- Email: [email protected]
Long-term benzodiazepine treatment reduces neuronal responsiveness to cholecystokinin: an electrophysiological study in the rat
European Journal of Pharmacology, 151 (1988) 135-138
135
Elsevier EJP 20145
Short communication
Long-term benzodi epine treatment reduces neuronal responsiveness to cholecystokinin: an electrophysiological study in the rat A l a i n Bouthillier a n d C l a u d e D e M o n t i g n y * Department of Psychiatry, McGill University, 1033 Pine Avenue West, MontrJal (QuJbec), Canada H3,4 1,41 Received 18 February 1988, accepted 19 April 1988
Acute benzodiazepine administration has been reported to antagonize the effect of cholecystokinin both in the periphery and in the central nervous system. A two-week treatment with either diazepam (5 mg/kg per day) or flurazepam (15 mg/kg per day) markedly reduced the excitatory effect of microiontophoretically applied sulphated cholecystokinin octapeptide-(26-33) on rat CA 3 hippocampal pyramidal neurons but not that of acetylcholine. In view of the sustained anxiolytic activity of benzodiazepines that contrasts with their transient sedative effect, the present results suggest that the reduction of neuronal responsiveness to cholecystokinin by these drugs might be related to their anxiolytic property. Cholecystokinin; Acetylcholine; Benzodiazepines; Sedation; Anxiety; Hippocampus
1. Introduction
Cholecystokinin (CCK) exists in high concentrations in the mammalian brain (Handelmann et al., 1981). Benzodiazepine receptor agonists have been reported to reduce the effect of CCK both in the periphery and in the central nervous system (Kubota et al., 1983; Bradwejn and De Montigny, 1984, 1985; Kubota et al., 1985). It is not yet proven that the antagonism by benzodiazepine of CCK in the central nervous system might be related to their anxiolytic or sedative activity. The clear dissociation between the transient sedative effect of benzodiazepine, due to a rapid development of tolerance, and their sustained anxiolytic activity (Haefely, 1978) were the basis for the present study. It was aimed to determine in vivo whether the reduction of neuronal responsiveness to CCK in the rat dorsal hippocampus would persist or vanish upon long-term administration of benzodiazepine.
* To whom all correspondence should be addressed.
2. Materials and methods
Male Sprague-Dawley rats weighing 250-300 g were used. Two groups were treated for 14 days with either diazepam (5 m g / k g per day i.p.) or flurazepam (15 m g / k g per day i.p.). The control groups were injected respectively with the diazepam vehicle (40% v / v propylene glycol, 10% v / v ethanol, 1.5% v / v benzylic alcohol, 48.5% v / v water and 5% sodium benzoate) or with a 0.9% saline solution. Each group comprised four or five rats. The rats were anesthetized with urethane (1.25 g / k g i.p.) 24 h after the last injection and were placed in a stereotaxic apparatus. Five-barrelled micropipettes were prepared in the conventional manner. The central barrel filled with 2 M NaC1 saturated with Fast Green F C F was used for extracellular recording of unitary activity. Recordings were obtained from pyramidal neurons of the CA 3 region of the dorsal hippocampus. These neurons were identified by their large amplitude (0.5-1.2 mV) long duration (0.8-1.2 ms) action potential, by the presence of complex spike
0014-2999/88/$03.50 © 1988 Elsevier Science Publishers B.V. (Biomedical Division)
136 discharges alternating with simple spike activity and by their tendency to exhibit rhythmic activity. The signal, amplified and displayed in the usual manner, was fed into a differential amplitude discriminator, generating square pulses from which integrated firing rate histograms were obtained. Three side barrels were used for ejection by microiontophoresis of the following substances: acetylcholine chloride (ACh) (20 m M in 200 m M NaC1, p H 4; Sigma), C C K (sulphated octapeptide-(26-33)) (10 ffM in 200 m M NaC1, p H 5; Sigma) and NaC1 (200 mM). This latter barrel was used to ascertain that the application of the same current used to eject C C K (50 nA) had no effect on neuronal activity. A leak or a small current (1-2 nA) of ACh was used during the electrode descent to activate and identify quiescent pyramidal neurons. The fourth side barrel, filled with 2 M NaC1, was used for automatic current balancing. Recording sites were marked by passing a - 27 ffA current through the central barrel for 20 min to make a Fast Green deposit, the location of which was verified histologically. The responsiveness of pyramidal neurons to C C K and ACh was assessed by determining the total number of spikes generated by 50 s microiontophoretic applications with constant currents (50 nA for C C K and 3 nA for ACh) in all rats. As reported from our laboratory and by others, some neurons showed a rapid tachyphylaxis to the excitatory effect of C C K (see fig. 1A). Hence, only the effect of the first application of C C K was taken into account. Most of the experiments were carried out in pairs of rats, each pair was comprised of a control rat and a treated rat, the experimenter being blind to the treatment. The same microelectrode was used to assess neuronal responsiveness in a pair of rats in order to minimize the possible contribution of fluctuations in transport numbers from one micropipette to another.
treatment with diazepam markedly reduced the number of pyramidal neurons which were activated by the standardized application of C C K (50 nA for 50 s): 24% were responsive to C C K in the diazepam group whereas 74% were responsive in the control group (table 1). The degree of neuronal activation by C C K was also markedly reduced by the diazepam treatment: the number of spikes generated by C C K application was reduced by 61% relative to the control group (fig. 1, table 1). The two-week treatment with flurazepam produced similar effects: only 36% of the pyramidal neurons responded to the standardized C C K application, as compared to 80% in the control group. The flurazepam treatment also diminished by 48% the degree of activation by C C K relative to the controls (table 1). Neither the diazepam nor the flurazepam treatments altered the effect of ACh
A - CONTROL ACh 3 E:3
o.
The spontaneous firing activity of hippocampal pyramidal neurons was similar in the control and benzodiazepine-treated rats (0.8 +_ 0.1 Hz (n = 43); 1.0 + 0.2 Hz (n = 43), respectively). The two-week
12]
I
I~
1'30
w xE oo
50 1 0
B - D I A Z E P A M (5 m g / k g / d a y × 14 days) ACh 3 r7
o LU OCO
[-i
CCK 50 II
I
1j j,j
IE3
100-
5O 0
3. Results
CCK -50 rm
2 min
Fig. 1. Integrated firing rate histograms showing the response of CA 3 hippocampal pyramidal neurons to microiontophoretic applications of acetylcholine (ACh) and sulphated octapeptide cholecystokinin (CCK) in a control rat (A) and in a rat treated with diazepam (B). Bars indicate the duration of the applications of the current shown in nA. The time base applies to both traces.
137 TABLE 1 Responsiveness of dorsal hippocampus CA 3 pyramidal neurons to microiontophoretic applications of acetylcholine (ACh) and cholecystokinin (CCK) following long-term diazepam or flurazepam treatment a. N u m b e r of neurons tested
Activated neurons (%)
Spikes g e n e r a t e d ± S.E.
ACh b
CCK b
ACh b
CCK b
Controls c Diazepam
23 21
23 (100) 21 (100)
17 (74) 5 (24)
353 ± 45 350_+ 49
425 _+52 167 _+29 e
Controls d Flurazepam
20 22
20 (100) 22 (100)
16 (80) 8 (36)
402 _+61 401 _+61
330 ± 42 174 ± 31 f
Treatment
Diazepam and flurazepam were administered i.p. for 14 days at a dose of 5 and 15 m g / k g per day, respectively. Electrophysiological experiments were carried out 24 h after the last injection, b A C h and C C K were applied for 50 s with currents of 3 and 50 nA, respectively. ~ These control rats received daily intraperitoneal injections of the diazepam vehicle for 14 days (see text), d These control rats received daily intraperitoneal injections of saline for 14 days. e p < 0.005 (compared with corresponding control value, two-tailed Student's t-test), f P < 0.05 (compared with corresponding control value, two-tailed Student's t-test). a
on the same neurons (table 1).
4. Discussion The present results show that a 14-day treatment with either diazepam or flurazepam reduced the responsiveness of rat dorsal hippocampus pyramidal neurons to CCK, whereas the effectiveness of ACh was unmodified. The data thus suggest that benzodiazepines do not lose their ability to attenuate the excitatory effect of CCK during long-term administration. Since both humans and rats show a rapid tolerance to the sedative effect of benzodiazepine (Haefely, 1978), it appears unlikely that this clinical effect of these drugs might be related to their ability to reduce the excitatory action of CCK. However, this neurobiological property of benzodiazepine might be related to their sustained anxiolytic activity. Interestingly, Gonsalves and Gallager (1987) have given an electrophysiological demonstration of complete tolerance to the potentiation of yaminobutyric acid (GABA) by benzodiazepine, following long-term treatment with diazepam without any consistent alteration in the number or the affinity of benzodiazepine receptors. Hence, the clear dissociation between this phenomenon and the persistence of the effects of benzodiazepine on the neuronal responsiveness to CCK would suggest that the effect of benzodiazepine on
CCK-induced activation might not be related to potentiation of GABAergic neurotransmission. Antidepressant drugs are also clinically efficacious anxiolytic agents (Kahn et al., 1987). Longterm admnistration of antidepressant drugs has been reported to reduce both the number of benzodiazepine receptors (Suranyi-Cadotte et al., 1984) and the electrophysiological effect of flurazepam (Bouthillier and De Montigny, 1987). However, the responsiveness of hippocampal neurons to CCK was unchanged in the latter study. This suggests that benzodiazepine and antidepressant drugs might produce their anxiolytic effect via different mechanisms.
Acknowledgements We thank D. Tardif for technical assistance, A. K h a n for typing the manuscript, and H o f f m a n n - L a R o c h e (Canada) for providing diazepam and flurazepam. A.B. was in receipt of a Canadian Medical Research Council (MRC) Studentship and C. De M. of an M R C Scientist Award. Supported in part by M R C Grant MT-6444.
References Bouthillier, A. and C. De Montigny, 1987, Long-term antidepressant treatment reduces neuronal responsiveness to flurazepam: an electrophysiological study in the rat, Neurosci. Lett. 73, 271.
138 Bradwejn, J. and C. De Montigny, 1984, Benzodiazepines antagonize cholecystokinin-induced activation of rat hippocampal neurons, Nature 312, 363. Bradwejn, J. and C. De Montigny, 1985, Effect of PK 8165, a partial benzodiazepine receptor agonist, on cholecystokinin-induced activation of hippocampal pyramidal neurons: a microiontophoretic study in the rat, European J. Pharmacol. 112, 415. Gonsalves, S.F. and D.W. Gallager, 1987, Time course for development of anticonvulsant tolerance and GABAergic subsensitivity after chronic diazepam, Brain Res. 405, 94. Haefely, W.E., 1978, Behavioral and neuropharmacological aspects of drugs used in anxiety and related states, in: Psychopharmacology: a Generation of Progress, M.A. Lipton, A. DiMascio and K.F. Killmann (eds.), (Raven Press, New York) p. 1359. Handelmann, G.~ D.K. Meyer, M.C. Beinfeld and W.H. Oertel, 1981, CCK-containing terminals in the hippocampus are
derived from intrinsic neurons: an immunohistochemical and radioimmunological study, Brain Res. 224, 180. Kahn, R.J., D.M. McNair and L.M. Frankenthaler, 1987, Tricychc treatment of generalized anxiety disorder, J. Affect. Dis. 13, 145. Kubota, K., K. Sugaya, M. Sunagane, I. Matsuda and Y. Matsuoka, 1985, Cholecystokinin antagonism by benzodiazepines in the contractile response of the isolated guineapig gallbladder, European J. Pharmacol. 110, 225. Kubota, K., N. Sunagane, K. Sugaya, T. Uruno and Y. Matsuoka, 1983, Competitive antagonism of cholecystokinin and some benzodiazepines at cholecystokinin receptors of smooth muscle, Jap. J. Pharmacol. 33, Suppl. 87P. Suranyi-Cadotte, B.E., T.V. Dam and R. Quirion, 1984, Antidepressant-anxiolytic interaction: decreased density of benzodiazepines receptors in rat brain following chronic administration of antidepressants, European J. Pharmacol. 106, 673.
135
Elsevier EJP 20145
Short communication
Long-term benzodi epine treatment reduces neuronal responsiveness to cholecystokinin: an electrophysiological study in the rat A l a i n Bouthillier a n d C l a u d e D e M o n t i g n y * Department of Psychiatry, McGill University, 1033 Pine Avenue West, MontrJal (QuJbec), Canada H3,4 1,41 Received 18 February 1988, accepted 19 April 1988
Acute benzodiazepine administration has been reported to antagonize the effect of cholecystokinin both in the periphery and in the central nervous system. A two-week treatment with either diazepam (5 mg/kg per day) or flurazepam (15 mg/kg per day) markedly reduced the excitatory effect of microiontophoretically applied sulphated cholecystokinin octapeptide-(26-33) on rat CA 3 hippocampal pyramidal neurons but not that of acetylcholine. In view of the sustained anxiolytic activity of benzodiazepines that contrasts with their transient sedative effect, the present results suggest that the reduction of neuronal responsiveness to cholecystokinin by these drugs might be related to their anxiolytic property. Cholecystokinin; Acetylcholine; Benzodiazepines; Sedation; Anxiety; Hippocampus
1. Introduction
Cholecystokinin (CCK) exists in high concentrations in the mammalian brain (Handelmann et al., 1981). Benzodiazepine receptor agonists have been reported to reduce the effect of CCK both in the periphery and in the central nervous system (Kubota et al., 1983; Bradwejn and De Montigny, 1984, 1985; Kubota et al., 1985). It is not yet proven that the antagonism by benzodiazepine of CCK in the central nervous system might be related to their anxiolytic or sedative activity. The clear dissociation between the transient sedative effect of benzodiazepine, due to a rapid development of tolerance, and their sustained anxiolytic activity (Haefely, 1978) were the basis for the present study. It was aimed to determine in vivo whether the reduction of neuronal responsiveness to CCK in the rat dorsal hippocampus would persist or vanish upon long-term administration of benzodiazepine.
* To whom all correspondence should be addressed.
2. Materials and methods
Male Sprague-Dawley rats weighing 250-300 g were used. Two groups were treated for 14 days with either diazepam (5 m g / k g per day i.p.) or flurazepam (15 m g / k g per day i.p.). The control groups were injected respectively with the diazepam vehicle (40% v / v propylene glycol, 10% v / v ethanol, 1.5% v / v benzylic alcohol, 48.5% v / v water and 5% sodium benzoate) or with a 0.9% saline solution. Each group comprised four or five rats. The rats were anesthetized with urethane (1.25 g / k g i.p.) 24 h after the last injection and were placed in a stereotaxic apparatus. Five-barrelled micropipettes were prepared in the conventional manner. The central barrel filled with 2 M NaC1 saturated with Fast Green F C F was used for extracellular recording of unitary activity. Recordings were obtained from pyramidal neurons of the CA 3 region of the dorsal hippocampus. These neurons were identified by their large amplitude (0.5-1.2 mV) long duration (0.8-1.2 ms) action potential, by the presence of complex spike
0014-2999/88/$03.50 © 1988 Elsevier Science Publishers B.V. (Biomedical Division)
136 discharges alternating with simple spike activity and by their tendency to exhibit rhythmic activity. The signal, amplified and displayed in the usual manner, was fed into a differential amplitude discriminator, generating square pulses from which integrated firing rate histograms were obtained. Three side barrels were used for ejection by microiontophoresis of the following substances: acetylcholine chloride (ACh) (20 m M in 200 m M NaC1, p H 4; Sigma), C C K (sulphated octapeptide-(26-33)) (10 ffM in 200 m M NaC1, p H 5; Sigma) and NaC1 (200 mM). This latter barrel was used to ascertain that the application of the same current used to eject C C K (50 nA) had no effect on neuronal activity. A leak or a small current (1-2 nA) of ACh was used during the electrode descent to activate and identify quiescent pyramidal neurons. The fourth side barrel, filled with 2 M NaC1, was used for automatic current balancing. Recording sites were marked by passing a - 27 ffA current through the central barrel for 20 min to make a Fast Green deposit, the location of which was verified histologically. The responsiveness of pyramidal neurons to C C K and ACh was assessed by determining the total number of spikes generated by 50 s microiontophoretic applications with constant currents (50 nA for C C K and 3 nA for ACh) in all rats. As reported from our laboratory and by others, some neurons showed a rapid tachyphylaxis to the excitatory effect of C C K (see fig. 1A). Hence, only the effect of the first application of C C K was taken into account. Most of the experiments were carried out in pairs of rats, each pair was comprised of a control rat and a treated rat, the experimenter being blind to the treatment. The same microelectrode was used to assess neuronal responsiveness in a pair of rats in order to minimize the possible contribution of fluctuations in transport numbers from one micropipette to another.
treatment with diazepam markedly reduced the number of pyramidal neurons which were activated by the standardized application of C C K (50 nA for 50 s): 24% were responsive to C C K in the diazepam group whereas 74% were responsive in the control group (table 1). The degree of neuronal activation by C C K was also markedly reduced by the diazepam treatment: the number of spikes generated by C C K application was reduced by 61% relative to the control group (fig. 1, table 1). The two-week treatment with flurazepam produced similar effects: only 36% of the pyramidal neurons responded to the standardized C C K application, as compared to 80% in the control group. The flurazepam treatment also diminished by 48% the degree of activation by C C K relative to the controls (table 1). Neither the diazepam nor the flurazepam treatments altered the effect of ACh
A - CONTROL ACh 3 E:3
o.
The spontaneous firing activity of hippocampal pyramidal neurons was similar in the control and benzodiazepine-treated rats (0.8 +_ 0.1 Hz (n = 43); 1.0 + 0.2 Hz (n = 43), respectively). The two-week
12]
I
I~
1'30
w xE oo
50 1 0
B - D I A Z E P A M (5 m g / k g / d a y × 14 days) ACh 3 r7
o LU OCO
[-i
CCK 50 II
I
1j j,j
IE3
100-
5O 0
3. Results
CCK -50 rm
2 min
Fig. 1. Integrated firing rate histograms showing the response of CA 3 hippocampal pyramidal neurons to microiontophoretic applications of acetylcholine (ACh) and sulphated octapeptide cholecystokinin (CCK) in a control rat (A) and in a rat treated with diazepam (B). Bars indicate the duration of the applications of the current shown in nA. The time base applies to both traces.
137 TABLE 1 Responsiveness of dorsal hippocampus CA 3 pyramidal neurons to microiontophoretic applications of acetylcholine (ACh) and cholecystokinin (CCK) following long-term diazepam or flurazepam treatment a. N u m b e r of neurons tested
Activated neurons (%)
Spikes g e n e r a t e d ± S.E.
ACh b
CCK b
ACh b
CCK b
Controls c Diazepam
23 21
23 (100) 21 (100)
17 (74) 5 (24)
353 ± 45 350_+ 49
425 _+52 167 _+29 e
Controls d Flurazepam
20 22
20 (100) 22 (100)
16 (80) 8 (36)
402 _+61 401 _+61
330 ± 42 174 ± 31 f
Treatment
Diazepam and flurazepam were administered i.p. for 14 days at a dose of 5 and 15 m g / k g per day, respectively. Electrophysiological experiments were carried out 24 h after the last injection, b A C h and C C K were applied for 50 s with currents of 3 and 50 nA, respectively. ~ These control rats received daily intraperitoneal injections of the diazepam vehicle for 14 days (see text), d These control rats received daily intraperitoneal injections of saline for 14 days. e p < 0.005 (compared with corresponding control value, two-tailed Student's t-test), f P < 0.05 (compared with corresponding control value, two-tailed Student's t-test). a
on the same neurons (table 1).
4. Discussion The present results show that a 14-day treatment with either diazepam or flurazepam reduced the responsiveness of rat dorsal hippocampus pyramidal neurons to CCK, whereas the effectiveness of ACh was unmodified. The data thus suggest that benzodiazepines do not lose their ability to attenuate the excitatory effect of CCK during long-term administration. Since both humans and rats show a rapid tolerance to the sedative effect of benzodiazepine (Haefely, 1978), it appears unlikely that this clinical effect of these drugs might be related to their ability to reduce the excitatory action of CCK. However, this neurobiological property of benzodiazepine might be related to their sustained anxiolytic activity. Interestingly, Gonsalves and Gallager (1987) have given an electrophysiological demonstration of complete tolerance to the potentiation of yaminobutyric acid (GABA) by benzodiazepine, following long-term treatment with diazepam without any consistent alteration in the number or the affinity of benzodiazepine receptors. Hence, the clear dissociation between this phenomenon and the persistence of the effects of benzodiazepine on the neuronal responsiveness to CCK would suggest that the effect of benzodiazepine on
CCK-induced activation might not be related to potentiation of GABAergic neurotransmission. Antidepressant drugs are also clinically efficacious anxiolytic agents (Kahn et al., 1987). Longterm admnistration of antidepressant drugs has been reported to reduce both the number of benzodiazepine receptors (Suranyi-Cadotte et al., 1984) and the electrophysiological effect of flurazepam (Bouthillier and De Montigny, 1987). However, the responsiveness of hippocampal neurons to CCK was unchanged in the latter study. This suggests that benzodiazepine and antidepressant drugs might produce their anxiolytic effect via different mechanisms.
Acknowledgements We thank D. Tardif for technical assistance, A. K h a n for typing the manuscript, and H o f f m a n n - L a R o c h e (Canada) for providing diazepam and flurazepam. A.B. was in receipt of a Canadian Medical Research Council (MRC) Studentship and C. De M. of an M R C Scientist Award. Supported in part by M R C Grant MT-6444.
References Bouthillier, A. and C. De Montigny, 1987, Long-term antidepressant treatment reduces neuronal responsiveness to flurazepam: an electrophysiological study in the rat, Neurosci. Lett. 73, 271.
138 Bradwejn, J. and C. De Montigny, 1984, Benzodiazepines antagonize cholecystokinin-induced activation of rat hippocampal neurons, Nature 312, 363. Bradwejn, J. and C. De Montigny, 1985, Effect of PK 8165, a partial benzodiazepine receptor agonist, on cholecystokinin-induced activation of hippocampal pyramidal neurons: a microiontophoretic study in the rat, European J. Pharmacol. 112, 415. Gonsalves, S.F. and D.W. Gallager, 1987, Time course for development of anticonvulsant tolerance and GABAergic subsensitivity after chronic diazepam, Brain Res. 405, 94. Haefely, W.E., 1978, Behavioral and neuropharmacological aspects of drugs used in anxiety and related states, in: Psychopharmacology: a Generation of Progress, M.A. Lipton, A. DiMascio and K.F. Killmann (eds.), (Raven Press, New York) p. 1359. Handelmann, G.~ D.K. Meyer, M.C. Beinfeld and W.H. Oertel, 1981, CCK-containing terminals in the hippocampus are
derived from intrinsic neurons: an immunohistochemical and radioimmunological study, Brain Res. 224, 180. Kahn, R.J., D.M. McNair and L.M. Frankenthaler, 1987, Tricychc treatment of generalized anxiety disorder, J. Affect. Dis. 13, 145. Kubota, K., K. Sugaya, M. Sunagane, I. Matsuda and Y. Matsuoka, 1985, Cholecystokinin antagonism by benzodiazepines in the contractile response of the isolated guineapig gallbladder, European J. Pharmacol. 110, 225. Kubota, K., N. Sunagane, K. Sugaya, T. Uruno and Y. Matsuoka, 1983, Competitive antagonism of cholecystokinin and some benzodiazepines at cholecystokinin receptors of smooth muscle, Jap. J. Pharmacol. 33, Suppl. 87P. Suranyi-Cadotte, B.E., T.V. Dam and R. Quirion, 1984, Antidepressant-anxiolytic interaction: decreased density of benzodiazepines receptors in rat brain following chronic administration of antidepressants, European J. Pharmacol. 106, 673.