Psychopharmacology (1990) 101 : 545-549
Psychopharmacology © Springer-Verlag 1990
Chronic morphine administration augments benzodiazepine binding and GABAA receptor function* Fred Lopez, Lawrence G. Miller, Michael L. Thompson, Andrew Schatzki, Shelley Chesley, David J. Greenblatt, and Richard I. Shader
Division of Clinical Pharmacology, Departments of Psychiatry and Pharmacology, Tufts University School of Medicine and New England Medical Center, Boston, MA, USA Received July 14, 1989 / Final version December 19, 1989
Abstract. Behavioral and neurochemical evidence indicates links between the opioid and GABA neurotransmitter systems. To assess effects of chronic opiates on the major site of postsynaptic GABAergic activity, the GABAA receptor, we administered chronic morphine and naltrexone to mice and evaluated binding at the benzodiazepine and t-butylbicyclophosphorothionate (TBPS) sites and GABA-dependent chloride uptake. After morphine (3 days), benzodiazepine receptor binding in vivo but not in vitro was increased in cortex compared to placebo-treated mice. TBPS binding was unchanged in cortex, but muscimol-stimulated chloride uptake was increased at low doses of muscimol. Benzodiazepine and TBPS binding and muscimol-stimulated chloride uptake were unchanged in naltrexone-(8 days) compared to placebo-treated mice. When naltrexone was administered previously to block opiate sites, the increases in benzodiazepine binding and chloride uptake observed with chronic morphine were reversed. These results indicate that chronic morphine but not naltrexone enhances benzodiazepine binding and GABAA receptor function, perhaps by an action at opioid receptors. Key words: Morphine - Naltrexone - G A B A - Benzodiazepine - Chloride
Several lines of evidence indicate interactions between the opioid and GABA neurotransmitter systems. Clinical studies report a substantial prevalence of benzodiazepine use among opiate abusers (Stitzer et al. 1981; Woods et al. 1987), perhaps indicating a convergence in effects on * Supported in part by Grants DA-05258 and MH-34223 from the US Public Health Service. Dr. Miller is the recipient of a Faculty Development Award in Clinical Pharmacology from the Pharmaceutical Manufacturers Association Foundation Offprint requests to: L.G. Miller, Division of Clinical Pharmacology, Box 1007, New England Medical Center, 171 Harrison Avenue, Boston, MA 02111, USA
central reward systems. In animal behavioral studies, antagonists in the two systems have overlapping effects. For example, opiate antagonists block benzodiazepineinduced hyperphagia in rats (Cooper et al. 1983a), while a benzodiazepine antagonist inhibits the respiratory depressant effects of opiates (Naughton et al. 1985). In addition, histochemical studies demonstrate that GABA and opiates co-localize to neurons in several brain regions (Guidotti et al. 1983; McEachern et al. 1985). These results indicate a possible neurochemical interaction between the opiate and GABA systems. In prior studies, benzodiazepine administration has been reported to modulate release of endogenous opiates in several brain regions (Harsing et al. 1982), while opiate administration may alter benzodiazepine binding at the GABAA receptor complex (Sivam and Ho 1982; Smith et al. 1984). In addition, endogenous peptides with antiopiate effects also interact with the GABAA receptor complex (Miller and Kastin 1987; Miller et al. 1987a, b, 1989). Limited information is available concerning the effects of chronic opiates on the GABA system, and in particular at the major site of post-synaptic GABA effect, the GABAA receptor. To evaluate these effects, we administered chronic morphine and naltrexone to mice and assessed benzodiazepine and chloride channel binding and receptor function in chloride uptake.
Materials and methods
Male CD1 mice, 6-8 weeks of age, were obtained from Charles River (Wilmington, MA). Morphine pellets (75 mg free base, 3 day duration), morphine placebo, naltrexone (30 mg base, 8 day duration) and naltrexone placebo pellets were obtained from NIDA. [3H]Ro15-1788 (specific activity 82 Ci/mmol), [3H]flunitrazepam (specific activity 78 Ci/mmol), [35S]TBPS (specific activity 81 Ci/ mmol), and [a6C1-] (specific activity 12.5 mCi/g),unlabeled TBPS, and Protosol were obtained from New England Nuclear (Boston, MA). Drug pellets were inserted subcutaneously under brief ether anesthesia. Naltrexone pellets were cut into quarters prior to implantation (i.e. 7.5 mg naltrexone per mouse). For morphine administration, receptor assays were performed at the end of the drug
546 release period (3 days). For naltrexone administration, assays were performed at the end of the drug release period (8 days), except in combined morphine/naltrexone studies where assays were performed after 3 days.
Benzodiazepine receptor bind±n9in vivo. Binding was determined by specific uptake of [3H]Rol 5-1788 as described previously (Miller et al. 1987c). Briefly, mice were injected via the tail vein with 3 pC± [3H]Ro15-1788 and sacrificed after 20 min. Brains were removed and dissected rapidly, solubilized for 24 h at 40 ° C in 2 ml Protosol and counted by scintillation spectrometry. Nonspecific binding was determined by injection of clonazepam, 5 mg/kg, 30 min before radioligand administration.
Benzodiazepine receptor bind±no in vitro. Benzodiazepine receptor binding was performed by the method of Squires and Braestrup (1977). Briefly, synaptosomal membranes (P2) were prepared as described. Mice were decapitated and brains rapidly removed on ice. Brains were dissected and weighed, and then placed in 30 vol 0.32 M sucrose. Tissue was homogenized in a Teflon-glass homogenizer for 20 s and homogenates were centrifuged at 1000 g for 4 ° C for 10 min. Pellets were discarded and supernatants were centrifuged at 19 000 g for 20 rain at 4 ° C. Pellets were resuspended with a Polytron (Brinkmann, Lucerne, Switzerland; setting 7,4 pulses) in 50 vol 50 mM TRIS-HC1 (pH 7.4) and centrifuged at 48000 g for 20 rain at 4 ° C. Washing was repeated 3 times, and pellets were then resuspended in 50 vol TRIS-HC1 as above and frozen at - 7 0 ° C at least overnight. Before use, membranes were thawed, resuspended as above and washed once with TRIS-HC1 as above. To determine benzodiazepine receptor binding, increasing concentrations of [3H]flunitrazepam were added to two sets of duplicate or triplicate tubes (0.1-10 nM). To one set of tubes was added a saturating concentration (10 - 5) of flurazepam, a central benzodiazepine receptor ligand. Brain membranes (0.1 ml, about 55 lag protein) were added to all tubes, along with 50 mM TRIS-HC1 buffer (pH 7.4) to achieve a final volume of 0.5 ml. After incubation for 45 min at 4 ° C, 3 ml ice-cold buffer was added to each tube and tubes were filtered on Whatman GF/B filters using a Brandel M30R apparatus (Gaithersburg, MD) and washed twice with buffer. Filters were counted by conventional scintillation spectrometry. Protein was determined according to Simpson and Sonne (1982).
Data analysis. Data from receptor binding studies were analyzed using the EBDA programs (McPherson 1983). Differences among groups were analyzed using analysis of variance with Dunnett's correction. For muscimol-stimulated chloride uptake, data were fitted to the function y = x a + b.
Results Benzodiazepine binding In vivo. After chronic morphine administration, benzodiazepine receptor binding assessed in vivo was significantly increased in cortex, but not in the other brain
150
zO m Q z
O L)
-
125 '
1OO
0. 75
50 CX
HI
HY
CB
P-M
BRAIN REGION
[35S]TBPS binding. Binding was performed as described previously (Supavilai and Karobath 1983). Membranes were prepared as described above. To a set of duplicate or triplicate tubes, a fixed amount of [35S]TBPS (2 nM) was added. Varying amounts of unlabeled TBPS (10-150 nM) were added along with 0.1 ml 2 M NaCl membranes (0.4 ml, approximately 0.1 mg protein) and buffer as above to a final volume of 1 ml. Tubes were incubated in a shaking water bath at 25 ° C for 90 min, and incubations were terminated by addition of 3 ml ice-cold buffer and filtration on Whatman GF/B filters as described above. Tubes were washed twice with buffer, and filters counted by conventional scintillation spectrometry. Nonspecific binding was determined by the addition of picrotoxinin, 10 -s M. 36Cl- uptake. Chloride uptake was performed using a modification of the method of Schwartz et al. (1986). Briefly, synaptoneurosomes were prepared as described previously. After centrifugation, tissue was resuspended gently in assay buffer (145 mM NaCI, 5 mM KCI, 1 mM MgCI~, 1 mM CaClz and 10 mM 4-(2-hydroxyethyl)-lpiperazine ethanesulfonic acid) with a glass rod and incubated for 15 min at 30 ° C. To 200 lal of membranes was added 200 lal of a solution containing [36C1-] (0.2 laCi/ml assay buffer) with or without muscimol (final concentration, 1-50 laM). Tubes were vortexed immediately and after 6 s the incubation was terminated by addition of 3 ml ice-cold assay buffer and filtration on Whatman G F / C filters as described above. Filters were counted by conventional scintillation spectrometry. The amount of [36C1-] in the absence of muscimol was subtracted from all values to eliminate non-GABA-related uptake.
Fig. 1. Effects of chronic morphine and naltrexone on bezodiazepine receptor binding in vivo. Binding was performed using specific uptake of [3H]Roa5-1788. Assays were performed in morphinetreated and naltrexone/morphine-treated mice after 3 days, and in naltrexone-treated mice after 8 days (CX= cortex, HY= hypothalamus, HI= hippocampus, CB = cerebellum, P - M = pons-medulla). Control binding in cortex was: Morphine placebo 1686+_ 86 fmol/g; naltrexone placebo 1500 + 132 fmol/g. ~ Naltrexone; N~ morphine; ~ naltrexone-morphine
Table 1. Effects of chronic morphine on benzodiazepine receptor binding and [aSS]TBPS binding in cortex K d
B m ax
(nM)
(pmot/mg protein)
[ 3H]flunitr azepam Morphine placebo Morphine
1.65 ± 0.07 1.69±0.15
1.03 ± 0.04 1.05±0.04
[35H]TBPS Morphine placebo Morphine
19.9 ± 1.2 24,8 ± 4.7
2.34 ± 0.12 2.32 ± 0.52
Binding was performed in synaptosomal membrane preparations (P2) from cerebral cortex. Results are m e a n + SEM, n = 3 for each group. There are no significant differences
547 Table 2. Effects of chronic naltrexone on benzodiazepine receptor binding and [asS]TBPS binding in cortex Kd (nM)
Bmax (pmol/mg protein)
[ 3H]flunitrazepam Naltrexone placebo Naltrexone
1.04±0.07 0.96 ± 0.03
t.35±0.05 1.48 ± 0.04
[35S]TBPS Naltrexone placebo Naltrexone
30.0 ± 6.3 35.3 ± 5.0
1.86 ± 0.64 2.75 ± 0.30
Binding was performed in synaptosomal membrane preparations (P2) from cerebral cortex. Results are m e a n + SEM, n = 3 for each group. There are no significant differences
30
25 ¸
ii J
~
..A
o D_ 20
15
•"
i /
, , j 10
i
.~
it..'f Z.f' ,g~e
,r"
.,
Al-;. ~'"
ii " l "j al
5
0 1
1
10 MUSCIMOL
100
[uM]
Fig. 2. Effects of chronic morphine and naltrexone on muscimolstimulated chloride uptake in cortical synaptoneurosomes. Studies were performed after 3 days in morphine-treated and naltrexone/ morphine-treated mice and after 8 days in naltrexone-treated mice. Results are means from two to three separate experiments performed in triplicate or quadruplicate. ECso values for each treatment are: morphine placebo, 10.9 pM ; morphine, 4.1 g M ; naltrexone placebo, 8.8 gM; naltrexone, 10.0 g M ; naltrexone morphine 8.1 gM - - o - Morphine placebo; - - o - morphine; . . . . zx--naltexone placebo; ..... • ..... naltrexone, ---. • . . . . naltrexone morphine
regions evaluated (hypothalamus, hippocampus, cerebellum, pons-medulla) (Fig. 1). No change in specific binding was observed after chronic naltrexone administration. Nonspecific binding was similar in mice treated with morphine, naltrexone, and their respective placebos (data no shown). In vitro. Receptor binding as determined in vitro was similar to placebo in morphine- (Table 1) and naltrexone-treated mice (Table 2). Drug administration did not affect apparent affinity or receptor number. TBPS binding and chloride uptake Binding at the TBPS site in cortex was similar in morphine and morphine placebo groups (Table 1) and in
naltrexone and naltrexone placebo groups (Table 2). No change was seen in apparent affinity or in receptor number after these treatments. GABAA receptor function was assessed by uptake of [36C1-] into cortical synaptoneurosomes stimulated by the GABA analog muscimol (Fig. 2). Uptake was similar in naltrexone and naltrexone placebo groups (naltrexone placebo ECso = 8.8 gM; naltrexone ECso = 10.0 gM, P < 0.40). However, chloride uptake in morphine-treated mice was significantly increased compared to morphine placebo-treated mice (morphine placebo ECso=10.9 gM; morphine EC5o=4.1 gM; P
Psychopharmacology © Springer-Verlag 1990
Chronic morphine administration augments benzodiazepine binding and GABAA receptor function* Fred Lopez, Lawrence G. Miller, Michael L. Thompson, Andrew Schatzki, Shelley Chesley, David J. Greenblatt, and Richard I. Shader
Division of Clinical Pharmacology, Departments of Psychiatry and Pharmacology, Tufts University School of Medicine and New England Medical Center, Boston, MA, USA Received July 14, 1989 / Final version December 19, 1989
Abstract. Behavioral and neurochemical evidence indicates links between the opioid and GABA neurotransmitter systems. To assess effects of chronic opiates on the major site of postsynaptic GABAergic activity, the GABAA receptor, we administered chronic morphine and naltrexone to mice and evaluated binding at the benzodiazepine and t-butylbicyclophosphorothionate (TBPS) sites and GABA-dependent chloride uptake. After morphine (3 days), benzodiazepine receptor binding in vivo but not in vitro was increased in cortex compared to placebo-treated mice. TBPS binding was unchanged in cortex, but muscimol-stimulated chloride uptake was increased at low doses of muscimol. Benzodiazepine and TBPS binding and muscimol-stimulated chloride uptake were unchanged in naltrexone-(8 days) compared to placebo-treated mice. When naltrexone was administered previously to block opiate sites, the increases in benzodiazepine binding and chloride uptake observed with chronic morphine were reversed. These results indicate that chronic morphine but not naltrexone enhances benzodiazepine binding and GABAA receptor function, perhaps by an action at opioid receptors. Key words: Morphine - Naltrexone - G A B A - Benzodiazepine - Chloride
Several lines of evidence indicate interactions between the opioid and GABA neurotransmitter systems. Clinical studies report a substantial prevalence of benzodiazepine use among opiate abusers (Stitzer et al. 1981; Woods et al. 1987), perhaps indicating a convergence in effects on * Supported in part by Grants DA-05258 and MH-34223 from the US Public Health Service. Dr. Miller is the recipient of a Faculty Development Award in Clinical Pharmacology from the Pharmaceutical Manufacturers Association Foundation Offprint requests to: L.G. Miller, Division of Clinical Pharmacology, Box 1007, New England Medical Center, 171 Harrison Avenue, Boston, MA 02111, USA
central reward systems. In animal behavioral studies, antagonists in the two systems have overlapping effects. For example, opiate antagonists block benzodiazepineinduced hyperphagia in rats (Cooper et al. 1983a), while a benzodiazepine antagonist inhibits the respiratory depressant effects of opiates (Naughton et al. 1985). In addition, histochemical studies demonstrate that GABA and opiates co-localize to neurons in several brain regions (Guidotti et al. 1983; McEachern et al. 1985). These results indicate a possible neurochemical interaction between the opiate and GABA systems. In prior studies, benzodiazepine administration has been reported to modulate release of endogenous opiates in several brain regions (Harsing et al. 1982), while opiate administration may alter benzodiazepine binding at the GABAA receptor complex (Sivam and Ho 1982; Smith et al. 1984). In addition, endogenous peptides with antiopiate effects also interact with the GABAA receptor complex (Miller and Kastin 1987; Miller et al. 1987a, b, 1989). Limited information is available concerning the effects of chronic opiates on the GABA system, and in particular at the major site of post-synaptic GABA effect, the GABAA receptor. To evaluate these effects, we administered chronic morphine and naltrexone to mice and assessed benzodiazepine and chloride channel binding and receptor function in chloride uptake.
Materials and methods
Male CD1 mice, 6-8 weeks of age, were obtained from Charles River (Wilmington, MA). Morphine pellets (75 mg free base, 3 day duration), morphine placebo, naltrexone (30 mg base, 8 day duration) and naltrexone placebo pellets were obtained from NIDA. [3H]Ro15-1788 (specific activity 82 Ci/mmol), [3H]flunitrazepam (specific activity 78 Ci/mmol), [35S]TBPS (specific activity 81 Ci/ mmol), and [a6C1-] (specific activity 12.5 mCi/g),unlabeled TBPS, and Protosol were obtained from New England Nuclear (Boston, MA). Drug pellets were inserted subcutaneously under brief ether anesthesia. Naltrexone pellets were cut into quarters prior to implantation (i.e. 7.5 mg naltrexone per mouse). For morphine administration, receptor assays were performed at the end of the drug
546 release period (3 days). For naltrexone administration, assays were performed at the end of the drug release period (8 days), except in combined morphine/naltrexone studies where assays were performed after 3 days.
Benzodiazepine receptor bind±n9in vivo. Binding was determined by specific uptake of [3H]Rol 5-1788 as described previously (Miller et al. 1987c). Briefly, mice were injected via the tail vein with 3 pC± [3H]Ro15-1788 and sacrificed after 20 min. Brains were removed and dissected rapidly, solubilized for 24 h at 40 ° C in 2 ml Protosol and counted by scintillation spectrometry. Nonspecific binding was determined by injection of clonazepam, 5 mg/kg, 30 min before radioligand administration.
Benzodiazepine receptor bind±no in vitro. Benzodiazepine receptor binding was performed by the method of Squires and Braestrup (1977). Briefly, synaptosomal membranes (P2) were prepared as described. Mice were decapitated and brains rapidly removed on ice. Brains were dissected and weighed, and then placed in 30 vol 0.32 M sucrose. Tissue was homogenized in a Teflon-glass homogenizer for 20 s and homogenates were centrifuged at 1000 g for 4 ° C for 10 min. Pellets were discarded and supernatants were centrifuged at 19 000 g for 20 rain at 4 ° C. Pellets were resuspended with a Polytron (Brinkmann, Lucerne, Switzerland; setting 7,4 pulses) in 50 vol 50 mM TRIS-HC1 (pH 7.4) and centrifuged at 48000 g for 20 rain at 4 ° C. Washing was repeated 3 times, and pellets were then resuspended in 50 vol TRIS-HC1 as above and frozen at - 7 0 ° C at least overnight. Before use, membranes were thawed, resuspended as above and washed once with TRIS-HC1 as above. To determine benzodiazepine receptor binding, increasing concentrations of [3H]flunitrazepam were added to two sets of duplicate or triplicate tubes (0.1-10 nM). To one set of tubes was added a saturating concentration (10 - 5) of flurazepam, a central benzodiazepine receptor ligand. Brain membranes (0.1 ml, about 55 lag protein) were added to all tubes, along with 50 mM TRIS-HC1 buffer (pH 7.4) to achieve a final volume of 0.5 ml. After incubation for 45 min at 4 ° C, 3 ml ice-cold buffer was added to each tube and tubes were filtered on Whatman GF/B filters using a Brandel M30R apparatus (Gaithersburg, MD) and washed twice with buffer. Filters were counted by conventional scintillation spectrometry. Protein was determined according to Simpson and Sonne (1982).
Data analysis. Data from receptor binding studies were analyzed using the EBDA programs (McPherson 1983). Differences among groups were analyzed using analysis of variance with Dunnett's correction. For muscimol-stimulated chloride uptake, data were fitted to the function y = x a + b.
Results Benzodiazepine binding In vivo. After chronic morphine administration, benzodiazepine receptor binding assessed in vivo was significantly increased in cortex, but not in the other brain
150
zO m Q z
O L)
-
125 '
1OO
0. 75
50 CX
HI
HY
CB
P-M
BRAIN REGION
[35S]TBPS binding. Binding was performed as described previously (Supavilai and Karobath 1983). Membranes were prepared as described above. To a set of duplicate or triplicate tubes, a fixed amount of [35S]TBPS (2 nM) was added. Varying amounts of unlabeled TBPS (10-150 nM) were added along with 0.1 ml 2 M NaCl membranes (0.4 ml, approximately 0.1 mg protein) and buffer as above to a final volume of 1 ml. Tubes were incubated in a shaking water bath at 25 ° C for 90 min, and incubations were terminated by addition of 3 ml ice-cold buffer and filtration on Whatman GF/B filters as described above. Tubes were washed twice with buffer, and filters counted by conventional scintillation spectrometry. Nonspecific binding was determined by the addition of picrotoxinin, 10 -s M. 36Cl- uptake. Chloride uptake was performed using a modification of the method of Schwartz et al. (1986). Briefly, synaptoneurosomes were prepared as described previously. After centrifugation, tissue was resuspended gently in assay buffer (145 mM NaCI, 5 mM KCI, 1 mM MgCI~, 1 mM CaClz and 10 mM 4-(2-hydroxyethyl)-lpiperazine ethanesulfonic acid) with a glass rod and incubated for 15 min at 30 ° C. To 200 lal of membranes was added 200 lal of a solution containing [36C1-] (0.2 laCi/ml assay buffer) with or without muscimol (final concentration, 1-50 laM). Tubes were vortexed immediately and after 6 s the incubation was terminated by addition of 3 ml ice-cold assay buffer and filtration on Whatman G F / C filters as described above. Filters were counted by conventional scintillation spectrometry. The amount of [36C1-] in the absence of muscimol was subtracted from all values to eliminate non-GABA-related uptake.
Fig. 1. Effects of chronic morphine and naltrexone on bezodiazepine receptor binding in vivo. Binding was performed using specific uptake of [3H]Roa5-1788. Assays were performed in morphinetreated and naltrexone/morphine-treated mice after 3 days, and in naltrexone-treated mice after 8 days (CX= cortex, HY= hypothalamus, HI= hippocampus, CB = cerebellum, P - M = pons-medulla). Control binding in cortex was: Morphine placebo 1686+_ 86 fmol/g; naltrexone placebo 1500 + 132 fmol/g. ~ Naltrexone; N~ morphine; ~ naltrexone-morphine
Table 1. Effects of chronic morphine on benzodiazepine receptor binding and [aSS]TBPS binding in cortex K d
B m ax
(nM)
(pmot/mg protein)
[ 3H]flunitr azepam Morphine placebo Morphine
1.65 ± 0.07 1.69±0.15
1.03 ± 0.04 1.05±0.04
[35H]TBPS Morphine placebo Morphine
19.9 ± 1.2 24,8 ± 4.7
2.34 ± 0.12 2.32 ± 0.52
Binding was performed in synaptosomal membrane preparations (P2) from cerebral cortex. Results are m e a n + SEM, n = 3 for each group. There are no significant differences
547 Table 2. Effects of chronic naltrexone on benzodiazepine receptor binding and [asS]TBPS binding in cortex Kd (nM)
Bmax (pmol/mg protein)
[ 3H]flunitrazepam Naltrexone placebo Naltrexone
1.04±0.07 0.96 ± 0.03
t.35±0.05 1.48 ± 0.04
[35S]TBPS Naltrexone placebo Naltrexone
30.0 ± 6.3 35.3 ± 5.0
1.86 ± 0.64 2.75 ± 0.30
Binding was performed in synaptosomal membrane preparations (P2) from cerebral cortex. Results are m e a n + SEM, n = 3 for each group. There are no significant differences
30
25 ¸
ii J
~
..A
o D_ 20
15
•"
i /
, , j 10
i
.~
it..'f Z.f' ,g~e
,r"
.,
Al-;. ~'"
ii " l "j al
5
0 1
1
10 MUSCIMOL
100
[uM]
Fig. 2. Effects of chronic morphine and naltrexone on muscimolstimulated chloride uptake in cortical synaptoneurosomes. Studies were performed after 3 days in morphine-treated and naltrexone/ morphine-treated mice and after 8 days in naltrexone-treated mice. Results are means from two to three separate experiments performed in triplicate or quadruplicate. ECso values for each treatment are: morphine placebo, 10.9 pM ; morphine, 4.1 g M ; naltrexone placebo, 8.8 gM; naltrexone, 10.0 g M ; naltrexone morphine 8.1 gM - - o - Morphine placebo; - - o - morphine; . . . . zx--naltexone placebo; ..... • ..... naltrexone, ---. • . . . . naltrexone morphine
regions evaluated (hypothalamus, hippocampus, cerebellum, pons-medulla) (Fig. 1). No change in specific binding was observed after chronic naltrexone administration. Nonspecific binding was similar in mice treated with morphine, naltrexone, and their respective placebos (data no shown). In vitro. Receptor binding as determined in vitro was similar to placebo in morphine- (Table 1) and naltrexone-treated mice (Table 2). Drug administration did not affect apparent affinity or receptor number. TBPS binding and chloride uptake Binding at the TBPS site in cortex was similar in morphine and morphine placebo groups (Table 1) and in
naltrexone and naltrexone placebo groups (Table 2). No change was seen in apparent affinity or in receptor number after these treatments. GABAA receptor function was assessed by uptake of [36C1-] into cortical synaptoneurosomes stimulated by the GABA analog muscimol (Fig. 2). Uptake was similar in naltrexone and naltrexone placebo groups (naltrexone placebo ECso = 8.8 gM; naltrexone ECso = 10.0 gM, P < 0.40). However, chloride uptake in morphine-treated mice was significantly increased compared to morphine placebo-treated mice (morphine placebo ECso=10.9 gM; morphine EC5o=4.1 gM; P
