- Email: [email protected]
The action of GABAB antagonists in the trigeminal nucleus of the rat
Neuropharmacology Vol. 31, No. 5, pp. 475480, Printed in Great Britain. All rights reserved
1992 Copyright
0
OOZS-3908/92 S5.00 + 0.00 I992 Pergamon Press plc
THE ACTION OF GABA, ANTAGONISTS IN THE TRIGEMINAL NUCLEUS OF THE RAT G. H. FROMM, K. SATO* and M. NAKATA? Department of Neurology, University of Pittsburgh, School of Medicine, Pittsburgh, PA 15261, U.S.A. (Accepred
7 November
1991)
iontophoretic administration of the GABA, antagonists (P-(3-aminopropyl)-P-diethoxymethylphosphinic acid (CGP 35348) and 2-hydroxy-saclofen blocked the action of iontophoretically applied L-baclofen on neurons in the trigeminal nucleus of rats, anesthetized with halothane. The Summary-The
substance CGP 35348 appeared to be more potent than 2-hydroxy-saclofen. The iontophoretic administration of GABA resembled L-baclofen in depressing excitatory transmission and facilitating segmental inhibition in the trigeminal nucleus. The depression of excitatory transmission was also blocked by CGP 35348 and the facilitation of segmental inhibition produced by GABA was partially blocked. These observations indicate that CGP 35348 is not only a baclofen antagonist but actually a GABA, receptor antagonist and the baclofen was acting at GABA, receptors in the trigeminal nucleus. The portion of the effect of GABA, not blocked by CGP 35348, was probably mediated by GABA, receptors, since it was previously found that segmental inhibition in the trigeminal nucleus could be modulated by GABA, agonists and antagonists as well. Kq words-baclofen, nucleus.
GABA,
GABA,
receptor
antagonists,
CGP 35348, 2-hydroxy-saclofen,
trigeminal
METHODS
It was previously found that neuronal activity in the spinal trigeminal nucleus could be modulated by the y-aminobutyric acid (GABA,) agonist, baclofen (Fromm, Terrence, Chattha and Glass, 1980; Fromm, Chattha, Terrence and Glass, 198 1; Fromm, Shibuya, Nakata and Terrence, 1990), as well as by GABA, agonists and antagonists (Fromm and Terrence, 1984; Nakata, Mukawa and Fromm, 1991). In the absence of confirmation by a GABAa antagonist, the question remained though as to whether baclofen actually exerts its action in the trigeminal nucleus at GABAs receptors. It has now been demonstrated that CGP 35348 (Olpe, Karlsson, Pozza, Brugger, Steinmann, Van Riezen, Fagg, Hall, Froestl and Bittiger, 1990) and 2-hydroxy-saclofen (Curtis, Gynther, Beattie, Kerr and Prager, 1988; Kerr, Ong, Johnston, Abbenante and Prager, 1988; Lambert, Harrison, Kerr, Ong, Prager and Teyler, 1989) are selective blockers of GABA, receptors, with CGP 35348 being the more potent. Therefore. the effect of these GABAa antagonists on the action of L-baclofen and of GABA on single neurons in the trigeminal nucleus was investigated in order to ascertain whether L-baclofen was indeed acting at GABA, receptors in the experimental model used.
The experiments were performed on 15 SpragueDawley rats, weighing 225-300 g, anesthetized with halothane, 3% for induction, 2% for surgery and 1% for recording, in 30-50% 0,/70-50% Nr. The rats were paralyzed with pancuronium bromide and artificially ventilated. Blood pressure, rectal temperature and blood gases were monitored throughout the experiment and maintained within physiological limits. The activity of single neurons in the spinal trigeminal nucleus oralis was recorded extracellularly with a glass micropipette (tip dia = 1.0-1.5 pm), glued along a five-barrelled micropipette (outer tip dia = 8-lOpm), so that the single tip of the barrel protruded 3-7pm beyond the tip of the five-barrel electrode (Fromm et al., 1990). A balancing channel was employed to alleviate current effects; this barrel and the recording micropipette were filled with 2 M NaCl, saturated with Fast Green-FCF. The four remaining barrels contained 0.01 M t-baclofen in 0.15 M NaCl, pH = 5.0 (kindly supplied by CibaGeigy), 0.01 M CGP 35348 (P-(3-aminopropyl)-Pdiethoxymethyl-phosphinic acid) in 0.15 M NaCI, pH = 6.7 (kindly supplied by Ciba-Geigy), 0.01 M 2-hydroxy-saclofen in 0.15 M NaCl, pH = 3 (Tokris) or 0.5 M GABA, pH = 3.5-4.0 (Sigma). Ejection, retention and balancing currents were produced by a constant-current source (Medical Systems Model BH-2). Stimuli were applied to the facial skin at 0.5 Hz with a pair of needle electrodes, using monophasic
*Present address: Department of Neuropsychiatry, Okayama University Medical School, 2-5-l Shikata-cho, Okayama 700, Japan. YPresent address: Department of Neurosurgery, University of the Ryukyus, Nishihara Uehara 207, Okinawa 903-01, Japan. 475
G. H. FROMM~~
476
squarewave pulses (2.2-7.7 V, 0.1 msec). The stimulus strength was adjusted to obtain the shortest possible latency of response for each neuron. Segmental or afferent inhibition was elicited by delivering a 100 msec train of conditioning stimuli (2.2-5.5 mA, 0.3 msec, 50 Hz) 100 msec before the test stimulus. Responses of neurons were monitored before, during and after the iontophoretic administration of the drugs on a cathode ray oscilloscope and stored on magnetic tape. Poststimulus histograms were obtained with a Neurolog NL 750/755 Signal Averager. The latency of the initial neuronal spike was measured with a time-interval measurement system (Quasitronics, Houston, Pennsylvania) and the mean latency and SD of 10 consecutive unconditioned responses and of 10 consecutive responses each following a conditioning train of stimuli was calculated. A value of 20 msec was assigned when the neuron did not respond at all. The location of the recording electrode was marked at the end of the experiments by passing a direct current of IO PA for 10 min through the electrode and then checked histologically, after staining with hematoxylin and eosin. RESULTS
The neurons used in this study were in the spinal trigeminal nucleus oralis, 0.9-2.6 mm posterior to the interaural line and 2.2-3.1 mm lateral to the midline (Paxinos and Watson, 1982) and were located by electrical stimulation of the face at 1 Hz, while slowly advancing the microelectrode. They responded to an unconditioned stimulus, with a mean latency of 2.2 msec (SD = 0.34). The segmental inhibition, elicited by delivering a conditioning train of stimuli 100 msec before the test stimulus, increased the mean latency of the response by 0.23 msec (SD = 0.10) and decreased the number of spikes if the neuron responded to each unconditioned stimulus with more than one spike (Figs 1 and 2). The iontophoretic administration of 60 nA CGP 35348 decreased the mean latency of the conditioned response in 10 of 20 neurons by 0.06-0.20 msec and this change was statistically significant in 7 (P < 0.05 by the Wilcoxon matched-pairs signed-ranks test). The administration of 30 nA CGP 35348 decreased the latency of the conditioned response in only 3 of the 20 neurons, with the change being smaller (0.0220.10 msec) and statistically significant in only 1 neuron. The iontophoretic administration of up to 200 nA 2-hydroxysaclofen decreased the mean latency of the conditioned response in only 1 of 5 neurons. As in the previous experiments (Fromm et al., 1990) the iontophoretic administration of 15-30 nA L-baclofen depressed the response of the neurons of the trigeminal nucleus to electrical stimulation of the face, increasing the mean latency of the initial spike and decreasing the total number of spikes (Figs 1 and 2). The iontophoresis of r.-baclofen also markedly
al.
enhanced the segmental inhibition, further increasing the mean latency of the initial spike and decreasing the number of spikes. The concomitant iontophoresis of 30-60 nA CGP 35348 (starting 1 min before the L-baclofen was applied), blocked the effect of Lbaclofen on both the unconditioned stimulus and segmental inhibition. Similarly, the iontophoresis of 2-hydroxy-saclofen blocked the action of iontophoretically applied L-baclofen on both the unconditioned stimulus and segmental inhibition. However, larger currents (100-200 nA) were required for 2-hydroxysaclofen to have a comparable effect (Fig. IB). The action of both CGP 35348 and 2-hydroxy-saclofen was tested in 4 neurons, while in I neuron only 2-hydroxy-saclofen could be tested and in 13 neurons, only CGP 35348 was tested. The iontophoretic administration of GABA resembled the action of L-baclofen in the trigeminal nucleus (Fig. 2). The concomitant iontophoresis of 30-60 nA CGP 35348 (starting I min before the GABA was applied) also blocked the effects of GABA. However, these doses of CGP 35348 were less effective in blocking the action of GABA than in blocking that of L-baclofen. The concomitant administration of 30-60 nA CGP 35348 almost completely blocked the facilitation of segmental inhibition produced by L-baclofen but only partially blocked the GABA-induced enhancement of segmental inhibition (Figs 2B and 3). The difference between the action of CGP 35348, against the effect of L-baclofen and against the effect of GABA on segmental inhibition was significant for both 30 nA and 60 nA CGP 35348 (U = 12.0, P = 0.019; and I/ = 12.5. P = 0.025. respectively. by the Mann-Whitney U-test). Moreover, while CGP 35348 only partially prevented the inhibition of the late discharge, induced by baclofen; it did not block the GABAergic depression of the late discharge at all (Fig. 2B). DISCUSSION
These experiments demonstrated that the actions of L-baclofen on single neurons in the trigeminal nucleus were blocked by the GABA, receptor antagonists CGP 35348 and 2-hydroxy-saclofen. The substance CGP 35348 appeared to be more effective than 2-hydroxy-saclofen as judged by the strength of the ejection current required to elicit an effect. However, the amount of current is, at best, only a relative measure of the amount of drug being delivered. It was also found that the iontophoretic administration of GABA resembled r.-baclofen in depressing excitatory transmission and facilitating segmental inhibition in the trigeminal nucleus and these actions were also blocked by CGP 35348. These observations indicate that CGP 35348 is not only a baclofen antagonist but a GABA, receptor antagonist and that baclofen was acting at GABA, receptors in the trigeminal nucleus. That CGP 35348 blocked the action of exogenously applied GABA or L-baclofen but had only a
GABA, antagonists
(4 CGP-35348
CONTROL
t
I
N = 16
cap-35348 -i-l
-BCF N=3
i
N= 16
@I SCF
CONTROL
SCF + l’-BCF
1-BCF
N=f6
N254
A . 3
8
3
3
6
N=O
N= 16
N=16
6
N=9
St
i
fr 3
6
3
Fig. 1. Poststimulus histograms (16 stimuli at 0.5 Hz, 0.1 msec bin width) of the response of a neuron in the trigeminal nucleus oralis. to an unconditioned stimulus (US) and to the same stimulus delivered 100 msec after a train of conditioning stimuli (Sl). Iv = total number of spikes. (A) Records obtained before administration of drug (CONTROL); during the iontophoretic application of 30 nA CGP-35348 (CGP 35348); during the jontophoretj~ apptication of 15 nA L-baciofen ft-BCF); and the iontopboretic application of 30 nA CGP 35348 + I5 nA L-baciofen (CGP 35348 i- L-BCF). (3) Records obtained before administration of drug (CONTROL); during the iontophoretic application of 200 nA 2-hydroxy-saclofen (SCP); during the iontophoretic application of IS nA r.-baclofen (L-BCF); and the iontophoretic application of 200 nA 2-hydroxy-saclofen + IS nA L-bactofen (S&F + L-BCF). L-Baclofen depressed the neuronal response to the unconditioned stimulus (US) and enhanced the segmental inhibition @I). CGP 35348 Blocked the effect of t,-baclofen on both the unconditioned stimulus and segmental inhibition. 2-Hydroxy-saciofen also blocked the action of t-baclofen but an ejection current of 200 nA was required to produce an effect comparable to that produced by 30nA CGP 35348. NP JfiS-E
6
G.
478
H.
FROMM er al.
(A) CONTROL N = 29
CGP-35348 + k-BCF
C-BCF
CGP-35348 N = 27 IIL
N = 21
N = 16
US
3
6
3
6
3
-7-h
6
N= 16
N--t6
.l--
I
6
3
6
3
3
6
MSEC
CGP-35348 + GABA
I
CONTROL
CGP-35348
N = 30
I
N-15
N = 31
3
6
N=t6
N-16
3
GABA
6
6
I_ 3
N=O
-L_ 3
3
N=17
6
3
US
6
N= 12
6
i’
3
MSEC Fig. 2. Poststimulus histograms of the responses of a neuron in the trigeminal nucleus oralis, recorded as in Fig, 1. (A) Records obtained before administration of drug (CONTROL); during the iontophoretic application of 30 nA CGP 35348 (CGP 35348); during the iontophoretic application of 15 nA L-baclofen (L-BCF); and the iontophoretic application of 30 nA CGP 35348 + 15 nA L-baclofen (CGP 35348 + LBCF). (B) Records obtained before administration of drug (CONTROL); during the iontophoretic application of 30 nA CGP 35348 (CGP 35348); during iontophoretic application of lOOnA GABA (GABA); and the iontophoretic application of 30nA CGP 35348 + tO0 nA GABA (CGP35348 + GABA). Like L-baclofen, GABA depressed the neuronal response to the unconditioned stimulus (US) and enhanced segmental inhibition (SI). The CGP 35348 also blocked these actions of GABA but was less effective than in blocking L-baclofen.
6
GABA, antagonists
small effect on segmental inhibition suggests that GABA, receptors are normally only weak modulators of neuronal activity in the trigeminal nucleus, in agreement with the observations of Karlsson and Olpe (1989) that GABA, receptors appear to be only weak modulators of hippocampal pyramidal neuronal activity, when activated by endogenous GABA. Similarly, Dutar and Nicoll (1988) found that the GABA,-mediated inhibitory postsynaptic potential (IPSP) in hippocampal pyramidal cells, was evoked only at high stimulus intensities. Since postsynaptic GABA, responses are more readily evoked in the dendrites (Nicoll, Malenka and Kauer, 1990), it could also be that the CGP 35348 did not reach these receptors in an adequate concentration, to block endogenously released GABA, since it was applied near the soma in these experiments. Segmental inhibition in the trigeminal nucleus appears to be mainly mediated by GABA, receptors, since it has been found (Nakata et al., 1991) that it was markedly depressed by the iontophoretic administration of bicuculline, as well as being enhanced by the GABA, agonists, muscimol and THIP (4,5,6,7tetrahydroisoxazolo-5,4-C-pyridine-3-01). This is probably why CGP 35348 only partially blocked the GABA-induced facilitation of segmental inhibition at doses that almost completely blocked the L-baclofeninduced facilitation (Figs 2 and 3). The paired-pulse
479
inhibition of hippocampal CA1 pyramidal cells is likewise mediated predominantly by GABA, receptors (Karlsson and Olpe, 1989). The depression of the initial response to the unconditioned stimulus in the trigeminal nucleus by Lbaclofen and by GABA is presumably mediated primarily by presynaptic GABA, receptors, decreasing the release of excitatory amino acids (Ault and Evans, 1981; Davies, 1981; Edwards, Harrison, Jack and Kullman, 1989; Fox, KrnjeviC, Morris, Puil and Werman, 1978; Johnston, Hailstone and Freeman, 1980; Olpe, Baudry, Fagni and Lynch, 1982; Potashner, 1979). There is a large concentration of GABA, sites in the spinal trigeminal tract (Bowery, Hudson and Price, 1987) and it was found that CGP 35348 blocked the action of GABA on the initial response to the unconditioned stimulus as effectively as that of L-baclofen. Furthermore, the GABA, agonists muscimol and THIP did not appear to affect the initial neuronal response to the unconditioned stimulus at the concentrations at which they facilitated segmental inhibition (Nakata et al., 1991). On the other hand, the fact that CGP 34358 blocked the GABAergic depression of the late discharge, less effectively than the L-baclofen-induced depression of this afterdischarge (Fig. 2B), suggests that this effect was mediated by both GABA, and GABA, receptors, in agreement with the previous
12-
10 -
6-
:
i E
IT F
Fig. 3. Graph showing the change in the mean latency of the response to electrical stimulation of the face, after a conditioning stimulus to the face, of 8 neurons in the trigeminal nucleus oralis, produced by: (A) iontophoretic application of L-baclofen; (B) iontophoretic application of 30 nA CGP 35348 + L-baclofen; (C) iontophoretic application of 60 nA CGP 35348 + L-baclofen; (D) iontophoretic application of GABA; (E) iontophoretic application of 30 nA CGP 35348 + GABA; (F) iontophoretic application of 60 nA CGP 35348 + GABA. The changes in mean latency are expressed as a ratio of the control value and the vertical lines indicate the standard deviation of the mean. The CGP 35348 was significantly less effective in blocking the GABA-induced, than the L-baclofen-induced facilitation, of segmental inhibition (SI) (U = 12.0, P = 0.019 for 30nA CGP 35348; and U = 12.5, P = 0.025 for 60 nA CGP 35348 by the Mann-Whitney U-test).
G. H. FP:ouu et 01.
480
observation that this late discharge was also reduced by muscimol and enhanced by bicuculline (Fromm and Terrence, 1984). Acknowledgements-We thank Professor N. G. Bowery and Dr H.-R. Glpe for critical review of the manuscript, Joyce Horner for technical assistance and Drs W. Froestl and H. Bittiger, Ciba-Geigy, for the CGP 35348 and the L-baclofen. This work was supported in part by a grant (NS 19889) from the National Institutes of Health.
REFERENCES
Ault R. and Evans R. H. (1981) The depressant action of baclofen on the isolated spinal cord of the neonatal rat. Eur. J. Pharmac. 11: 351-364. Bowery N. G., Hudson A. L. and Price G. W. (1987) GABA, and GABA, receptor site distribution in the rat central nervous system. Neuroscience 20: 365-383. Curtis D. R., Gynther B. D., Beattie D. T., Kerr D. I. B. and Prager R. H. (1988) Baclofen antagonism by 2-hydroxy-saclofen in the cat spinal cord. Neurosri. Lerr. 92: 97-101. Davies J. (1981) Selective depression of synaptic excitation in cat spinal neurones by baclofen: an iontophoretic study. Br. J. Pharmac. 12: 373-384. Dutar P. and Nicoll R. A. (1988) A physiological role for GABA, receptors in the central nervous system. Nature 332: 156-158. Edwards F. R., Harrison P. J., Jack J. J. B. and Kullmann D. M. (1989) Reduction by baclofen of monosynaptic EPSPs in lumbosacral motoneurones of the anaesthetized cat. J. Physiol. 416: 539-556. Fox S., Krnjevic K., Morris M. E., Puil E. and Werman R. (1978) Action of baclofen on mammalian synaptic transmission. Neuroscience 3: 495-5 15.
Fromm G. H., Chattha A. S.. Terrence C. F. and Glass J. D. (1981) Role of inhibitory mechanisms in trigeminal neuralgia. Neurology 31: 683497. Fromm G. H., Shibuya T.. Nakata M. and Terrence C. F. (1990) Effects of o-baclofen and L-baclofen on the trigeminal nucleus. Neuropharmrrcology 29: 249-254.
Fromm G. H. and Terrence C. F. (1984) What is the role of GABA in the mechanism of action of anticonvulsant drugs? In: Neurotransmitters, Seizures, and Epilepsy, II (Fariello R. G., Morselli P. L., Lloyd K. G., Quesney L. F. and Engel J. Jr, Eds), pp. 81-91. Raven Press, New York. Fromm G. H., Terrence C. F., Chattha A. S. and Glass J. D. (1980) Baclofen in trigeminal neuralgia: its effect on the spinal trigeminal nucleus: a pilot study. Archs Neural. 37: 768-771. Johnston G. A. R., Hailstone M. H. and Freeman C. G. (1980) Baclofen: stereoselective inhibition of excitant amino acid release. J. Pharm. Pharmac. 32: 230-23 1. Karlsson G. and Olpe H.-R. (1989) Inhibitory processes in normal and epileptic-like rat hippocampal slices: the role of GABA, receptors. Eur. J. Pharmac. 163: 285-290. Kerr D. I. B., Ong J., Johnston G. A. R., Abbenante J. and Prager R. H. (1988) 2-hydroxy-saclofen: an improved antagonist at central and peripheral GABA, receptors. Neurosci. Leti. 92: 92-96. Lambert N. A., Harrison N. L., Kerr D. 1. B., Ong J., Prager R. H. and Teyler T. J. (1989) Blockade of the late IPSP in rat CAI hippocampal neurons by 2-hydroxy-saclofen. Neurosci. Lett. 107: 125-128. Nakata M., Mukawa J. and Fromm G. H. (1991) Why are most GABA agonists not effective antiepileptic medications? Jap. J. Psych&. Neural. 45: 391-394. Nicoll R. A., Malenka R. C. and Kauer J. A. (1990) Functional comparison of neurotransmitter receptor subtypes in mammalian central nervous system. Physiol. Rec. 70: 513-565. Olpe H.-R., Baudry M., Fagni L. and Lynch G. (1982) The blocking action of baclofen on excitatory transmission in the rat hippocampal slice. J. Neurosci. 2: 698-703. Olpe H.-R.; _Karl&on G., Pozza M. F.. Brugger F.. Steinmann M.. Van Riezen H.. Faae G.. Hall R. G.. Froestl W. and Bittiger H. (1990) CGI35348: a centrally active blocker of GABA, receptors. Eur. J. Pharmac. 187: 27-38. Paxinos G. and Watson C. (1982) The Rut Brain in Stereotoxic Coordinates. Academic Press, Sydney. Potashner S. J. (1979) Baclofen: effects on amino acid release and metahlism in slices of guinea pig cerebral cortex. J. Neurochem. 32, 103 -109.
1992 Copyright
0
OOZS-3908/92 S5.00 + 0.00 I992 Pergamon Press plc
THE ACTION OF GABA, ANTAGONISTS IN THE TRIGEMINAL NUCLEUS OF THE RAT G. H. FROMM, K. SATO* and M. NAKATA? Department of Neurology, University of Pittsburgh, School of Medicine, Pittsburgh, PA 15261, U.S.A. (Accepred
7 November
1991)
iontophoretic administration of the GABA, antagonists (P-(3-aminopropyl)-P-diethoxymethylphosphinic acid (CGP 35348) and 2-hydroxy-saclofen blocked the action of iontophoretically applied L-baclofen on neurons in the trigeminal nucleus of rats, anesthetized with halothane. The Summary-The
substance CGP 35348 appeared to be more potent than 2-hydroxy-saclofen. The iontophoretic administration of GABA resembled L-baclofen in depressing excitatory transmission and facilitating segmental inhibition in the trigeminal nucleus. The depression of excitatory transmission was also blocked by CGP 35348 and the facilitation of segmental inhibition produced by GABA was partially blocked. These observations indicate that CGP 35348 is not only a baclofen antagonist but actually a GABA, receptor antagonist and the baclofen was acting at GABA, receptors in the trigeminal nucleus. The portion of the effect of GABA, not blocked by CGP 35348, was probably mediated by GABA, receptors, since it was previously found that segmental inhibition in the trigeminal nucleus could be modulated by GABA, agonists and antagonists as well. Kq words-baclofen, nucleus.
GABA,
GABA,
receptor
antagonists,
CGP 35348, 2-hydroxy-saclofen,
trigeminal
METHODS
It was previously found that neuronal activity in the spinal trigeminal nucleus could be modulated by the y-aminobutyric acid (GABA,) agonist, baclofen (Fromm, Terrence, Chattha and Glass, 1980; Fromm, Chattha, Terrence and Glass, 198 1; Fromm, Shibuya, Nakata and Terrence, 1990), as well as by GABA, agonists and antagonists (Fromm and Terrence, 1984; Nakata, Mukawa and Fromm, 1991). In the absence of confirmation by a GABAa antagonist, the question remained though as to whether baclofen actually exerts its action in the trigeminal nucleus at GABAs receptors. It has now been demonstrated that CGP 35348 (Olpe, Karlsson, Pozza, Brugger, Steinmann, Van Riezen, Fagg, Hall, Froestl and Bittiger, 1990) and 2-hydroxy-saclofen (Curtis, Gynther, Beattie, Kerr and Prager, 1988; Kerr, Ong, Johnston, Abbenante and Prager, 1988; Lambert, Harrison, Kerr, Ong, Prager and Teyler, 1989) are selective blockers of GABA, receptors, with CGP 35348 being the more potent. Therefore. the effect of these GABAa antagonists on the action of L-baclofen and of GABA on single neurons in the trigeminal nucleus was investigated in order to ascertain whether L-baclofen was indeed acting at GABA, receptors in the experimental model used.
The experiments were performed on 15 SpragueDawley rats, weighing 225-300 g, anesthetized with halothane, 3% for induction, 2% for surgery and 1% for recording, in 30-50% 0,/70-50% Nr. The rats were paralyzed with pancuronium bromide and artificially ventilated. Blood pressure, rectal temperature and blood gases were monitored throughout the experiment and maintained within physiological limits. The activity of single neurons in the spinal trigeminal nucleus oralis was recorded extracellularly with a glass micropipette (tip dia = 1.0-1.5 pm), glued along a five-barrelled micropipette (outer tip dia = 8-lOpm), so that the single tip of the barrel protruded 3-7pm beyond the tip of the five-barrel electrode (Fromm et al., 1990). A balancing channel was employed to alleviate current effects; this barrel and the recording micropipette were filled with 2 M NaCl, saturated with Fast Green-FCF. The four remaining barrels contained 0.01 M t-baclofen in 0.15 M NaCl, pH = 5.0 (kindly supplied by CibaGeigy), 0.01 M CGP 35348 (P-(3-aminopropyl)-Pdiethoxymethyl-phosphinic acid) in 0.15 M NaCI, pH = 6.7 (kindly supplied by Ciba-Geigy), 0.01 M 2-hydroxy-saclofen in 0.15 M NaCl, pH = 3 (Tokris) or 0.5 M GABA, pH = 3.5-4.0 (Sigma). Ejection, retention and balancing currents were produced by a constant-current source (Medical Systems Model BH-2). Stimuli were applied to the facial skin at 0.5 Hz with a pair of needle electrodes, using monophasic
*Present address: Department of Neuropsychiatry, Okayama University Medical School, 2-5-l Shikata-cho, Okayama 700, Japan. YPresent address: Department of Neurosurgery, University of the Ryukyus, Nishihara Uehara 207, Okinawa 903-01, Japan. 475
G. H. FROMM~~
476
squarewave pulses (2.2-7.7 V, 0.1 msec). The stimulus strength was adjusted to obtain the shortest possible latency of response for each neuron. Segmental or afferent inhibition was elicited by delivering a 100 msec train of conditioning stimuli (2.2-5.5 mA, 0.3 msec, 50 Hz) 100 msec before the test stimulus. Responses of neurons were monitored before, during and after the iontophoretic administration of the drugs on a cathode ray oscilloscope and stored on magnetic tape. Poststimulus histograms were obtained with a Neurolog NL 750/755 Signal Averager. The latency of the initial neuronal spike was measured with a time-interval measurement system (Quasitronics, Houston, Pennsylvania) and the mean latency and SD of 10 consecutive unconditioned responses and of 10 consecutive responses each following a conditioning train of stimuli was calculated. A value of 20 msec was assigned when the neuron did not respond at all. The location of the recording electrode was marked at the end of the experiments by passing a direct current of IO PA for 10 min through the electrode and then checked histologically, after staining with hematoxylin and eosin. RESULTS
The neurons used in this study were in the spinal trigeminal nucleus oralis, 0.9-2.6 mm posterior to the interaural line and 2.2-3.1 mm lateral to the midline (Paxinos and Watson, 1982) and were located by electrical stimulation of the face at 1 Hz, while slowly advancing the microelectrode. They responded to an unconditioned stimulus, with a mean latency of 2.2 msec (SD = 0.34). The segmental inhibition, elicited by delivering a conditioning train of stimuli 100 msec before the test stimulus, increased the mean latency of the response by 0.23 msec (SD = 0.10) and decreased the number of spikes if the neuron responded to each unconditioned stimulus with more than one spike (Figs 1 and 2). The iontophoretic administration of 60 nA CGP 35348 decreased the mean latency of the conditioned response in 10 of 20 neurons by 0.06-0.20 msec and this change was statistically significant in 7 (P < 0.05 by the Wilcoxon matched-pairs signed-ranks test). The administration of 30 nA CGP 35348 decreased the latency of the conditioned response in only 3 of the 20 neurons, with the change being smaller (0.0220.10 msec) and statistically significant in only 1 neuron. The iontophoretic administration of up to 200 nA 2-hydroxysaclofen decreased the mean latency of the conditioned response in only 1 of 5 neurons. As in the previous experiments (Fromm et al., 1990) the iontophoretic administration of 15-30 nA L-baclofen depressed the response of the neurons of the trigeminal nucleus to electrical stimulation of the face, increasing the mean latency of the initial spike and decreasing the total number of spikes (Figs 1 and 2). The iontophoresis of r.-baclofen also markedly
al.
enhanced the segmental inhibition, further increasing the mean latency of the initial spike and decreasing the number of spikes. The concomitant iontophoresis of 30-60 nA CGP 35348 (starting 1 min before the L-baclofen was applied), blocked the effect of Lbaclofen on both the unconditioned stimulus and segmental inhibition. Similarly, the iontophoresis of 2-hydroxy-saclofen blocked the action of iontophoretically applied L-baclofen on both the unconditioned stimulus and segmental inhibition. However, larger currents (100-200 nA) were required for 2-hydroxysaclofen to have a comparable effect (Fig. IB). The action of both CGP 35348 and 2-hydroxy-saclofen was tested in 4 neurons, while in I neuron only 2-hydroxy-saclofen could be tested and in 13 neurons, only CGP 35348 was tested. The iontophoretic administration of GABA resembled the action of L-baclofen in the trigeminal nucleus (Fig. 2). The concomitant iontophoresis of 30-60 nA CGP 35348 (starting I min before the GABA was applied) also blocked the effects of GABA. However, these doses of CGP 35348 were less effective in blocking the action of GABA than in blocking that of L-baclofen. The concomitant administration of 30-60 nA CGP 35348 almost completely blocked the facilitation of segmental inhibition produced by L-baclofen but only partially blocked the GABA-induced enhancement of segmental inhibition (Figs 2B and 3). The difference between the action of CGP 35348, against the effect of L-baclofen and against the effect of GABA on segmental inhibition was significant for both 30 nA and 60 nA CGP 35348 (U = 12.0, P = 0.019; and I/ = 12.5. P = 0.025. respectively. by the Mann-Whitney U-test). Moreover, while CGP 35348 only partially prevented the inhibition of the late discharge, induced by baclofen; it did not block the GABAergic depression of the late discharge at all (Fig. 2B). DISCUSSION
These experiments demonstrated that the actions of L-baclofen on single neurons in the trigeminal nucleus were blocked by the GABA, receptor antagonists CGP 35348 and 2-hydroxy-saclofen. The substance CGP 35348 appeared to be more effective than 2-hydroxy-saclofen as judged by the strength of the ejection current required to elicit an effect. However, the amount of current is, at best, only a relative measure of the amount of drug being delivered. It was also found that the iontophoretic administration of GABA resembled r.-baclofen in depressing excitatory transmission and facilitating segmental inhibition in the trigeminal nucleus and these actions were also blocked by CGP 35348. These observations indicate that CGP 35348 is not only a baclofen antagonist but a GABA, receptor antagonist and that baclofen was acting at GABA, receptors in the trigeminal nucleus. That CGP 35348 blocked the action of exogenously applied GABA or L-baclofen but had only a
GABA, antagonists
(4 CGP-35348
CONTROL
t
I
N = 16
cap-35348 -i-l
-BCF N=3
i
N= 16
@I SCF
CONTROL
SCF + l’-BCF
1-BCF
N=f6
N254
A . 3
8
3
3
6
N=O
N= 16
N=16
6
N=9
St
i
fr 3
6
3
Fig. 1. Poststimulus histograms (16 stimuli at 0.5 Hz, 0.1 msec bin width) of the response of a neuron in the trigeminal nucleus oralis. to an unconditioned stimulus (US) and to the same stimulus delivered 100 msec after a train of conditioning stimuli (Sl). Iv = total number of spikes. (A) Records obtained before administration of drug (CONTROL); during the iontophoretic application of 30 nA CGP-35348 (CGP 35348); during the jontophoretj~ apptication of 15 nA L-baciofen ft-BCF); and the iontopboretic application of 30 nA CGP 35348 + I5 nA L-baciofen (CGP 35348 i- L-BCF). (3) Records obtained before administration of drug (CONTROL); during the iontophoretic application of 200 nA 2-hydroxy-saclofen (SCP); during the iontophoretic application of IS nA r.-baclofen (L-BCF); and the iontophoretic application of 200 nA 2-hydroxy-saclofen + IS nA L-bactofen (S&F + L-BCF). L-Baclofen depressed the neuronal response to the unconditioned stimulus (US) and enhanced the segmental inhibition @I). CGP 35348 Blocked the effect of t,-baclofen on both the unconditioned stimulus and segmental inhibition. 2-Hydroxy-saciofen also blocked the action of t-baclofen but an ejection current of 200 nA was required to produce an effect comparable to that produced by 30nA CGP 35348. NP JfiS-E
6
G.
478
H.
FROMM er al.
(A) CONTROL N = 29
CGP-35348 + k-BCF
C-BCF
CGP-35348 N = 27 IIL
N = 21
N = 16
US
3
6
3
6
3
-7-h
6
N= 16
N--t6
.l--
I
6
3
6
3
3
6
MSEC
CGP-35348 + GABA
I
CONTROL
CGP-35348
N = 30
I
N-15
N = 31
3
6
N=t6
N-16
3
GABA
6
6
I_ 3
N=O
-L_ 3
3
N=17
6
3
US
6
N= 12
6
i’
3
MSEC Fig. 2. Poststimulus histograms of the responses of a neuron in the trigeminal nucleus oralis, recorded as in Fig, 1. (A) Records obtained before administration of drug (CONTROL); during the iontophoretic application of 30 nA CGP 35348 (CGP 35348); during the iontophoretic application of 15 nA L-baclofen (L-BCF); and the iontophoretic application of 30 nA CGP 35348 + 15 nA L-baclofen (CGP 35348 + LBCF). (B) Records obtained before administration of drug (CONTROL); during the iontophoretic application of 30 nA CGP 35348 (CGP 35348); during iontophoretic application of lOOnA GABA (GABA); and the iontophoretic application of 30nA CGP 35348 + tO0 nA GABA (CGP35348 + GABA). Like L-baclofen, GABA depressed the neuronal response to the unconditioned stimulus (US) and enhanced segmental inhibition (SI). The CGP 35348 also blocked these actions of GABA but was less effective than in blocking L-baclofen.
6
GABA, antagonists
small effect on segmental inhibition suggests that GABA, receptors are normally only weak modulators of neuronal activity in the trigeminal nucleus, in agreement with the observations of Karlsson and Olpe (1989) that GABA, receptors appear to be only weak modulators of hippocampal pyramidal neuronal activity, when activated by endogenous GABA. Similarly, Dutar and Nicoll (1988) found that the GABA,-mediated inhibitory postsynaptic potential (IPSP) in hippocampal pyramidal cells, was evoked only at high stimulus intensities. Since postsynaptic GABA, responses are more readily evoked in the dendrites (Nicoll, Malenka and Kauer, 1990), it could also be that the CGP 35348 did not reach these receptors in an adequate concentration, to block endogenously released GABA, since it was applied near the soma in these experiments. Segmental inhibition in the trigeminal nucleus appears to be mainly mediated by GABA, receptors, since it has been found (Nakata et al., 1991) that it was markedly depressed by the iontophoretic administration of bicuculline, as well as being enhanced by the GABA, agonists, muscimol and THIP (4,5,6,7tetrahydroisoxazolo-5,4-C-pyridine-3-01). This is probably why CGP 35348 only partially blocked the GABA-induced facilitation of segmental inhibition at doses that almost completely blocked the L-baclofeninduced facilitation (Figs 2 and 3). The paired-pulse
479
inhibition of hippocampal CA1 pyramidal cells is likewise mediated predominantly by GABA, receptors (Karlsson and Olpe, 1989). The depression of the initial response to the unconditioned stimulus in the trigeminal nucleus by Lbaclofen and by GABA is presumably mediated primarily by presynaptic GABA, receptors, decreasing the release of excitatory amino acids (Ault and Evans, 1981; Davies, 1981; Edwards, Harrison, Jack and Kullman, 1989; Fox, KrnjeviC, Morris, Puil and Werman, 1978; Johnston, Hailstone and Freeman, 1980; Olpe, Baudry, Fagni and Lynch, 1982; Potashner, 1979). There is a large concentration of GABA, sites in the spinal trigeminal tract (Bowery, Hudson and Price, 1987) and it was found that CGP 35348 blocked the action of GABA on the initial response to the unconditioned stimulus as effectively as that of L-baclofen. Furthermore, the GABA, agonists muscimol and THIP did not appear to affect the initial neuronal response to the unconditioned stimulus at the concentrations at which they facilitated segmental inhibition (Nakata et al., 1991). On the other hand, the fact that CGP 34358 blocked the GABAergic depression of the late discharge, less effectively than the L-baclofen-induced depression of this afterdischarge (Fig. 2B), suggests that this effect was mediated by both GABA, and GABA, receptors, in agreement with the previous
12-
10 -
6-
:
i E
IT F
Fig. 3. Graph showing the change in the mean latency of the response to electrical stimulation of the face, after a conditioning stimulus to the face, of 8 neurons in the trigeminal nucleus oralis, produced by: (A) iontophoretic application of L-baclofen; (B) iontophoretic application of 30 nA CGP 35348 + L-baclofen; (C) iontophoretic application of 60 nA CGP 35348 + L-baclofen; (D) iontophoretic application of GABA; (E) iontophoretic application of 30 nA CGP 35348 + GABA; (F) iontophoretic application of 60 nA CGP 35348 + GABA. The changes in mean latency are expressed as a ratio of the control value and the vertical lines indicate the standard deviation of the mean. The CGP 35348 was significantly less effective in blocking the GABA-induced, than the L-baclofen-induced facilitation, of segmental inhibition (SI) (U = 12.0, P = 0.019 for 30nA CGP 35348; and U = 12.5, P = 0.025 for 60 nA CGP 35348 by the Mann-Whitney U-test).
G. H. FP:ouu et 01.
480
observation that this late discharge was also reduced by muscimol and enhanced by bicuculline (Fromm and Terrence, 1984). Acknowledgements-We thank Professor N. G. Bowery and Dr H.-R. Glpe for critical review of the manuscript, Joyce Horner for technical assistance and Drs W. Froestl and H. Bittiger, Ciba-Geigy, for the CGP 35348 and the L-baclofen. This work was supported in part by a grant (NS 19889) from the National Institutes of Health.
REFERENCES
Ault R. and Evans R. H. (1981) The depressant action of baclofen on the isolated spinal cord of the neonatal rat. Eur. J. Pharmac. 11: 351-364. Bowery N. G., Hudson A. L. and Price G. W. (1987) GABA, and GABA, receptor site distribution in the rat central nervous system. Neuroscience 20: 365-383. Curtis D. R., Gynther B. D., Beattie D. T., Kerr D. I. B. and Prager R. H. (1988) Baclofen antagonism by 2-hydroxy-saclofen in the cat spinal cord. Neurosri. Lerr. 92: 97-101. Davies J. (1981) Selective depression of synaptic excitation in cat spinal neurones by baclofen: an iontophoretic study. Br. J. Pharmac. 12: 373-384. Dutar P. and Nicoll R. A. (1988) A physiological role for GABA, receptors in the central nervous system. Nature 332: 156-158. Edwards F. R., Harrison P. J., Jack J. J. B. and Kullmann D. M. (1989) Reduction by baclofen of monosynaptic EPSPs in lumbosacral motoneurones of the anaesthetized cat. J. Physiol. 416: 539-556. Fox S., Krnjevic K., Morris M. E., Puil E. and Werman R. (1978) Action of baclofen on mammalian synaptic transmission. Neuroscience 3: 495-5 15.
Fromm G. H., Chattha A. S.. Terrence C. F. and Glass J. D. (1981) Role of inhibitory mechanisms in trigeminal neuralgia. Neurology 31: 683497. Fromm G. H., Shibuya T.. Nakata M. and Terrence C. F. (1990) Effects of o-baclofen and L-baclofen on the trigeminal nucleus. Neuropharmrrcology 29: 249-254.
Fromm G. H. and Terrence C. F. (1984) What is the role of GABA in the mechanism of action of anticonvulsant drugs? In: Neurotransmitters, Seizures, and Epilepsy, II (Fariello R. G., Morselli P. L., Lloyd K. G., Quesney L. F. and Engel J. Jr, Eds), pp. 81-91. Raven Press, New York. Fromm G. H., Terrence C. F., Chattha A. S. and Glass J. D. (1980) Baclofen in trigeminal neuralgia: its effect on the spinal trigeminal nucleus: a pilot study. Archs Neural. 37: 768-771. Johnston G. A. R., Hailstone M. H. and Freeman C. G. (1980) Baclofen: stereoselective inhibition of excitant amino acid release. J. Pharm. Pharmac. 32: 230-23 1. Karlsson G. and Olpe H.-R. (1989) Inhibitory processes in normal and epileptic-like rat hippocampal slices: the role of GABA, receptors. Eur. J. Pharmac. 163: 285-290. Kerr D. I. B., Ong J., Johnston G. A. R., Abbenante J. and Prager R. H. (1988) 2-hydroxy-saclofen: an improved antagonist at central and peripheral GABA, receptors. Neurosci. Leti. 92: 92-96. Lambert N. A., Harrison N. L., Kerr D. 1. B., Ong J., Prager R. H. and Teyler T. J. (1989) Blockade of the late IPSP in rat CAI hippocampal neurons by 2-hydroxy-saclofen. Neurosci. Lett. 107: 125-128. Nakata M., Mukawa J. and Fromm G. H. (1991) Why are most GABA agonists not effective antiepileptic medications? Jap. J. Psych&. Neural. 45: 391-394. Nicoll R. A., Malenka R. C. and Kauer J. A. (1990) Functional comparison of neurotransmitter receptor subtypes in mammalian central nervous system. Physiol. Rec. 70: 513-565. Olpe H.-R., Baudry M., Fagni L. and Lynch G. (1982) The blocking action of baclofen on excitatory transmission in the rat hippocampal slice. J. Neurosci. 2: 698-703. Olpe H.-R.; _Karl&on G., Pozza M. F.. Brugger F.. Steinmann M.. Van Riezen H.. Faae G.. Hall R. G.. Froestl W. and Bittiger H. (1990) CGI35348: a centrally active blocker of GABA, receptors. Eur. J. Pharmac. 187: 27-38. Paxinos G. and Watson C. (1982) The Rut Brain in Stereotoxic Coordinates. Academic Press, Sydney. Potashner S. J. (1979) Baclofen: effects on amino acid release and metahlism in slices of guinea pig cerebral cortex. J. Neurochem. 32, 103 -109.