J Neural Transm [GenSect] (1992) 88:199-221
__ Journal o f Neural Transmission 9 Springer-Verlag 1992 Printed in Austria
Cholinergic mechanisms in physical dependence on barbiturates, ethanol and benzodiazepines A. NordbergI and G. WahlstriJm2 1Department of Pharmacology, Uppsala University, Uppsala, and 2Department of Pharmacology, Ume~ University, Ume& Sweden Accepted January 14, 1992
Summary. The aim of this review is to summarize the effects of acute and chronic treatment with barbiturates, ethanol and benzodiazepines on cholinergic mechanisms in the brains of experimental animals. A single dose of each of these substances reduces the turnover of ACh in the brain. Long-term treatment has the opposite effect; complicated interactions including decreased content of ACh are induced. Barbiturates have been shown to bind stereospecifically to muscarinic and nicotinic receptors in the brain, but this has not been observed for ethanol or the benzodiazepines. The effects on the cholinergic system are affected by the length of treatment and choice of treatment regimen. No effect on cholinergic parameters, such as muscarinic receptors, in the brain is observed on withdrawal of ethanol or barbiturate treatment when the animals are still tolerant towards the substances. The increase in the number of muscarinic receptors observed in several brain regions on withdrawal is seen as a sign of cholinergic supersensitivity. The number of receptors returns to normal when abstinence convulsions have occurred. The assumption of a cholinergic influence is supported by the finding that atropine, given as a single dose on the day of withdrawal of barbital, can prevent the muscarinic receptor changes. Furthermore, long-term barbital or ethanol treatment can induce permanent persistent changes in the cholinergic system in the brain. Cognitive defects and a significant permanent reduction in the content of ACh can be measured in rats which have had long-term barbital treatment. Similarly, a reduced number of muscarinic receptors has been measured in different brain regions of chronic alcoholics. Accumulating data support the role of the cholinergic system in expressing symptoms of physical dependence on barbiturates, ethanol and benzodiazepines as well as in the permanent long-term effects observed after end of treatment. Keywords: Physical dependence, ethanol, barbiturates, benzodiazepines, cholinergic mechanisms, muscarinic receptors, nicotinic receptors, acetylcholine, convulsions, abstinence, chronic treatment.
200
A. Nordberg and G. Wahlstr6m
Introduction
Physical dependence was defined by Himmelsbach (1943) as "an arbitrary term used to denote the presence of an acquired abnormal state wherein the regular administration of adequate amounts of a drug has, through previous prolonged use, become requisite to physiological equilibrium". According to this definition, physical dependence is associated with tolerance phenomena. However, more recent investigations have clearly established that a number of processes are involved in the development of tolerance (Hunt, 1985; Tabakoff and Hoffman, 1987), and it has been put forward that tolerance and physical dependence are caused by distinctly different mechanisms (Wahlstr6m, 1979; Koob and Bloom, 1988). The diagnosis of physical dependence incorporates a withdrawal syndrome. When investigating the mechanisms involved in physical dependence in the CNS, a search for changes in neurotransmitter systems must be related to symptoms found in vivo. A schematic model for this concept, based on changes in an ideal transmitter system, is illustrated in Fig. 1. As shown in the figure, possible reactive components in the system include the transmitter activity, the receptor plasticity and the effector system, all of which might change during the development of physical dependence. To elucidate these mechanisms in experimental animals, a large number of
BEFORE DRUG
DRUG ACTION
/.s E R
R
TOLERANCE a.
b.
T/~
\
.s m,.E
~E
\ \X
T
WITHDRAWAL a.
b.
R
R .S.~E
Fig. 1. Schematic model of possible changes in an ideal transmitter system with tolerance and withdrawal of a dependence-producing drug. T transmitter; R receptor; E effect; X drug. Changes in different part of the system are indicated by the size of the letters
Cholinergic mechanisms in physical dependence
201
treatment models have been used by different research groups, including repeated parenteral administration, pellet implantation, oral fluid intake or vapour administrations. The duration of treatment is of primary importance in the production of dependence. Barbiturates and ethanol can both induce functional tolerance and physical dependence in experimental animals and man. In recent years, barbiturates have been replaced by benzodiazepines in clinical use mainly because of the decreased toxicity of the latter group but also with the idea that they may be less prone to induce dependence (Mark, 1978). Recent experience, however, indicates that the benzodiazepines do, in fact, induce dependence (Cappell et al., 1986; Haefely, 1986; Lader, 1987, 1988; Wood et al., 1987; Murphy and Tyrer, 1988). Since the benzodiazepines are now used extensively for various indications in man, the problem of dependence will increase in magnitude. The effects of barbiturates on the brain bovine interaction with many transmitters in the CNS such as acetylcholine (ACh), noradrenaline (NA), dopamine (DA) and gammaaminobutyric acid (GABA) (Ho and Harris, 1981). Ethanol is known to induce a wide range of effects in the brain such as reinforcement, cognitive deficits and intoxication. Ethanol can interact with neural membranes and thereby change membrane fluidity. It has been suggested that this interaction is not non-specific but occurs via specific membrane-bound proteins at "receptive sites" (Tabakoff and Hoffman, 1987). Acute and chronic administration of ethanol in animals induces changes in the catecholaminergic, serotonergic, cholinergic. GABAergic and glutaminergic systems in the brain (Hoffman and Tabakoff, 1985; Engel and Liljequist, 1983; Hunt, 1985; Deitrich etal., 1989; Hoffman etal., 1990). Recently, the N-methyl-D-aspartate (NMDA) receptorgated ion channel has been reported to interact with ethanol in electrophysiological and biochemical experiments (Hoffman etal., 1989; Lovinger etal., 1989). The opioid and peptidergic systems have also been reported to be changed after withdrawal of ethanol treatment (Arakisainen and Peura, 1988). The mechanism of action of ethanol in the brain has recently been reviewed by Deitrich et al. (1989). The mechanisms of action of the benzodiazepines appear mainly to involve the GABA system in the brain (Costa, 1979; Haefely et al., 1985) but the possibility that the cholinergic system play a role has been discussed (Skolnick and Paul, 1981). Although multiple neurotransmitters are involved in the mechanisms of action of barbiturates, ethanol and benzodiazepines in the brain, this review will focus solely on the cholinergic system and its role in the expression of symptoms of physical dependence on these substances. As will be illustrated below, accumulating data in the literature indicate that the cholinergic mechanisms in the brain play an important role in acquired tolerance phenomena and abstinence mechanisms. Learning is a contributing factor in the development of tolerance (Kalant, 1985). Since cholinergic mechanisms are involved in learning processes (Drachman and Leavitt, 1974) the importance of the cholinergic system in tolerance and withdrawal phenomena is further stressed.
202
A. Nordberg and G. Wahlstr6m Barbiturates
A number of the acute effects of different barbiturates can be explained as interactions with the cholinergic system in the brain. A single anaesthetic dose of pentobarbital, injected intravenously into mice, induces an increased content of endogenous ACh in the cortex and hippocampus while no effect is observed in the striatum (Nordberg and Sundwall, 1975). Similarly, the uptake of choline (Ch) via the high-affinity uptake system (Atweh et al., 1975) and the turnover of ACh (Nordberg and Sundwall, 1977) are decreased in the hippocampus and cortex. It has been observed that hexobarbital is stereospecifically bound to muscarinic receptors in the rat brain (Nordberg and Wahlstr6m, 1984). In concentrations clearly within those reached when inducing anaesthesia in vivo, in vitro hexobarbital displaces both muscarinic agonists (3 H-oxotremorine) and antagonists (3 H-quinuclidinyl benzilate) from their binding sites in various brain regions of the rat (Nordberg and Wahlstr6m, 1984). (-)-R-hexobarbital, the stereoisomer with less anaesthetic effect, is the more potent displacer. Cholinergic agonists are more readily displaced by hexobarbital than are antagonists. Amobarbital has been shown to have an affinity for nicotinic receptors in the brain (Dodson etal., 1987) and pentobarbital for nicotinic receptors in the electrical eel (Miller et al., 1982). Furthermore, (+)-pentobarbital was found to be more potent than (-)-pentobarbital in binding to nicotinic receptors on membranes isolated from Torpedo electroplaques (Roth etal., 1989). Recent data obtained by Dodson et al. (1990) in the study of ten different barbiturates support the existence of a barbiturate binding site on the nicotinic receptor with specific structural requirements. Although the nicotinic receptor and the GABA receptor belong to the same superfamily of receptors (Barnard et al., 1988), pentobarbital shows different stereoselectivity for the two receptor types (Roth et al., 1989; Ticku and Rastogi, 1986). Furthermore, a number of effects of barbiturates can be modified by drugs acting on the cholinergic system. Cholinergic agonists such as pilocarpine (Wahlstr6m, 1976 a) and the cholinesterase inhibitor physostigmine (Wahlstr6m, 1977) increase the dose of hexobarbital needed to induce anaesthesia in rats. Antagonists like atropine decrease the required dose of hexobarbital (Wahlstr6m, 1976b). This interaction with cholinergic agonists and antagonists cannot be observed for all lipid soluble barbiturates (Bolander and Wahlstr6m, 1984), indicating that there are differences within the barbiturate group in this resepect. Differences in the mechanism of action behind the anaesthetic effect of different barbiturates has also been substantiated by the lack of additative effects in direct studies of the interaction between barbiturates (Wahlstr6m and Norberg, 1984; Norberg and Wahlstr6m, 1986). Barbital has been used in chronic treatment studies of barbiturates in rodents (Wahlstr6m, 1968; McBridge and Turnbull, 1970). Barbital has an advantage over other barbiturates in that it is metabolized only to a small extent and that it has a half-life that is long enough for significant amounts to accumulate
Cholinergic mechanisms in physical dependence
203
during a day of barbital drinking (Wahlstr6m, 1979). Immediately after 4 weeks of barbital treatment in rats, McBridge and Turnbull (1970) found no change in the content of endogenous ACh in the whole brain. Similar results were also obtained in rats when the endogenous ACh content was analysed immediately following much longer treatment periods (25-30 weeks) (Nordberg and Wahlstr6m, 1977). During the abstinence after forced oral long-term consumption of barbital by rats convulsions and decreased sensitivity to hexobarbital threshold doses for EEG burst supression are recorded (Wahlstr6m, 1974). The maximal tolerance to the hexobarbital threshold and the maximal number of spontaneous convulsions are measured 3 days after withdrawal of barbital. An increased sensitivity to the temperature-reducing effect of pilocarpine is seen (Wahlstr6m and Ekwall, 1976) as a sign of supersensitivity of the cholinergic system on the third day of abstinence. The assumption that cholinergic mechanisms are involved in the withdrawal syndrome is supported by the finding of decreased amounts of endogenous ACh in the brain during abstinence following 30 weeks of oral barbital treatment of rats (Nordberg and Wahlstr6m, 1977). Decreased amounts of endogenous ACh have been measured in the striata of animals abstinent for 3 days and with a maximal frequency of spontaneous convulsions (Nordberg and Wahlstr6m, 1977). The content of ACh is still reduced on the 12th day of withdrawal when the convulsions have almost disappeared (Nordberg and Wahlstr6m, 1977). On the same day, a significant decrease in endogenous ACh has also been measured in the cerebellum + medulla oblongata + midbrain (Nordberg and Wahlstr6m, 1977). Interestingly, the biosynthesis of ACh (3 H-ACh) from a tracer dose of radioactive choline (3 H-Ch) was observed to be significantly enhanced in the hippocampus + cortex and cerebellum + medulla oblongata + midbrain of rats abstinent for 3 days, while no significant effect was found in the striatum (Nordberg and Wahlstr6m, 1979). When the ratio of 3H_ACh/3 H-Ch was calculated as a rough measure of ACh turnover, the ratio was significantly increased in the hippocampus + cortex and cerebellum + midbrain + medulla oblongata in rats abstinent for 3 days. The muscarinic receptors seem to be involved in the production of convulsions in the withdrawal period following barbital treatment (Nordberg et al., 1980). An increased number of muscarinic binding sites has been measured in the striatum at the end of treatment and during the first part of abstinence following long-term barbital treatment (Nordberg etal., 1980; Nordberg and Wahlstr6m, 1982 a). Similar changes but with different time courses have also been observed in other areas of the brain (Nordberg and Wahlstr6m, 1982 a). Thus, a variable increase in the number of muscarinic receptors was measured from days 2-9 of abstinence in the cortex whereas an increase was measured in the midbrain on day 9 (Nordberg and Wahlstr6m, 1982a). The change in muscarinic receptors could be traced to an increase in the proportion of muscarinic agonist low-affinity binding sites (Nordberg and Wahlstr6m, 1982b). As indicated in Fig. 2 the presence of convulsions during withdrawal seems to
204
A. Nordberg and G. WahlstrSm STRIATUM
% 150
BARBITAL
ETHANOL
A_
('j) 100 1 III 50
I
(.9 Z I
Z
m
Convulsions
+
0
0
CORTEX
0 Z o/~
__ Journal o f Neural Transmission 9 Springer-Verlag 1992 Printed in Austria
Cholinergic mechanisms in physical dependence on barbiturates, ethanol and benzodiazepines A. NordbergI and G. WahlstriJm2 1Department of Pharmacology, Uppsala University, Uppsala, and 2Department of Pharmacology, Ume~ University, Ume& Sweden Accepted January 14, 1992
Summary. The aim of this review is to summarize the effects of acute and chronic treatment with barbiturates, ethanol and benzodiazepines on cholinergic mechanisms in the brains of experimental animals. A single dose of each of these substances reduces the turnover of ACh in the brain. Long-term treatment has the opposite effect; complicated interactions including decreased content of ACh are induced. Barbiturates have been shown to bind stereospecifically to muscarinic and nicotinic receptors in the brain, but this has not been observed for ethanol or the benzodiazepines. The effects on the cholinergic system are affected by the length of treatment and choice of treatment regimen. No effect on cholinergic parameters, such as muscarinic receptors, in the brain is observed on withdrawal of ethanol or barbiturate treatment when the animals are still tolerant towards the substances. The increase in the number of muscarinic receptors observed in several brain regions on withdrawal is seen as a sign of cholinergic supersensitivity. The number of receptors returns to normal when abstinence convulsions have occurred. The assumption of a cholinergic influence is supported by the finding that atropine, given as a single dose on the day of withdrawal of barbital, can prevent the muscarinic receptor changes. Furthermore, long-term barbital or ethanol treatment can induce permanent persistent changes in the cholinergic system in the brain. Cognitive defects and a significant permanent reduction in the content of ACh can be measured in rats which have had long-term barbital treatment. Similarly, a reduced number of muscarinic receptors has been measured in different brain regions of chronic alcoholics. Accumulating data support the role of the cholinergic system in expressing symptoms of physical dependence on barbiturates, ethanol and benzodiazepines as well as in the permanent long-term effects observed after end of treatment. Keywords: Physical dependence, ethanol, barbiturates, benzodiazepines, cholinergic mechanisms, muscarinic receptors, nicotinic receptors, acetylcholine, convulsions, abstinence, chronic treatment.
200
A. Nordberg and G. Wahlstr6m
Introduction
Physical dependence was defined by Himmelsbach (1943) as "an arbitrary term used to denote the presence of an acquired abnormal state wherein the regular administration of adequate amounts of a drug has, through previous prolonged use, become requisite to physiological equilibrium". According to this definition, physical dependence is associated with tolerance phenomena. However, more recent investigations have clearly established that a number of processes are involved in the development of tolerance (Hunt, 1985; Tabakoff and Hoffman, 1987), and it has been put forward that tolerance and physical dependence are caused by distinctly different mechanisms (Wahlstr6m, 1979; Koob and Bloom, 1988). The diagnosis of physical dependence incorporates a withdrawal syndrome. When investigating the mechanisms involved in physical dependence in the CNS, a search for changes in neurotransmitter systems must be related to symptoms found in vivo. A schematic model for this concept, based on changes in an ideal transmitter system, is illustrated in Fig. 1. As shown in the figure, possible reactive components in the system include the transmitter activity, the receptor plasticity and the effector system, all of which might change during the development of physical dependence. To elucidate these mechanisms in experimental animals, a large number of
BEFORE DRUG
DRUG ACTION
/.s E R
R
TOLERANCE a.
b.
T/~
\
.s m,.E
~E
\ \X
T
WITHDRAWAL a.
b.
R
R .S.~E
Fig. 1. Schematic model of possible changes in an ideal transmitter system with tolerance and withdrawal of a dependence-producing drug. T transmitter; R receptor; E effect; X drug. Changes in different part of the system are indicated by the size of the letters
Cholinergic mechanisms in physical dependence
201
treatment models have been used by different research groups, including repeated parenteral administration, pellet implantation, oral fluid intake or vapour administrations. The duration of treatment is of primary importance in the production of dependence. Barbiturates and ethanol can both induce functional tolerance and physical dependence in experimental animals and man. In recent years, barbiturates have been replaced by benzodiazepines in clinical use mainly because of the decreased toxicity of the latter group but also with the idea that they may be less prone to induce dependence (Mark, 1978). Recent experience, however, indicates that the benzodiazepines do, in fact, induce dependence (Cappell et al., 1986; Haefely, 1986; Lader, 1987, 1988; Wood et al., 1987; Murphy and Tyrer, 1988). Since the benzodiazepines are now used extensively for various indications in man, the problem of dependence will increase in magnitude. The effects of barbiturates on the brain bovine interaction with many transmitters in the CNS such as acetylcholine (ACh), noradrenaline (NA), dopamine (DA) and gammaaminobutyric acid (GABA) (Ho and Harris, 1981). Ethanol is known to induce a wide range of effects in the brain such as reinforcement, cognitive deficits and intoxication. Ethanol can interact with neural membranes and thereby change membrane fluidity. It has been suggested that this interaction is not non-specific but occurs via specific membrane-bound proteins at "receptive sites" (Tabakoff and Hoffman, 1987). Acute and chronic administration of ethanol in animals induces changes in the catecholaminergic, serotonergic, cholinergic. GABAergic and glutaminergic systems in the brain (Hoffman and Tabakoff, 1985; Engel and Liljequist, 1983; Hunt, 1985; Deitrich etal., 1989; Hoffman etal., 1990). Recently, the N-methyl-D-aspartate (NMDA) receptorgated ion channel has been reported to interact with ethanol in electrophysiological and biochemical experiments (Hoffman etal., 1989; Lovinger etal., 1989). The opioid and peptidergic systems have also been reported to be changed after withdrawal of ethanol treatment (Arakisainen and Peura, 1988). The mechanism of action of ethanol in the brain has recently been reviewed by Deitrich et al. (1989). The mechanisms of action of the benzodiazepines appear mainly to involve the GABA system in the brain (Costa, 1979; Haefely et al., 1985) but the possibility that the cholinergic system play a role has been discussed (Skolnick and Paul, 1981). Although multiple neurotransmitters are involved in the mechanisms of action of barbiturates, ethanol and benzodiazepines in the brain, this review will focus solely on the cholinergic system and its role in the expression of symptoms of physical dependence on these substances. As will be illustrated below, accumulating data in the literature indicate that the cholinergic mechanisms in the brain play an important role in acquired tolerance phenomena and abstinence mechanisms. Learning is a contributing factor in the development of tolerance (Kalant, 1985). Since cholinergic mechanisms are involved in learning processes (Drachman and Leavitt, 1974) the importance of the cholinergic system in tolerance and withdrawal phenomena is further stressed.
202
A. Nordberg and G. Wahlstr6m Barbiturates
A number of the acute effects of different barbiturates can be explained as interactions with the cholinergic system in the brain. A single anaesthetic dose of pentobarbital, injected intravenously into mice, induces an increased content of endogenous ACh in the cortex and hippocampus while no effect is observed in the striatum (Nordberg and Sundwall, 1975). Similarly, the uptake of choline (Ch) via the high-affinity uptake system (Atweh et al., 1975) and the turnover of ACh (Nordberg and Sundwall, 1977) are decreased in the hippocampus and cortex. It has been observed that hexobarbital is stereospecifically bound to muscarinic receptors in the rat brain (Nordberg and Wahlstr6m, 1984). In concentrations clearly within those reached when inducing anaesthesia in vivo, in vitro hexobarbital displaces both muscarinic agonists (3 H-oxotremorine) and antagonists (3 H-quinuclidinyl benzilate) from their binding sites in various brain regions of the rat (Nordberg and Wahlstr6m, 1984). (-)-R-hexobarbital, the stereoisomer with less anaesthetic effect, is the more potent displacer. Cholinergic agonists are more readily displaced by hexobarbital than are antagonists. Amobarbital has been shown to have an affinity for nicotinic receptors in the brain (Dodson etal., 1987) and pentobarbital for nicotinic receptors in the electrical eel (Miller et al., 1982). Furthermore, (+)-pentobarbital was found to be more potent than (-)-pentobarbital in binding to nicotinic receptors on membranes isolated from Torpedo electroplaques (Roth etal., 1989). Recent data obtained by Dodson et al. (1990) in the study of ten different barbiturates support the existence of a barbiturate binding site on the nicotinic receptor with specific structural requirements. Although the nicotinic receptor and the GABA receptor belong to the same superfamily of receptors (Barnard et al., 1988), pentobarbital shows different stereoselectivity for the two receptor types (Roth et al., 1989; Ticku and Rastogi, 1986). Furthermore, a number of effects of barbiturates can be modified by drugs acting on the cholinergic system. Cholinergic agonists such as pilocarpine (Wahlstr6m, 1976 a) and the cholinesterase inhibitor physostigmine (Wahlstr6m, 1977) increase the dose of hexobarbital needed to induce anaesthesia in rats. Antagonists like atropine decrease the required dose of hexobarbital (Wahlstr6m, 1976b). This interaction with cholinergic agonists and antagonists cannot be observed for all lipid soluble barbiturates (Bolander and Wahlstr6m, 1984), indicating that there are differences within the barbiturate group in this resepect. Differences in the mechanism of action behind the anaesthetic effect of different barbiturates has also been substantiated by the lack of additative effects in direct studies of the interaction between barbiturates (Wahlstr6m and Norberg, 1984; Norberg and Wahlstr6m, 1986). Barbital has been used in chronic treatment studies of barbiturates in rodents (Wahlstr6m, 1968; McBridge and Turnbull, 1970). Barbital has an advantage over other barbiturates in that it is metabolized only to a small extent and that it has a half-life that is long enough for significant amounts to accumulate
Cholinergic mechanisms in physical dependence
203
during a day of barbital drinking (Wahlstr6m, 1979). Immediately after 4 weeks of barbital treatment in rats, McBridge and Turnbull (1970) found no change in the content of endogenous ACh in the whole brain. Similar results were also obtained in rats when the endogenous ACh content was analysed immediately following much longer treatment periods (25-30 weeks) (Nordberg and Wahlstr6m, 1977). During the abstinence after forced oral long-term consumption of barbital by rats convulsions and decreased sensitivity to hexobarbital threshold doses for EEG burst supression are recorded (Wahlstr6m, 1974). The maximal tolerance to the hexobarbital threshold and the maximal number of spontaneous convulsions are measured 3 days after withdrawal of barbital. An increased sensitivity to the temperature-reducing effect of pilocarpine is seen (Wahlstr6m and Ekwall, 1976) as a sign of supersensitivity of the cholinergic system on the third day of abstinence. The assumption that cholinergic mechanisms are involved in the withdrawal syndrome is supported by the finding of decreased amounts of endogenous ACh in the brain during abstinence following 30 weeks of oral barbital treatment of rats (Nordberg and Wahlstr6m, 1977). Decreased amounts of endogenous ACh have been measured in the striata of animals abstinent for 3 days and with a maximal frequency of spontaneous convulsions (Nordberg and Wahlstr6m, 1977). The content of ACh is still reduced on the 12th day of withdrawal when the convulsions have almost disappeared (Nordberg and Wahlstr6m, 1977). On the same day, a significant decrease in endogenous ACh has also been measured in the cerebellum + medulla oblongata + midbrain (Nordberg and Wahlstr6m, 1977). Interestingly, the biosynthesis of ACh (3 H-ACh) from a tracer dose of radioactive choline (3 H-Ch) was observed to be significantly enhanced in the hippocampus + cortex and cerebellum + medulla oblongata + midbrain of rats abstinent for 3 days, while no significant effect was found in the striatum (Nordberg and Wahlstr6m, 1979). When the ratio of 3H_ACh/3 H-Ch was calculated as a rough measure of ACh turnover, the ratio was significantly increased in the hippocampus + cortex and cerebellum + midbrain + medulla oblongata in rats abstinent for 3 days. The muscarinic receptors seem to be involved in the production of convulsions in the withdrawal period following barbital treatment (Nordberg et al., 1980). An increased number of muscarinic binding sites has been measured in the striatum at the end of treatment and during the first part of abstinence following long-term barbital treatment (Nordberg etal., 1980; Nordberg and Wahlstr6m, 1982 a). Similar changes but with different time courses have also been observed in other areas of the brain (Nordberg and Wahlstr6m, 1982 a). Thus, a variable increase in the number of muscarinic receptors was measured from days 2-9 of abstinence in the cortex whereas an increase was measured in the midbrain on day 9 (Nordberg and Wahlstr6m, 1982a). The change in muscarinic receptors could be traced to an increase in the proportion of muscarinic agonist low-affinity binding sites (Nordberg and Wahlstr6m, 1982b). As indicated in Fig. 2 the presence of convulsions during withdrawal seems to
204
A. Nordberg and G. WahlstrSm STRIATUM
% 150
BARBITAL
ETHANOL
A_
('j) 100 1 III 50
I
(.9 Z I
Z
m
Convulsions
+
0
0
CORTEX
0 Z o/~
