- Email: [email protected]
Diazepam physical dependence and withdrawal in rats is associated with alteration in GABAA receptor function
Life Sciences, Vol. 59, No. 19, pp. 1631-1641, 1996 Copyright0 1996 Ekevier Science Inc. Printed in the USA. All rights resewed 0024-32m/% $15.00 t .lm
ELSEVIER
PII SOO24-3205(96)00494-S
DIAZEPAM PHYSICAL DEPENDENCE AND WITHDRAWAL IN RATS IS ASSOCIATED WITH ALTERATION IN GABAA RECEPTOR FUNCTION Sadamu Toki, Toshikazu
Saito, Shinichi
Hattal,
and Naohiko Takahata
Departments of Neuropsychiatry and Pharmacology’, School of Medicine, Sapporo Medical University, Chuo-ku, S.l, W.16, Sapporo, 060, Japan. (Received in final form August 28, 1996)
Summary Alteration in the function of the GABAA receptor complex and its relation to changes in withdrawal signs in diazepam (DZP)-dependent rats were studied. Physical dependence on DZP was induced in male F344 rats by using the drug-admixed food method. After cessation of treatment, withdrawal signs such as spontaneous convulsions were observed and withdrawal scores were maximal at 39 - 45 hr after the DZP withdrawal. Furthermore, these withdrawal signs almost disappeared by 159 - 168 hr after the DZP withdrawal. GABA-stimulated 3sCI- influx into cerebral cortical membrane vesicles was significantly decreased in rats 0 hr after DZP withdrawal and significantly increased in rats 42 hr after DZP withdrawal compared with control rats. Flunitrazepam (FZ)-induced potentiation and an antagonistic effect of Rol5-1788 on GABA-stimulated 360 influx were observed in control rats. No FZ-potentiated GABAstimulated 36CI- influx was observed in rats 0 hr after DZP withdrawal; however, such an effect of FZ was recognized in rats 42 hr and 162 hr after DZP withdrawal. No antagonistic effect of Ro15-1788 on the FZinduced stimulation was recognized in rats 0 hr and 42 hr after DZP withdrawal, but was recognized at 162 hr after DZP treatment, although it was not significant. In a [3H]FZ assay of binding to benzodiazepine (BZ) receptors, Bmax values were significantly decreased in rats 0 hr after DZP withdrawal, but increased at 42 hr after DZP withdrawal, compared with control rats. Bmax had almost returned to the control level at 162 hr after DZP treatment rats. In conclusion, these results indicate that functional changes in the GABAA/BZ receptor/Cl- channel complex, i.e. increased sensitivity in GABAA receptors and impairment in the functional coupling between BZ receptors and GABAA receptors, may possibly be involved in the biochemical mechanism of the severe withdrawal symptoms appearing after chronic treatment with DZP. Key Words: diazepam, physical dependence, withdrawal symptoms, %I- influx, GABA, receptor Benzodiazepines (BZs) are widely used as hypnotics, anxiolytics and anticonvulsants. BZs act via a receptor that is part of the y-aminobutyric acidA (GABAA) receptor to potentiate GABA-stimulated 3%I- influx (1). The binding of BZ agonists Corresponding 3041
Author:
Sadamu
Toki, M.D., Tel.’ (8111) 611-2111;
Fax: (8111)
644-
1632
Diazepam Dependence and GABA* Receptor
Vol. 59, No. 19, 1996
results in frequent opening of GABA-operated chloride ion channels, generally facilitating the inhibitory actions of GABA (2). Previous studies in rodents have shown that chronic exposure to BZ produces tolerance and physical dependence (3, 4). Clinical studies have also demonstrated that long-term use of BZ often results in tolerance to and dependence on many of the therapeutic actions of BZ (56). Furthermore, severe discontinuation syndrome has been reported with symptoms such as seizures and even mortality in humans (7,8). Thus, the mechanisms by which BZ tolerance and dependence are mediated have been the focus of much recent interest. Whereas the functional changes of the GABA* receptor have been suggested as a mechanism to explain BZ tolerance (9,10,11,12), only a few studies have reported the role of GABAA receptor function in the development of withdrawal symptoms such as alteration in motor activity (13, 14), and there has been no report on the relation of GABAA receptor function to severe withdrawal symptoms such as spontaneous convulsions. One of the reasons for this is the lack of an animal model of overt BZ dependence which develops severe withdrawal symptoms including tremors and convulsions after cessation of BZ treatment. Recently, Suzuki and co-workers have developed an escalating-dose-schedule method (a drug-admixed food [DAF] method) to produce physical dependence upon diazepam (DZP) (15). With this procedure, rats become severely dependent upon DZP, and subsequent withdrawal convulsions can be induced. In the present study, we employed the DAF method to produce physical dependence and withdrawal signs in rats as an animal model, and investigated alterations in the function of the GABAA receptor complex and their relation to changes in withdrawal signs in DZP-dependent rats. The results presented here indicate that the function of the GABAA/BZ receptor/Cl- channel complex was altered in withdrawal rats after chronic DZP treatment, and suggest that this functional alteration may be associated with BZ withdrawal symptoms, including spontaneous convulsions.
Materials
and
Methods
DZP Treatment Male Fischer 344 (F344) rats, 6 weeks of age at the beginning of the experiment, were obtained from Charles River Japan, Inc. (Atsugi, Japan). Rats were housed individually under standard laboratory conditions of lighting (12-hr light-dark cycle) and temperature (24 + 20°C). They were fed the DZP-admixed food for 30 days according to the method of Suzuki et al. (15) and drank tap water ad libitum. The concentration of DZP in the food was gradually increased from 1 to 12 mg per 1 g of food. Control rats were fed a standard diet and housed in the same laboratory conditions. Rats were sacrificed at 0 hr (i.e. immediately after the 30-day DZP treatment; tolerant rats), 42 hr and 162 hr after the final DZP treatment. Estimation of Withdrawal Scores Withdrawal was conducted by substituting normal food for DZP-admixed the last day of the treatment at 4:00 p.m. Withdrawal signs were evaluated at times after termination of DZP-treatment as described by Suzuki et al. (15). To the intensity of physical dependence upon DZP, withdrawal signs were scored point scale as summarized in Table 1.
food on various quantify on a 5-
Blood DZP Concentration DZP concentrations in plasma of DZP-treated rats were determined by highperformance liquid chromatography (SRL Inc., Japan) according to the method of Brodie et al. (16).
Vol. 59, No. 19, 1996
Diazepam Dependence and GABA* Receptor
1633
Preparation of Membrane Vesicles The cerebral cortices were rapidly dissected and membrane vesicles were prepared as described by Obata and Yamamura (17). Briefly, the cerebral cortices obtained were homogenized by hand (12 strokes) using a glass-glass homogenizer in 10 volumes of ice-cold 10 mM HEPES buffer (pH 7.5) containing 145 mM NaCI, 5 mM KCI, 1 mM MgCl2, 10 mM D-CJhCOSe, and 1 mM CaCl2. The homogenates were then centrifuged at 1,000 x g for 15 min at 4°C. Pellets were resuspended in 10 mM HEPES buffer and washed three times under the same conditions. The final pellet was resuspended in 10 mM HEPES buffer with a final protein content of 7 - 8 mg protein per ml of suspension. Preparation of Svnaptic Membrane Synaptic membrane fractions were prepared from the cerebral cortex as described by Zukin et al. (18). Cerebral cortices were homogenated in 10 volumes of ice-cold 0.32 M sucrose buffer and the homogenates were centrifuged at 1,000 x g for 10 min at 4°C. Supernatants were collected and centrifuged at 20,000 x g for 20 min at 4°C. Pellets were suspended in distilled water and centrifuged at 8,000 x g for 20 min at 4°C. Supernatants obtained were diluted with 20 ml of 50 mM Tris-citrate buffer (pH 7.1) and centrifuged three times at 48,000 x g for 20 min at 4°C. The final pellet was resuspendend in the above buffer and stored at -80°C for at least 18 hr before the receptor binding assay as described (18). Measurement of 360 -- Influx The influx of 36CI- into membrane vesicles was measured by the method of Obata and Yamamura (17) with a minor modification. The membrane vesicle suspensions (100 ~1) were preincubated with each drug for 10 min at 30°C. After preincubation, 360 influx was initiated by the addition of 100 ml of buffer containing 36CI- (0.2 PCi) and the indicated concentrations of GABA. The influx was terminated after 10 set by the addition of 3 ml of ice-cold buffer followed by rapid vacuum filtration through a GF/C glass fiber filter pretreated with 0.05% polyethylenimine. The filters were washed three times with 3 ml of ice-cold 10 mM HEPES buffer, and bound radioactivity was quantified in a Beckman LS 5801 counter. GABA-stimulated 36CIinflux was calculated as the difference between basal uptake in the absence of GABA and total uptake in the presence of GABA. Receptor Binding Assay [3H]Flunitrazepam (FZ) was used for the BZ receptor binding assay in the concentration range of 1 to 40 nM in saturation binding studies. The final reaction mixture volume was 1 ml of 50 mM Tris-citrate buffer containing 0.2-0.5 mg of membrane preparations. Reactions were initiated by adding [3H]FZ (approximately 2 x 1 O5 dpm) and performed at 2°C for 60 min. They were quenched by the addition of icecold 50 mM Tris-citrate buffer. Bound and free ligands were separated by rapid vacuum filtration (Brandel Cell Harvester M 24R) on Whatman GF/B filters. The filter was washed three times with 3 ml of the same ice-cold buffer, and bound radioactivity was quantified in a Beckman LS 6000LL counter. Specific binding was defined as the difference between [3H]FZ binding in the absence and presence of 10 mM FZ. Binding parameters were analyzed to obtain the values of Kd and Bmax by using Scatchard transformation. Protein Determination and Data Analysis Protein content was determined by the Coomassie Blue binding method using bovine serum albumin as a standard (19). Statistical analyses of data were performed by one-way ANOVA or Student’s ttest. P values of less than 0.05 were taken to indicate a significant difference in the mean values being compared.
1634
Diazepam Dependence and GABA, Receptor
Vol. 59, No. 19, 1996
Materials [3H]FIunitrazepam (3.0 TBq/mmol) was purchased from New England Nuclear. Na36CI- (4.0 MBq/nmol) was from Amersham. Diazepam and flunitrazepam were generous gifts from Dr. Tsutomu Suzuki and Esai Pharmaceutical Co., Ltd., respectively. All other reagents were of analytical grade. Results DZP Treatment and Withdrawal Siuns At the first session of the treatment, rats were given two kinds of drug-admixed food containing 1 mg of DZP per lg of food and 2 mg of DZP per lg of food for five days according to the protocol described previously (15). The mean DZP intake of rats in the first session was 73.3 f 5.6 mg/kg/day. The concentration of DZP in the food was gradually increased. The mean DZP intake during feeding with the final DZP concentration (12 mg/g food, for six days) was 701.1 f 14.1 mg/kg/day. At the time of the last treatment, DZP concentrations in plasma reached 249.1 f 24.9 ng/ml. After termination of DZP treatment, withdrawal signs such as weight loss, vocalization, irritability, muscle rigidity, ear-twitching, and piloerection were observed in all eight DZP-treated animals. Straub’s tail, nosebleed and tremors were observed in six of the eight animals. Fascicular twitches and spontaneous convulsions were observed in four of eight rats. The total withdrawal score observed during the seven days after withdrawal was 21.3 rl: 1.8 (Table 1). Withdrawal scores of rats were maximal at 39 - 45 hr after the final treatment and they had almost disappeared at 162 hr after the final treatment (Fig. 1) Effect of Chronic DZP Treatment on GABA-stimulated 36CI- Influx In control rats, the addition of GABA to the membrane vesicle suspension resulted in stimulation of scCl- influx in a dose-dependent manner (5 PM, 11.7 * 2.3 TABLE Maximal
Behavioral
Changes
I
During Diazepam
Positive Animals /Total Withdrawal
Score
Weight loss 5-l 0% 1o-1 5% Vocalization Irritability Muscle rigidity Ear-twitching Piloerection Straub’s tail Fascicular twitch Nosebleed Agression Tremor Spontaneous convulsion Death Withdrawal
scores
;;I (2) (2) (2) (2) (3) (3) I:; I;;
Withdrawal Number of Animals
Control
DZP treated
O/8 O/8 O/8 O/8 O/8 O/8 O/8 018 O/8 O/8 O/8 018
2J8 618 818 818 818 818 818 618 418 618 O/8 618
018 O/8
418 O/8
0
21.3k1.8
Vol. 59, No. 19, 1996
Diazepam Dependence and GABA* Receptor
1635
20
18 $’
16
8
14
0
12
24
36
48
60
72
84
96
Time after withdrawal
108120132144156168
(hours)
Fig. 1. Changes in withdrawal scores after DZP treatment. Withdrawal was conducted by substituting normal food for diazepam-admixed food. Withdrawal signs were scored on the five-point scale shown in Table 1 as described by Suzuki et al. (15). Each point represents the mean + SE of eight animals.
nmol/mg protein/l 0 set; 10, 19.8+1.6; 20, 27.3 + 2.1; 40, 35.9 f 3.8; 200, 42.0+1.9). The addition of 10 PM GABA resulted in the nearly half maximal stimulation of %Iinflux and maximal stimulation was observed with 200 PM GABA. In tolerant rats (0 hr), although 10 PM GABA-stimulated 36CI- influx into membrane vesicles of rats was not different from in control rats, 200 PM GABA-stimulated %I- influx was significantly decreased in comparison with control rats (Fig. 2). At 42 hr after DZP treatment of rats, both 10 /JM and 200 PM GABA-stimulated %I- influxes were significantly increased in comparison with control rats. At 162 hr after DZP treatment of rats, the increases in GABA-stimulated 3sCI- influx observed at 42 hr were reversed to levels similar to those in control animals (Fig. 2). The stimulatory effect of GABA on 36CI- influx observed in each group was effectively inhibited by the addition of the GABA antagonist, bicuculline (SIC) or the chloride channel blocker, picrotoxinin (PIC) (Fig. 3). These results confirmed that GABA-stimulated %I- influx reflected the functional coupling of the GABA,+, receptor with the chloride channel. On the other hand, no blocking action of Ro15-1788 (10 PM), a BZ antagonist, on 10 PM and 200 PM GABA-stimulated 3sC1influx was observed in control and DZP-treated rats (results not shown), indicating that there was no contamination by DZP, which might affect the measurement of the GABAstimulated 3sCI- influx, in the membrane vesicle preparation after the DZP treatment of rats. As shown in Fig. 4, the addition of 1 PM FZ to the assay mixture significantly potentated 10 PM GABA-stimulated 36CI- influx in the control rats and, similar to earlier studies in which this assay procedure was used (20, 21), it was suppressed by the BZ
Diazepam Dependence and GABA* Receptor
1636
so-
*
Vol. 59, No. 19, 1996
.
GABA 200 /JM
0
GABA 10 pM
5040-
I
30 PO-
P
IOO-lControl
dhr
162 hr
42 hr
Time after withdrawal Fig. 2. Effects of 10 PM and 200 PM GABA on 3sCI- influx in rats at 0 hr (tolerant rats), 42 hr and 162 hr after DZP withdrawl. Each point represents the mean + SE of 10 to 12 experiments. *Significantly different from each control value. (pcO.05)
Control
0 hr
0 0
GABA 10j1M PIG
0
BIC
42 hr
162 hr
Time after withdrawal Fig. 3. Effects of 100 /JM PIC and 1 mM BIC on 10 PM GABA-stimulated 3%I- influx in rats at 0 hr (tolerant rats), 42 hr and 162 hr after DZP withdrawal. Each point represents the mean +: SE of 10 to 12 experiments. *Significantly different from the corresponding value with 10 PM GABA alone.
Diazepam Dependence and GABAA Receptor
Vol. 59, No. 19, 1996
200
0
*
0
FZ FZ + Rol5-
1637
1788
175 z I
5(d 150 i
2 5
125 1 +
E g_ 100 i 75-l
7 Control
O’hr
I
42 hr
162 hr
Time after withdrawal Fig. 4 Enhancement of 10 PM GABA-stimulated WI- influx by the addition of 1 /.fM FZ and inhibition by the addition of 10 PM Rol51788 in rats at 0 hr (tolerant rats), 42 hr and 162 hr after DZP withdrawal. Data are expressed as a percentage of the value of 10 PM GABA-stimulated 36CI- influx. Each point *Significantly different represents the mean 2 SE of 10 to 12 experiments. from the corresponding value with 10 PM GABA alone. +Significantly different from the value with 10 PM GABA plus 1 PM FZ.
antagonist, Rol5-1788 (10 PM). However, FZ-induced potentiation and an antagonistic effect of Rol5-1788 on GABA-stimulated 36CI- influx were not observed in tolerant (0 hr) rats. In contrast, FZ significantly enhanced GABA-stimulated 3sCI- influx into membrane vesicles at 42 hr after DZP treatment of rats, but it was not effectively suppressed by the addition of Ro15-1788. Furthermore, at 162 hr after DZP treatment, FZ significantly WI- influx and an antagonistic effect of Ro15-1788 potentiated the GABA-stimulated on the FZ-induced stimulation was recognized, although it was not significant (Fig. 4). Effects of Chronic DZP Treatment on the Receptor Binding Assav Analysis of [sH] FZ binding to BZ receptors in cortical membrane preparations showed a significant decrease in binding sites (Bmax) in the tolerant (0 hr) animals compared with the control. In contrast, the Bmax value in the 42-hr-after-DZP-treatment group was significantly higher than that in the control, and the increase in Bmax had almost returned to the control level in 162-hr-after-DZP-treatment rats. On the other hand, there were no significant differences in the dissociation constant (Kd) among the four groups (Table 2). Discussion Several previous studies using experimental animals have shown that behavioral tolerance to BZ is associated with down-regulation of BZ receptors (9), and reduced functional coupling of the BZ recognition site with the GABAA receptor/Clchannel (10). However, the relationship between withdrawal symptoms and alteration of GABAA receptor functions has not been described in detail, although some reports
1638
Diazepam Dependence and GABA, Receptor
TABLE [sH]Flunitrazepam
Control 0 hr 42 hr 162 hr
II
binding to benzodiazepine membranes Kd (nM) 3.47 3.49 3.81 3.61
+ 0.06 -+ 0.19 +0.12 r 0.11
Vol. 59, No. 19, 1996
receptors
in rat synaptic
Bmax (pmol/mg protein) 1.84 1.30 2.18 1.90
+ f + f
0.05 0.03* 0.02* 0.01
Benzodiazepine receptor binding in DZP-dependent rats at 0 hr (tolerant rats), 42 hr and 162 hr after DZP withdrawal. Binding was determined in cerebral cortical synaptic membrane using the ligand [sH]flunitrazepam. Data represent the mean + SE of six experiments. *Significantly different from control (~~0.05).
have demonstrated the relationship between the open-field activity and alteration in the function of GABA* receptors during the BZ-withdrawal state in mice (13,14). In contrast to animal studies, various withdrawal symptoms were recognized in patients with BZ dependence, not only alteration in motor activity but also severe withdrawal symptoms such as tremors, delirium and spontaneous convulsions (7, 8). However, studies examining the molecular mechanism for severe withdrawal signs, including convulsions, have been difficult to perform due to the lack of an animal model presenting severe withdrawal signs after BZ is withdrawn. Thus, the mechanisms underlying serious physical dependence have been studied less extensively. A recent report described how the OAF method can produce physical dependence on DZP and subsequent withdrawal convulsions in rats (15). Therefore, in the present study, to investigate the role of GABAA receptor function in development of physical dependence on DZP and subsequent severe withdrawal signs, we employed the OAF method for inducing physical dependence on DZP. Although the mean DZP intake (701 .l mg/kg/day) during feeding with the final DZP concentration was a little lower than that described by Suzuki et al. (15) (773.3 mg/kg/day), the blood concentration of DZP in DZP-treated rats was essentially the same as that reported in their study (15). After cessation of DZP treatment, various withdrawal signs could be observed (Table l), and the withdrawal scores reached the maximum at 39 - 45 hr after the final DZP treatment (Fig. 1). These withdrawal signs had almost disappeared at 159 - 168 hr after the final DZP treatment. Therefore, we examined the function of GABAA receptors immediately after the final DZP treatment (0 hr, tolerant rats), and at 42 hr and 162 hr after the final DZP treatment, In tolerant rats (0 hr after DZP withdrawal), stimulation of 360 influx by 200 PM GABA, which produced maximal stimulation of ssCI- influx, was significantly decreased compared with that in control rats, although 10 PM (a half-maximally effective concentration for WI- influx) GABA-stimulated %I- influx was not changed (Fig. 2). This result is in accord with the findings of Miller et al. (13) and Galpern et al. (14) of a decrease in 3W- influx caused by a saturation level of muscimol and the lack of change in 360 influx caused by the ECsO level of the agonist in BZ-treated rats. The reduction in GABA (200 PM)-stimulated WI- influx seemed to indicate that the GABAA receptors or chloride channels decreased in tolerant rats. In our previous study, the number of binding sites of GABAA receptors was not altered in DZP-tolerant rats (22).
Vol. 59, No. 19, 1996
Diazepam Dependence and GABA* Receptor
1639
[=S]T BPS (ta trend toward decreased Bmax of Furthermore, butylbicyclophosphorothionate) binding to chloride channels after BZ treatment has been demonstrated (13, 14). Therefore, reduction in the saturation level of GABAstimulated 360 influx might have been due to a decrease in the number of chloride channels after chronic DZP treatment. The reason for lack of reduction in 10 PM GABAstimulated 36CI- influx in tolerant rats is unclear. We have previously observed a slight increase in the affinity of the GABAA receptor, although it was insignificant, in DZPtolerant rats (22). Such alteration in the affinity of the GABAA receptor might have an influence on the modulation of 36Cl- influx by stimulation with the EC& level of GABA in tolerant rats. Further studies are needed to resolve this question. Some previous studies have indicated that there is no significant change in GABA-stimulated 36Cl- influx after chronic administration of BZ to rats (10,ll). The apparent discrepancy may be explained by differences in the methods of drug administration. The chronic treatment with BZ used in earlier studies, in which withdrawal signs and the blood concentrations of BZ were not described, might not have been sufficient to cause an alteration in the effect of GABA on 36CI- influx or severe withdrawal signs. Miller and his co-workers (13,14) have observed decreased muscimol-stimulated 360 influx before the appearance of withdrawal signs in chronically BZ-treated rats. Therefore, it is assumed that decreased sensitivity of GABAA receptors to GABA may be required for the initial biochemical changes causing the appearance of BZ withdrawal signs. On the other hand, in withdrawal rats at 42 hr after the cessation of DZP treatment, GABA (10 I.IM and 200 PM) -stimulated 36CI- influx was increased in comparison with control rats (Fig. 2). Our previous study indicated that the affinity of high affinity sites for GABAA receptors to GABA, estimated by [3H]muscimol binding, was increased at 42 hr after DZP withdrawn (22). It may be suggested, therefore, that potentiation in the stimulatory effect of GABA on 36CI- influx results from the increased sensitivity of GABAA receptors to GABA, which may lead to enhancement in the coupling of the GABAA receptor with the chloride channel. FZ effectively enhanced GABA-stimulated 360 influx in control rats (Fig. 4). However, potentiation of GABA-stimulated 36CI- influx by FZ was not observed in tolerant rats (0 hr after DZP withdrawal). In contrast, FZ significantly potentiated GABAstimulated 36CI- influx at 42 hr and 162 hr after the final DZP treatment. It seems that the alteration in the ability of FZ to potentiate GABA-stimulated 36CI- influx observed in rats after withdrawal may not be explained only by the changes in the BZ receptors. There was no change in the dissociation constant (Kd) of BZ receptors between control rats and rats after withdrawn of DZP (Table 2). Furthermore, although the number of binding sites (Bmax) of BZ receptors was increased at 42 hr compared with that in the control, the extent of FZ-induced enhancement of GABA-stimulated 36CI- influx was not as great as that in the control (control, 168.7 f 16.0%; 42 hr, 142.7 f 14.7%) (Table 2 and Fig. 4). In addition, whereas the increase in the number of binding sites of BZ receptors observed at 42 hr returned to the control level at 162 hr after the DZP withdrawal, enhancement of FZ in GABA-stimulated 36CI- influx did not completely return to the control level (Fig. 4). Thus, alterations in the number of binding sites of BZ receptors did not appear to be consistent with the effect of FZ on GABA-stimulated 36CI- influx in rats after DZP withdrawal. Alternatively, previous studies have suggested reduction of functional coupling of the BZ receptors with the GABAA receptors in chronically BZtreated rats (10, 12). It may be possible, therefore, that the functional coupling between BZ receptors and GABAA receptors was impaired by chronic treatment of rats with DZP, which was related to behavioral tolerance to BZ in tolerant rats. This impairment of the coupling appeared to last after the DZP withdrawal and not only the changes in the BZ receptors but also the impairment of the coupling might be involved in the appearance of severe withdrawal signs as indicated in the present study (Table 1). This might be
1640
Diazepam Dependence and GABAA Receptor
Vol. 59, No. 19, 1996
supported by the findings that the effect of the BZ antagonist, Rol5-1788, was not observed in tolerant rats (0 hr after DZP withdrawal) and in rats at 42 hr after the withdrawal, whereas it effectively suppressed FZ-induced potentiation of GABAstimulated ssCI- influx in controls (Fig. 4). At 162 hr after the DZP withdrawal, the antagonistic effect of Rol5-1788 was recognized, although it was not statistically significant (Fig. 4). This may suggest that impairment in the functional coupling between BZ receptors and GABAA receptors tended to recover toward the control level at 162 hr after the withdrawal, but this recovery was still insufficient, which might relate to the appearance of the few withdrawal signs observed at 162 hr (Fig. 1). In summary, using DZP-dependent rats produced by the DAF method, we have demonstrated that functional changes in the GABAA/BZ receptor/Cl- channel complex, i.e. increased sensitivity in GABAA receptors, and impairment in the functional coupling between BZ receptors and GABAA receptors, may possibly be involved in the manifestation of severe withdrawal symptoms such as tremors and convulsions after chronic treatment with DZP. The current understanding of the molecular structure of the GABA* receptor is that it is a heteropentameric glycoprotein composed of multiple polypeptide subunits, i.e. al-6, pl-4, yl-3,, 6 and ~1-2 (23). Previous studies have reported that chronic benzodiazepine treatment decreases the al (24), a5 and y2 (25) subunit mRNA levels in rat cerebral cortex. Furthermore, previous studies have suggested that [sH]FZ is bound to the a subunit of the GABAA receptor (26) Therefore, these changes of subunits may be concerned with down-regulation of BZ receptors (Table 2) and the reduced functional coupling of BZ receptors with GABAA receptors observed in this study in tolerant rats. Alternatively, it is known that the function of Nmethyl-D-aspartate (NMDA) receptors is related to convulsions or other seizures of epilepsy. Recent evidence has suggested that the NMDA antagonist, 3-[(&)-2carboxypiperazin-4-yl]-propyl-1-phosphonate prevents the withdrawal symptoms of DZP and that NMDA-dependent mechanisms may contribute to the development of tolerance to DZP and to expression of withdrawal signs (27). Thus, the behavioral appearance of signs of withdrawal in BZ dependence might be caused by alteration in the functional interaction between the GABAA receptor complex and the NMDA receptor. Further studies with respect to these aspects are required to more definitively establish the molecular mechanism of BZ dependence. Acknowledgments This work was partly supported by a research grant from the Ministry of Health and Welfare, Japan. The authors thank Drs. Tsutomu Suzuki and Hirokazu Mizoguchi for helpful discussions and Dr. Akemi Nabeshima for technical assistance. We also thank Esai Pharmaceutical Co., Ltd. for the generous gift of flunitrazepam. References 1. 2. 3. 4. 5. 6. 7. 8.
W. HAEFELY, E. KYBURZ, M. GERECKE and H. MCHLER, Adv. Drug Res. 14 165-322 (1985). R.E. STUDY and J.L. BARKER, J. Am. Med. Ass. 247 2147-2151 (1982). SE. FILE, Neurosci. Biobehav. Rev. 9 113-121 (1985). J.P. GONZALEZ, A.J. McCULLOCH, P.J. NICHOLLS, R.D. SEWELL and A. TEKLE, Alcohol Alcohol. 19 325-332 (1984). K. RICKELS, W.G. CASE, R. W. Downing and A. Winokur, J. Am. Med. Ass. 250 767-771 (1983). C. HALLSTOM and M. LADER, Int. Pharmacopsychiatry 16 235-244 (1981). R.T. OWEN and P. TYRER, Drugs 25 385-398 (1983). H. PETURSSON and M.H. LADER, Br. J. Addict. 76 133-145 (1981).
Vol. 59, No. 19, 1996
9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.
21. 22. 23. 24. 25. 26. 27.
Diazepam Dependence and GABA* Receptor
1641
L.G. MILLER, D.J. GREENBLATT, J.G. BARNHILL and R.I. SHADER, J. Pharmacol. Exp. Ther. 246 170-l 76 (1988). M. Li, H.C. ROSENBERG and T.H. CHIU, Eur. J. Pharmacol. 247 313-318 (1993). D.O. NGUR, H.C. ROSENBERG and T.H. CHIU, Eur. J. Pharmacol. 176 351-356 (1990). 0. YU, T.H. CHIU and H.C. ROSENBERG, J. Pharmacol. Exp. Ther. 246 107-l 13 (1988). L.G. MILLER, D.J. GREENBLATT, R.B. ROY, W.R. SUMMER and R.I. SHADER, J. Pharmacol. Exp. Ther. 246 177-181 (1988). W.R. GALPERN, M. LUMPKIN, D.J. GREENBLATT, R.I. SHADER and L.G. MILLER, Psycopharmacology. 104 225-230 (1991). T. SUZUKI, M.-S. LU, H. MOTEGI, T. YOSHIDA and M. MISAWA, Pharmacol. Biochem. Behav. 43 387-393 (1992). RR BRODIE, L.F. CHASSEAUD and T. TAYLOR, J. Chromatogr. 150 361-366 (1978). T. OBATA and H.I. YAMAMURA, Life Sci. $2 659-665 (1988). S.R. ZUKIN, A.B. YOUNG and S.H. SNYDER, Proc. Natl. Acad. Sci. USA 71 4802-4807 ( 1974). M.M. BRADFORD, Anal. Biochem. 72 248-254 (1976). T. OBATA, M. MORELLI, A. CONCAS, M. SERRA and H.I. YAMAMURA, Chloride Channels and Their Modulation bv Neurotransmitters and Drugs. Giovanni Biggio and E. Costa (Eds), 175-l 87, Raven Press, New York (1988). M. HIROUCHI, J. TAGUCHI, T. UEHA and K. KURIYAMA, Biochem. Biophys. Res. Commun. 146 1471-1477 (1987). S. TOKI, T. SAITO, A. NABESHIMA, S. HATTA, M. WATANABE and N. TAKAHATA, Alcohol. Clin. Exp. Res. 20 40A-44A (1996). R.L. MACDONALD and R.W. OLSEN, Annu. Rev. Neurosci. 17 569-602 (1994). E.I. TIETZ, X. HUANG, X. WENG, H.C. ROSENBERG and T.H. CHIU, J. Mol. Neurosci. 2 277-292 (1993). Y. WU, H.C. ROSENBERG, T.H. CHIU and T.J. ZHAO, J. Mol. Neurosci. 5 105. 120 (1994). K. FUCHS, H. MCHLER and W. SfEGHART, Neurosci. Lett. $Q 314-319 (1988). K.G. STEPPUHN and L. TURSKI, Proc. Natl. Acad. Sci. USA 90 6889-6893 (1993).
ELSEVIER
PII SOO24-3205(96)00494-S
DIAZEPAM PHYSICAL DEPENDENCE AND WITHDRAWAL IN RATS IS ASSOCIATED WITH ALTERATION IN GABAA RECEPTOR FUNCTION Sadamu Toki, Toshikazu
Saito, Shinichi
Hattal,
and Naohiko Takahata
Departments of Neuropsychiatry and Pharmacology’, School of Medicine, Sapporo Medical University, Chuo-ku, S.l, W.16, Sapporo, 060, Japan. (Received in final form August 28, 1996)
Summary Alteration in the function of the GABAA receptor complex and its relation to changes in withdrawal signs in diazepam (DZP)-dependent rats were studied. Physical dependence on DZP was induced in male F344 rats by using the drug-admixed food method. After cessation of treatment, withdrawal signs such as spontaneous convulsions were observed and withdrawal scores were maximal at 39 - 45 hr after the DZP withdrawal. Furthermore, these withdrawal signs almost disappeared by 159 - 168 hr after the DZP withdrawal. GABA-stimulated 3sCI- influx into cerebral cortical membrane vesicles was significantly decreased in rats 0 hr after DZP withdrawal and significantly increased in rats 42 hr after DZP withdrawal compared with control rats. Flunitrazepam (FZ)-induced potentiation and an antagonistic effect of Rol5-1788 on GABA-stimulated 360 influx were observed in control rats. No FZ-potentiated GABAstimulated 36CI- influx was observed in rats 0 hr after DZP withdrawal; however, such an effect of FZ was recognized in rats 42 hr and 162 hr after DZP withdrawal. No antagonistic effect of Ro15-1788 on the FZinduced stimulation was recognized in rats 0 hr and 42 hr after DZP withdrawal, but was recognized at 162 hr after DZP treatment, although it was not significant. In a [3H]FZ assay of binding to benzodiazepine (BZ) receptors, Bmax values were significantly decreased in rats 0 hr after DZP withdrawal, but increased at 42 hr after DZP withdrawal, compared with control rats. Bmax had almost returned to the control level at 162 hr after DZP treatment rats. In conclusion, these results indicate that functional changes in the GABAA/BZ receptor/Cl- channel complex, i.e. increased sensitivity in GABAA receptors and impairment in the functional coupling between BZ receptors and GABAA receptors, may possibly be involved in the biochemical mechanism of the severe withdrawal symptoms appearing after chronic treatment with DZP. Key Words: diazepam, physical dependence, withdrawal symptoms, %I- influx, GABA, receptor Benzodiazepines (BZs) are widely used as hypnotics, anxiolytics and anticonvulsants. BZs act via a receptor that is part of the y-aminobutyric acidA (GABAA) receptor to potentiate GABA-stimulated 3%I- influx (1). The binding of BZ agonists Corresponding 3041
Author:
Sadamu
Toki, M.D., Tel.’ (8111) 611-2111;
Fax: (8111)
644-
1632
Diazepam Dependence and GABA* Receptor
Vol. 59, No. 19, 1996
results in frequent opening of GABA-operated chloride ion channels, generally facilitating the inhibitory actions of GABA (2). Previous studies in rodents have shown that chronic exposure to BZ produces tolerance and physical dependence (3, 4). Clinical studies have also demonstrated that long-term use of BZ often results in tolerance to and dependence on many of the therapeutic actions of BZ (56). Furthermore, severe discontinuation syndrome has been reported with symptoms such as seizures and even mortality in humans (7,8). Thus, the mechanisms by which BZ tolerance and dependence are mediated have been the focus of much recent interest. Whereas the functional changes of the GABA* receptor have been suggested as a mechanism to explain BZ tolerance (9,10,11,12), only a few studies have reported the role of GABAA receptor function in the development of withdrawal symptoms such as alteration in motor activity (13, 14), and there has been no report on the relation of GABAA receptor function to severe withdrawal symptoms such as spontaneous convulsions. One of the reasons for this is the lack of an animal model of overt BZ dependence which develops severe withdrawal symptoms including tremors and convulsions after cessation of BZ treatment. Recently, Suzuki and co-workers have developed an escalating-dose-schedule method (a drug-admixed food [DAF] method) to produce physical dependence upon diazepam (DZP) (15). With this procedure, rats become severely dependent upon DZP, and subsequent withdrawal convulsions can be induced. In the present study, we employed the DAF method to produce physical dependence and withdrawal signs in rats as an animal model, and investigated alterations in the function of the GABAA receptor complex and their relation to changes in withdrawal signs in DZP-dependent rats. The results presented here indicate that the function of the GABAA/BZ receptor/Cl- channel complex was altered in withdrawal rats after chronic DZP treatment, and suggest that this functional alteration may be associated with BZ withdrawal symptoms, including spontaneous convulsions.
Materials
and
Methods
DZP Treatment Male Fischer 344 (F344) rats, 6 weeks of age at the beginning of the experiment, were obtained from Charles River Japan, Inc. (Atsugi, Japan). Rats were housed individually under standard laboratory conditions of lighting (12-hr light-dark cycle) and temperature (24 + 20°C). They were fed the DZP-admixed food for 30 days according to the method of Suzuki et al. (15) and drank tap water ad libitum. The concentration of DZP in the food was gradually increased from 1 to 12 mg per 1 g of food. Control rats were fed a standard diet and housed in the same laboratory conditions. Rats were sacrificed at 0 hr (i.e. immediately after the 30-day DZP treatment; tolerant rats), 42 hr and 162 hr after the final DZP treatment. Estimation of Withdrawal Scores Withdrawal was conducted by substituting normal food for DZP-admixed the last day of the treatment at 4:00 p.m. Withdrawal signs were evaluated at times after termination of DZP-treatment as described by Suzuki et al. (15). To the intensity of physical dependence upon DZP, withdrawal signs were scored point scale as summarized in Table 1.
food on various quantify on a 5-
Blood DZP Concentration DZP concentrations in plasma of DZP-treated rats were determined by highperformance liquid chromatography (SRL Inc., Japan) according to the method of Brodie et al. (16).
Vol. 59, No. 19, 1996
Diazepam Dependence and GABA* Receptor
1633
Preparation of Membrane Vesicles The cerebral cortices were rapidly dissected and membrane vesicles were prepared as described by Obata and Yamamura (17). Briefly, the cerebral cortices obtained were homogenized by hand (12 strokes) using a glass-glass homogenizer in 10 volumes of ice-cold 10 mM HEPES buffer (pH 7.5) containing 145 mM NaCI, 5 mM KCI, 1 mM MgCl2, 10 mM D-CJhCOSe, and 1 mM CaCl2. The homogenates were then centrifuged at 1,000 x g for 15 min at 4°C. Pellets were resuspended in 10 mM HEPES buffer and washed three times under the same conditions. The final pellet was resuspended in 10 mM HEPES buffer with a final protein content of 7 - 8 mg protein per ml of suspension. Preparation of Svnaptic Membrane Synaptic membrane fractions were prepared from the cerebral cortex as described by Zukin et al. (18). Cerebral cortices were homogenated in 10 volumes of ice-cold 0.32 M sucrose buffer and the homogenates were centrifuged at 1,000 x g for 10 min at 4°C. Supernatants were collected and centrifuged at 20,000 x g for 20 min at 4°C. Pellets were suspended in distilled water and centrifuged at 8,000 x g for 20 min at 4°C. Supernatants obtained were diluted with 20 ml of 50 mM Tris-citrate buffer (pH 7.1) and centrifuged three times at 48,000 x g for 20 min at 4°C. The final pellet was resuspendend in the above buffer and stored at -80°C for at least 18 hr before the receptor binding assay as described (18). Measurement of 360 -- Influx The influx of 36CI- into membrane vesicles was measured by the method of Obata and Yamamura (17) with a minor modification. The membrane vesicle suspensions (100 ~1) were preincubated with each drug for 10 min at 30°C. After preincubation, 360 influx was initiated by the addition of 100 ml of buffer containing 36CI- (0.2 PCi) and the indicated concentrations of GABA. The influx was terminated after 10 set by the addition of 3 ml of ice-cold buffer followed by rapid vacuum filtration through a GF/C glass fiber filter pretreated with 0.05% polyethylenimine. The filters were washed three times with 3 ml of ice-cold 10 mM HEPES buffer, and bound radioactivity was quantified in a Beckman LS 5801 counter. GABA-stimulated 36CIinflux was calculated as the difference between basal uptake in the absence of GABA and total uptake in the presence of GABA. Receptor Binding Assay [3H]Flunitrazepam (FZ) was used for the BZ receptor binding assay in the concentration range of 1 to 40 nM in saturation binding studies. The final reaction mixture volume was 1 ml of 50 mM Tris-citrate buffer containing 0.2-0.5 mg of membrane preparations. Reactions were initiated by adding [3H]FZ (approximately 2 x 1 O5 dpm) and performed at 2°C for 60 min. They were quenched by the addition of icecold 50 mM Tris-citrate buffer. Bound and free ligands were separated by rapid vacuum filtration (Brandel Cell Harvester M 24R) on Whatman GF/B filters. The filter was washed three times with 3 ml of the same ice-cold buffer, and bound radioactivity was quantified in a Beckman LS 6000LL counter. Specific binding was defined as the difference between [3H]FZ binding in the absence and presence of 10 mM FZ. Binding parameters were analyzed to obtain the values of Kd and Bmax by using Scatchard transformation. Protein Determination and Data Analysis Protein content was determined by the Coomassie Blue binding method using bovine serum albumin as a standard (19). Statistical analyses of data were performed by one-way ANOVA or Student’s ttest. P values of less than 0.05 were taken to indicate a significant difference in the mean values being compared.
1634
Diazepam Dependence and GABA, Receptor
Vol. 59, No. 19, 1996
Materials [3H]FIunitrazepam (3.0 TBq/mmol) was purchased from New England Nuclear. Na36CI- (4.0 MBq/nmol) was from Amersham. Diazepam and flunitrazepam were generous gifts from Dr. Tsutomu Suzuki and Esai Pharmaceutical Co., Ltd., respectively. All other reagents were of analytical grade. Results DZP Treatment and Withdrawal Siuns At the first session of the treatment, rats were given two kinds of drug-admixed food containing 1 mg of DZP per lg of food and 2 mg of DZP per lg of food for five days according to the protocol described previously (15). The mean DZP intake of rats in the first session was 73.3 f 5.6 mg/kg/day. The concentration of DZP in the food was gradually increased. The mean DZP intake during feeding with the final DZP concentration (12 mg/g food, for six days) was 701.1 f 14.1 mg/kg/day. At the time of the last treatment, DZP concentrations in plasma reached 249.1 f 24.9 ng/ml. After termination of DZP treatment, withdrawal signs such as weight loss, vocalization, irritability, muscle rigidity, ear-twitching, and piloerection were observed in all eight DZP-treated animals. Straub’s tail, nosebleed and tremors were observed in six of the eight animals. Fascicular twitches and spontaneous convulsions were observed in four of eight rats. The total withdrawal score observed during the seven days after withdrawal was 21.3 rl: 1.8 (Table 1). Withdrawal scores of rats were maximal at 39 - 45 hr after the final treatment and they had almost disappeared at 162 hr after the final treatment (Fig. 1) Effect of Chronic DZP Treatment on GABA-stimulated 36CI- Influx In control rats, the addition of GABA to the membrane vesicle suspension resulted in stimulation of scCl- influx in a dose-dependent manner (5 PM, 11.7 * 2.3 TABLE Maximal
Behavioral
Changes
I
During Diazepam
Positive Animals /Total Withdrawal
Score
Weight loss 5-l 0% 1o-1 5% Vocalization Irritability Muscle rigidity Ear-twitching Piloerection Straub’s tail Fascicular twitch Nosebleed Agression Tremor Spontaneous convulsion Death Withdrawal
scores
;;I (2) (2) (2) (2) (3) (3) I:; I;;
Withdrawal Number of Animals
Control
DZP treated
O/8 O/8 O/8 O/8 O/8 O/8 O/8 018 O/8 O/8 O/8 018
2J8 618 818 818 818 818 818 618 418 618 O/8 618
018 O/8
418 O/8
0
21.3k1.8
Vol. 59, No. 19, 1996
Diazepam Dependence and GABA* Receptor
1635
20
18 $’
16
8
14
0
12
24
36
48
60
72
84
96
Time after withdrawal
108120132144156168
(hours)
Fig. 1. Changes in withdrawal scores after DZP treatment. Withdrawal was conducted by substituting normal food for diazepam-admixed food. Withdrawal signs were scored on the five-point scale shown in Table 1 as described by Suzuki et al. (15). Each point represents the mean + SE of eight animals.
nmol/mg protein/l 0 set; 10, 19.8+1.6; 20, 27.3 + 2.1; 40, 35.9 f 3.8; 200, 42.0+1.9). The addition of 10 PM GABA resulted in the nearly half maximal stimulation of %Iinflux and maximal stimulation was observed with 200 PM GABA. In tolerant rats (0 hr), although 10 PM GABA-stimulated 36CI- influx into membrane vesicles of rats was not different from in control rats, 200 PM GABA-stimulated %I- influx was significantly decreased in comparison with control rats (Fig. 2). At 42 hr after DZP treatment of rats, both 10 /JM and 200 PM GABA-stimulated %I- influxes were significantly increased in comparison with control rats. At 162 hr after DZP treatment of rats, the increases in GABA-stimulated 3sCI- influx observed at 42 hr were reversed to levels similar to those in control animals (Fig. 2). The stimulatory effect of GABA on 36CI- influx observed in each group was effectively inhibited by the addition of the GABA antagonist, bicuculline (SIC) or the chloride channel blocker, picrotoxinin (PIC) (Fig. 3). These results confirmed that GABA-stimulated %I- influx reflected the functional coupling of the GABA,+, receptor with the chloride channel. On the other hand, no blocking action of Ro15-1788 (10 PM), a BZ antagonist, on 10 PM and 200 PM GABA-stimulated 3sC1influx was observed in control and DZP-treated rats (results not shown), indicating that there was no contamination by DZP, which might affect the measurement of the GABAstimulated 3sCI- influx, in the membrane vesicle preparation after the DZP treatment of rats. As shown in Fig. 4, the addition of 1 PM FZ to the assay mixture significantly potentated 10 PM GABA-stimulated 36CI- influx in the control rats and, similar to earlier studies in which this assay procedure was used (20, 21), it was suppressed by the BZ
Diazepam Dependence and GABA* Receptor
1636
so-
*
Vol. 59, No. 19, 1996
.
GABA 200 /JM
0
GABA 10 pM
5040-
I
30 PO-
P
IOO-lControl
dhr
162 hr
42 hr
Time after withdrawal Fig. 2. Effects of 10 PM and 200 PM GABA on 3sCI- influx in rats at 0 hr (tolerant rats), 42 hr and 162 hr after DZP withdrawl. Each point represents the mean + SE of 10 to 12 experiments. *Significantly different from each control value. (pcO.05)
Control
0 hr
0 0
GABA 10j1M PIG
0
BIC
42 hr
162 hr
Time after withdrawal Fig. 3. Effects of 100 /JM PIC and 1 mM BIC on 10 PM GABA-stimulated 3%I- influx in rats at 0 hr (tolerant rats), 42 hr and 162 hr after DZP withdrawal. Each point represents the mean +: SE of 10 to 12 experiments. *Significantly different from the corresponding value with 10 PM GABA alone.
Diazepam Dependence and GABAA Receptor
Vol. 59, No. 19, 1996
200
0
*
0
FZ FZ + Rol5-
1637
1788
175 z I
5(d 150 i
2 5
125 1 +
E g_ 100 i 75-l
7 Control
O’hr
I
42 hr
162 hr
Time after withdrawal Fig. 4 Enhancement of 10 PM GABA-stimulated WI- influx by the addition of 1 /.fM FZ and inhibition by the addition of 10 PM Rol51788 in rats at 0 hr (tolerant rats), 42 hr and 162 hr after DZP withdrawal. Data are expressed as a percentage of the value of 10 PM GABA-stimulated 36CI- influx. Each point *Significantly different represents the mean 2 SE of 10 to 12 experiments. from the corresponding value with 10 PM GABA alone. +Significantly different from the value with 10 PM GABA plus 1 PM FZ.
antagonist, Rol5-1788 (10 PM). However, FZ-induced potentiation and an antagonistic effect of Rol5-1788 on GABA-stimulated 36CI- influx were not observed in tolerant (0 hr) rats. In contrast, FZ significantly enhanced GABA-stimulated 3sCI- influx into membrane vesicles at 42 hr after DZP treatment of rats, but it was not effectively suppressed by the addition of Ro15-1788. Furthermore, at 162 hr after DZP treatment, FZ significantly WI- influx and an antagonistic effect of Ro15-1788 potentiated the GABA-stimulated on the FZ-induced stimulation was recognized, although it was not significant (Fig. 4). Effects of Chronic DZP Treatment on the Receptor Binding Assav Analysis of [sH] FZ binding to BZ receptors in cortical membrane preparations showed a significant decrease in binding sites (Bmax) in the tolerant (0 hr) animals compared with the control. In contrast, the Bmax value in the 42-hr-after-DZP-treatment group was significantly higher than that in the control, and the increase in Bmax had almost returned to the control level in 162-hr-after-DZP-treatment rats. On the other hand, there were no significant differences in the dissociation constant (Kd) among the four groups (Table 2). Discussion Several previous studies using experimental animals have shown that behavioral tolerance to BZ is associated with down-regulation of BZ receptors (9), and reduced functional coupling of the BZ recognition site with the GABAA receptor/Clchannel (10). However, the relationship between withdrawal symptoms and alteration of GABAA receptor functions has not been described in detail, although some reports
1638
Diazepam Dependence and GABA, Receptor
TABLE [sH]Flunitrazepam
Control 0 hr 42 hr 162 hr
II
binding to benzodiazepine membranes Kd (nM) 3.47 3.49 3.81 3.61
+ 0.06 -+ 0.19 +0.12 r 0.11
Vol. 59, No. 19, 1996
receptors
in rat synaptic
Bmax (pmol/mg protein) 1.84 1.30 2.18 1.90
+ f + f
0.05 0.03* 0.02* 0.01
Benzodiazepine receptor binding in DZP-dependent rats at 0 hr (tolerant rats), 42 hr and 162 hr after DZP withdrawal. Binding was determined in cerebral cortical synaptic membrane using the ligand [sH]flunitrazepam. Data represent the mean + SE of six experiments. *Significantly different from control (~~0.05).
have demonstrated the relationship between the open-field activity and alteration in the function of GABA* receptors during the BZ-withdrawal state in mice (13,14). In contrast to animal studies, various withdrawal symptoms were recognized in patients with BZ dependence, not only alteration in motor activity but also severe withdrawal symptoms such as tremors, delirium and spontaneous convulsions (7, 8). However, studies examining the molecular mechanism for severe withdrawal signs, including convulsions, have been difficult to perform due to the lack of an animal model presenting severe withdrawal signs after BZ is withdrawn. Thus, the mechanisms underlying serious physical dependence have been studied less extensively. A recent report described how the OAF method can produce physical dependence on DZP and subsequent withdrawal convulsions in rats (15). Therefore, in the present study, to investigate the role of GABAA receptor function in development of physical dependence on DZP and subsequent severe withdrawal signs, we employed the OAF method for inducing physical dependence on DZP. Although the mean DZP intake (701 .l mg/kg/day) during feeding with the final DZP concentration was a little lower than that described by Suzuki et al. (15) (773.3 mg/kg/day), the blood concentration of DZP in DZP-treated rats was essentially the same as that reported in their study (15). After cessation of DZP treatment, various withdrawal signs could be observed (Table l), and the withdrawal scores reached the maximum at 39 - 45 hr after the final DZP treatment (Fig. 1). These withdrawal signs had almost disappeared at 159 - 168 hr after the final DZP treatment. Therefore, we examined the function of GABAA receptors immediately after the final DZP treatment (0 hr, tolerant rats), and at 42 hr and 162 hr after the final DZP treatment, In tolerant rats (0 hr after DZP withdrawal), stimulation of 360 influx by 200 PM GABA, which produced maximal stimulation of ssCI- influx, was significantly decreased compared with that in control rats, although 10 PM (a half-maximally effective concentration for WI- influx) GABA-stimulated %I- influx was not changed (Fig. 2). This result is in accord with the findings of Miller et al. (13) and Galpern et al. (14) of a decrease in 3W- influx caused by a saturation level of muscimol and the lack of change in 360 influx caused by the ECsO level of the agonist in BZ-treated rats. The reduction in GABA (200 PM)-stimulated WI- influx seemed to indicate that the GABAA receptors or chloride channels decreased in tolerant rats. In our previous study, the number of binding sites of GABAA receptors was not altered in DZP-tolerant rats (22).
Vol. 59, No. 19, 1996
Diazepam Dependence and GABA* Receptor
1639
[=S]T BPS (ta trend toward decreased Bmax of Furthermore, butylbicyclophosphorothionate) binding to chloride channels after BZ treatment has been demonstrated (13, 14). Therefore, reduction in the saturation level of GABAstimulated 360 influx might have been due to a decrease in the number of chloride channels after chronic DZP treatment. The reason for lack of reduction in 10 PM GABAstimulated 36CI- influx in tolerant rats is unclear. We have previously observed a slight increase in the affinity of the GABAA receptor, although it was insignificant, in DZPtolerant rats (22). Such alteration in the affinity of the GABAA receptor might have an influence on the modulation of 36Cl- influx by stimulation with the EC& level of GABA in tolerant rats. Further studies are needed to resolve this question. Some previous studies have indicated that there is no significant change in GABA-stimulated 36Cl- influx after chronic administration of BZ to rats (10,ll). The apparent discrepancy may be explained by differences in the methods of drug administration. The chronic treatment with BZ used in earlier studies, in which withdrawal signs and the blood concentrations of BZ were not described, might not have been sufficient to cause an alteration in the effect of GABA on 36CI- influx or severe withdrawal signs. Miller and his co-workers (13,14) have observed decreased muscimol-stimulated 360 influx before the appearance of withdrawal signs in chronically BZ-treated rats. Therefore, it is assumed that decreased sensitivity of GABAA receptors to GABA may be required for the initial biochemical changes causing the appearance of BZ withdrawal signs. On the other hand, in withdrawal rats at 42 hr after the cessation of DZP treatment, GABA (10 I.IM and 200 PM) -stimulated 36CI- influx was increased in comparison with control rats (Fig. 2). Our previous study indicated that the affinity of high affinity sites for GABAA receptors to GABA, estimated by [3H]muscimol binding, was increased at 42 hr after DZP withdrawn (22). It may be suggested, therefore, that potentiation in the stimulatory effect of GABA on 36CI- influx results from the increased sensitivity of GABAA receptors to GABA, which may lead to enhancement in the coupling of the GABAA receptor with the chloride channel. FZ effectively enhanced GABA-stimulated 360 influx in control rats (Fig. 4). However, potentiation of GABA-stimulated 36CI- influx by FZ was not observed in tolerant rats (0 hr after DZP withdrawal). In contrast, FZ significantly potentiated GABAstimulated 36CI- influx at 42 hr and 162 hr after the final DZP treatment. It seems that the alteration in the ability of FZ to potentiate GABA-stimulated 36CI- influx observed in rats after withdrawal may not be explained only by the changes in the BZ receptors. There was no change in the dissociation constant (Kd) of BZ receptors between control rats and rats after withdrawn of DZP (Table 2). Furthermore, although the number of binding sites (Bmax) of BZ receptors was increased at 42 hr compared with that in the control, the extent of FZ-induced enhancement of GABA-stimulated 36CI- influx was not as great as that in the control (control, 168.7 f 16.0%; 42 hr, 142.7 f 14.7%) (Table 2 and Fig. 4). In addition, whereas the increase in the number of binding sites of BZ receptors observed at 42 hr returned to the control level at 162 hr after the DZP withdrawal, enhancement of FZ in GABA-stimulated 36CI- influx did not completely return to the control level (Fig. 4). Thus, alterations in the number of binding sites of BZ receptors did not appear to be consistent with the effect of FZ on GABA-stimulated 36CI- influx in rats after DZP withdrawal. Alternatively, previous studies have suggested reduction of functional coupling of the BZ receptors with the GABAA receptors in chronically BZtreated rats (10, 12). It may be possible, therefore, that the functional coupling between BZ receptors and GABAA receptors was impaired by chronic treatment of rats with DZP, which was related to behavioral tolerance to BZ in tolerant rats. This impairment of the coupling appeared to last after the DZP withdrawal and not only the changes in the BZ receptors but also the impairment of the coupling might be involved in the appearance of severe withdrawal signs as indicated in the present study (Table 1). This might be
1640
Diazepam Dependence and GABAA Receptor
Vol. 59, No. 19, 1996
supported by the findings that the effect of the BZ antagonist, Rol5-1788, was not observed in tolerant rats (0 hr after DZP withdrawal) and in rats at 42 hr after the withdrawal, whereas it effectively suppressed FZ-induced potentiation of GABAstimulated ssCI- influx in controls (Fig. 4). At 162 hr after the DZP withdrawal, the antagonistic effect of Rol5-1788 was recognized, although it was not statistically significant (Fig. 4). This may suggest that impairment in the functional coupling between BZ receptors and GABAA receptors tended to recover toward the control level at 162 hr after the withdrawal, but this recovery was still insufficient, which might relate to the appearance of the few withdrawal signs observed at 162 hr (Fig. 1). In summary, using DZP-dependent rats produced by the DAF method, we have demonstrated that functional changes in the GABAA/BZ receptor/Cl- channel complex, i.e. increased sensitivity in GABAA receptors, and impairment in the functional coupling between BZ receptors and GABAA receptors, may possibly be involved in the manifestation of severe withdrawal symptoms such as tremors and convulsions after chronic treatment with DZP. The current understanding of the molecular structure of the GABA* receptor is that it is a heteropentameric glycoprotein composed of multiple polypeptide subunits, i.e. al-6, pl-4, yl-3,, 6 and ~1-2 (23). Previous studies have reported that chronic benzodiazepine treatment decreases the al (24), a5 and y2 (25) subunit mRNA levels in rat cerebral cortex. Furthermore, previous studies have suggested that [sH]FZ is bound to the a subunit of the GABAA receptor (26) Therefore, these changes of subunits may be concerned with down-regulation of BZ receptors (Table 2) and the reduced functional coupling of BZ receptors with GABAA receptors observed in this study in tolerant rats. Alternatively, it is known that the function of Nmethyl-D-aspartate (NMDA) receptors is related to convulsions or other seizures of epilepsy. Recent evidence has suggested that the NMDA antagonist, 3-[(&)-2carboxypiperazin-4-yl]-propyl-1-phosphonate prevents the withdrawal symptoms of DZP and that NMDA-dependent mechanisms may contribute to the development of tolerance to DZP and to expression of withdrawal signs (27). Thus, the behavioral appearance of signs of withdrawal in BZ dependence might be caused by alteration in the functional interaction between the GABAA receptor complex and the NMDA receptor. Further studies with respect to these aspects are required to more definitively establish the molecular mechanism of BZ dependence. Acknowledgments This work was partly supported by a research grant from the Ministry of Health and Welfare, Japan. The authors thank Drs. Tsutomu Suzuki and Hirokazu Mizoguchi for helpful discussions and Dr. Akemi Nabeshima for technical assistance. We also thank Esai Pharmaceutical Co., Ltd. for the generous gift of flunitrazepam. References 1. 2. 3. 4. 5. 6. 7. 8.
W. HAEFELY, E. KYBURZ, M. GERECKE and H. MCHLER, Adv. Drug Res. 14 165-322 (1985). R.E. STUDY and J.L. BARKER, J. Am. Med. Ass. 247 2147-2151 (1982). SE. FILE, Neurosci. Biobehav. Rev. 9 113-121 (1985). J.P. GONZALEZ, A.J. McCULLOCH, P.J. NICHOLLS, R.D. SEWELL and A. TEKLE, Alcohol Alcohol. 19 325-332 (1984). K. RICKELS, W.G. CASE, R. W. Downing and A. Winokur, J. Am. Med. Ass. 250 767-771 (1983). C. HALLSTOM and M. LADER, Int. Pharmacopsychiatry 16 235-244 (1981). R.T. OWEN and P. TYRER, Drugs 25 385-398 (1983). H. PETURSSON and M.H. LADER, Br. J. Addict. 76 133-145 (1981).
Vol. 59, No. 19, 1996
9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.
21. 22. 23. 24. 25. 26. 27.
Diazepam Dependence and GABA* Receptor
1641
L.G. MILLER, D.J. GREENBLATT, J.G. BARNHILL and R.I. SHADER, J. Pharmacol. Exp. Ther. 246 170-l 76 (1988). M. Li, H.C. ROSENBERG and T.H. CHIU, Eur. J. Pharmacol. 247 313-318 (1993). D.O. NGUR, H.C. ROSENBERG and T.H. CHIU, Eur. J. Pharmacol. 176 351-356 (1990). 0. YU, T.H. CHIU and H.C. ROSENBERG, J. Pharmacol. Exp. Ther. 246 107-l 13 (1988). L.G. MILLER, D.J. GREENBLATT, R.B. ROY, W.R. SUMMER and R.I. SHADER, J. Pharmacol. Exp. Ther. 246 177-181 (1988). W.R. GALPERN, M. LUMPKIN, D.J. GREENBLATT, R.I. SHADER and L.G. MILLER, Psycopharmacology. 104 225-230 (1991). T. SUZUKI, M.-S. LU, H. MOTEGI, T. YOSHIDA and M. MISAWA, Pharmacol. Biochem. Behav. 43 387-393 (1992). RR BRODIE, L.F. CHASSEAUD and T. TAYLOR, J. Chromatogr. 150 361-366 (1978). T. OBATA and H.I. YAMAMURA, Life Sci. $2 659-665 (1988). S.R. ZUKIN, A.B. YOUNG and S.H. SNYDER, Proc. Natl. Acad. Sci. USA 71 4802-4807 ( 1974). M.M. BRADFORD, Anal. Biochem. 72 248-254 (1976). T. OBATA, M. MORELLI, A. CONCAS, M. SERRA and H.I. YAMAMURA, Chloride Channels and Their Modulation bv Neurotransmitters and Drugs. Giovanni Biggio and E. Costa (Eds), 175-l 87, Raven Press, New York (1988). M. HIROUCHI, J. TAGUCHI, T. UEHA and K. KURIYAMA, Biochem. Biophys. Res. Commun. 146 1471-1477 (1987). S. TOKI, T. SAITO, A. NABESHIMA, S. HATTA, M. WATANABE and N. TAKAHATA, Alcohol. Clin. Exp. Res. 20 40A-44A (1996). R.L. MACDONALD and R.W. OLSEN, Annu. Rev. Neurosci. 17 569-602 (1994). E.I. TIETZ, X. HUANG, X. WENG, H.C. ROSENBERG and T.H. CHIU, J. Mol. Neurosci. 2 277-292 (1993). Y. WU, H.C. ROSENBERG, T.H. CHIU and T.J. ZHAO, J. Mol. Neurosci. 5 105. 120 (1994). K. FUCHS, H. MCHLER and W. SfEGHART, Neurosci. Lett. $Q 314-319 (1988). K.G. STEPPUHN and L. TURSKI, Proc. Natl. Acad. Sci. USA 90 6889-6893 (1993).