Address correspondence and reprint requests to Dr M Naassila, Equipe Région INSERM 24 (ERI 24), Groupe de Recherche sur l’Alcool et les Pharmacodépendances (GRAP), Université de Picardie Jules Verne, Faculté de Pharmacie, 1 rue des Louvels, Amiens 80000, France. E-mail: email@example.com
As the contribution of cannabinoid (CB1) receptors in the neuroadaptations following chronic alcohol exposure is unknown, we investigated the neuroadaptations induced by chronic alcohol exposure on both NMDA and GABAA receptors in CB1−/− mice. Our results show that basal levels of hippocampal [3H]MK-801 ((1)-5-methyl-10,11-dihydro-5Hdibenzo[a,d]cyclohepten-5,10-imine) binding sites were decreased in CB1−/− mice and that these mice were also less sensitive to the locomotor effects of MK-801. Basal level of both hippocampal and cerebellar [3H]muscimol binding was lower and sensitivity to the hypothermic effects of diazepam and pentobarbital was increased in CB1−/− mice. GABAAα1, β2, and γ2 and NMDA receptor (NR) 1 and 2B subunit mRNA levels were altered in striatum of CB1−/− mice. Our results also showed that [3H]MK-801 binding sites were increased in cerebral cortex and hippocampus after chronic ethanol ingestion only in wild-type mice. Chronic ethanol ingestion did not modify the sensitivity to the locomotor effects of MK-801 in both genotypes. Similarly, chronic ethanol ingestion reduced the number of [3H]muscimol binding sites in cerebral cortex, but not in cerebellum, only in CB1+/+ mice. We conclude that lifelong deletion of CB1 receptors impairs neuroadaptations of both NMDA and GABAA receptors after chronic ethanol exposure and that the endocannabinoid/CB1 receptor system is involved in alcohol dependence.
Alcoholism is a devastating illness with a profound public health impact and which is influenced by both genes and environment. The specific genes involved have not yet been identified and many genes, each with a small effect, are likely to predispose an individual to abuse alcohol. Some promising new targets are ermerging by using rodent genetic animal models such as transgenics and null mutants (Crabbe and Phillips 2004). Thus the generation of transgenic mice over-expressing or knockout mice deleted of a selected gene involved in neurotransmission has proven valuable for studying its implication in ethanol-mediated behaviors. Among the numerous targets that have been identified, endocannabinoid (CB1) receptors are thought to mediate several behavioral effects of ethanol.
There is a substantial and growing literature implicating the brain endocannabinoid system in alcohol and other drugs addiction (see for review Maldonado et al. 2006). In humans, some studies have suggested that genetic variants of the CNR1 gene might be associated with susceptibility to alcohol or drug dependence (Comings et al. 1997; Schmidt et al. 2002), although not all agree (Preuss et al. 2003). Recent advances in the understanding of the neurobiological basis of alcohol dependence support the view that the endocannabinoid system represents a new candidate for the control of alcohol rewarding properties. Indeed, CB1 receptors have been implicated in the pharmacological and behavioral effects of ethanol (Hungund et al. 2002; Mechoulam and Parker 2003). Interestingly, the main endocannabinoid-degrading enzyme fatty acid amide hydrolase has also been shown to be involved in alcohol drinking behavior (Basavarajappa et al. 2006; Hansson et al. 2007).
CB1 receptors are abundant in the brain reward circuitry and participate in the addictive properties induced by different drugs of abuse. Activity of the dopaminergic neurons of the mesocorticolimbic pathway are controlled by excitatory and inhibitory inputs that are modulated by CB1 receptors. In vivo microdialysis studies revealed that alcohol did not enhance extracellular levels of dopamine in the nucleus accumbens in CB1 knockout mice or after CB1 antagonist pre-treatment (Hungund et al. 2003). Recent in vivo electrophysiological data indicate that the alcohol-induced increase in dopaminergic neurons firing rate is blocked by CB1 receptor antagonist SR141716A or by fatty acid amide hydrolase antagonism (Cohen et al. 2002; Perra et al. 2005). This suggests that both endocannabinoids and CB1 receptors are involved in the cellular effects of alcohol in the mesolimbic reward circuit.
Converging evidence suggests that the CB1 receptor signaling system could play an important role in modulating alcohol-reinforcing effects and alcohol drinking behavior. Different studies have shown that SR141716A reduces alcohol intake (Arnone et al. 1997; Colombo et al. 1998; Rodriguez de Fonseca et al. 1999), the alcohol deprivation effect (Serra et al. 2002), and the motivation to consume alcohol in a progressive ratio paradigm (Gallate and McGregor 1999) in rats, while a CB1 receptor agonist increased the motivation to consume alcohol in a progressive ratio paradigm (Gallate et al. 1999). In addition, ethanol (0.5–2.0 g/kg) has been shown to decrease operant responding to a greater extent in CB1−/− mice than in wild-type mice, suggesting a possible role of CB1 receptor in the rate disruptive effects of ethanol (Baskfield et al. 2004). Recently, we and others have shown that ethanol consumption and/or preference are decreased in CB1−/− mice generated on a CD1 background (Naassila et al. 2004) or a C57BL/6J background (Hungund et al. 2003; Poncelet et al. 2003; Wang et al. 2003). Importantly, our recent study has clearly demonstrated that this decreased ethanol consumption is associated with a decreased sensitivity to the rewarding effects of alcohol measured in the conditioned place preference paradigm (Houchi et al. 2005). Furthermore, our previous study has shown that this decrease in voluntary ethanol intake and preference observed in CB1−/− mice was associated with an increased ethanol sensitivity (hypothermia, sedation, and locomotion) and ethanol withdrawal severity (Naassila et al. 2004).
Although chronic alcohol exposure has been shown to down-regulate CB1 receptors and increases the content of endogenous CBs in cultured cells (Basavarajappa et al. 2000), little is known about the implication of endocannabinoids and CB1 receptors in the neuronal adaptations induced by chronic alcohol exposure. We have previously shown that CB1−/− mice display enhanced severity of alcohol withdrawal-induced convulsions (Naassila et al. 2004) and this enhanced severity could be due, at least in part, to a difference in neuroadaptation of neurotransmitter systems involved in alcohol withdrawal.
Given the established role of both glutamatergic and GABAergic systems in sensitivity to acute alcohol exposure and adaptation after chronic exposure (De Witte 2004), the present study was designed to investigate the effect of chronic ethanol exposure on NMDA and GABAA receptor systems in mice lacking the CB1 receptors. As we observed difference in basal level of [3H]muscimol binding sites between genotypes, the sensitivity to positive allosteric modulators of GABAA receptors was investigated and the basal level of expression of several GABAA receptor subunits (α1, α2, α6, β2, and γ2) potentially involved in ethanol action was also determined in hippocampus, striatum, and cerebellum. These brain regions have been shown to be important for the behavioral responses to ethanol and for both NMDA and GABAA neuroadaptations following chronic exposure to ethanol (see Faingold et al. 1998).
Almost half of all GABAA receptors in the brain is that containing the αl subunit in combination with the β2 and γ2 subunits. Recent evidence suggests that GABAA receptors containing an α1 and an β2 subunit mediate the sedative effect of benzodiazepines and of anesthetics, respectively (McKernan et al. 2000; Reynolds et al. 2003), whereas receptors with an α2 subunit mediate benzodiazepine’s anxiolytic effect (Low et al. 2000). Similarly, as we observed difference in basal level of [3H]MK-801 ((1)-5-methyl-10,11-dihydro-5Hdibenzo[a,d]cyclohepten-5,10-imine) binding sites between genotypes, the sensitivity to locomotor effect of MK-801 was measured and the level of expression of the NMDA receptor subunit (NR1) (ubiquitously expressed in the CNS and obligatory for the receptor functioning) and the NR2 subunit that confer increased sensitivity of the receptor to ethanol (Sucher et al. 1996) were determined.
Materials and methods
CB1 null-mutant mice were generated by homologous recombination as described (Ledent et al. 1999). Briefly, a phosphoglycerokinase-Neo cassette was inserted between AvrII and SfiI sites located 1073 base pairs apart, replacing the first 233 codons of the gene. Homologous recombination in R1 cells and aggregation with CD1 eight-cell stage embryos were performed. A recombinant line was used to generate chimeras allowing germ line transmission of the mutant gene. Heterozygous mice were bred for 15 generations on a CD1 background before generating the wild-type and CB1 null littermates used in this study. The F14 generation of homozygous mice was genotyped and therefore used to produce the F15 generation that has been used for the experiments. Adult male wild-type and knockout mice (8 to 14-week old) weighing 20–30 g were used. All animals used in a given experiment were derived from the same breeding series and were matched for age and weight. Mice were housed in groups of 10 in clear plastic cages and maintained in a temperature (∼20°C) and humidity-controlled room on a 12 h light/dark cycle. The number of animals was kept to a minimum and all efforts were made to avoid animal suffering. Experiments were carried out in strict accordance with both the Guide for the Care and Use of Laboratory Animals (National Institutes of Health) and the E.C. regulations for animal use in research (CEE No 86/609). All experiments were performed under blind conditions.
Ethanol at 20% (w/v) was prepared for i.p. injection in 0.9% saline from 96.2% ethyl alcohol (Carlo Erba réactifs, Val de Reuil, France). (+)-MK-801, sodium pentobarbital, and diazepam were obtained from Sigma (Paris, France) and prepared in 0.9% saline.
Chronic alcohol exposure
Male mice were individually housed in plastic mouse cages with ad libitum access to standard rodent chow and were habituated in their home cage to drinking from one bottle containing plain water for 1 week. Mice were exposed to forced and gradual (single bottle exposure, starting with 3% ethanol and increasing to 12% ethanol) ethanol exposure for 21 days. Another group of mice, the control group, was not exposed to alcohol and had free access to water also for 21 days. Ethanol intake and body weight were assessed regularly throughout the experiments. To obtain a measure of ethanol consumption that corrected for individual differences in mouse size, grams of ethanol consumed per kilogram of body weight per day were calculated for each mouse.
Determination of blood ethanol levels
Three times during the last week ethanol exposure period, 12 h after the incorporation of a newly fresh ethanol solution (0700 h, lights on), blood from the tip of the tail was obtained in heparinized tubes from each one of the animals. The blood ethanol levels (BELs) were determined by the NAD/NADH enzymatic method (sensibility ≥ 0.1 mg/dL), as previously described (Allali-Hassani et al. 1997).
Locomotor activity was assessed in the LE 8811 IR motor activity monitor (BIOSEB, Chaville, France) as previously described (Naassila et al. 2004). Animals were confined to 45 cm2 clear acrylic plastic chamber, in which horizontal locomotion was measured from photocell beam interruptions. Photocell beams transected the chamber 2 cm above the floor at 16 sites along each side. Test chambers were shielded from external noise and light, but each test field was illuminated with a white fluorescent light and was fully ventilated. Male (n = 6–8/group) mice were injected i.p. with saline or 0.03, 0.1, 0.3, and 0.5 mg/kg MK-801 and were immediately placed in activity monitors for a 30 min test duration. To study the sensitivity to the locomotor effect of MK-801 (0.3 mg/kg) after chronic ethanol ingestion, mice were placed in the test cages for a 30-min habituation period before drug injection, in order to analyze the influence of chronic ethanol ingestion on the basal activity. Their activity was also measured for 30 min after MK-801 injection.
Sensitivity to drug-induced hypothermia
To measure hypothermia induced by diazepam (5.0 and 10 mg/kg body weight) or pentobarbital (25 and 50 g/kg body weight) administration, rectal temperature was measured using a KJT thermocouple (BIOSEB, Paris, France) at 20°C before and after an i.p. drug injection, as previously described (Naassila et al. 2002). Rectal temperature was assessed every 30 min after drug administration (n = 10/group). Two-way 2 × 6 (genotype × time) repeated measures analyzes of variance (RM-anovas) and Tukey’s post hoc test were used for statistical analysis. In order to investigate the development of ethanol tolerance after chronic ethanol exposure, we measured the sensitivity to ethanol (3.0 g/kg) -induced hypothermia 60 min after injection, the first day of the chronic ethanol exposure (day 1) and the last day of this chronic exposure (day 21). Different groups of mice (n = 10/group) were used for the test at day 1 and 21.
mRNA levels measured with semiquantitative RT-PCR
The mRNA levels for the GABAA receptors β2 and γ2 subunits were measured in hippocampus and the α6 subunit mRNA levels were measured in cerebellum where the α6 subunit is specifically expressed. Total RNAs were extracted from hippocampus, striatum, and cerebellum (n = 8 mice/genotype) using the RNA Insta-Pure System according to the manufacturer’s procedure (Eurogentec, Belgium). The mRNA levels for the various subunits were measured using the semiquantitative RT-PCR technique and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA was co-amplied and used as a standard, as previously described (Naassila and Daoust 2002). Two micrograms of total RNAs were converted to cDNAs using 5 U of reverse transcriptase from Moloney-Murine leukemia virus for 2 h at 37°C in 30 μL of 50 mmol/L Tris–HCl, pH 8.3, containing 10 U of RNAse inhibitor, 250 pmol of random hexamers (pd(N)6), 1 mmol/L of each deoxynucleoside tri-phosphate (d-ATP, d-CTP, d-GTP, and d-TTP), 75 mmol/L KCl, 3 mmol/L MgCl2, and 10 mmol/L dithiothreitol. Amplification was then performed in 50 μL of 10 mmol/L Tris–HCl, pH 8.3, containing 2.5 U of Red-Taq polymerase, 2 mmol/L MgCl2, 35 mmol/L KCl, 0.1% Triton X-100, 0.2 mmol/L of each deoxynucleoside triphosphate, and 200 nmol/L of each primer. The amplification profile involved four linked procedures as follows: 5 min at 94°C for one cycle; 1 min at 94°C, 1 min at the specific annealing temperature and 1 min at 72°C for various numbers of cycles and, finally, 8 min at 72°C for one cycle. PCR was carried out at different annealing temperatures and numbers of cycle depending on the subunits, using specific primers. GABAα1, forward 5′-TGCCCATGCCTGCCCACTAAAAA-3′ reverse 5′-GCCATCCCACGCATACCCTCTCT-3′; GABAα2, forward 5′-AAAAGAGGATGGGCTTGGGA-3′ reverse 5′-ACGGGATGTTTTCTGCCTGTAT-3′; GABAα6, forward 5′-CAAGCTCAACTTGAAGATGAAGG-3′ reverse 5′-TCCATCCATAGGGAAGTTAACC-3′; GABAβ2, forward 5′-GGAGTGACAAAGATTGAGCTTCCT-3′ reverse 5′-GTCTCCAAGTCCCATTACTGCTTC-3′; GABAγ2, forward 5′-GTGAAGACAACTTCTGGTGACTATGTGGT-3′ reverse 5′-CATATTCTTCATCCCTCTCTTGAAGGTG-3′; NR1, forward 5′-AACCTGCAGAACCGCAAG-3′ reverse 5′-GCTTGATGAGCAGGTCTATGC-3′; NR2B, forward 5′-TGCACAATTACTCCTCGACG-3′ reverse 5′-TCCGATTCTTCTTCTGAGCC-3′; and GAPDH, forward 5′-TGAAGGTCGGTGTGAACGGATTTG-3′ reverse 5′-CATGTAGGCCATGAGGTCCACCAC-3′. The amplified fragment sizes in bp were 511, 563, 420, 564, 484, 334, 222, and 983 for GABAα1, GABAα2, GABAα6, GABAβ2, GABAγ2, NR1, NR2B, and GAPDH, respectively.
Ten microliters of PCR products were electrophoresed in 2% agarose gel stained with ethidium bromide (0.5 μg/mL) and quantified with The ImagerTM gel analyser (Appligene Oncor, Illkirch, France) and NIH Image 1.44 analysis software (National Institutes of Health, Bethesda, MD, USA).
Membrane preparation and [3H]MK-801, [3H]muscimol binding studies
Membrane preparation and binding experiments were performed as previously described (Naassila and Daoust 2002). Briefly, hippocampi and cerebellum from individual mouse (n = 7–9/genotype) were rapidly removed on ice and thawed in five volumes of 0.32 mol/L sucrose using an Elvejhem-type potter (Fisher Scientific Bioblock, Illkirch, France). After the first centrifugation (3000 g, 4°C, 15 min), the supernatant was further centrifuged (48 000 g, 4°C, 15 min). The pellet was carefully rinsed at least 10 more times using five volumes of Tris–HCl buffer (pH 7.4, 20°C). The final pellet was frozen (−18°C) until use, and a 10 μL of aliquot was used for protein measurement by the method of Lowry et al. (1951). [3H]MK-801 binding assays were performed using well-washed membranes. Then, 0.15 mg of protein was incubated in 0.5 mL of 10 mmol/L Tris–HCl buffer (pH 7.4, 20°C) containing 2.5 nmol/L [3H]MK-801, 100 μmol/L glutamate, and 30 μmol/L glycine at 20°C for 2 h. Non-specific binding was defined using 100 μmol/L unlabeled MK-801. The contents of the tubes were rapidly filtered on Whatman glass fibre filters (GF/B, 45-μm pore size; Fisher Scientific Bioblock) and rinsed with 2 × 5 mL of cold Tris buffer. Radioactivity was determined using 5 mL of ACS scintillation fluid (Fisher Scientific Bioblock) and counted in a Wallac 1414 Winspectral liquid scintillation counter (Perkin Elmer, Courtaboeuf, France; 60% efficiency for [3H]). The same procedure was used for the [3H]muscimol binding and incubation were carried out with 5 nmol/L [3H]muscimol (as described by Mehta and Ticku 1998) without both glutamate and glycine and the non-specific binding was defined using 2 mmol/L GABA. The concentrations of both [3H]MK-801 and [3H]muscimol have been determined from pilot saturation experiments and correspond to the Kd value for each radioligand.
Statistical analyzes were conducted using SigmaStat software (SPSS Inc., Erkrath, Deutschland). Radioligand binding experiments were analyzed using a two-way analysis of variance (anova) followed by a Tuckey’s post hoc test (factors genotype × treatment). For the locomotor activity, the effect of genotype and drug was analyzed using two-way anova (genotype × dose) or (genotype × treatment) and Tuckey’s post hoc test where appropriate. Gene expression of the GABAA subunits was analyzed with Mann–Whitney tests, each subunit was analyzed separately. For the experiments on the sensitivity to drug-induced hypothermia, a two-way (genotype × time) RM-anovas and Tukey’s post hoc test were used for statistical analysis. A significance level of 0.05 was used for all tests.
Effects of chronic ethanol administration on both [3H]MK801 and [3H]muscimol binding
Regarding NMDA receptors in cerebral cortex (Fig. 1a, left panel), two-way anova showed a significant interaction between factors treatment and genotype (F3.32 = 6.77, p < 0.05). Tuckey’s post hoc test revealed that CB1+/+ mice exposed to ethanol displayed an increased number of [3H]MK-801 binding sites versus non-ethanol exposed CB1+/+ mice (p < 0.05) and versus CB1−/− exposed to ethanol (p < 0.01). However, chronic ethanol ingestion failed to change the number of [3H]MK-801 binding sites in CB1−/− mice. In hippocampus (Fig. 1a, right panel), basal levels of [3H]MK-801 binding sites were lower in CB1−/− mice compared with CB1+/+ mice (F3.16 = 6.47, p < 0.05). Chronic ethanol exposure induced an increase in the number of [3H]MK-801 binding sites in CB1−/− mice, with a main effect of genotype (F3.32 = 7.85, p < 0.01) and a main effect of treatment (F3.35 = 7.09, p < 0.05) but no significant interaction between these two factors (F3.32 = 3.03, p > 0.05). Tuckey’s post hoc test revealed that CB1+/+ mice exposed to ethanol displayed an increased number of [3H]MK-801 binding sites versus non-ethanol exposed CB1+/+ mice (p < 0.05) and versus CB1−/− mice exposed to ethanol (p < 0.05).
Regarding GABAA receptors (Fig. 1b), basal levels of [3H]muscimol binding sites were lower in CB1−/− mice compared with CB1+/+ mice in cerebral cortex (F3.32 = 11.94, p < 0.01; Fig. 1b, left panel) and in cerebellum (F3.32 = 6.01, p < 0.05; Fig. 1b, right panel). In cerebral cortex, two-way anova showed a main effect of treatment (F3.32 = 15.75, p < 0.001; Fig. 1b, left panel), genotype (F3.32 = 3.95, p < 0.05), and also a significant interaction between factors treatment and genotype (F3.32 = 3.41, p < 0.05). Tuckey’s post hoc test revealed that CB1+/+ mice exposed to ethanol displayed a decreased number of [3H]muscimol binding sites versus non-ethanol exposed CB1+/+ mice (p < 0.01). However, chronic ethanol ingestion failed to change the number of [3H]muscimol binding sites in CB1−/− mice. In contrast to cerebral cortex, chronic ethanol ingestion did not alter [3H]muscimol binding in cerebellum in both genotypes, with no main effect of treatment (F3.32 = 0.24, p > 0.05; Fig. 1b, right panel).
Chronic ethanol administration, BELs and tolerance to ethanol-induced hypothermia
No significant difference between genotypes were observed for consumption of ethanol during forced chronic administration (Fig. 2a; F7.206 = 3.66, p = 0.06; genotype × ethanol concentration: F7.206 = 0.40, p = 0.75). Body weights were not significantly altered in both genotypes at the end of the ethanol administration (33.1 ± 0.73 g vs. 33.9 ± 1.11 g and 27.2 ± 1.25 g vs. 29.1 ± 0.89 g in CB1+/+ and CB1−/− mice, respectively). Two-way anova revealed no main effect of ethanol administration (F3.35 = 2.11, p = 0.16).
Chronic ethanol exposure did not induce tolerance to ethanol metabolism after injection of ethanol 3 g/kg in both genotypes (Fig. 2b) when BELs were measured 60 min after ethanol injection [anova: no main effect of ethanol exposure (F3.40 = 3.37, p = 0.08) and of genotype (F3.40 = 0.0002, p = 0.98)]. In contrast, this chronic ethanol exposure paradigm induced a significant tolerance to the ethanol (3 g/kg) -induced hypothermia in both genotypes (Fig. 2c). Two-way anova revaled a main effect of genotype (F3.40 = 5.88, p = 0.01) and of ethanol exposure (F3.40 = 22.11, p < 0.001). Tuckey’s post hoc test also revealed that CB1−/− were more sensitive to the hypothermic effects of ethanol (p < 0.05), both before and after chronic ethanol exposure. We also checked in few animals if our pattern of chronic ethanol exposure can provoke any behavioral sign of ethanol withdrawal. We did not observe any sign of ethanol withdrawal.
The BELs determined in ethanol-exposed animals during the last week of the chronic ethanol exposure period were: 72.0 ± 10.2 mg/dL (range 26.5–136.4 mg/dL).
Sensitivity to the hypothermic effects of diazepam and pentobarbital
As differences in basal levels of [3H]muscimol binding were found between genotypes, mice were tested for their sensitivity to the hypothermic effects of two positive allosteric modulators of GABAA receptors, diazepam and pentobarbital (Fig. 3).
CB1−/− mice were more sensitive to diazepam-induced hypothermia with a more pronounced effect at the 5 mg/kg dose [Fig. 3a, significant interaction between factors time and genotype (F9.99 = 11.12, p < 0.001)]. No genotypic difference was observed at the 10 mg/kg dose [no main effect of genotype (F9.99 = 0.04, p > 0.05 and no significant interaction (F9.99 = 1.39, p > 0.05)]. CB1−/− were also more sensitive to pentobarbital-induced hypothermia with a significant interaction between factors time and genotype (Fig. 3b, F7.79 = 3.48, p < 0.05) at the 25 mg/kg dose and a main effect of genotype at the 50 mg/kg dose (F7.79 = 20.74, p < 0.001).
Expression of GABAA and NR
No genotypic difference was observed for hippocampal α1 (p = 0.49), α2 (p = 0.44), β2 (p = 0.52), and γ2 subunits (p = 0.91) (Fig. 4a). In contrast, both NR were significantly increased by 24% for NR1 (p = 0.006) and by 59% for NR2B (p = 0.035) (Fig. 4a). In striatum, no genotypic difference was observed for α2 subunit (p = 0.504) but a significant increase was observed for α1 (31%, p = 0.003) and NR2B (245%, p = 0.01) in CB1−/− mice compared with wild-type mice (Fig. 4b). In addition, a significant decrease was observed for NR1 (60%, p = 0.01), β2 (35%, p = 0.04), and γ2 (42%, p = 0.045) in CB1−/− mice compared with wild-type mice (Fig. 4b). No genotypic difference was observed for the α6 subunit in the cerebellum (p = 0.76).
Sensitivity to the locomotor effects of MK-801 before and after chronic ethanol ingestion
Male CB1−/− mice showed an increase in basal locomotor activity compared with wild-type mice [(Fig. 5a insert, p = 0.02, Mann–Whitney; as previously described (Ledent et al. 1999; Naassila et al. 2004; Houchi et al. 2005)]. CB1−/− were also less sensitive to the locomotor effects of MK-801 than wild-type mice [main effect of dose and genotype (Fig. 5a), F9.64 = 7.52, p < 0.001 and F9.64 = 12.35, p < 0.001, respectively]. Significant increase of locomotion was observed at the 0.1 mg/kg dose in CB1+/+ mice (Tuckey’s post hoc test, p < 0.05), whereas a significant effect was seen only at the highest dose (0.5 mg/kg) in CB1−/− mice (Tuckey’s post hoc test, p < 0.05).
A genotypic difference was observed for the basal locomotor activity after chronic ethanol ingestion (Fig. 5b, significant interaction between factors treatment and genotype, F3.32 = 10.02, p < 0.005). Two-way anova revealed a main effect of genotype in the locomotor response to MK-801 after chronic ethanol ingestion (F3.32 = 5.87, p < 0.05). Chronic ethanol ingestion significantly increased the basal locomotor activity in CB1+/+ mice (Tuckey’s post hoc test, p < 0.001) but not in CB1−/− mice (Tuckey’s post hoc test, p > 0.05). MK-801 (0.3 mg/kg) increased locomotor activity only in CB1+/+ mice (Fig. 5b, Tuckey’s post hoc test, p < 0.05) but chronic ethanol ingestion did not increase the sensitivity to the locomotor effect of MK-801 (0.3 mg/kg) in CB1+/+ (data not shown). In contrast, CB1−/− mice failed to respond to the locomotor effect of MK-801 after chronic ethanol ingestion (Tuckey’s post hoc test, p > 0.05), as observed before chronic ethanol ingestion.
The overall finding in this study is that mice lacking the CB1 receptors do not display both NMDA and GABAA receptors neuroadaptations occurring after chronic ethanol ingestion. Our results also show that lifelong deletion of CB1 gene-induced compensatory alterations of these two receptors at the pharmacological and behavioral levels.
The chronic ethanol exposure paradigm used in the present study induced tolerance to ethanol-induced hypothermia and this effect does not appear to be secondary to differences in acute clearance of ethanol, because BELs at 1 h after ethanol administration did not differ between groups of animals. Moreover, alcohol consumption produced pharmacologically significant BELs even when measured 12 h after the incorporation of a newly fresh ethanol solution (0700 h, lights on), thus indicating that mice consumed pharmacologically relevant doses of ethanol. This point is very important as rodents tend not drinking to pharmacologically significants levels of ethanol that have been defined as BELs >100 mg/dL in C57BL/6J mice, by Crabbe and colleagues (Rhodes et al. 2005). Rodents seem to limit their voluntary drinking to amounts that can be readily metabolized, possibly in an attempt to avoid reaching intoxicating BELs.
Prolonged exposure to ethanol results in the development of dependence and of tolerance to its behavioral actions that are associated with modifications of both GABAA and NMDA receptors. These modifications have been suggested to play a crucial role in the alcohol withdrawal syndrome (De Witte 2004). In our study, chronic ethanol exposure induced an increase in [3H]MK-801 binding sites in both cerebral cortex and hippocampus, but only in wild-type mice. After chronic ethanol exposure, NMDA receptors appear to increase in number and/or function (Gulya et al. 1991 and see for review Ron 2004). Our results also show that the basal level of locomotor activity was increased only in wild-type after chronic ethanol ingestion reaching then, the level of activity of that of knockout mice. This increased level of locomotor activity is correlated with the observed increased [3H]MK-801 binding sites and decreased [3H]muscimol binding sites. These results agree with the well-established implication of the imbalance between the function of GABAA and NMDA receptors observed during the hyperexcitability of ethanol withdrawal (De Witte 2004). Moreover, after chronic ethanol ingestion, MK-801 (0.3 mg/kg) significantly increased locomotor activity of wild-type mice but not that of knockout mice. This reveals that neither NMDA receptor levels nor sensitivity to the locomotor effects of MK-801 are altered by chronic ethanol exposure in knockout mice. Unexpectingly, the increased number of [3H]MK-801 binding sites in both cerebral cortex and hippocampus of wild-type mice after chronic ethanol exposure was not associated with an increased sensitivity to the locomotor effects of MK-801. Other studies have shown that behavioral differences in response to MK-801 administration in several lines of mice selected for differing sensitivities to actions of ethanol do not display differences in brain [3H]MK-801 binding density or affinity (Nakki et al. 1995; Velardo et al. 1998). These findings suggest that differences in density or affinity of NMDA receptors cannot explain differential responses to MK-801. The neural mechanisms underlying the behaviorally activating effects of MK-801, including increased locomotion and stereotypies, may also involve activation of dopaminergic neurotransmission (Criswell et al. 1993). The blunted locomotor response to MK-801 observed in CB1−/− may thus involve alteration of dopamine receptors. In this regard, we have previously shown that CB1−/− mice display an increase in D2/D3 receptors, as determined by [3H]raclopride binding (Houchi et al. 2005). The present results also show that CB1−/− mice are less sensitive to the locomotor effects of MK-801 that could explain, at least in part, the decreased sensitivity to the hyperlocomotor effect of ethanol that has already been shown in CB1−/− mice (Naassila et al. 2004). Unexpectingly, the present findings show that the decreased number of hippocampal [3H]MK-801 binding sites is associated with an increased level of expression of both NR1 and NR2B subunits. As the presence of specific splice variants of the NR1 subunit has been shown to alter the sensitivity of the receptor to MK-801 (Rodriguez-Paz et al. 1995), this could explain how an increased level of expression of specific NR1 splice variant could alter the level of [3H]MK-801 binding. In the present study, the primers used for measuring the NR1 subunit expression did not distinguish between the eight splice variants of this subunit that is a constitutive component of the NMDA receptor and is uniformly distributed in the brain. In contrast, the level of NR1 subunit mRNA was decreased by 60% in striatum and the mRNA level of NR2B subunit was enhanced by 245% in this same brain area.
The decreased number of [3H]muscimol binding sites in cerebral cortex observed after chronic ingestion in wild-type mice is absent in knockout mice, suggesting that, as observed for NMDA receptors, the neuroadaptations of GABAA receptors following chronic ethanol are blunted in mice lacking CB1 receptors. A decreased number of high-affinity sites for [3H]muscimol has already been demonstrated in rat brain cortex after chronic ethanol treatment (Negro et al. 1995).
We previously demonstrated that ethanol withdrawal severity was increased in CB1 knockout mice compared with wild-type mice (Naassila et al. 2004). This increased withdrawal severity does not seem to be due to an alteration of NMDA and GABAA receptors as showed by the present results, at least in the analyzed brain areas. Several hypotheses can be generated to explain the discrepancy between increased alcohol withdrawal severity and unchanged ligand binding. One possibility is that genotypic difference previously observed for ethanol withdrawal severity is not important enough to be associated with an alteration in [3H]muscimol binding sites number. An alternative explanation might be that the increased severity of withdrawal syndrome may be due to modifications of subunit composition of GABAA receptor, thus altering receptor functioning, without modification of [3H]muscimol binding sites number (Mehta and Ticku 2005).
As we found a decrease in the basal level number of [3H]muscimol binding sites in cerebral cortex and cerebellum, we also tested the sensitivity of mice to the hypothermic effects of GABAA positive allosteric modulators and measured the expression levels of different GABAA receptor subunits. The basal level of expression of five GABAA receptor subunits (α1, α2, α6, β2, and γ2) that have already been found to be important in the behavioral responses to ethanol or benzodiazepines (Mehta and Ticku 2005) were measured in hippocampus, striatum, and cerebellum. The alterations of basal levels of [3H]muscimol binding sites and the increased sensitivity to the GABAA positive allosteric modulators were not associated with alteration of both α2 and α6 subunits expression. In contrast, compensatory increase in α1 and decrease in β2 and γ2 subunit expression were observed in striatum of CB1−/− mice. This increase in α1 subunit expression is consistent with the increased sensitivity to the hypothermic/sedative effects of ethanol previously described in CB1−/− mice (Naassila et al. 2004) and also with the decreased sensitivity to the sedative effects of ethanol observed in α1 knockout mice (Blednov et al. 2003). Interestingly the α2 subunit that has been shown to be involved in anxiolytic action of benzodiazepines is not changed in striatum and hippocampus of CB1−/− mice. It has been previously shown that CB1−/− mice do not display alteration in their sensitivity to the anxiolytic effects of ethanol (Houchi et al. 2005) and that they were insensitive to the anxiolytic effect of benzodiazepine (Urigüen et al. 2004). Expression of both α1 and γ2 subunits is altered in CB1−/−. In addition, these two subunits influence the affinity and efficacy at the benzodiazepine site (Wingrove et al. 1997) and the presence of γ2 confers the classical benzodiazepine pharmacology to GABAA receptors (Pritchett et al. 1989). It is therefore possible to argue that this alteration could at least in part explain the increase sensitivity of CB1−/− to the hypothermic effects of benzodiazepines. In addition, heterozygous γ2 knockout mice have been shown to display enhanced sensitivity to the anxiolytic effects of diazepam (Crestani et al. 1999).
Interestingly, our results demonstrate that the NR2B subunit mRNA level is dramatically increased in both hippocampus and striatum of CB1−/− mice. As this subunit has been reported to be critical in age-dependent plasticity and memory formation (Tang et al. 1999), this increased level of NR2B expression observed in CB1−/− is in agreement with studies showing that CB1 receptor invalidation enhances learning and memory (Reibaud et al. 1999; Varvel and Lichtman 2002). Previous findings have shown that chronic ethanol exposure may specifically enhance functioning of the NR2B-containing receptors in amygdala, a key neural structure involved in emotional and cognitive behaviors, as well as in drug abuse and dependence (Floyd et al. 2003). Thus suggesting that NR2B neuroadaptations may influence the expression of ethanol withdrawal anxiety. The enhanced level of NR2B subunit expression could also be involved in the difference of behavioral responses to ethanol observed between wild-type and CB1−/− mice. In this regard, studies on recombinant NMDARs revealed that ethanol inhibition of NR2A- or NR2B-containing NMDARs coexpressed with NR1 is greater than ethanol inhibition of NR2C- or NR2D-containing NMDARs (Sucher et al. 1996). In addition, NR2B subunit has been shown to be involved in the hypnotic effects of ethanol and in ethanol withdrawal (Malinowska et al. 1999), therefore indicating that increased expression of this NR in CB1−/− may explain the increased sensitivity of CB1−/− to the depressant effects of ethanol.
Benzodiazepine and barbiturate GABAA receptor, facilitating hypnotic drugs as well as ethanol produce similar behavioral effects, have additive or even synergistic interactive neurobehavioral effects, produce cross-tolerance with ethanol, and suppress ethanol withdrawal syndrome (Krystal et al. 2006). Our results show that CB1−/− mice display an increased sensitivity to the hypothermic effects of both compounds diazepam and pentobarbital suggesting that the previously described (Naassila et al. 2004) increased sensitivity to the hypothermic effects of ethanol of CB1−/− mice may involve an increased sensitivity of GABAA receptors to ethanol. This increased sensitivity of CB1−/− mice to benzodiazepine, pentobarbital, and ethanol may be due to alteration of GABAA subunits expression. In this regard, depending on their subunit composition, GABAA receptors display different affinities for various ligands, including barbiturates, benzodiazepines, ethanol, and neurosteroids (reviewed in Johnston 1996). For example, a polymorphism of the γ2 subunit has been associated with genetic susceptibility to ethanol-induced motor incoordination and hypothermia, conditioned taste aversion, and withdrawal (Buck and Hood 1998). In contrast to this increased sensitivity to ethanol-induced hypothermia observed in CB1−/− mice, we have also previously shown that deletion of the CB1 gene did not modify sensitivity to ethanol’s anxiolytic effects, revealing that the ethanol phenotype of CB1−/− mice is not simply due to a global, unidirectional change in acute sensitivity, but is behavior specific (Houchi et al. 2005). Thus, different receptor subtypes may contribute to the selective effects of drugs such as ethanol and benzodiazepines on certain types of behavior.
The mechanism by which absence of CB1 receptors impairs neuroadaptations of both NMDA and GABAA receptors following chronic ethanol exposure is unknown. Recent study in mice indicated that chronic ethanol exposure induces a down-regulation of CB1 receptors and CB1 receptor-stimulated G protein activation in cortex, hippocampus, striatum, and cerebellum (Vinod et al. 2006). This down-regulation may involve desensitization of receptors associated with the increased levels of endocannabinoids that result from endocannabinoids transporter inhibition by chronic ethanol exposure (Basavarajappa et al. 2003). A significant role of the endocannabinoid system has been suggested in modulating dopaminergic transmission in mesocorticolimbic neurons via complex pathways involving GABAergic and glutamatergic neurotransmissions (see for review Gardner 2005). Indeed, CB1 receptors are found on GABAergic and glutamatergic axons terminals and could contribute to alter the balance of excitatory/inhibitory inputs to dopaminergic neurons from the ventral tegmental area and thus modulate dopaminergic tone of the reward system (Maldonado et al. 2006). Our results show that basal levels of [3H]muscimol and [3H]MK-801 binding sites are decreased in CB1−/− mice. As endocannabinoids act pre-synaptically to inhibit transmitter release, the absence of CB1 receptors in knockout mice may promote a developmental compensatory decrease in NMDA and GABAA receptors to counteract the overstimulation of post-synaptic receptor due to lack of retrocontrol of synaptic activity by CB1 receptors.
The behavioral responses to ethanol of CB1−/− mice may reflect not only the mutation but also consequent compensatory/adaptive processes and/or developmental defects. It has been shown for example that deletion of the CB1 receptor resulted in pronounced alterations in the HPA axis and the opioidergic system in the caudate putamen (Urigüen et al. 2005). Because both NMDA and GABAA receptors have been strongly implicated in both alcohol self-administration (Hodge et al. 1995) and in amygdala-dependent anxiety behaviors, adaptations of amygdala NMDA and GABAA receptors to chronic ethanol may be key for the behavioral consequences of long-term alcohol self-administration. The amygdala is intimately associated with affective behaviors that contribute to the abuse of drugs, including ethanol. As it is well established that experimental manipulation of these receptors has profound behavioral consequences, from altering affective states to regulating drug discrimination (Hodge and Cox 1998), it would be interesting to investigate potential alterations of both receptors in this limbic forebrain region after lifelong deletion of CB1 receptor.
Collectively our results and above mentioned studies, suggest that deletion of CB1 receptors and/or pronounced neuroadaptations of numerous systems (glutamate, GABA, opioid, and HPA axis) could participate to the behavioral responses to ethanol and alter the neuroadaptations occurring after chronic ethanol exposure. These data also suggest that functional alterations in CB1 receptors may affect the efficacy of anxiolytic, anticonvulsivant, and anticraving drugs in the treatment of psychiatric disorders such as anxiety, depression, schizophrenia, Parkinson’s disease, Alzheimer’s disease, and addiction.
In summary, the present findings show the involvement of the endocannabinoids/CB1 receptor system in the alcohol-induced neuroadaptive changes of both glutamatergic and GABAergic systems in the brain and indicate that compensations have taken place in the CB1 knockout mice. Present data and also previous work reveal some neuroadaptations induced by the CB1 gene knockout such as increased striatal dopamine D2 receptors, decreased hippocampal and cerebellar levels of [3H]muscimol binding sites, and decreased hippocampal levels of [3H]MK-801 binding sites. These consequent compensatory/adaptive processes and/or developmental defects may contribute to adult phenotype and may influence complex behaviors as the behavioral responses to ethanol that involve a large number of neurotransmitter systems.
VW is supported by a doctoral Fellowship from INSERM-Mildt, EB is supported by a doctoral Fellowship from the Ministry of National Education and Research/Technology (MENRT), and HH is supported by a doctoral Fellowship from Conseil régional de Picardie. We thank Jenny Molet, Lucille Foyard, Brian Lavaury, and Christopher Iger for excellent technical assistance. The results of the present study have been presented at the 2006 ISBRA meeting.