the head to a focussed beam o f microwave irradiation o f power density 0.7 kW for 1.1 s. The left and right cerebral cortex and cerebellum were dissected, weighed and stored at - 2 0 ° C until analysis. Tissue was homogenized in 0.5 ml o f perchloric acid (0,1 M), centrifuged for 3 rain and supernatant diluted 1:5 in borate buffer (pH 9.7). The supernatant was derivatised by reacting with o-phthalaldehyde in ethanol for H P L C analysis [21]. The reaction mixture (20ill) was injected onto the column (Spherisorb ODS-2, 25 cm, 5 j~m). G A B A was determined by the m e t h o d o f Lindroth and M o p p e r [23] with minor modifications. The mobile phase was 0.1 M sodium dihydrogen phosphate containing EDTA (0 5 m M ) and methanol to 13% v/v to improve separation o f G A B A and the internal standard (flaminobutyric acid) and to minimise analysis run time. Five min following the right M C A occlusion the concentration o f G A B A in the left (control)cortex was similar to that seen in the cortex taken from an unoperated g r o u p o f animals (Table II). The G A B A concentration in the right cortex (that is, the ischaemic side) was similar to that o f the left side and tissue from the unoperated animals (Table II). Accumulation o f G A B A following A O A A injection
TABLE II THE CONCENTRATION IN LEFT AND RIGHT CORTEX AND CEREBELLUM OF UNTREATED (CONTROL) MICE AND 5 MIN AND 4 HOURS AFTER A RIGHT MIDDLE CEREBRAL ARTERY OCCLUSION TOGETHER WITH THE GABA CONCENTRATION 40 MIN FOLLOWING INJECTION OF AMINO-OXYACETIC ACID (AOAA; 30 mg/kg) Results shown as mean + S.E.M. (n), N.D. not determined, Results in/~mol/g. Brain region
Time after AOAA (rain)
Control
MCA occlusion 5 min
4h
Left cortex
0 40
N.D. N.D.
2.32 _+0.16(7) 3.76 + 0.26(7)
1.61 -+0.17(6) 3.22 + 0.26(6)
Right cortex
0 40
2.44 _+0.28(8) 4.21 _+0.42(7)
2.49 + 0.18(8) 2.96 _+0.19(8)*
1.86 + 0.18(6) 3.50 _+0.23(6)
Cerebellum
0 40
1.18 _+0.05(6) 2.48 + 0.24(5)
1.25 _+0.09(6) 2.17 + 0.40(8)
1.00 + 0.05(6) 2.44 _+(I.35(6)
*Different from left cortex or control tissue at same time P
141
NSL 08544
The immediate consequences of middle cerebral artery occlusion on G A B A synthesis in mouse cortex and cerebellum A.R. Green, A.J. Cross, M.F. Snape and R.J. De Souza Astra Neuroscience Research Unit, London (UK) (Received 14 October 1991; Revised version received 12 January 1992; Accepted 21 January 1992)
Key words:
7-Aminobutyric acid synthesis; Ischaemia; Middle cerebral artery occlusion
The effect on y-aminobutyric acid (GABA) synthesis of focal ischaemia in the right cortex of the mouse was investigated by performing a right middle cerebral artery (MCA) occlusion. Synthesis of GABA was determined by measurement of the rate of GABA accumulation in tissue following injection of amino oxyacetic acid (AOAA; 30 mg/kg, i.p.). Five min following the MCA occlusion, the rate of GABA synthesis in the right (ischaemic) cortex was decreased by approximately 70% compared to either the left cortex or the right cortex of untreated controls. The basal GABA concentration was however unaffected. Four hours after the occlusion the rate of GABA synthesis was similar in the right and left cortex. The rate of GABA accumulation in the cerebellum was unchanged at both times after the right MCA occlusion compared with untreated control mice. The data suggest that there is a rapid but short lasting decrease in GABA synthesis following an ischaemic insult and it is suggested that this might be associated with the EEG spiking activity that occurs at this time.
There is now substantial evidence that competitive and non-competitive NMDA antagonists are neuroprotective in various models of stroke and cerebral ischaemic damage. Dizocilpine (MK-801), for example, prevents the hippocampal damage that normally occurs after a brief occlusion of the cerebral arteries of the gerbil [9, 10]. Recently, we have demonstrated that chlormethiazole is also neuroprotective in this model, even when the drug has been given 3 h after the period of occlusion [4]. There is, however, no evidence for chlormethiazole binding to any site on the N-methyl-o-aspartate (NMDA) receptor complex [5]. What has been established is that chlormethiazole can potentiate the action of GABA at the GABAA receptor and modulate channel opening [7, 15, 19, 20, 25]. This raises the possibility that chlormethiazole may act by enhancing GABA function and that this enhancement results in decreased NMDA function, an hypothesis which is given credence by the observation that chlormethiazole is an effective inhibitor of the convulsive behaviour induced by injection of N-methyl-DLaspartate (NMDLA) [5]. Following a global ischaemic episode 'spiking' occurs in the EEG, suggesting convulsive-like changes in the Correspondence: A.R. Green, Astra Neuroscience Research Unit, 1 Wakefield Street, London WCIN IPJ, UK.
brain [3]. Since previous studies have found that GABA synthesis in the brain is markedly inhibited after a convulsion elicited either chemically or electrically [13], it seemed possible that cerebral ischaemia might also result in a decrease in GABA synthesis. The current study has investigated this possibility. The model used to study the consequences of ischaemia on GABA synthesis was that of middle cerebral artery (MCA) occlusion in the mouse. This focal model was chosen both for its relative simplicity and also because it induces ischaemic damage on one side of the cortex whilst sparing the other. The study could therefore be conducted by investigating the intact and damaged sides of the cortex in the same animal. In addition, the consequences of the unilateral right MCA occlusion on GABA synthesis in the cerebellum, a region remote from the lesion, could also be examined. Male C57 B 1 / 6 0 l a mice (20-25 g) obtained from Harlan Olac (Bicester, UK) were used throughout the study. They were housed in groups of 6 in a 12 h light/12 h dark cycle (lights on 07,00 h) with food and water freely available. Middle cerebral artery occlusion was performed by anaesthetising the animals with 4% halothane in O2/N20 (ratio 2:1). The distal portion of the right middle cerebral artery was permanently occluded as follows. The animals underwent a right subtemporal craniotomy after partial
142 TABLE I BINDING OF PK 11195 IN MOUSE RIGHT CEREBRAL CORTEX 7 DAYS AFTER A MIDDLE CEREBRAL ARTERY OCCLUSION Results expressed as mean _+S.E.M.
Sham Ischaemic
PK 11195 Binding (fmol/mg protein)
n
% increase in binding
164_+15 379_+58
8 8
231 _+19
removal o f the temporalis muscle. The exposed M C A was electrocauterised and severed. The w o u n d was then sutured and anaesthesia discontinued. The animals were conscious and displaying l o c o m o t o r activity within another 2 min. The procedure used resulted in the formation o f an ischaemic infarct extending across the posterior frontal cortex and anterior parietal cortex, in the territory o f the middle, cerebral artery. Infarcts when visualised using ex vivo tetrazolium salt histochemistry were approximately 3 m m in diameter. The extent o f variability in size o f the infarct produced was assessed by the use o f P K 11 195 binding [8] following dissection o f the left and right cerebral cortex. Results obtained using this assay were similar to those previously reported [1] and indicate the reproducibility o f the operative procedure (Table I). G A B A synthesis was determined by measurement o f the rate o f accumulation o f G A B A following administration o f the G A B A transaminase inhibitor amino-oxyacetic acid ( A O A A ) , a m e t h o d originally described by Walters et al. [19] and recently validated [13]. Accumula-
tion o f G A B A after A O A A has been shown to be linear for at least 60 min [13]. A single measure o f G A B A concentration 40 min after A O A A is therefore justified, and has been used previously [7, 13] to calculate the rate o f G A B A synthesis. Results using this method [17] have been essentially identical with those obtained by other methods including mass f r a g m e n t o g r a p h y [16]. Animals were injected with A O A A (30 mg/kg, i.p.) either 5 rain or 4 h after the completion o f the surgical procedure. F o r t y rain after A O A A injection the mice were killed by exposure o f the head to a focussed beam o f microwave irradiation o f power density 0.7 kW for 1.1 s. The left and right cerebral cortex and cerebellum were dissected, weighed and stored at - 2 0 ° C until analysis. Tissue was homogenized in 0.5 ml o f perchloric acid (0,1 M), centrifuged for 3 rain and supernatant diluted 1:5 in borate buffer (pH 9.7). The supernatant was derivatised by reacting with o-phthalaldehyde in ethanol for H P L C analysis [21]. The reaction mixture (20ill) was injected onto the column (Spherisorb ODS-2, 25 cm, 5 j~m). G A B A was determined by the m e t h o d o f Lindroth and M o p p e r [23] with minor modifications. The mobile phase was 0.1 M sodium dihydrogen phosphate containing EDTA (0 5 m M ) and methanol to 13% v/v to improve separation o f G A B A and the internal standard (flaminobutyric acid) and to minimise analysis run time. Five min following the right M C A occlusion the concentration o f G A B A in the left (control)cortex was similar to that seen in the cortex taken from an unoperated g r o u p o f animals (Table II). The G A B A concentration in the right cortex (that is, the ischaemic side) was similar to that o f the left side and tissue from the unoperated animals (Table II). Accumulation o f G A B A following A O A A injection
TABLE II THE CONCENTRATION IN LEFT AND RIGHT CORTEX AND CEREBELLUM OF UNTREATED (CONTROL) MICE AND 5 MIN AND 4 HOURS AFTER A RIGHT MIDDLE CEREBRAL ARTERY OCCLUSION TOGETHER WITH THE GABA CONCENTRATION 40 MIN FOLLOWING INJECTION OF AMINO-OXYACETIC ACID (AOAA; 30 mg/kg) Results shown as mean + S.E.M. (n), N.D. not determined, Results in/~mol/g. Brain region
Time after AOAA (rain)
Control
MCA occlusion 5 min
4h
Left cortex
0 40
N.D. N.D.
2.32 _+0.16(7) 3.76 + 0.26(7)
1.61 -+0.17(6) 3.22 + 0.26(6)
Right cortex
0 40
2.44 _+0.28(8) 4.21 _+0.42(7)
2.49 + 0.18(8) 2.96 _+0.19(8)*
1.86 + 0.18(6) 3.50 _+0.23(6)
Cerebellum
0 40
1.18 _+0.05(6) 2.48 + 0.24(5)
1.25 _+0.09(6) 2.17 + 0.40(8)
1.00 + 0.05(6) 2.44 _+(I.35(6)
*Different from left cortex or control tissue at same time P
