Cannabidiol regulates behavioral and brain alterations induced by spontaneous alcohol withdrawal

The main goal of this study was to evaluate if the administration of cannabidiol (CBD) regulates behavioral and gene expression alterations induced by spontaneous alcohol withdrawal (SAW) in mice. Increasing doses of ethanol were administered to C57BL/6J male mice for 15 days (2.5, 3 and 3.5 g/kg/12 h, p. o.), and SAW was studied at 6, 12, 24, and 72 h after the last ethanol administration. The efficacy of acute CBD (10, 20, and 40 mg/ kg, i. p.) to regulate behavioral changes induced by SAW was explored at 6 h. Gene expression analyses of cannabinoid receptors 1 ( Cnr1 ) and 2 ( Cnr2 ), mu-opioid receptor ( Opmr1 ), and proopiomelanocortin ( Pomc ) in the nucleus accumbens (NAcc), and Pomc and tyrosine hydroxylase ( Th ) in the ventral tegmental area (VTA), were carried out by real time-PCR. Pearson correlation was used to identify potential associations between the gene expression data and the anxiety-like behaviors. Biostatistical studies suggest associations between gene expression data and the anxiogenic behaviors in mice exposed to the SAW model and treated with VEH and 40 mg/kg of CBD. Mice exposed to the SAW model showed significant somatic withdrawal signs, anxiety-like behaviors, and remarkable changes in the gene expression of all brain targets at 6 h. CBD dose-dependently normalized the behavioral, somatic withdrawal signs and anxiety- like behaviors and modulated gene expression changes in the NAcc, but not in the VTA. The results of this study suggest that CBD may regulate specific alcohol withdrawal-associated alterations. However, further studies are required to explore the possible mechanisms involved.


Introduction
Alcohol use disorders (AUD) are among the most critical health problems worldwide, with complex therapeutic management and devastating personal, familial, social, and economic impact (Cohen et al., 2022). Patients develop alcohol withdrawal syndrome (AWS) after the cessation or reduction of chronic alcohol consumption, presenting hyperexcitability, anxiety, insomnia, and seizures in more than 50% of cases (APA, 2013). Pharmacological treatment of AWS is nonspecific and associated with poor response, deterioration of the patient's recovery, and higher early relapse rates (around 60%) (Onaivi et al., 2008;Serrano et al., 2012;Sanchez-Marin et al., 2017;Vinod et al., 2010). This early relapse is associated with worsened treatment response and chronic psychosocial consequences (Sanchez-Marin et al., 2017;Neumeister et al., 2012;Hirvonen et al., 2013), while long-lasting abstinence is associated with better patient recovery (Neumeister et al., 2012;Hirvonen et al., 2013;Ceccarini et al., 2014;Agudelo et al., 2013). Therefore, developing specific and pharmacological agents modulating the AWS-associated alterations has become one of the most critical objectives in treating patients with AUD.
The implementation of animal models that simulate, at least in part, the main characteristics of AWS could be an essential tool to increase the understanding of the neurobiological mechanisms involved in this clinical situation and to identify new therapeutic options. Several animal models of alcohol withdrawal have been reported by administering ethanol orally (gavage or voluntary consumption) or intraperitoneally for several days or months and evaluating ethanol withdrawal-induced alterations at different time points (Perez and De Biasi, 2015;Silva and Madeira, 2012;Muschen et al., 2019). The behavioral disturbances in these models are heterogenous and do not simulate all the complex clinical symptoms in patients. Limited pharmacological studies have been conducted with animal models of spontaneous alcohol withdrawal (SAW).
Cannabidiol (CBD) is one of the main compounds of the Cannabis sativa plant that shows anxiolytic, antidepressant, and antipsychotic actions in animal and clinical studies (Elsaid et al., 2019). In the last few years, our group has studied the role of CBD in the different components of alcohol dependence. Our laboratory demonstrated that CBD lacks activity as a drug of abuse in different behavioral paradigms (conditioned place preference, self-administration, and spontaneous withdrawal) (Viudez-Martinez et al., 2018a). The therapeutic potential of CBD in alcohol dependence has been studied in different aspects of addiction, such as reinforcement, motivation, reinstatement, and binge drinking. Indeed, CBD reduced voluntary ethanol consumption and the motivation and relapse in the oral alcohol self-administration paradigm (Viudez-Martinez et al., 2018a;Gonzalez-Cuevas et al., 2018). More recently, our group reported that early administration of CBD also regulates behavioral and brain alterations induced by ethanol exposure during gestation and lactation (Gasparyan et al., 2023). These CBD effects were associated with gene expression changes in brain targets of the endocannabinoid, opioid, and dopaminergic systems. However, the potential role of CBD in regulating behavioral and brain alterations induced by an alcohol withdrawal paradigm is still unknown.
Therefore, the main goals of this study were to evaluate: 1) the time course of the presentation of somatic withdrawal signs (rearings, rubbings, and groomings) and anxiogenic behaviors in mice exposed to the SAW model, 2) the effects of CBD on several withdrawal signs and anxiety-like behaviors at the time point of their greatest presentation, and 2) changes in the relative gene expression of cannabinoid receptors 1 (Cnr1) and 2 (Cnr2), mu-opioid receptor (Opmr1) and proopiomelanocortin (Pomc) in the nucleus accumbens (NAcc), and tyrosine hydroxylase (Th) and Pomc in the ventral tegmental area (VTA) in mice exposed to the SAW model and their modulation with CBD. Correlation studies were conducted to find associations between the gene expression data and anxiety-like behaviors.

Mice
Male C57BL/6J mice were purchased from Charles River laboratories (Lille, France) and were 5 weeks old at the time of their arrival. Mice were housed in groups of 5 per cage, in the Service of Animal Facilities of the Universidad Miguel Hernandez de Elche, under controlled environmental conditions of temperature (21 ± 2 • C), relative humidity (60 ± 10%), and light-dark cycle (lights from 08:00 to 20:00h). Food and water were available at libitum. After one week of acclimatization, the experimental procedures began. One hundred and thirty-six C57BL/6J male mice were used to carry out all the experiments. The Ethics Committee of Universidad Miguel Hernandez approved the study.

Drugs
Cannabidiol (CBD) was acquired from THC Pharm (Frankfurt, Germany). It was dissolved in ethanol, cremophor, and saline at a 1:1:18 proportion to obtain the required doses of 10, 20, and 40 mg/kg of weight (Navarrete et al., 2022;Gasparyan et al., 2021a). The combination of ethanol, cremophor, and saline in a ratio of 1:1:18 is commonly utilized for dissolving CBD powder. Moreover, given that the total administration volume for mice is 0.3 ml, the amount of ethanol (0.015 ml) does not affect the response in the model. The drug was prepared for each experiment immediately before its intraperitoneal (i.p.) administration.

2.3.1 Light-dark box (LDB)
It is a widely used test to evaluate anxiety-like behaviors in rodents, described for the first time by Crawley and Goodwin (Crawley and Goodwin, 1980;Gasparyan et al., 2021b). LDB test was carried out using a methacrylate apparatus with two compartments (20x20x15 cm). One is transparent, and the other is dark and opaque, separated by an opaque tunnel. In addition, the light compartment is illuminated with a lamp (60 W) placed 25 cm above it. Mice were individually placed in the lighted compartment facing the tunnel. The time spent in this compartment and the number of transitions between both compartments were recorded for 5 min.

2.3.2 Motor activity and somatic withdrawal signs
Mice were placed into individual methacrylate boxes (25x25x25 cm) and videotaped for 15 min to evaluate somatic withdrawal signs (number of rearings, rubbings, and groomings) and the motor response (traveled distance), as described previously (Gasparyan et al., 2021b).

Experimental design
SAW model was induced at different time points after cessation of increasing ethanol doses of 2.5, 3, and 3.5 g/kg every 5 days, corresponding to a range of 20-29% v/v ethanol solution. Ethanol solutions were prepared daily immediately after administration via oral gavage, twice a day (every 12 h), for 15 days. For the control group (CT), mice were treated with water twice daily for 15 days.

Experiment 1. Time course of the behavioral alterations induced by SAW
The SAW-induced somatic withdrawal signs and anxiety-like behaviors were evaluated 6, 12, 24, and 72 h after the last ethanol administration ( Fig. 1) (N = 7 rodents/group). This time course allowed identifying the time point of the most significant behavioral alterations to perform the subsequent pharmacological study.

Experiment 2. Evaluation of CBD effects on SAW model-induced alterations
The efficacy of CBD (10, 20, and 40 mg/kg, i. p.) in modulating the SAW model-associated alterations was evaluated at the peak of their most significant presentation (detected in the previous time-course evaluation). Considering the pharmacokinetics of CBD, the drug was administered 1 h and a half before the behavioral evaluation (Deiana et al., 2012) (Fig. 3). Two sets of mice were used to analyze CBD's effects on anxiety-like behaviors (LDB) and somatic withdrawal signs and motor activity (OF). Mice exposed to the LDB were used to conduct the gene expression studies, as described in the next section. All the behavioral analyses were conducted under blinded conditions (N = 9-10 rodents/group).

Gene expression analyses
Relative gene expression analyses of Cnr1 and Cnr2, Opmr1 and Pomc in the Nacc, and Th and Pomc in the VTA were analyzed using the brain samples obtained from mice exposed to the LDB test. Briefly, mice were killed by cervical dislocation 150 min after administering CBD (or vehicle) and brains were removed from the skull and stored at − 80 • C.
Coronal sections of 500 μm were obtained containing the brain regions by slicing the samples in the cryostat (− 10 • C) according to the Paxinos and Franklin's atlas (Paxinos and Franklin, 2004). Brain nuclei were micro-dissected following the method described by Palkovits and previously modified by our group (Navarrete et al., 2012). Total RNA was extracted using TRI Reagent and was retrotranscribed to complementary DNA (cDNA) (High-Capacity cDNA Reverse Transcription Kit with RNase Inhibitor, Applied Biosystems, Madrid). Quantitative analyses of and Th (Mm00447546_m1) were performed on the StepOne Sequence Detector System. The ribosomal 18S rRNA was used as the reference gene, and the fold change in target gene expression was calculated using the e 2 ΔΔ− Ct method (Livak and Schmittgen, 2001).

Statistical analyses
Statistical analyses were carried out using one-way ANOVA followed by Student-Newman-Keul's post hoc test when comparing different experimental groups. Differences were considered significant if the type-I error or alpha probability was less than 5%. SigmaPlot 11 software (Systat Software Inc., Chicago, IL, USA) was used for all statistical analyses. Pearson correlation was used to explore the correlations between gene expression data and the time in the lighted box of the LDB test using self-coding in the phyton.

Somatic withdrawal signs
Mice exposed to the animal model of SAW showed an increased number of rearings at 6, 12, and 24 h after the last ethanol administration ( Fig. 2A, One-way ANOVA, F (45,4) = 7.902, P < 0.001), achieving values similar to the control group at the time point of 72 h. Interestingly, the exposure to the SAW model increased rubbing (Fig. 2B, Oneway ANOVA, F (4,45) = 3.501, P = 0.015) and reduced grooming (Fig. 2C, One-way ANOVA, F (4,45) = 7.125, P < 0.001) numbers in all the evaluation points. However, at 24 and 72 h, the number of groomings tended to increase without achieving complete normalization.

Anxiogenic behavior
The light-dark box test revealed a reduced time spent in the lighted box at 6 h in mice exposed to the SAW model (Fig. 2D, Student's t-test, t = 3.652, P = 0.002, 18 d. f.). However, no changes were found at the time points of 12 h (Student's t-test, t = − 0.110, P = 0.914, 18 d. f.), 24 h (Student's t-test, t = − 0.920, P = 0.370, 18 d. f.) or 72 h (Student's t-test, t = '0.328, P = 0.747, 18 d. f.). No differences were found in the number of transitions at all evaluation times (data not shown).

Modulation of somatic withdrawal signs and motor activity
In mice exposed to the SAW model, CBD normalized the increase in the number of rearings (Fig. 4A, One-way ANOVA, F (4,39) = 5.624, P < 0.001) and tends to normalize the number of groomings (Fig. 4C, Oneway ANOVA, F (4,39) = 12.715, P < 0.001) at the highest dose of 40 mg/kg. Interestingly, only the low and intermediate doses of CBD modulated the increase in the number of rubbings (Fig. 4B, One-way ANOVA, F (4,39) = 9.573, P < 0.001). No statistical differences in total distance traveled were found in mice exposed to the SAW model and treated with vehicle or CBD.

Anxiolytic actions of CBD
Mice exposed to the animal model of SAW showed reduced time in the lighted box in the LDB test, and CBD administration normalized this anxiogenic behavior at the highest dose of 40 mg/kg (Fig. 4E, One-way ANOVA, F (4,48) = 7.951, P < 0.001). No differences were found in the number of transitions (Fig. 4F, One-way ANOVA, F (4,48) = 1.291, P = 0.288).

Fig. 1.
A schematic presentation of the timeline diagram followed to evaluate the time course of the presentation of the behavioral alterations induced by the animal model of spontaneous alcohol withdrawal (SAW). Mice were treated with increasing doses of ethanol for 15 days (2.5, 3 and 3.5 g/kg, p. o., every 5 days) or water (for the control group), evaluating somatic withdrawal signs and anxiety-like behaviors using the OF and LDB tests, respectively, at different time points (6, 12, 24 and 72 h) after the last ethanol administration. EtOH, ethanol; p. o., oral administration; OF, open field; LDB, light-dark box.

CBD effects on gene expression studies in the NAcc
Cannabinoid targets. SAW model exposure increased the gene expression of Cnr2. Cnr1 gene expression was reduced in the NAcc of the SAW model exposed and VEH-treated mice. CBD administration normalizes this reduction at all three doses (Fig. 5A, One-way ANOVA, F (4,47) = 5.816, P < 0.001). Interestingly, the three doses of CBD induce an additional up-regulation of the gene expression of Cnr2 (Fig. 5B, F (4,47) = 24.776, P < 0.001).
Opioid targets. Mice exposed to the SAW model showed an increase in the gene expression of Opmr1, which was not modulated with CBD administration (Fig. 5C, One-way ANOVA, F (4,47) = 9.251, P < 0.001). Pomc gene expression was significantly reduced in mice exposed to the SAW model and treated with VEH. Interestingly, CBD administration modulates this reduction and tends to increase it more at the highest dose of 40 mg/kg (Fig. 5D, One-way ANOVA, F (4,47) = 18.004, P < 0.001).

Correlation analyses
The analyses were conducted considering the group of SAW modelexposed and VEH-treated mice and those mice exposed to the SAW model and treated with the most anxiolytic dose of CBD in the LDB test (40 mg/kg, CBD40) (Fig. 6). Correlation studies were performed to find associations between the gene expression results and the time in the lighted box of the LDB test. In the SAW-VEH group, a positive association was found between Cnr1 and Opmr1 gene expressions (r = 0.68, P < 0.05). Interestingly, in the SAW-CBD40 group, the time spent in the lighted box of the LDB test was negatively associated with the gene expression of Cnr1 in the NAcc (r = − 0.75, P < 0.05) and positively related to the Pomc gene expression in the VTA (r = 0.74, P < 0.05). In the same group of mice, negative associations were found between Cnr2 and Opmr1 (r = − 0.67, P < 0.05) and between Cnr2 and Th (r = − 0.68, P < 0.05) gene expressions.

Discussion
The results of this study revealed CBD's efficacy in regulating anxiety-like behaviors and somatic withdrawal signs associated with the Fig. 3. Schematic presentation of the timeline diagram followed to evaluate the effects of different doses of CBD (10, 20 and 40 mg/kg, i. p.) on SAW model-induced somatic withdrawal signs and anxietylike behaviors 6 h after the last ethanol administration, being the time point of the most significant behavioral alterations according to the previous experiment. One hour after anxiety-like behavioral evaluation, mice were killed by cervical dislocation, and brain samples were obtained to carry out the gene expression studies of different brain targets. SAW, spontaneous alcohol withdrawal; CT, control group; CBD, cannabidiol; OF, open field; LDB, lightdark box; i. p., intraperitoneal administration. Fig. 4. Evaluation of CBD effects (10, 20 and 40 mg/kg, i. p.) on the somatic withdrawal signs, anxiety-like behaviors and motor activity 6 h after the last ethanol administration in mice exposed to the animal model of SAW. The number of rearings (A), rubbings (B), groomings (C) and total distance traveled (D) were evaluated using the OF test, and the time in the lighted box (E) and the number of transitions (F) using the LDB test. CT, control; SAW, spontaneous alcohol withdrawal; CT, control group; i. p., intraperitoneal administration. * vs. CT, # vs. SAW-VEH and/or SAW -CBD treated mice. SAW model in C57BL/6J male mice. The following observations support this assumption: (1) Mice exposed to the SAW model showed anxiogenic behaviors, increased rearings and rubbings, and reduced groomings 6 h after the last ethanol administration, (2) CBD administration induced anxiolytic actions at the highest dose of 40 mg/kg (i.p.) and modulated the somatic withdrawal signs (rearings, rubbings, and groomings), (3) CBD normalized the reduction found in the Cnr1 gene expression, modulated the decrease of Pomc and induced an additional upregulation of Cnr2 gene expression in the NAcc and (4) CBD failed to modulate Opmr1 gene expression in the NAcc, and Th and Pomc gene expression changes in the VTA.
Implementing animal models simulating some of the main features of alcohol withdrawal is critical to identify new targets for its therapeutic management. In this study, an animal model of SAW was carried out by ethanol gavage administration twice a day for 15 days and increasing the dose of ethanol every 5 days to avoid tolerance (Days 1-5, 2.5 g/kg/ 12h; Days 6-10, 3 g/kg/12h; Days 11-15, 3.5 g/kg/12h; p. o.). This type of administration accurately controls the amount of ethanol administered to induce homogenous behavioral alterations, which is complicated to achieve with vaporized ethanol or liquid dietary alcohol exposure protocols (Alberto et al., 2023;Fathi et al., 2021;Williams et al., 2012). To our knowledge, this is the first study evaluating the time course of the SAW model-induced behavioral alterations, including somatic withdrawal signs and anxiogenic-like behaviors, allowing the identification of the most significant behavioral alterations. In the first part of the study, anxiety and somatic withdrawal signs were evaluated at 6, 12, 24, and 72 h after the last ethanol administration. Rearings significantly increased in mice exposed to the SAW model at 6, 12, and 24 h, and no differences were found at 72 h after the last ethanol administration. However, the number of rubbings was higher in mice exposed to the SAW model at all time points. In contrast, the number of groomings was significantly reduced in mice exposed to the SAW model. Both withdrawal signs, rubbings and groomings, are related to self-care. However, rubbings tend to be more pronounced under anxious   6. Pearson correlation studies between gene expression data and the time spent in the lighted box of the LDB test in mice exposed to the SAW model and treated with VEH (SAW-VEH) (A) and those exposed to the SAW model and treated with 40 mg/kg of CBD, i. p. (SAW-CBD40). Statistically significant associations were highlighted, with P < 0.05. conditions and groomings under basal non-stressed conditions. Only a few studies have evaluated these somatic signs associated with alcohol withdrawal, finding increased withdrawal scores a few hours after the last ethanol exposure. However, the exact alterations of all these somatic signs and the time course of their presentation have yet to be further described (Perez and De Biasi, 2015). Another important aspect of alcohol withdrawal is anxiety. This behavior was described in some previous studies 10-24 h after the last ethanol exposure (Perez and De Biasi, 2015;Junqueira-Ayres et al., 2017). In our study, mice exposed to the SAW model showed anxiogenic-like behaviors only 6 h after the last ethanol administration in the LDB test but not at the following evaluation time points. These differences in this behavioral alteration could be related to the type and duration of ethanol exposure, which varies between studies (Perez and De Biasi, 2015;Muschen et al., 2019;Fathi et al., 2021;Williams et al., 2012). Therefore, the time point of 6 h presenting significant somatic withdrawal signs and anxiogenic behaviors was the time selected to conduct the studies with CBD.
The second part of this study evaluated if CBD administration could modulate the behavioral and gene expression alterations associated with the SAW model. The results showed that CBD modulates the somatic withdrawal signs at different doses, consistent with previous studies with other drugs of abuse (Gasparyan et al., 2021a). Interestingly, CBD induced an anxiolytic action only at the highest dose of 40 mg/kg. This is the first study where the CBD administration modulated the somatic withdrawal signs in mice exposed to a SAW model.
Relative gene expression studies were carried out in two brain regions of the reward circuit, NAcc and VTA, to explore the brain targets related to the potential therapeutic actions of CBD. Mice exposed to the SAW model showed reduced Cnr1 and increased Cnr2 gene expressions. Cannabinoid receptor 1 (CB1r) involvement in alcohol withdrawal has been described previously. In addition, the blockade of this receptor reduced anxiety-like behaviors in a rat model of alcohol withdrawal (Rubio et al., 2008). Some authors reported reduced CB1r protein levels in the hippocampus during early withdrawal (Mitrirattanakul et al., 2007). However, in another study, no changes were observed in the optical density of CB1r in the caudate putamen (Rubio et al., 2008).
Recent advances suggest that CBD can act as a negative allosteric modulator of CB1r or an indirect agonist by inhibiting anandamide reuptake (Bisogno et al., 2001;Tham et al., 2019). Interestingly, mice exposed to the SAW model reduced gene expression of Cnr1 in the Nacc, which was normalized in all three doses with CBD administration. Therefore, considering the available information, the modulation of Cnr1 gene expression may be responsible, at least in part, for the behavioral improvement induced by CBD administration. The exposure to the SAW model also increased Cnr2 gene expression in the NAcc. This increase was found in previous withdrawal studies from other drugs of abuse and could be related to functional damage induced by drug exposure (Navarrete et al., 2022;Gasparyan et al., 2021a). Interestingly, CBD induced an additional upregulation of Cnr2 gene expression, consistent with its activity as an antagonist of the CB2r (Thomas et al., 2007).
The interaction between CBD and the opioid system has been scarcely explored. In vitro studies suggest CBD could act as an allosteric modulator of mu-opioid receptors (MOR) (Kathmann et al., 2006). Mice exposed to the SAW model showed an increased gene expression of Opmr1 in the NAcc, which was not altered by CBD administration. The Pomc gene encodes the precursor of the endogenous opioid peptide beta-endorphin. The expression of this gene in Nacc and VTA is closely involved in the reward circuit (Grauerholz et al., 1998;Andino et al., 2011).
The interaction between CBD and Pomc has been poorly explored. Interestingly, mice exposed to the SAW model significantly reduced Pomc gene expression in NAcc and VTA. CBD administration modulated this reduction in the NAcc but not in the VTA. In a model of stress, CBD administration reduced the gene expression of this target in the arcuate nucleus in mice exposed to acute stress (Viudez-Martinez et al., 2018b).
In addition, in a recent study published by our group, CBD modulated Pomc gene expression in mice exposed to an animal model of heroin withdrawal (Navarrete et al., 2022). Therefore, the ability of CBD to modulate Pomc gene expression in the NAcc may be related, at least in part, to its anxiolytic action and modulation of somatic withdrawal signs.
The dopaminergic system is closely related to the reward circuit. Tyrosine hydroxylase (TH) is the rate-limiting enzyme of dopamine synthesis in the VTA's body cells of the dopaminergic neurons. To date, there is no information about changes in the gene expression of this target during alcohol withdrawal. Interestingly, exposure to this SAW model reduced Th gene expression, but CBD administration failed to modulate this change. These results are inconsistent with those reported previously regarding CBD's induced modulation of Th gene expression in animal models of heroin or cocaine withdrawal (Navarrete et al., 2022;Gasparyan et al., 2021a). These discrepancies could be related to the dose of CBD used during the evaluation and the type of drug of abuse used in each study (heroin, cocaine, or alcohol). Further studies are needed to explore CBD's actions on this target in mice exposed to this animal model of SAW.
Correlation studies were carried out using the Pearson correlation to explore further the potential associations between the gene expression results and anxiety-like behaviors. All of these results were from the same animals. The group of mice exposed to the SAW model and treated with vehicle (SAW-VEH) and the group exposed to the model and treated with 40 mg/kg of CBD (SAW-CBD40) were selected to evaluate the potential associations induced by the exposure to the SAW model and to analyze possible associations in the group treated with the most effective dose of CBD in the LDB test.
The interaction between cannabinoid and opioid systems has been described previously in other experimental conditions (Manzanares et al., 1999(Manzanares et al., , 2005Tanda et al., 1997;Corchero et al., 1999;Milligan et al., 2020;Welch, 2009). Mice exposed to the SAW model and treated with VEH showed a positive association between Cnr1 and Opmr1 gene expressions. Indeed, synergistic action between CB1r and MOR receptors was reported in pain control and negative allosteric modulation of CB1r suppressing opioid-mediated reward (Navarro et al., 2001;Yu et al., 2022;Iyer et al., 2022). Therefore, it is possible to hypothesize a positive association between Cnr1 and Opmr, closely involved, at least in part, in the behavioral effects found in mice exposed to the SAW model. In the SAW-CBD40, a negative association was found between the time spent in the lighted box and the gene expression of Cnr1. This result is consistent with previous reports involving the CB1r in anxiety-like behaviors (Lisboa et al., 2015). In addition, the pretreatment with the CB1r receptor antagonist rimonabant blocked the anxiolytic actions induced by CBD (Austrich-Olivares et al., 2022). Therefore, these results suggest that the behavioral modulation induced by CBD could be related to its ability to interact with the CB1r. In the same group, the time spent in the lighted box is positively associated with the Pomc gene expression in the VTA. There is little information about CBD's effects on Pomc gene expression in this brain region. However, CBD could modulate Pomc gene expression through its interaction with CB1r. Further studies are required to explore this potential interaction and CBD's induced modulation of Pomc gene expression in the NAcc.
The correlation studies revealed a negative association between the Cnr2 and Opmr1 and Cnr2 and Th gene expressions. The interaction between CB2r and MOR receptors has been described previously in animal models of alcohol addiction (Manzanares et al., 2005;Ortega-Alvaro et al., 2015;Navarrete et al., 2018) and other experimental conditions (Grenald et al., 2017). In addition, the available literature supports the synergistic action between CB2r and MOR (Grenald et al., 2017;Paldy et al., 2008). Therefore, considering CBD's mechanism of action, the negative association between Cnr2 and Opmr1 gene expression could be related to CBD's ability to block CB2r. Only a few studies evaluated the interaction between CB2r and TH targets (Ortega-Alvaro et al., 2015; Navarrete et al., 2018). In a recent study carried out with the patch-clamp technique, a reduction of VTA dopaminergic neurons' excitability through the activation of cannabinoid CB2r was observed, indicating a possible direct interaction between this cannabinoid receptor and dopamine release (Ma et al., 2019). Therefore, CBD could probably modulate these dopaminergic neurons' activity, perhaps through its ability to block CB2r cannabinoid receptors (Thomas et al., 2007).
In summary, the results obtained in this study support the utility of CBD in modulating the somatic withdrawal signs and anxiety-like behaviors associated with the animal model of SAW. The administration of CBD modulated Cnr1 and Pomc gene expression changes in mice exposed to the SAW model. Correlation studies suggest CBD could regulate opioid and dopaminergic systems by interacting with the cannabinoid receptors CB1r and CB2r. Further studies are required to explore these interactions and their potential significance in alcohol withdrawal. One of the limitations of the present study is the use of only males to minimize the impact of hormonal influences on the observed behaviors in mice exposed to the model and treated with CBD. Nonetheless, examining the model-induced alterations in females represents an intriguing avenue for future research.

Funding
This work was supported by the Institute de Salud Carlos III PI21/ 00488 and RD21/0009/0008 to JM, and "Fundación Bienvenida Navarro-Luciano Tripodi" to AG.

Declaration of competing interest
None.

Data availability
Data will be made available on request.