- Email: [email protected]
Pipequaline acts as a partial agonist of benzodiazepine receptors: An electrophysiological study in the hippocampus of the rat
Neuropharmacology Vol. 26, No. 9, pp. 1337-1342, Printed in Great Britain. All rights reserved
1987 Copyright 0
0028-3908/87 $3.00 + 0.00 1987 Pergamon Journals Ltd
PIPEQUALINE ACTS AS A PARTIAL AGONIST OF BENZODIAZEPINE RECEPTORS: AN ELECTROPHYSIOLOGICAL STUDY IN THE HIPPOCAMPUS OF THE RAT G. DEBONNEL* and C. DE MONTIGNY Institut P. Pine1 de Montreal and Centre de recherche en sciences neurologiques, Faculte de m&mine, Universitt de Montreal, C.P. 6128, Succursale “A”, Montreal, Qu&ec, Canada H3C 357 (Accepted 26 February 1987) Summary-Pipequaline (PK 8169, a quinoline derivative and a ligand of the benzodiazepine binding site, is a clinically-effective anxiolytic, which is devoid of sedative and anticonvulsant properties. Several biochemical and behavioral studies have indicated that this molecule shares some properties with both agonists and antagonists of benzodiaxepine receptors. The present in vivo electrophysiological studies were undertaken to determine the effects of microiontophoretic applications and of intravenous injections of pipequaline on hippocampal pyramidal neurons, activated by kainate, glutamate or acetylcholine and to characterize the effects of pipequaline on the action of benzodiaxepines. Intravenously administered pipequaline exerted a partial suppression of activations by kainate, glutamate and acetylcholine. Microiontophoretic applications of pipequaline reduced the neuronal activation by kainate. This effect was blocked by RO 15-1788. In small intravenous doses, pipequaline potentiated the effect of microiontophoretically-applied flurazepam whereas, in larger doses, it suppressed the effects of microiontophoretically-applied fluraxepam and of intravenously administered lorazepam on kainateinduced activation. Similarly, microiontophoretic applications of pipequaline blocked the suppressant effect of microiontophoretically-applied fluraxepam on kainate-induced activation. These results constitute further evidence that the selective anxiolytic activity of pipequaline might be ascribed to its partial agonistic action on benzodiazepine receptors. Key words: pipequaline (PK 8169, benzodiaxepine receptors, hippocampal pyramidal neurons, kainate.
Pipequaline, a phenylquinoline derivative, exerts an anticonflict effect in the rat, without inducing any sedation (Le Fur, Mizoule, Burgevin, Ferris, Heaulme, Gauthier, Gueremy and Uzan, 1981). Several clinical studies have confirmed the anxiolytic effect of pipequaline and its lack of sedative property in humans (Poggioli, Bonnet and Von Frenckell, 1985; Von Frenckell, Ansseau and Bonnet, 1986; Willer, Von Frenckell, Bonnet and Le Fur, 1986; Bradwejn and de Montigny, unpublished observations). However, pipequaline appears to be devoid of any “anxiolytic” activity in the social interaction model in the rat (File and Lister, 1983) and to decrease motor activity in the rat (File, 1983); effects which are not blocked by the benzodiazepine antagonist ethyl 8-fluoro-5,6-dihydro-5-methyl-6-oxo4H-imidazo[l,S-a][l,4]benzodiazepine-3-carboxylate (RO 15-1788) (File and Pellow, 1984). That the pharmacological profile of this molecule clearly differs from that of benzodiazepines is further indicated by its ability to protentiate the effects of sub-convulsant doses of picrotoxin, pentetrazol and 1,3-dihydro-5-methyl-2H-1,4-benzodiazepine-2-one (RO 5-3663) (File, 1984; File and Simmons, 1984; File and Wilks, 1985) and to reduce the protective *Address all correspondence
to this author.
effect of diazepam against bicuculine-induced seizures (Gee, Brinton and Yamamura, 1983; Gee and Yamamura, 1983), although it does not potentiate the proconvulsant activity of 7-chloro-1,3-dihydrol-methyl-5-(p-chlorophenyl)2H-l,Cbenzodiazepine2-one (RO 5-4864) (Pellow and File, 1985). Radioligand binding studies in vitro have shown that pipequaline displaces [‘Hldiazepam and [3H]flunitrazepam in membranes from the brain of the rat (Le Fur et al., 1981; Le Fur, 1982; Gee et al., 1983; Gee and Yamamura, 1983; Benavides, Malgouris, Flamier, Tur, Quarteronet, Begassat, Camelin, Uzan, Gueremy and Le Fur, 1984; Keane, Simiand and Morre, 1984). The binding of pipequaline in the cerebral cortex and in the cerebellum is enhanced by y-aminobutyric acid (GABA) but its GABA ratio (“GABA-shift”) falls between that of agonists and antagonists (Morelli, Gee and Yamamura, 1982; Skolnick, Schweri, Williams, Moncada and Paul, 1982; Gee et al., 1983; Wood, Loo, Braunwalder, Yokoyama and Cheney, 1984), suggesting that it may act as a partial agonist at the benzodiazepine receptor site. Furthermore, the failure of pipequaline to decrease levels of cGMP in the cerebellar cortex suggests that some of its pharmacological effects could be independent of GABA (Le Fur et al., 1981; Benavides et al., 1984).
1337
G.
1338
DEBONNELand
In the light of these behavioral and biochemical data, it was of interest to determine electrophysiologically in z&o the effect of pipequaline on be~odia~pine receptors, in an attempt to further our understanding of the clinical anxiolytic property of this molecule. It has previously been reported that benzodiazepines antagonize the activation of pyramidal neurons induced by kainate in the CA, region of the dorsal hippocampus in the rat through activation of benzodiazepine receptors (Debonnel and de Montigny, 1983; de Montigny, Debonnel and Tardif, 1983). This paradigm was used in the present study to determine the effect of pipequaline on the firing activity of pyramidal neurons in the hippocampus and to study its interactions with betrzodiazepines. METHODS
Male Sprague-Dawley rats (200-300 g) were anesthetized with urethane (1.25 g/kg, i.p.) and were mounted in a stereotaxic apparatus. An indwelling femoral catheter was installed for intravenous injections. Five-ba~elIed glass micropipettes were prepared in a conventional manner (Haigler and Aghajanian, 1974) and their tips broken to 8-10 PM under microscopic control. The central barrel, used for extracellular recording of hippocampal pyramidal neurons, was filled with a 2 M NaCl solution saturated with Fast Green (FCF). A deposit of Fast Green was left at the last recording site at the end of each experiment for subsequent histological verification on cryostat sections. One side barrel, used for automatic current balancing, was filled with a 2 M NaCl solution. The remaining barrels, used for microiontophoresis, were filled with three of the following solutions: kainate 1 mM in 400 mM NaCl, pH: 8; glutamate. NC1 50 mM in 50 mM NaCl, pH: 8; acetylcholine. Cl (ACh) 20 mM in 20 mM NaCl, pH: 4; pipequaline 10 mM, pH: 4; flurazepam .2 HCl IOmM, pH: 3. Extracellular recordings of unit activity were obtained in the CA1 region of the dorsal hippocampus. Pyramidal neurons were identified by their long duration (0.8-l .5 msec), large amplitude (0.5-2 mV) action potentials and by the presence of “complex spike” discharges. The signal, amplified and displayed on an oscilloscope, was fed to a differential amplitude discriminator, generating square pulses, from which integrated firing rate histograms were obtained. When a stable recording was obtained, the intensity of the currents of kainate (42.4 f 2.7 nA), glutamate (9.4 + 0.7 nA) and/or ACh (30.2 + 3.2 nA) were adjusted to induce a similar degree of sustained activation (i.e. a firing rate of lO-2OHz). The suppressant effects of microiontophoretic applications of pipequaline or flurazepam on these activations were assessed by comparing the mean firing rate of the neuron during the 50 set preceding the microiontophoretic applications to that during the 50 set application. Data were discarded if there
C.
DE MONTIGNY
was not a recovery of at least 60% after the cessation of the application. For most neurons, several applications of pipequaline or flurazepam with different currents were carried out. The results are expressed as the mean degree of suppression produced by each current used. For intravenous injections, pipequaline was dissolved in water at pH 4, and lorazepam and RO 15-1788 in 40% propylene glycol, 10% ethyl alcohol, 1.5% benzyl alcohol, 5% Na benzoate and 43.5% water, at pH4. These drugs were administered through a femoral catheter in a volume of 0.1-0.3 ml.
The suppressant effects of pipequaline, lorazepam or flurazepam were determined by comparing the degree of activation before and 3-5min after their intravenous administration. The effect of RO 15-1788, a selective benzodiazepine receptor antagonist (Hunkeler, Mohler, Pieri, Pole, Bonetti, Cumin, Shaffner and Haefely, 1981) on the activity of pipequaline, was measured by comparing the degree of suppression induced by microiontophoretic applications of pipequaline before and 3-5 min after the administration of RO 15-1788 (3.5 mg/kg, iv.). In these series of experiments, the rats received either a single dose of pipequaline, lorazepam or RO 15-1788 or an injection of pipequaline 3-5 min before administering one dose of lorazepam. In these experiments, only one cell per rat was tested. RESULTS
(1) The agonistic activity ofpipequaline (a) Intravenous injections of pipequaline. In small doses (from 0.4 to 1.2 mg/kg) pipequaline did not produce any consistent effect on activations induced by kainate, glutamate or ACh: 18 of the 26 neurons tested were not, or only slightly (C 10%) affected by these injections. At doses ranging from 6 to 10 mg/kg, pipequaline reduced activations induced by kainate, glutamate or ACh in 60% of the 75 neurons tested.
DOSE (mg/kg. i.v,) OF PK 8165
Fig. 1. Suppression of kainate @CA)-induced activation of CA, hippoeampal neurons by intravenous injections of pipequaline (PK 8165). The degree of suppression was measured 3-5 min after the injection. Note that the intervals on the X axis are not proportional to the doses given. Only one neuron was tested in each rat at each dose. The number of neurons tested is given in parentheses.
PK 8165, a partial benzodiazepine
5
10
15
20
30
MICROIONTOPHORETIC CURRENl (nA) OF PK 8165
RO 15-1788 (3.5 mglkg)
Fig. 2. Suppression of kainate-induced activation of CA, hippocampal pyramidal neurons by microiontophoretic applications of pipequaline. The degree of suppression was measured by comparing the number of spikes generated by kainate before and during 50 set applications of pipequaline. The number of neurons tested with each current is given in parentheses.
It was only at the largest doses of 12 and 15 mg/kg that pipequaline produced a clear suppressive effect on all neurons (n = 26). However, the degree of suppression obtained with the 15 mg/kg dose appeared to be smaller than that produced by the 12 mg/kg dose. The dose-response relationship for the effect of pipequaline on kainate-induced activation is illustrated in Figure 1. The effects of pipequaline on glutamate- and ACh-induced activations were similar to that on kainate-induced activation (data not shown). The effect of pipequaline was very short lasting, as there was a full recovery of the neuronal activation within 15 min after the injection, even after the large doses. (b) Microiontophoretic applications of pipequaline.
As intravenously injected pipequaline did not show any selectivity, microiontophoretic applications of the drug were performed during activation induced by kainate only; pipequaline reduced kainate-induced activation in all but three neurons (n = 44). The degree of suppression by microiontophoretic applications of pipequaline, as was the case with its
KA-7 PKg
g
g&I
20 20 20 0000~
20
20
200 ;
150
I 5
100
E
500
!
1339
RO 1$-l 760 (3.5 mg/kg, I.v.)
Fig. 3. Integrated firing rate histogram of a pyramidal neuron of the CA, region of the dorsal hippocampus, showing the suppression by pipequaline of kainate-induced activation and the block of this suppression by intravenous injection of RO 15-1788. Bars indicate the duration of microiontophoretic applications for which currents are given in nA.
Fig. 4. Suppression of kainate-induced activations of CA, hippocampal pyramidal neurons by microiontophoretic applications of pipequaline (16.5 +_1.5 nA) before and after intravenous administration of RO 15-1788 (3.5 mg/kg). The same neurons were tested before and after RO 15-1788, using the same current for pipequaline *P < 0.001 (paired Student’s r-test).
intravenous administration, was relatively small, as the maximum decrease in the firing rate in most neurons was less than 50%. There was a dose-response relationship at the smaller currents used, but the response plateaued at about 50% from 15 nA on (Fig. 2). In order to determine if this effect of pipequaline was mediated through activation of the benzodiazepine receptor, the benzodiazepine antagonist RO 15-1788 was used. The degree of suppression produced by microiontophoretic applications of pipequaline was compared before and 3-5 min after intravenous administration of RO 15-1788 (3.5mg/kg); RO 15-1788 reduced by 65% the effect of pipequaline (Figs 3 and 4). (2) The antagonistic activity of pipequaline (a) Intravenous
administration
of
pipequaline.
Microiontophoretic applications of flurazepam, with a current of 5 nA, induced a mean suppression of 30% of the firing rate of pyramidal neurons, activated by kainate. At a small dose (400pg/kg, i.v.), pipequaline markedly potentiated the effect of flurazepam which was increased by nearly 90% (Figs 5A and 6). However, in larger doses (800 and 1200 p g/kg), pipequaline markedly reduced the efficacy of flurazepam (Figs 5B and 6). To provide further evidence for the antagonistic activity of pipequaline on benzodiazepine receptors, the effect of the intravenous administration of lorazepam (500pg/kg) was compared in naive rats and in rats pretreated with a large dose of pipequaline (10 mg/kg, i.v.) administered 3-5 min prior to the injection of lorazepam. Consistent with the data presented above (Fig. l), in the present series of experiments, this dose of pipequaline produced a 20% decrease of kainate-induced activity. Lorazepam produced a 60% suppression of activation induced by kainate in naive rats. The pretreatment with pipequaline reduced the effect of lorazepam by more than 90% (Fig. 7).
G.
1340
DEBONNEL
and C.
DE
MONTICNY
KA-6
A 250,
FLU 5 0
5 n
5
5
0
0
PK 8165 IO 4 mg/kg.
B
I Y)
KA -13
FLU 5 5 c3oc1
5
5 no
5
PK 8165 (I 2 mg/kg.
1.x)
Fig. 5. Integrated firing rate histogram of CA, hippocampal pyramidal neurons showing the potentiatioin of the effect of flurazepam (FLU) on kainate-induced activation by a small dose of pipequaline (PK 8165) (A) and the blockade of the effect of flurazepam by a larger dose of pipequaline (B). (b) Microiontophoretic application of pipequaline. To ascertain whether this antagonistic activity of pipequaline resulted from a direct effect of the drug in the dorsal hippocampus, the efficacy of microiontophoretic applications of flurazepam was measured before and during microiontophoretic application of pipequaline. As illustrated in Figure 8, pipequaline antagonized the effect of flurazepam. The application of flurazepam alone produced a 60% suppression of kainate induced activation and that of pipequaline applied with similar currents, produced only a 20% suppression (Fig. 9). The effectiveness of flurazepam was reduced by 60% during concomittant application of pipequaline (Fig. 9). In these experiments, no potentiation of the effect of flurazepam was
0
BEFORE
m
AFTER PK 8165
2
microiontophoretically-applied pipequaline (PK 8165) of the effect of flurazepam (FLU) on kainate (KA)-induced activation and the recovery of the effect for flurazepam following the cessation of the application of pipequaline.
LU DOSE (mglkg)
OF PKS165
Fig. 6. Suppression of kainate-induced activation of CA, hippocampal pyramidal neurons by small current (5 nA) applications of flurazepam alone and after the intravenous administration of pipequaline (PK 8165). *P c 0.005;
FLU during PK 6165 Ez5
Fig. 9. Suppression of kainate-induced activation of CA, hippocampal pyramidal neurons by microiontophoretic applications of flurazepam (FLU) (11.5 + 2.4 nA), of pipequaline (PK 8165) (10.3 + 1.5 nA) and of flurazepam during the concomitant application of pipequaline. *P < 0.001 (two-tailed Student’s r-test comparing the degree of suppression by flurazepam alone to that produced by fiurazepam during the application of pipequaline).
PK 8165,
a partial benzodiazepine
DISCUSSION
Pipequaline decreased the firing rate of activated pyramidal cells, but was devoid of any selectivity since it exerted a similar effect on activations induced by kainate, glutamate or ACh, whereas benzodiazepines act preferentially on kainate-induced activation (Debonnel and de Montigny, 1983). This suppression by pipequaline was observed either after its intravenous administration or during microiontophoretic application. The blockade of the effect of pipequaline by RO 15-1788, a selective benzodiazepine receptor antagonist, indicates that its effect, in this model, was mediated through the activation of benzodiazepine receptors (Fig. 4). This is in keeping with previous demonstrations of a reversal of the electrophysiological and of some of the behavioral effects of pipequaline by RO 15-1788 (File and Pellow, 1984; File and Simmonds, 1984; Bradwejn and de Montigny, 1985a, b), and with the high affinity binding of pipequaline on benzodiazepine receptors in vitro (Le Fur et al., 1981; Le Fur, 1982; Gee et al., 1983; Gee and Yamamura, 1983; Benavides et al., 1984; Keane et al., 1984). The apparent discrepancy between the present results and the failure of intravenously administered pipequaline to reduce in vivo, the binding of [‘Hlflunitrazepam (Keane et al., 1984) may be due to the long time interval (30 min) between the injection of pipequaline and that of the radioligand in the latter study. Indeed in the present study, the effect of pipequaline on the benzodiazepine receptor was very short-lasting as was observed previously (Bradwejn and de Montigny, 1985a), in keeping with the extremely short halflife of pipequaline in the rat (Le Fur, personal communication). Large intravenous doses of pipequaline (12 and 15 mg/kg, i.v.) were required to obtain a consistent suppression of the firing activity of hippocampal pyramidal neurons (Fig. 1) compared to those of lorazepam and diazepam (Debonnel and de Montigny, 1983). Similarly, large microiontophoretic currents of pipequaline were needed to produce a consistent effect (Fig. 2), as compared to flurazepam and chlordiazepoxide (Debonnel and de Montigny, 1983). The lower potency of pipequaline might be related to its lower affinity for benzodiazepine receptors compared to benzodiazepines (Le Fur et al., 1981; Yamamura, Mikami, Yamamura, Horst, Morelli, Bautz and O’Brien, 1982; Trifiletti and Snyder, 1984; Wood et al., 1984). However, the fact that the maximal suppression produced by pip equaline, injected intravenously, as well as applied microiontophoretically, plateaued at approximately 50%, would suggest that it possesses a weak intrinsic activity. This interpretation is in agreement with the previous suggestions that pipequaline acts as a partial agonist at benzodiazepine receptors (Gee et al., 1983; Gee and Yamamura, 1983; Benavides et al., 1984; Mizoule, Rataud, Uzan, Mazadier, Daniel, Gauthier, Ollat, Gueremy, Renault, Dubroeucq and Le Fur, 1984).
1341
In the present study, the intravenous administration of pipequaline (from 0.8 mg/kg) blocked both the effect of flurazepam applied microiontophoretically and that of lorazepam injected intravenously (Figs 5B, 6 and 7). Furthermore, microiontophoretic application of pipequaline also blocked the effect of flurazepam (Figs 8 and 9). This is in keeping with the ability of pipequaline to reduce the potentiation by flurazepam of muscimol-induced depolarization in slices of cuneate nucleus in the rat (File and Simmonds, 1984; Simmonds, 1985). Similarly, Gee et al. (1983) and Gee and Yamamura (1983) demonstrated the antagonistic activity of large doses of pipequaline on the anticonvulsant action of diazepam on bicuculline-induced seizures. The ability of pipequaline to suppress, almost completely, the effects of both flurazepam and lorazepam in the present study, contrasting with its weak activity as an agonist, suggests that pipequaline is closer to the antagonistic pole on the agonist-antagonist spectrum of benzodiazepine receptor ligands. This is consistent with the fact that the GABA ratio of pipequaline is closer to that of the antagonist RO 15-1788 than that of the full agonist diazepam (Morelli et al., 1982; Skolnick et al., 1982; Gee et al., 1983; Benavides et al., 1984; Wood et al., 1984). An unexpected observation in the present study was the potentiation of the effect of flurazepam by a small dose (400 pg/kg, i.v.) of pipequaline (Fig. 6). The large amplitude of this effect, as well as the fact that it was produced by a dose of pipequaline which had no or little effect by itself, suggests that it was a real potentiation. Mizoule et al. (1984) have reported that pipequaline increased the anticonvulsant, myorelaxant and sedative effects of diazepam and Benavides et al. (1984) have described a potentiation by pipequaline of the effect of diazepam on levels of cGMP in the cerebellar cortex of the rat. However, the doses required to induce a significant potentiation (12.5-25mg/kg) were 30-60 times larger in these experiments than in the present ones. Methodological differences such as the route of administration (intraperitoneal versus intravenous), the species used (mice compared to rats) and the interval between the administration of pipequaline and the test (30min compared to 3 min) could well account for this apparent discrepancy, particularly in the light of the extremely short half-life of the drug in rats. The present failure to reveal a potentiation of the flurazepam-induced suppression by microiontophoretic applications of pipequaline could be due to the fact that, even with the smallest currents used, the local concentration of the drug was in the range of its antagonist activity. This interpretation is in keeping with the blockade of the effect of flurazepam on muscimol-induced depolarization by a concentration as small as 100 nM of pipequaline (Simmonds, 1985). However, it cannot be ruled out that the potentiation of the effect of flurazepam observed after the intravenous administration of pipequaline was not due to a direct effect of the drug on the neuron recorded.
1342
G. DEBONNEL and C. DE MONTIGNY
In conclusion, the present results suggest that pipequaline increases, _ at very small d&e, the neuronal responsiveness to benzodiazepines whereas, at larger doses (6_15mg/kg, i.v.) it exerts a weak agonistic and a potent antagonistic effect on benzodiazepine receptors. These findings provide further evidence that pi~qualine is a partial agonist at benzodiazepine receptors. This property of pipequaline might be related to its ability to reduce anxiety without producing sedation in humans. Acknowledgements--This
study was supported in part by grant MT-6444 (to C. de M.); C. de M. is in receipt of an MRC Scientist Award. We thank D. Cyr and G. Filosi for preparing the illustrations and L. Perrault for typing the manuscript. MRC
REFERENCES
Benavides J., Malgouris C., Flamier A., Tur C., Quarteronet D.. Benassat F.. Camelin J. C.. Uzan A., Gueremv C. anh L; Fur &. (1984) Biochemical evidence -that 2-phenyl-4[2-(4-piperidinyl) ethyl] quinoline, a quinoline derivative with pure anticonflict properties, is a partial agonist of benzodiazepine receptors. Neuropharmacology 23: 1129-l I36. Bradwejn J. and de Montigny C. (1985a) Effects of PK 8165, a partial benzodiazepine receptor agonist, on cholecystokinin-induced activation of hippocampal pyramidal neurons: A microiontonhoretic studv in the rat. Eur. J. Pharmac. 112: 415-418: Bradwejn J. and de Montigny C. (198Sb) Antagonism of cholecystokinin-induced activation by benzodiazepine receptor agonists. In: Neuronal Cholecystokinin (van Der Haegen J.-J. and Crawley J. N., Eds),.pp. 57si80. New York Academy of Sciences, New York. Debonnel G. and de Montigny C. (1983) ~~~ia~pines selectively antagonize kainate-indu~d activation in the rat hippocampus. Eur. J. Pharmac. 93: 45-54. de Montigny C., Debonnel G. and Tardif D. (1983) Electrophysiological and pharmacological studies on kainic acid-induced neuronal activation. In: Excitotoxins (Fuxe K., Roberts R. and Schwartz R., Eds), pp. 157-169. MacMillan Press, London. File S. (1983) Sedative effects of PK 9084 and PK 8165, alone and in combination with chlordiazepoxide. Br. J. Pharmac. 79: 219-223.
File S. (1984) Modi~cation of seizures elicited by the benzodiazepine RO S-3663 a comparison with picrotoxin. J. Pharm. Pharmac. 36: 837-840.
File S. and Lister R. G. (1983) Quinolines and anxiety: Anxionenic effects of CGS 8216 and partial anxiolvtic profile- of PK 9084. Pharmac. Biochem. Behav. -18: 185-188. File S. and Pellow S. (1984) Behavioural effects of PK 8165 that are not mediated by benzodiazepines binding sites. Neurosci. L&t. 50: 197-201.
File S. and Simmonds M. A. (1984) Interactions of two phenylquinolines with picrotoxin and benzodiazepines in vivo and in vitro. Eur. J. Pharmac. W: 295-300. File S. and Wilks L. (1985) Effect of acute and chronic treatment on the pro- and anti-convulsant actions of CL 218,872, PK 8165 and PK 9084, putative ligands for the benzodiazepine receptor. J. Pharm. Pharmuc. 37: 252-256.
Gee K. W., B&ton R. E. and Yamamura H. I. (1983) PK 8165 and PK 9084, two quinoline derivatives with anxiolytic properties, antagonize the ant~con~~nt effects of diazepam. Brain Res. 264: 168-172.
Gee K. W. and Yamamura H. I. (1983) Selective anxiolytics: Are the actions related to partial “agonist” activity or a preferential affinity for benzodiazepine receptor subtypes? In: Benzodiazepine Recognition Site Ligunds: Biochemistry and Pharmacology (Binnio G. and Costa E., Eds), pp. I-9. Raven Press, New q&k. Haimrler H. J. and Aehaianian G. K. (1974) Lvsereic acid d~thylamide and lerotonin: a comparisoh ok e&cts on serotonergic neurons and neurons receiving a serotonergic input. J. Pharmac. exp. Ther. 168: 688-699. Hunkeler W., Miihler H., Pieri L., Pole P., Bonetti E. P., Cumin R., Schaffner R. and Haefely W. (1981) Selective antagonists of benzodiazepines. Nature 290: 514-516. Keane P. E., Simiand J. and Morre M. (1984) The quinolines PK 8165 and PK 9084 possess benzodiazepines like activity in vitro but not in vivo. Neurosci. Lett. 45: 89.-93. Le Fur G. (1982) RCcepteurs de I’anxiCtt: Nouvelle approche pha~acologique. L’~nc~p~ia~e 8: 145-150. Le Fur G., Mizoule .I., Burgevin M. C., Ferris O., Heaulme M., Gauthier A., Gueremy C. and Uzan A. (1981) Multiple benzodiazepine receptors: Evidence of a dissociation between anticonfict and anticonvulsant properties bv PK 8165 and PK 9084 (two ouinoline derivatives). Life .$ci. 28: 1439-1448. * Mizoule J.. Rataud J.. Uzan A.. Mazadier M.. Daniel M., Gauihier A., Oilat C., Gueremy C., Renault C., Dubroeucq M. C. and Le Fur G. (1984) Pharmacological evidence that PK 8165 behaves as a partial agonist of brain type ~nzodi~epine receptors. Archs int. Pharmacodyn. Th&. 271: 189-197.
Moreili M., Gee K. W. and Yamamura H. I. (1982) The effects of GABA on in uitro binding of two novel non-benzodiazepines, PK 8165 and CGS 8216, to benzodiazepine receptors in the rat brain. Life Sci. 31: 77-81. Pellow S. and File S. E. (1985) Pro and anti-convulsant drug effects in combination with the convuIsant benzodiazepine RO 5-4864. J. Pharm. Pharmac. 37: 560-563. Poggioli J. A., Bonnet D. and Von Frenckell R. (1985) Activity of PR 8165 in anxiety induced by dental surgery: A dose ranging double blind study. Curr. ther. Res. 38: 423-43 1. Simmonds M. A. (1985) Antagonism of flurazepam and other effects of RO 15-1788, PK 8165 and RO 5-4864 on the GABA-A receptor complex in rat cuneate nucleus. Eur. J. Pharmac. 117: 51-61. Skolnick P., Schweri M. M., Williams E. F., Moncada V. Y. and Paul S. M. (1982) An in vitro binding assay which differentiates benzodiazepine agonists and antagonists. Eur. J. Phurmuc. 78: 13%136.Trifiletti R. R. and Snvder S. H. 11984) Anxiolvtic cvclopyrrolones zopiclone+and suricldne bind to a*novei site linked allosterically to benzodiazepine receptors. Molec. Pharmac. 26: 458469.
Von Frenckell R., Ansseau M. and Bonnet D. (1986) Evaluation of the sedative properties of PK 8165 (Pipequaline), a benzodiazepine partial agonist in normal subjects. Int. clin. Psychopharmac. 1: 24-35. Wille; J. C., Von Frenckell’ R., Bonnet D. and Le Fur G. (1986) The ability of PK 8165, a quinoline derivative to reduce responses to a stressful situation in a double-blind study in man. Neuropha~aco~agy 25: 275-281. Wood P. L.. Loo P.. Braunwalder A.. Yokovama N. and Cheney d. L. (1984) In vitro characterization of benzodiazepines receptor agonists, antagonists, inverse agonists and agonist/antagonist J. Pharmac. exp. Ther. 231: 572-576.
Yamamura H. I., Mikami T., Yamamura S. H., Horst W. D., Morelli M., Bautz G. and O’Brien R. A. (1982) i3H]C1 218,872, a novel triazolop~idazine which labels the ~~odiazepine receptor in rat brain. Eur. J. Pharmac. 77: 351-354.
1987 Copyright 0
0028-3908/87 $3.00 + 0.00 1987 Pergamon Journals Ltd
PIPEQUALINE ACTS AS A PARTIAL AGONIST OF BENZODIAZEPINE RECEPTORS: AN ELECTROPHYSIOLOGICAL STUDY IN THE HIPPOCAMPUS OF THE RAT G. DEBONNEL* and C. DE MONTIGNY Institut P. Pine1 de Montreal and Centre de recherche en sciences neurologiques, Faculte de m&mine, Universitt de Montreal, C.P. 6128, Succursale “A”, Montreal, Qu&ec, Canada H3C 357 (Accepted 26 February 1987) Summary-Pipequaline (PK 8169, a quinoline derivative and a ligand of the benzodiazepine binding site, is a clinically-effective anxiolytic, which is devoid of sedative and anticonvulsant properties. Several biochemical and behavioral studies have indicated that this molecule shares some properties with both agonists and antagonists of benzodiaxepine receptors. The present in vivo electrophysiological studies were undertaken to determine the effects of microiontophoretic applications and of intravenous injections of pipequaline on hippocampal pyramidal neurons, activated by kainate, glutamate or acetylcholine and to characterize the effects of pipequaline on the action of benzodiaxepines. Intravenously administered pipequaline exerted a partial suppression of activations by kainate, glutamate and acetylcholine. Microiontophoretic applications of pipequaline reduced the neuronal activation by kainate. This effect was blocked by RO 15-1788. In small intravenous doses, pipequaline potentiated the effect of microiontophoretically-applied flurazepam whereas, in larger doses, it suppressed the effects of microiontophoretically-applied fluraxepam and of intravenously administered lorazepam on kainateinduced activation. Similarly, microiontophoretic applications of pipequaline blocked the suppressant effect of microiontophoretically-applied fluraxepam on kainate-induced activation. These results constitute further evidence that the selective anxiolytic activity of pipequaline might be ascribed to its partial agonistic action on benzodiazepine receptors. Key words: pipequaline (PK 8169, benzodiaxepine receptors, hippocampal pyramidal neurons, kainate.
Pipequaline, a phenylquinoline derivative, exerts an anticonflict effect in the rat, without inducing any sedation (Le Fur, Mizoule, Burgevin, Ferris, Heaulme, Gauthier, Gueremy and Uzan, 1981). Several clinical studies have confirmed the anxiolytic effect of pipequaline and its lack of sedative property in humans (Poggioli, Bonnet and Von Frenckell, 1985; Von Frenckell, Ansseau and Bonnet, 1986; Willer, Von Frenckell, Bonnet and Le Fur, 1986; Bradwejn and de Montigny, unpublished observations). However, pipequaline appears to be devoid of any “anxiolytic” activity in the social interaction model in the rat (File and Lister, 1983) and to decrease motor activity in the rat (File, 1983); effects which are not blocked by the benzodiazepine antagonist ethyl 8-fluoro-5,6-dihydro-5-methyl-6-oxo4H-imidazo[l,S-a][l,4]benzodiazepine-3-carboxylate (RO 15-1788) (File and Pellow, 1984). That the pharmacological profile of this molecule clearly differs from that of benzodiazepines is further indicated by its ability to protentiate the effects of sub-convulsant doses of picrotoxin, pentetrazol and 1,3-dihydro-5-methyl-2H-1,4-benzodiazepine-2-one (RO 5-3663) (File, 1984; File and Simmons, 1984; File and Wilks, 1985) and to reduce the protective *Address all correspondence
to this author.
effect of diazepam against bicuculine-induced seizures (Gee, Brinton and Yamamura, 1983; Gee and Yamamura, 1983), although it does not potentiate the proconvulsant activity of 7-chloro-1,3-dihydrol-methyl-5-(p-chlorophenyl)2H-l,Cbenzodiazepine2-one (RO 5-4864) (Pellow and File, 1985). Radioligand binding studies in vitro have shown that pipequaline displaces [‘Hldiazepam and [3H]flunitrazepam in membranes from the brain of the rat (Le Fur et al., 1981; Le Fur, 1982; Gee et al., 1983; Gee and Yamamura, 1983; Benavides, Malgouris, Flamier, Tur, Quarteronet, Begassat, Camelin, Uzan, Gueremy and Le Fur, 1984; Keane, Simiand and Morre, 1984). The binding of pipequaline in the cerebral cortex and in the cerebellum is enhanced by y-aminobutyric acid (GABA) but its GABA ratio (“GABA-shift”) falls between that of agonists and antagonists (Morelli, Gee and Yamamura, 1982; Skolnick, Schweri, Williams, Moncada and Paul, 1982; Gee et al., 1983; Wood, Loo, Braunwalder, Yokoyama and Cheney, 1984), suggesting that it may act as a partial agonist at the benzodiazepine receptor site. Furthermore, the failure of pipequaline to decrease levels of cGMP in the cerebellar cortex suggests that some of its pharmacological effects could be independent of GABA (Le Fur et al., 1981; Benavides et al., 1984).
1337
G.
1338
DEBONNELand
In the light of these behavioral and biochemical data, it was of interest to determine electrophysiologically in z&o the effect of pipequaline on be~odia~pine receptors, in an attempt to further our understanding of the clinical anxiolytic property of this molecule. It has previously been reported that benzodiazepines antagonize the activation of pyramidal neurons induced by kainate in the CA, region of the dorsal hippocampus in the rat through activation of benzodiazepine receptors (Debonnel and de Montigny, 1983; de Montigny, Debonnel and Tardif, 1983). This paradigm was used in the present study to determine the effect of pipequaline on the firing activity of pyramidal neurons in the hippocampus and to study its interactions with betrzodiazepines. METHODS
Male Sprague-Dawley rats (200-300 g) were anesthetized with urethane (1.25 g/kg, i.p.) and were mounted in a stereotaxic apparatus. An indwelling femoral catheter was installed for intravenous injections. Five-ba~elIed glass micropipettes were prepared in a conventional manner (Haigler and Aghajanian, 1974) and their tips broken to 8-10 PM under microscopic control. The central barrel, used for extracellular recording of hippocampal pyramidal neurons, was filled with a 2 M NaCl solution saturated with Fast Green (FCF). A deposit of Fast Green was left at the last recording site at the end of each experiment for subsequent histological verification on cryostat sections. One side barrel, used for automatic current balancing, was filled with a 2 M NaCl solution. The remaining barrels, used for microiontophoresis, were filled with three of the following solutions: kainate 1 mM in 400 mM NaCl, pH: 8; glutamate. NC1 50 mM in 50 mM NaCl, pH: 8; acetylcholine. Cl (ACh) 20 mM in 20 mM NaCl, pH: 4; pipequaline 10 mM, pH: 4; flurazepam .2 HCl IOmM, pH: 3. Extracellular recordings of unit activity were obtained in the CA1 region of the dorsal hippocampus. Pyramidal neurons were identified by their long duration (0.8-l .5 msec), large amplitude (0.5-2 mV) action potentials and by the presence of “complex spike” discharges. The signal, amplified and displayed on an oscilloscope, was fed to a differential amplitude discriminator, generating square pulses, from which integrated firing rate histograms were obtained. When a stable recording was obtained, the intensity of the currents of kainate (42.4 f 2.7 nA), glutamate (9.4 + 0.7 nA) and/or ACh (30.2 + 3.2 nA) were adjusted to induce a similar degree of sustained activation (i.e. a firing rate of lO-2OHz). The suppressant effects of microiontophoretic applications of pipequaline or flurazepam on these activations were assessed by comparing the mean firing rate of the neuron during the 50 set preceding the microiontophoretic applications to that during the 50 set application. Data were discarded if there
C.
DE MONTIGNY
was not a recovery of at least 60% after the cessation of the application. For most neurons, several applications of pipequaline or flurazepam with different currents were carried out. The results are expressed as the mean degree of suppression produced by each current used. For intravenous injections, pipequaline was dissolved in water at pH 4, and lorazepam and RO 15-1788 in 40% propylene glycol, 10% ethyl alcohol, 1.5% benzyl alcohol, 5% Na benzoate and 43.5% water, at pH4. These drugs were administered through a femoral catheter in a volume of 0.1-0.3 ml.
The suppressant effects of pipequaline, lorazepam or flurazepam were determined by comparing the degree of activation before and 3-5min after their intravenous administration. The effect of RO 15-1788, a selective benzodiazepine receptor antagonist (Hunkeler, Mohler, Pieri, Pole, Bonetti, Cumin, Shaffner and Haefely, 1981) on the activity of pipequaline, was measured by comparing the degree of suppression induced by microiontophoretic applications of pipequaline before and 3-5 min after the administration of RO 15-1788 (3.5 mg/kg, iv.). In these series of experiments, the rats received either a single dose of pipequaline, lorazepam or RO 15-1788 or an injection of pipequaline 3-5 min before administering one dose of lorazepam. In these experiments, only one cell per rat was tested. RESULTS
(1) The agonistic activity ofpipequaline (a) Intravenous injections of pipequaline. In small doses (from 0.4 to 1.2 mg/kg) pipequaline did not produce any consistent effect on activations induced by kainate, glutamate or ACh: 18 of the 26 neurons tested were not, or only slightly (C 10%) affected by these injections. At doses ranging from 6 to 10 mg/kg, pipequaline reduced activations induced by kainate, glutamate or ACh in 60% of the 75 neurons tested.
DOSE (mg/kg. i.v,) OF PK 8165
Fig. 1. Suppression of kainate @CA)-induced activation of CA, hippoeampal neurons by intravenous injections of pipequaline (PK 8165). The degree of suppression was measured 3-5 min after the injection. Note that the intervals on the X axis are not proportional to the doses given. Only one neuron was tested in each rat at each dose. The number of neurons tested is given in parentheses.
PK 8165, a partial benzodiazepine
5
10
15
20
30
MICROIONTOPHORETIC CURRENl (nA) OF PK 8165
RO 15-1788 (3.5 mglkg)
Fig. 2. Suppression of kainate-induced activation of CA, hippocampal pyramidal neurons by microiontophoretic applications of pipequaline. The degree of suppression was measured by comparing the number of spikes generated by kainate before and during 50 set applications of pipequaline. The number of neurons tested with each current is given in parentheses.
It was only at the largest doses of 12 and 15 mg/kg that pipequaline produced a clear suppressive effect on all neurons (n = 26). However, the degree of suppression obtained with the 15 mg/kg dose appeared to be smaller than that produced by the 12 mg/kg dose. The dose-response relationship for the effect of pipequaline on kainate-induced activation is illustrated in Figure 1. The effects of pipequaline on glutamate- and ACh-induced activations were similar to that on kainate-induced activation (data not shown). The effect of pipequaline was very short lasting, as there was a full recovery of the neuronal activation within 15 min after the injection, even after the large doses. (b) Microiontophoretic applications of pipequaline.
As intravenously injected pipequaline did not show any selectivity, microiontophoretic applications of the drug were performed during activation induced by kainate only; pipequaline reduced kainate-induced activation in all but three neurons (n = 44). The degree of suppression by microiontophoretic applications of pipequaline, as was the case with its
KA-7 PKg
g
g&I
20 20 20 0000~
20
20
200 ;
150
I 5
100
E
500
!
1339
RO 1$-l 760 (3.5 mg/kg, I.v.)
Fig. 3. Integrated firing rate histogram of a pyramidal neuron of the CA, region of the dorsal hippocampus, showing the suppression by pipequaline of kainate-induced activation and the block of this suppression by intravenous injection of RO 15-1788. Bars indicate the duration of microiontophoretic applications for which currents are given in nA.
Fig. 4. Suppression of kainate-induced activations of CA, hippocampal pyramidal neurons by microiontophoretic applications of pipequaline (16.5 +_1.5 nA) before and after intravenous administration of RO 15-1788 (3.5 mg/kg). The same neurons were tested before and after RO 15-1788, using the same current for pipequaline *P < 0.001 (paired Student’s r-test).
intravenous administration, was relatively small, as the maximum decrease in the firing rate in most neurons was less than 50%. There was a dose-response relationship at the smaller currents used, but the response plateaued at about 50% from 15 nA on (Fig. 2). In order to determine if this effect of pipequaline was mediated through activation of the benzodiazepine receptor, the benzodiazepine antagonist RO 15-1788 was used. The degree of suppression produced by microiontophoretic applications of pipequaline was compared before and 3-5 min after intravenous administration of RO 15-1788 (3.5mg/kg); RO 15-1788 reduced by 65% the effect of pipequaline (Figs 3 and 4). (2) The antagonistic activity of pipequaline (a) Intravenous
administration
of
pipequaline.
Microiontophoretic applications of flurazepam, with a current of 5 nA, induced a mean suppression of 30% of the firing rate of pyramidal neurons, activated by kainate. At a small dose (400pg/kg, i.v.), pipequaline markedly potentiated the effect of flurazepam which was increased by nearly 90% (Figs 5A and 6). However, in larger doses (800 and 1200 p g/kg), pipequaline markedly reduced the efficacy of flurazepam (Figs 5B and 6). To provide further evidence for the antagonistic activity of pipequaline on benzodiazepine receptors, the effect of the intravenous administration of lorazepam (500pg/kg) was compared in naive rats and in rats pretreated with a large dose of pipequaline (10 mg/kg, i.v.) administered 3-5 min prior to the injection of lorazepam. Consistent with the data presented above (Fig. l), in the present series of experiments, this dose of pipequaline produced a 20% decrease of kainate-induced activity. Lorazepam produced a 60% suppression of activation induced by kainate in naive rats. The pretreatment with pipequaline reduced the effect of lorazepam by more than 90% (Fig. 7).
G.
1340
DEBONNEL
and C.
DE
MONTICNY
KA-6
A 250,
FLU 5 0
5 n
5
5
0
0
PK 8165 IO 4 mg/kg.
B
I Y)
KA -13
FLU 5 5 c3oc1
5
5 no
5
PK 8165 (I 2 mg/kg.
1.x)
Fig. 5. Integrated firing rate histogram of CA, hippocampal pyramidal neurons showing the potentiatioin of the effect of flurazepam (FLU) on kainate-induced activation by a small dose of pipequaline (PK 8165) (A) and the blockade of the effect of flurazepam by a larger dose of pipequaline (B). (b) Microiontophoretic application of pipequaline. To ascertain whether this antagonistic activity of pipequaline resulted from a direct effect of the drug in the dorsal hippocampus, the efficacy of microiontophoretic applications of flurazepam was measured before and during microiontophoretic application of pipequaline. As illustrated in Figure 8, pipequaline antagonized the effect of flurazepam. The application of flurazepam alone produced a 60% suppression of kainate induced activation and that of pipequaline applied with similar currents, produced only a 20% suppression (Fig. 9). The effectiveness of flurazepam was reduced by 60% during concomittant application of pipequaline (Fig. 9). In these experiments, no potentiation of the effect of flurazepam was
0
BEFORE
m
AFTER PK 8165
2
microiontophoretically-applied pipequaline (PK 8165) of the effect of flurazepam (FLU) on kainate (KA)-induced activation and the recovery of the effect for flurazepam following the cessation of the application of pipequaline.
LU DOSE (mglkg)
OF PKS165
Fig. 6. Suppression of kainate-induced activation of CA, hippocampal pyramidal neurons by small current (5 nA) applications of flurazepam alone and after the intravenous administration of pipequaline (PK 8165). *P c 0.005;
FLU during PK 6165 Ez5
Fig. 9. Suppression of kainate-induced activation of CA, hippocampal pyramidal neurons by microiontophoretic applications of flurazepam (FLU) (11.5 + 2.4 nA), of pipequaline (PK 8165) (10.3 + 1.5 nA) and of flurazepam during the concomitant application of pipequaline. *P < 0.001 (two-tailed Student’s r-test comparing the degree of suppression by flurazepam alone to that produced by fiurazepam during the application of pipequaline).
PK 8165,
a partial benzodiazepine
DISCUSSION
Pipequaline decreased the firing rate of activated pyramidal cells, but was devoid of any selectivity since it exerted a similar effect on activations induced by kainate, glutamate or ACh, whereas benzodiazepines act preferentially on kainate-induced activation (Debonnel and de Montigny, 1983). This suppression by pipequaline was observed either after its intravenous administration or during microiontophoretic application. The blockade of the effect of pipequaline by RO 15-1788, a selective benzodiazepine receptor antagonist, indicates that its effect, in this model, was mediated through the activation of benzodiazepine receptors (Fig. 4). This is in keeping with previous demonstrations of a reversal of the electrophysiological and of some of the behavioral effects of pipequaline by RO 15-1788 (File and Pellow, 1984; File and Simmonds, 1984; Bradwejn and de Montigny, 1985a, b), and with the high affinity binding of pipequaline on benzodiazepine receptors in vitro (Le Fur et al., 1981; Le Fur, 1982; Gee et al., 1983; Gee and Yamamura, 1983; Benavides et al., 1984; Keane et al., 1984). The apparent discrepancy between the present results and the failure of intravenously administered pipequaline to reduce in vivo, the binding of [‘Hlflunitrazepam (Keane et al., 1984) may be due to the long time interval (30 min) between the injection of pipequaline and that of the radioligand in the latter study. Indeed in the present study, the effect of pipequaline on the benzodiazepine receptor was very short-lasting as was observed previously (Bradwejn and de Montigny, 1985a), in keeping with the extremely short halflife of pipequaline in the rat (Le Fur, personal communication). Large intravenous doses of pipequaline (12 and 15 mg/kg, i.v.) were required to obtain a consistent suppression of the firing activity of hippocampal pyramidal neurons (Fig. 1) compared to those of lorazepam and diazepam (Debonnel and de Montigny, 1983). Similarly, large microiontophoretic currents of pipequaline were needed to produce a consistent effect (Fig. 2), as compared to flurazepam and chlordiazepoxide (Debonnel and de Montigny, 1983). The lower potency of pipequaline might be related to its lower affinity for benzodiazepine receptors compared to benzodiazepines (Le Fur et al., 1981; Yamamura, Mikami, Yamamura, Horst, Morelli, Bautz and O’Brien, 1982; Trifiletti and Snyder, 1984; Wood et al., 1984). However, the fact that the maximal suppression produced by pip equaline, injected intravenously, as well as applied microiontophoretically, plateaued at approximately 50%, would suggest that it possesses a weak intrinsic activity. This interpretation is in agreement with the previous suggestions that pipequaline acts as a partial agonist at benzodiazepine receptors (Gee et al., 1983; Gee and Yamamura, 1983; Benavides et al., 1984; Mizoule, Rataud, Uzan, Mazadier, Daniel, Gauthier, Ollat, Gueremy, Renault, Dubroeucq and Le Fur, 1984).
1341
In the present study, the intravenous administration of pipequaline (from 0.8 mg/kg) blocked both the effect of flurazepam applied microiontophoretically and that of lorazepam injected intravenously (Figs 5B, 6 and 7). Furthermore, microiontophoretic application of pipequaline also blocked the effect of flurazepam (Figs 8 and 9). This is in keeping with the ability of pipequaline to reduce the potentiation by flurazepam of muscimol-induced depolarization in slices of cuneate nucleus in the rat (File and Simmonds, 1984; Simmonds, 1985). Similarly, Gee et al. (1983) and Gee and Yamamura (1983) demonstrated the antagonistic activity of large doses of pipequaline on the anticonvulsant action of diazepam on bicuculline-induced seizures. The ability of pipequaline to suppress, almost completely, the effects of both flurazepam and lorazepam in the present study, contrasting with its weak activity as an agonist, suggests that pipequaline is closer to the antagonistic pole on the agonist-antagonist spectrum of benzodiazepine receptor ligands. This is consistent with the fact that the GABA ratio of pipequaline is closer to that of the antagonist RO 15-1788 than that of the full agonist diazepam (Morelli et al., 1982; Skolnick et al., 1982; Gee et al., 1983; Benavides et al., 1984; Wood et al., 1984). An unexpected observation in the present study was the potentiation of the effect of flurazepam by a small dose (400 pg/kg, i.v.) of pipequaline (Fig. 6). The large amplitude of this effect, as well as the fact that it was produced by a dose of pipequaline which had no or little effect by itself, suggests that it was a real potentiation. Mizoule et al. (1984) have reported that pipequaline increased the anticonvulsant, myorelaxant and sedative effects of diazepam and Benavides et al. (1984) have described a potentiation by pipequaline of the effect of diazepam on levels of cGMP in the cerebellar cortex of the rat. However, the doses required to induce a significant potentiation (12.5-25mg/kg) were 30-60 times larger in these experiments than in the present ones. Methodological differences such as the route of administration (intraperitoneal versus intravenous), the species used (mice compared to rats) and the interval between the administration of pipequaline and the test (30min compared to 3 min) could well account for this apparent discrepancy, particularly in the light of the extremely short half-life of the drug in rats. The present failure to reveal a potentiation of the flurazepam-induced suppression by microiontophoretic applications of pipequaline could be due to the fact that, even with the smallest currents used, the local concentration of the drug was in the range of its antagonist activity. This interpretation is in keeping with the blockade of the effect of flurazepam on muscimol-induced depolarization by a concentration as small as 100 nM of pipequaline (Simmonds, 1985). However, it cannot be ruled out that the potentiation of the effect of flurazepam observed after the intravenous administration of pipequaline was not due to a direct effect of the drug on the neuron recorded.
1342
G. DEBONNEL and C. DE MONTIGNY
In conclusion, the present results suggest that pipequaline increases, _ at very small d&e, the neuronal responsiveness to benzodiazepines whereas, at larger doses (6_15mg/kg, i.v.) it exerts a weak agonistic and a potent antagonistic effect on benzodiazepine receptors. These findings provide further evidence that pi~qualine is a partial agonist at benzodiazepine receptors. This property of pipequaline might be related to its ability to reduce anxiety without producing sedation in humans. Acknowledgements--This
study was supported in part by grant MT-6444 (to C. de M.); C. de M. is in receipt of an MRC Scientist Award. We thank D. Cyr and G. Filosi for preparing the illustrations and L. Perrault for typing the manuscript. MRC
REFERENCES
Benavides J., Malgouris C., Flamier A., Tur C., Quarteronet D.. Benassat F.. Camelin J. C.. Uzan A., Gueremv C. anh L; Fur &. (1984) Biochemical evidence -that 2-phenyl-4[2-(4-piperidinyl) ethyl] quinoline, a quinoline derivative with pure anticonflict properties, is a partial agonist of benzodiazepine receptors. Neuropharmacology 23: 1129-l I36. Bradwejn J. and de Montigny C. (1985a) Effects of PK 8165, a partial benzodiazepine receptor agonist, on cholecystokinin-induced activation of hippocampal pyramidal neurons: A microiontonhoretic studv in the rat. Eur. J. Pharmac. 112: 415-418: Bradwejn J. and de Montigny C. (198Sb) Antagonism of cholecystokinin-induced activation by benzodiazepine receptor agonists. In: Neuronal Cholecystokinin (van Der Haegen J.-J. and Crawley J. N., Eds),.pp. 57si80. New York Academy of Sciences, New York. Debonnel G. and de Montigny C. (1983) ~~~ia~pines selectively antagonize kainate-indu~d activation in the rat hippocampus. Eur. J. Pharmac. 93: 45-54. de Montigny C., Debonnel G. and Tardif D. (1983) Electrophysiological and pharmacological studies on kainic acid-induced neuronal activation. In: Excitotoxins (Fuxe K., Roberts R. and Schwartz R., Eds), pp. 157-169. MacMillan Press, London. File S. (1983) Sedative effects of PK 9084 and PK 8165, alone and in combination with chlordiazepoxide. Br. J. Pharmac. 79: 219-223.
File S. (1984) Modi~cation of seizures elicited by the benzodiazepine RO S-3663 a comparison with picrotoxin. J. Pharm. Pharmac. 36: 837-840.
File S. and Lister R. G. (1983) Quinolines and anxiety: Anxionenic effects of CGS 8216 and partial anxiolvtic profile- of PK 9084. Pharmac. Biochem. Behav. -18: 185-188. File S. and Pellow S. (1984) Behavioural effects of PK 8165 that are not mediated by benzodiazepines binding sites. Neurosci. L&t. 50: 197-201.
File S. and Simmonds M. A. (1984) Interactions of two phenylquinolines with picrotoxin and benzodiazepines in vivo and in vitro. Eur. J. Pharmac. W: 295-300. File S. and Wilks L. (1985) Effect of acute and chronic treatment on the pro- and anti-convulsant actions of CL 218,872, PK 8165 and PK 9084, putative ligands for the benzodiazepine receptor. J. Pharm. Pharmuc. 37: 252-256.
Gee K. W., B&ton R. E. and Yamamura H. I. (1983) PK 8165 and PK 9084, two quinoline derivatives with anxiolytic properties, antagonize the ant~con~~nt effects of diazepam. Brain Res. 264: 168-172.
Gee K. W. and Yamamura H. I. (1983) Selective anxiolytics: Are the actions related to partial “agonist” activity or a preferential affinity for benzodiazepine receptor subtypes? In: Benzodiazepine Recognition Site Ligunds: Biochemistry and Pharmacology (Binnio G. and Costa E., Eds), pp. I-9. Raven Press, New q&k. Haimrler H. J. and Aehaianian G. K. (1974) Lvsereic acid d~thylamide and lerotonin: a comparisoh ok e&cts on serotonergic neurons and neurons receiving a serotonergic input. J. Pharmac. exp. Ther. 168: 688-699. Hunkeler W., Miihler H., Pieri L., Pole P., Bonetti E. P., Cumin R., Schaffner R. and Haefely W. (1981) Selective antagonists of benzodiazepines. Nature 290: 514-516. Keane P. E., Simiand J. and Morre M. (1984) The quinolines PK 8165 and PK 9084 possess benzodiazepines like activity in vitro but not in vivo. Neurosci. Lett. 45: 89.-93. Le Fur G. (1982) RCcepteurs de I’anxiCtt: Nouvelle approche pha~acologique. L’~nc~p~ia~e 8: 145-150. Le Fur G., Mizoule .I., Burgevin M. C., Ferris O., Heaulme M., Gauthier A., Gueremy C. and Uzan A. (1981) Multiple benzodiazepine receptors: Evidence of a dissociation between anticonfict and anticonvulsant properties bv PK 8165 and PK 9084 (two ouinoline derivatives). Life .$ci. 28: 1439-1448. * Mizoule J.. Rataud J.. Uzan A.. Mazadier M.. Daniel M., Gauihier A., Oilat C., Gueremy C., Renault C., Dubroeucq M. C. and Le Fur G. (1984) Pharmacological evidence that PK 8165 behaves as a partial agonist of brain type ~nzodi~epine receptors. Archs int. Pharmacodyn. Th&. 271: 189-197.
Moreili M., Gee K. W. and Yamamura H. I. (1982) The effects of GABA on in uitro binding of two novel non-benzodiazepines, PK 8165 and CGS 8216, to benzodiazepine receptors in the rat brain. Life Sci. 31: 77-81. Pellow S. and File S. E. (1985) Pro and anti-convulsant drug effects in combination with the convuIsant benzodiazepine RO 5-4864. J. Pharm. Pharmac. 37: 560-563. Poggioli J. A., Bonnet D. and Von Frenckell R. (1985) Activity of PR 8165 in anxiety induced by dental surgery: A dose ranging double blind study. Curr. ther. Res. 38: 423-43 1. Simmonds M. A. (1985) Antagonism of flurazepam and other effects of RO 15-1788, PK 8165 and RO 5-4864 on the GABA-A receptor complex in rat cuneate nucleus. Eur. J. Pharmac. 117: 51-61. Skolnick P., Schweri M. M., Williams E. F., Moncada V. Y. and Paul S. M. (1982) An in vitro binding assay which differentiates benzodiazepine agonists and antagonists. Eur. J. Phurmuc. 78: 13%136.Trifiletti R. R. and Snvder S. H. 11984) Anxiolvtic cvclopyrrolones zopiclone+and suricldne bind to a*novei site linked allosterically to benzodiazepine receptors. Molec. Pharmac. 26: 458469.
Von Frenckell R., Ansseau M. and Bonnet D. (1986) Evaluation of the sedative properties of PK 8165 (Pipequaline), a benzodiazepine partial agonist in normal subjects. Int. clin. Psychopharmac. 1: 24-35. Wille; J. C., Von Frenckell’ R., Bonnet D. and Le Fur G. (1986) The ability of PK 8165, a quinoline derivative to reduce responses to a stressful situation in a double-blind study in man. Neuropha~aco~agy 25: 275-281. Wood P. L.. Loo P.. Braunwalder A.. Yokovama N. and Cheney d. L. (1984) In vitro characterization of benzodiazepines receptor agonists, antagonists, inverse agonists and agonist/antagonist J. Pharmac. exp. Ther. 231: 572-576.
Yamamura H. I., Mikami T., Yamamura S. H., Horst W. D., Morelli M., Bautz G. and O’Brien R. A. (1982) i3H]C1 218,872, a novel triazolop~idazine which labels the ~~odiazepine receptor in rat brain. Eur. J. Pharmac. 77: 351-354.