Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by SAVANNAHRIVNATLABBF on 11/21/14 For personal use only.
Amino acids and the synaptic pharmacology of granule cells in the dentate gyrus of the rat T. $. HICKSAND M.MCLENNAN Deprbment of Physiology, U~iversit~1 of British Columbia, Vu~mcouver,B.C., Cataadu V6T 11/5
Received March 5, 1979 HICKS,T. P., and MCEENNAN, H. 1979. Amino acids and the synaptic pharmacology of granule cells in the dentate gyrus of the rat. Can. J . Physiol. Pharmacol. 57, 973-978. Granule cells of the dentate gyrus in the hippocampi of anaesthetized rats were excited by stimulation of the contralateral hippocampus (the commissural input) and the ipsilateral entorhinal cortex (the perforant path). The cells were also activated by the electrophoretic administration of various amino acids. A selective antagonism of glutamate and perforant path excitations was obtaiiled with glutamic acid diethylester, and of aspartate and other amino acid induced and commissural excitations with D- or DL-a-aminoadipate.An excitatory etTect of a-aminoadipate which was som~etitnesobserved was prevented by the r-aminobutyric acid antagonist bicuculline, and may be a disinhibitory phenomenon. The results lend support to the proposition that the transmitter of the perforant path is glutanlate while that of the connnaissuraH fibres is aspartate.
Introdwetism Among the compounds believed to function as chemical mediators of synaptic excitation in the vertebrate central nervous system are the dicarbsxylic amino acids GLUT and ASP (McLennam 1975; Johnson 1978; Watkins 1978). One of thc major criteria needed to substantiate this belief at any specific site is the demonstration of an identity of pharmacological action (Werman % 966), i.e. that the effect of a suspected transmitter exogenously administered and that of a known synaptic input are affected in the same way by a given pharmacological agent. Until recently this criterion has proved difficult to satisfy for the excitatory arnino acids, due to the lack of effective and pharmacologically specific antagonists of their actions, particularly in the rat (McLennan and Wheal 1976). However the parallel testing of two antagonists of amino acid induced excitations, although neither has complete specificity of action, has lcd to the proposition that at least two receptors for the amino acids exist, and has further indicated that the use of these compounds as pharmacological tools may be profitable (Hicks et a&. 1978; McLennan and Lodge 1979). The two connpounds, GBEE and D-a-aminoadipic acid, have been shown to have a consistent pattern of effect against amino acid induced firings of thalamic and spinal neurones (Haldeman and McLennan 1972; Biscoe et al. 1978; Evans and Watkins ABBREVIATIONS: GLUT, L-glutamate; ASP, L-aspartate; GBEE, L-glutamie acid diethylester; PP, perforant path: COMM, corninissura% pathway; DEH, DL-hornocysteate; NMA, N-methy 1-DL-aspartate;GABA, y-minobutyric acid; BICUC, bica~cullinemethochloride; aAA, a-arninoadipate, EPSP, excitatory postsynaptic potential.
1978; Hicks et al. 1978; McLennan and Hall 1978; Hall et al. 1979; McLennan and Lodge 1979), in that GBEE is more eflective against GLUT while ASP aisd other amino acid induced excitations are more sensitive to D-a-aminoadipic acid. Thus the existence of two groups of excitatory receptors which have distinguishable cl-aaracteristicsmay be proposed, one preferentially activated by GLUT and blocked by GDEE and the other more affected by ASP and particularly by N-methylaspartate which is antagonized by D-a-aininoadipic acid. However there is overlap in the pharmacological behaviour of the two groups, for sufficiently high doses of either of the antagonists can reduce the effects of most amino acid excitants. If there are some central synapscs mediated by GLUT and others which employ ASP as a transmitter, then it might be reasonabIe to expect that tlnose ncurones which receive one or other input would be relatively more sensitive to the appropriate a~niinoacid. This has been claimed for Renshaw cclls and dorsal horn interneurones in the cat spinal cord (Duggan 1974; McCulloch et a/. 1974), where tlss suggestion was made that the dorsal root fibres might liberate GLUT while spinal interneurones were aspartergic; however this differential sensitivity could not be confirmed in the rat (Hutchinson et al. 1978). The well-defined neuronal organization of the hippocampal formation provides an excellent testing ground for comparing the actions of amino acids and their antagonists. In the dentate gyrus the primary cell type is the granule cell which is distributed within this tissue in a thin horseshoe-shaped layer. The synaptic inputs to these neurones occur in nonovcrlapping zones which have been studied electrophys-
0008-4212/79/090973-06$01 .OQ/8 ,@ 1979 National Research Council s f CanadaiGonsell national de recherches du Canada
974
CAW. 1. PNYSHOL. PHARMACOL. VOL. 57. 1979
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by SAVANNAHRIVNATLABBF on 11/21/14 For personal use only.
iologically and anatomically (Blackstad 1956, 1958; Andersen et nb. 1966; L@mo 197 1 ) . Of particular interest is the likelihood that GLUT anad ASP mediate the synaptic inputs from different groups of afferent fibres onto the dendrites of the granule cells. Nadler et al. ( 1976) and White ut al. (1977) have reported that a calciunn-dependent efflux of GLUT and ASP c~ccurredfollowing potassium-induced depolarization of slices of rat dentate gyms and regis superior. Surgical interniption of the entorhinal input by way of the PP reduced the release of GLUT from the dentate gyrus while lesions of the CBMM from the contralateral hippocanlpus diminished the release of ASP (Nadler et al. 1978). Wheal and Miller ( 1979 3 have shown that tlne "lutamate-preferring' antagonist GDEE can prevent the PP-evoked activation of dentate granule cells, which also suggests that the synapses fornned by the PP fibres are glutarnate mediated. In other regions of the hippscampus the excitation of pyramidal cells by amins acids has been reported (Biscoe and Straughan 1966), and GDEE is effective in preventing the amino acid induced firing (Segal 1976; Spencer et al. 1976). This action of glutamate probabl y occurs in the dendritic region of the neuromes, for this has been shown to be the most sensitive site of action for GLUT (Budar 1974) and is the locus of the synaptic inputs. However there has been no analysis of the phssible involvement of the amino acids in the dentate gyms using a comparison of their relative potencies on granule cells and differential antagonism of the amino acid effects. It is the intention of the present study to provide such an analysis of the possible involvement of the amino acids in the synaptic organization of the dentate gyrus.
TABLE1. Relative potencies of G L U T and ASP tested dentate granule cells
01-1
Ratio lASP :Ic;LT;T No. of cells
ASP ASP ASP
> GLUT GLUT < GLUT =
9 3
L
4
kange
Mean -t SD
1.2-8 - 0 1 .0 0.7-0.9
2.8rt2.2 -
eB.8k0.1
pH 7 ) . N o differences between the actions of the isolated D-isomer and of DL-a-aminoadipatewere found; the results have therefore been pooled and the abbreviation aiSA will hereafter be lased. Responses were displayed on an oscilloscope and photographed. and action potentials were integrated and the output recorded on a paper chart. Bipolar stimulating electrodes were placed in the contra1.9; KGnig and lateral hippocamptls ( A 3.7, L 1.5, %' Klippel (1974)) to activate the crossed fibre projection comnaonly termed the COPITM, and in the ipsilateral angular 1.3) for the P P emanating from bundle ( P 0.5, L4.4, %' the entorhinal cortex. Cells in the dentate gyrus could be activated synaptisalIy by low intensity (20-100 ,uA) single square-wave pulses of 0. l ms duration delivered through the electrodes. The micropipettes were lowered through the cerebral cortex at stereotaxic coordinates A 3.4 and 1, 1.5. and the field responses of CAI pyramidal cells and the extracellularly recorded EPSPs of these and of the granule cell dendrites observed (H,@mo 197 1 9. Somata of the granule cells were situated approximately 0.1 rnm below the point of reversal of the latter EPSP.
+
+
Results
Neurones lying in the upper blade of the dentate gyrus and which were synaptically activated by stimulation of PP and CBMM were readily excited by the electrophoretic administration of the excitatory amino acids. Since the pipette assembly was presumably situated near the cell bodies any difference in the apparent potency between GLUT and ASP could reflect the distribution of receptors affected by the Materials and Methods ejected materials relative to that location. The relaThe excitatory effects of G L U T and ASP and of certain other amino acids on neurones of the dentate gyrus have tive apparent potencies of GLUT and ASP were been examined in rats anaesthetized with urethane (1.5 obtained by comparing the magnitudes of the ejectg/kg, ip). The animals were positioned in a stereotaxic ing currents required to elicit stable, equal, and subframe and the calvarium widely removed. T h e aura was maximal levels of the faring frequencies of granule reflected and the exposed cortex covered with warm Locke's cells, a valid comparison since the 'transport numsolution or parafin oil. The animals' body temperatures were maintained at 37°C. Extracellular recording of neu- bers' for these two substances are equivalent (Hall ronaI activity via the central 4 M NaCl containing barrel of et al. 1979). The results of such determinations on a mailtipipette assembly and the electrophoretic ejection of cells where reproducible tests were made on a minsubstances from the surrounding barrels w-ere standard imum of three occasions, are presented in Table 1. procedures. One peripheral barrel always contained Pontamine sky blue for histological confirmation of electrode Although not true for every cell, on average for all 15 cells ASP was twice as effective as GLUT ( p < placements; the others contained some of the following: GT,UT, ASP, or DL-a-aminoadipate ( I C N Pharmaceuticals) 0.05, Student's t test for the comparison of a single (all 0.5 M, p H 8 ) ; DLH, NPITA (both from Calbiochem) or small sample with a known standard). Since the PP D-a-aminoadipatc (all 0.2 M, pH 8 9 ; G D E E (Koch-Light and Mann Research) or GABA (ICN Pharn~aceuticaIs) input to the granule cells, which has been suggested (both 0.5 ikf, p H 3.5); or BICUC (10 mM in 0.15 NaCI, to be GLUT mediated (Wheal and Miller B979), is
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by SAVANNAHRIVNATLABBF on 11/21/14 For personal use only.
HICKS AND MeLENNAN
Recovery ( 3 mln 1
0
1NMA
GLUT12
FIG. 1. The actions of CYAA (administered with a current of 12 nA) and GDEE (40 nA) on the excitation of a granule cell by NMA (10 nA) and GLUT (12 nA) (left column), and synaptic activation via the cornmissural input (central column) and the PP (right column). The rows of records were obtained sequentially from above downwards, and are photographs of four superimposed sweeps.
distally located on the dendrites this result might indicate the existence of an ASP-mediated innervation which is more proximally situated. The effects of the antagonists GBEE and a A A were examined on amino acid induced and synaptic activations of 30 granule cells. GDEE administered with ejecting currents of 20-50 nA attenuated considerably the effects of GLUT but not those of NMA (Fig. 1 C-E), DLH, or ASP. The action of aAA, on the other hand, always spared excitations produced by GLUT at doses up to 30 nA but was effective in abolishing or reducing markedly those of the other three amino acids (e.g. Fig. 1 A-C). This selectivity had been anticipated in light of the findings of Hicks et al. ( 1978) and of McLennan and Lodge (1979) on thalamic and spinal neurones, respec-
tively. However it was not always possible to test the antagonistic action of a A A since on some occasions (10 of 30 cells) a marked excitatory effect of this conapound itself was found. This property has not been previously examined pharinacologically. although Lodge et al. (1 979) did report that DaAA could raise the background firing rate and enhance the effects of GLUT on spinal interneurones. The matter will be further discussed below. When selective antagonism of the amino acid induced excitations had beela established the synaptic responses of the cells were examined. In the presence of sufficient ainounts of GDEE to antagonize the effects of GLUT, the response to stimulation of PP but not of COMM was also blocked. Conversely a dose of a A A which selectively blocked ASP or NMA
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by SAVANNAHRIVNATLABBF on 11/21/14 For personal use only.
996
CAN. J. PHYSIOL. HBHARlMACOL. VOL. 57, 1979
excitations concomitantly reduced the COMM activation of the neurorses while the response t s PP was unaltered. Recovery to control levels of both the synaptic and the amino acid effects was observed after cessation of the currents ejecting the antagonists (Fig. 1 ) . Similar cornpletc series of antagonisms a~adrecoveries using both blocking agents were obtained on six other cells. It was earlier renaarked that when aAA was tested for its ability to antagonize amino acid induced firing, an incrcase in the background firing rate of some cells was observed. This heightened activity was noted even when the synaptic response to COMM stimulation was redarced by aAA. Although L-a-aminoadipate is a weak excitant (Curtis and Watkins 1960; Hall et aE. 1979) the D-isonaer has usually been reported to be without effect (Biscoe et a%.1977; Hall et a&.1977; McLennan and Hall 1978; but see Lodge et ab. 19791, yet the present effect was obtained both with the racernic mixture and with the separated Disomer. It seenaed likely therefore that some meehanism other than one sf direct excitation might be involved. and a 'disinhibitory' praccss is one possibility. To examine this, experiments of the type illustrated by Fig. 2 were performed. In this instance a granule ccll was made to fire at a rate of 35-50 sgikes/s by the conkin~aousejection of GLUT (25 nA). Superimposed upon this background the inhibitory and excitatory effects elicited by ejections of GABA and of aAA, respectively, were elicited. Thereafter by reducing the GLUT current the background firing frequency ccruld be maintained at a roughly constant level during ejection of the GABA antagonist BICUC. When the inhibitory response to GABA was largely abolished ( 1-2 min after onmset of BSLCUC. 25 nA, Fig. 2B), thc excitation pre-~iously produced by a A A was considerably attenuated, and reappeared following cessation of the BICUC current coincidentally with the return (3% the inhibitory response to GABA. Identical results were obtained with five other cells.
GLUT 2%
A
....;
Amino acids and the synaptic pharmacology of granule cells in the dentate gyrus of the rat T. $. HICKSAND M.MCLENNAN Deprbment of Physiology, U~iversit~1 of British Columbia, Vu~mcouver,B.C., Cataadu V6T 11/5
Received March 5, 1979 HICKS,T. P., and MCEENNAN, H. 1979. Amino acids and the synaptic pharmacology of granule cells in the dentate gyrus of the rat. Can. J . Physiol. Pharmacol. 57, 973-978. Granule cells of the dentate gyrus in the hippocampi of anaesthetized rats were excited by stimulation of the contralateral hippocampus (the commissural input) and the ipsilateral entorhinal cortex (the perforant path). The cells were also activated by the electrophoretic administration of various amino acids. A selective antagonism of glutamate and perforant path excitations was obtaiiled with glutamic acid diethylester, and of aspartate and other amino acid induced and commissural excitations with D- or DL-a-aminoadipate.An excitatory etTect of a-aminoadipate which was som~etitnesobserved was prevented by the r-aminobutyric acid antagonist bicuculline, and may be a disinhibitory phenomenon. The results lend support to the proposition that the transmitter of the perforant path is glutanlate while that of the connnaissuraH fibres is aspartate.
Introdwetism Among the compounds believed to function as chemical mediators of synaptic excitation in the vertebrate central nervous system are the dicarbsxylic amino acids GLUT and ASP (McLennam 1975; Johnson 1978; Watkins 1978). One of thc major criteria needed to substantiate this belief at any specific site is the demonstration of an identity of pharmacological action (Werman % 966), i.e. that the effect of a suspected transmitter exogenously administered and that of a known synaptic input are affected in the same way by a given pharmacological agent. Until recently this criterion has proved difficult to satisfy for the excitatory arnino acids, due to the lack of effective and pharmacologically specific antagonists of their actions, particularly in the rat (McLennan and Wheal 1976). However the parallel testing of two antagonists of amino acid induced excitations, although neither has complete specificity of action, has lcd to the proposition that at least two receptors for the amino acids exist, and has further indicated that the use of these compounds as pharmacological tools may be profitable (Hicks et a&. 1978; McLennan and Lodge 1979). The two connpounds, GBEE and D-a-aminoadipic acid, have been shown to have a consistent pattern of effect against amino acid induced firings of thalamic and spinal neurones (Haldeman and McLennan 1972; Biscoe et al. 1978; Evans and Watkins ABBREVIATIONS: GLUT, L-glutamate; ASP, L-aspartate; GBEE, L-glutamie acid diethylester; PP, perforant path: COMM, corninissura% pathway; DEH, DL-hornocysteate; NMA, N-methy 1-DL-aspartate;GABA, y-minobutyric acid; BICUC, bica~cullinemethochloride; aAA, a-arninoadipate, EPSP, excitatory postsynaptic potential.
1978; Hicks et al. 1978; McLennan and Hall 1978; Hall et al. 1979; McLennan and Lodge 1979), in that GBEE is more eflective against GLUT while ASP aisd other amino acid induced excitations are more sensitive to D-a-aminoadipic acid. Thus the existence of two groups of excitatory receptors which have distinguishable cl-aaracteristicsmay be proposed, one preferentially activated by GLUT and blocked by GDEE and the other more affected by ASP and particularly by N-methylaspartate which is antagonized by D-a-aininoadipic acid. However there is overlap in the pharmacological behaviour of the two groups, for sufficiently high doses of either of the antagonists can reduce the effects of most amino acid excitants. If there are some central synapscs mediated by GLUT and others which employ ASP as a transmitter, then it might be reasonabIe to expect that tlnose ncurones which receive one or other input would be relatively more sensitive to the appropriate a~niinoacid. This has been claimed for Renshaw cclls and dorsal horn interneurones in the cat spinal cord (Duggan 1974; McCulloch et a/. 1974), where tlss suggestion was made that the dorsal root fibres might liberate GLUT while spinal interneurones were aspartergic; however this differential sensitivity could not be confirmed in the rat (Hutchinson et al. 1978). The well-defined neuronal organization of the hippocampal formation provides an excellent testing ground for comparing the actions of amino acids and their antagonists. In the dentate gyrus the primary cell type is the granule cell which is distributed within this tissue in a thin horseshoe-shaped layer. The synaptic inputs to these neurones occur in nonovcrlapping zones which have been studied electrophys-
0008-4212/79/090973-06$01 .OQ/8 ,@ 1979 National Research Council s f CanadaiGonsell national de recherches du Canada
974
CAW. 1. PNYSHOL. PHARMACOL. VOL. 57. 1979
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by SAVANNAHRIVNATLABBF on 11/21/14 For personal use only.
iologically and anatomically (Blackstad 1956, 1958; Andersen et nb. 1966; L@mo 197 1 ) . Of particular interest is the likelihood that GLUT anad ASP mediate the synaptic inputs from different groups of afferent fibres onto the dendrites of the granule cells. Nadler et al. ( 1976) and White ut al. (1977) have reported that a calciunn-dependent efflux of GLUT and ASP c~ccurredfollowing potassium-induced depolarization of slices of rat dentate gyms and regis superior. Surgical interniption of the entorhinal input by way of the PP reduced the release of GLUT from the dentate gyrus while lesions of the CBMM from the contralateral hippocanlpus diminished the release of ASP (Nadler et al. 1978). Wheal and Miller ( 1979 3 have shown that tlne "lutamate-preferring' antagonist GDEE can prevent the PP-evoked activation of dentate granule cells, which also suggests that the synapses fornned by the PP fibres are glutarnate mediated. In other regions of the hippscampus the excitation of pyramidal cells by amins acids has been reported (Biscoe and Straughan 1966), and GDEE is effective in preventing the amino acid induced firing (Segal 1976; Spencer et al. 1976). This action of glutamate probabl y occurs in the dendritic region of the neuromes, for this has been shown to be the most sensitive site of action for GLUT (Budar 1974) and is the locus of the synaptic inputs. However there has been no analysis of the phssible involvement of the amino acids in the dentate gyms using a comparison of their relative potencies on granule cells and differential antagonism of the amino acid effects. It is the intention of the present study to provide such an analysis of the possible involvement of the amino acids in the synaptic organization of the dentate gyrus.
TABLE1. Relative potencies of G L U T and ASP tested dentate granule cells
01-1
Ratio lASP :Ic;LT;T No. of cells
ASP ASP ASP
> GLUT GLUT < GLUT =
9 3
L
4
kange
Mean -t SD
1.2-8 - 0 1 .0 0.7-0.9
2.8rt2.2 -
eB.8k0.1
pH 7 ) . N o differences between the actions of the isolated D-isomer and of DL-a-aminoadipatewere found; the results have therefore been pooled and the abbreviation aiSA will hereafter be lased. Responses were displayed on an oscilloscope and photographed. and action potentials were integrated and the output recorded on a paper chart. Bipolar stimulating electrodes were placed in the contra1.9; KGnig and lateral hippocamptls ( A 3.7, L 1.5, %' Klippel (1974)) to activate the crossed fibre projection comnaonly termed the COPITM, and in the ipsilateral angular 1.3) for the P P emanating from bundle ( P 0.5, L4.4, %' the entorhinal cortex. Cells in the dentate gyrus could be activated synaptisalIy by low intensity (20-100 ,uA) single square-wave pulses of 0. l ms duration delivered through the electrodes. The micropipettes were lowered through the cerebral cortex at stereotaxic coordinates A 3.4 and 1, 1.5. and the field responses of CAI pyramidal cells and the extracellularly recorded EPSPs of these and of the granule cell dendrites observed (H,@mo 197 1 9. Somata of the granule cells were situated approximately 0.1 rnm below the point of reversal of the latter EPSP.
+
+
Results
Neurones lying in the upper blade of the dentate gyrus and which were synaptically activated by stimulation of PP and CBMM were readily excited by the electrophoretic administration of the excitatory amino acids. Since the pipette assembly was presumably situated near the cell bodies any difference in the apparent potency between GLUT and ASP could reflect the distribution of receptors affected by the Materials and Methods ejected materials relative to that location. The relaThe excitatory effects of G L U T and ASP and of certain other amino acids on neurones of the dentate gyrus have tive apparent potencies of GLUT and ASP were been examined in rats anaesthetized with urethane (1.5 obtained by comparing the magnitudes of the ejectg/kg, ip). The animals were positioned in a stereotaxic ing currents required to elicit stable, equal, and subframe and the calvarium widely removed. T h e aura was maximal levels of the faring frequencies of granule reflected and the exposed cortex covered with warm Locke's cells, a valid comparison since the 'transport numsolution or parafin oil. The animals' body temperatures were maintained at 37°C. Extracellular recording of neu- bers' for these two substances are equivalent (Hall ronaI activity via the central 4 M NaCl containing barrel of et al. 1979). The results of such determinations on a mailtipipette assembly and the electrophoretic ejection of cells where reproducible tests were made on a minsubstances from the surrounding barrels w-ere standard imum of three occasions, are presented in Table 1. procedures. One peripheral barrel always contained Pontamine sky blue for histological confirmation of electrode Although not true for every cell, on average for all 15 cells ASP was twice as effective as GLUT ( p < placements; the others contained some of the following: GT,UT, ASP, or DL-a-aminoadipate ( I C N Pharmaceuticals) 0.05, Student's t test for the comparison of a single (all 0.5 M, p H 8 ) ; DLH, NPITA (both from Calbiochem) or small sample with a known standard). Since the PP D-a-aminoadipatc (all 0.2 M, pH 8 9 ; G D E E (Koch-Light and Mann Research) or GABA (ICN Pharn~aceuticaIs) input to the granule cells, which has been suggested (both 0.5 ikf, p H 3.5); or BICUC (10 mM in 0.15 NaCI, to be GLUT mediated (Wheal and Miller B979), is
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by SAVANNAHRIVNATLABBF on 11/21/14 For personal use only.
HICKS AND MeLENNAN
Recovery ( 3 mln 1
0
1NMA
GLUT12
FIG. 1. The actions of CYAA (administered with a current of 12 nA) and GDEE (40 nA) on the excitation of a granule cell by NMA (10 nA) and GLUT (12 nA) (left column), and synaptic activation via the cornmissural input (central column) and the PP (right column). The rows of records were obtained sequentially from above downwards, and are photographs of four superimposed sweeps.
distally located on the dendrites this result might indicate the existence of an ASP-mediated innervation which is more proximally situated. The effects of the antagonists GBEE and a A A were examined on amino acid induced and synaptic activations of 30 granule cells. GDEE administered with ejecting currents of 20-50 nA attenuated considerably the effects of GLUT but not those of NMA (Fig. 1 C-E), DLH, or ASP. The action of aAA, on the other hand, always spared excitations produced by GLUT at doses up to 30 nA but was effective in abolishing or reducing markedly those of the other three amino acids (e.g. Fig. 1 A-C). This selectivity had been anticipated in light of the findings of Hicks et al. ( 1978) and of McLennan and Lodge (1979) on thalamic and spinal neurones, respec-
tively. However it was not always possible to test the antagonistic action of a A A since on some occasions (10 of 30 cells) a marked excitatory effect of this conapound itself was found. This property has not been previously examined pharinacologically. although Lodge et al. (1 979) did report that DaAA could raise the background firing rate and enhance the effects of GLUT on spinal interneurones. The matter will be further discussed below. When selective antagonism of the amino acid induced excitations had beela established the synaptic responses of the cells were examined. In the presence of sufficient ainounts of GDEE to antagonize the effects of GLUT, the response to stimulation of PP but not of COMM was also blocked. Conversely a dose of a A A which selectively blocked ASP or NMA
Can. J. Physiol. Pharmacol. Downloaded from www.nrcresearchpress.com by SAVANNAHRIVNATLABBF on 11/21/14 For personal use only.
996
CAN. J. PHYSIOL. HBHARlMACOL. VOL. 57, 1979
excitations concomitantly reduced the COMM activation of the neurorses while the response t s PP was unaltered. Recovery to control levels of both the synaptic and the amino acid effects was observed after cessation of the currents ejecting the antagonists (Fig. 1 ) . Similar cornpletc series of antagonisms a~adrecoveries using both blocking agents were obtained on six other cells. It was earlier renaarked that when aAA was tested for its ability to antagonize amino acid induced firing, an incrcase in the background firing rate of some cells was observed. This heightened activity was noted even when the synaptic response to COMM stimulation was redarced by aAA. Although L-a-aminoadipate is a weak excitant (Curtis and Watkins 1960; Hall et aE. 1979) the D-isonaer has usually been reported to be without effect (Biscoe et a%.1977; Hall et a&.1977; McLennan and Hall 1978; but see Lodge et ab. 19791, yet the present effect was obtained both with the racernic mixture and with the separated Disomer. It seenaed likely therefore that some meehanism other than one sf direct excitation might be involved. and a 'disinhibitory' praccss is one possibility. To examine this, experiments of the type illustrated by Fig. 2 were performed. In this instance a granule ccll was made to fire at a rate of 35-50 sgikes/s by the conkin~aousejection of GLUT (25 nA). Superimposed upon this background the inhibitory and excitatory effects elicited by ejections of GABA and of aAA, respectively, were elicited. Thereafter by reducing the GLUT current the background firing frequency ccruld be maintained at a roughly constant level during ejection of the GABA antagonist BICUC. When the inhibitory response to GABA was largely abolished ( 1-2 min after onmset of BSLCUC. 25 nA, Fig. 2B), thc excitation pre-~iously produced by a A A was considerably attenuated, and reappeared following cessation of the BICUC current coincidentally with the return (3% the inhibitory response to GABA. Identical results were obtained with five other cells.
GLUT 2%
A
....;
