TYPE Methods PUBLISHED 27 March 2023 DOI 10.3389/fphar.2023.1143365 A male mouse model of WIN OPEN ACCESS 55,212–2 self-administration to EDITED BY Paola Brusa, University of Turin, Italy study cannabinoid addiction REVIEWED BY Marsida Kallupi, María del Mar Cajiao-Manrique 1, Rafael Maldonado 1,2*† and University of California, San Diego, United States Elena Martín-García 1,2,3*† Benjamin B Land, 1 Laboratory of Neuropharmacology-Neurophar, Department of Medicine and Life Sciences, Universitat University of Washington, United States Pompeu Fabra (UPF), Barcelona, Spain, 2Hospital del Mar Medical Research Institute (IMIM), Barcelona, *CORRESPONDENCE Spain, 3Departament de Psicobiologia i Metodologia de les Ciències de la Salut, Universitat Autònoma de Rafael Maldonado, Barcelona (UAB), Barcelona, Spain rafael.maldonado@upf.edu Elena Martín-García, elena.martin@upf.edu † These authors have contributed equally We have established for the first time a mouse model of cannabinoid addiction to this work using WIN 55,212–2 intravenous self-administration (0.0125 mg/kg/infusion) in SPECIALTY SECTION C57Bl/6J mice. This model allows to evaluate the addiction criteria by grouping This article was submitted to Experimental Pharmacology and Drug them into 1) persistence of response during a period of non-availability of the drug, Discovery, 2) motivation for WIN 55,212–2 with a progressive ratio, and 3) compulsivity when a section of the journal the reward is associated with a punishment such as an electric foot-shock, in Frontiers in Pharmacology agreement with the Diagnostic and Statistical Manual of Mental Disorders 5th RECEIVED 12 January 2023 edition (DSM-5). This model also allows to measure two parameters that have ACCEPTED 14 March 2023 PUBLISHED 27 March 2023 been related with the DSM-5 diagnostic criteria of craving, resistance to extinction and reinstatement, and two phenotypic traits suggested as predisposing factors, CITATION Cajiao-Manrique MdM, Maldonado R and impulsivity and sensitivity to reward. We found that 35.6% of mice developed the Martín-García E (2023), A male mouse criteria of cannabinoid addiction, allowing to differentiate between resilient and model of WIN 55,212–2 self- vulnerable mice. Therefore, we have established a novel and reliable model to administration to study cannabinoid addiction. study the neurobiological correlates underlying the resilience or vulnerability to Front. Pharmacol. 14:1143365. develop cannabinoid addiction. This model included the chemogenetic inhibition doi: 10.3389/fphar.2023.1143365 of neuronal activity in the medial prefrontal cortex to the nucleus accumbens COPYRIGHT pathway to assess the neurobiological substrate of cannabinoid addiction. This © 2023 Cajiao-Manrique, Maldonado and model will shed light on the neurobiological substrate underlying cannabinoid Martín-García. This is an open-access article distributed under the terms of the addiction. Creative Commons Attribution License (CC BY). The use, distribution or KEYWORDS reproduction in other forums is permitted, provided the original author(s) cannabinoid addiction, mouse model, WIN 55,212-2 self-administration, persistence of and the copyright owner(s) are credited response, motivation, compulsive-like behavior and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted 1 Introduction which does not comply with these terms. Cannabis sativa derivatives are the most used illicit drugs worldwide, with an increased consumption over the recent years. In Europe, cannabis use has enhanced from 5.7% in 2015 to 7.7% in 2022 in adults aged from 15 to 64 (European Monitoring Centre for Drugs and Drug Addiction, 2015; European Monitoring Centre for Drugs and Drug Addiction, 2022). However, attitudes toward cannabis use have softened since there is a growing social perception that cannabis is relatively harmless (Weiss and Volkow, 2022). This lack of risk perception has led to an increase in the prevalence of cannabis use disorder, previously defined as cannabis dependence (Zehra et al., 2018). Cannabis use disorder is defined as a chronically relapsing neuropsychiatric disorder diagnosed by applying the criteria defined in the 5th edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5). In this edition, the term addiction is synonymous with Frontiers in Pharmacology 01 frontiersin.org Cajiao-Manrique et al. 10.3389/fphar.2023.1143365 severe substance-use disorder and requires the accomplishment of six chemogenetic manipulations to decipher the neurobiological out of eleven diagnostic criteria (Koob and Volkow, 2016). mechanisms involved in cannabinoid addiction. For this purpose, Cannabis addiction results from the interaction between gene we chemogenetically silenced the prelimbic (PL) -nucleus networks, epigenetics, and socio-environmental factors (Piazza and accumbens (NAc) pathway in our mouse model through a Deroche-Gamonet, 2013; Hamilton and Nestler, 2019; Lüscher et al., Designer Receptors Exclusively Activated by Designer Drugs 2020; Maldonado et al., 2021). Not all individuals repeatedly exposed to (DREADD) approach, a pathway closely involved in the the drug make this transition to addiction (Piazza and Deroche- development of addictive behavior (Domingo-Rodríguez et al., Gamonet, 2013), raising the question of why some vulnerable 2004; Compton et al., 2022). individuals become addicted while others are resilient. Although multiple neuroadaptations induced by cannabinoids administration have been described, the precise neurobiological mechanisms 2 Materials and equipment underlying cannabinoid addiction in vulnerable individuals remain largely unknown. At the present moment, animal models of 2.1 Animals cannabinoid addiction have not been yet developed, and there is an urgent need of an accurate animal model to disentangle the Eight weeks old male C57BL/6J mice (n = 30) (Charles River, neurobiological correlates of addiction to cannabinoids. France) were housed individually with food and water available ad Animal models of drug exposure allow to investigate brain long- libitum in controlled laboratory conditions (21oC ± 1°C, 55% ± lasting changes produced by drugs of abuse. Non-contingent drug 10%). Mice were tested during the first hours of the dark phase of a administration animal models were firstly developed to evaluate reversed light/dark cycle (lights off at 8:00 a.m. and on at 20:00 p.m.). short and long-lasting changes after exposure to a drug (Panlilio Body weight and food intake were monitored throughout the entire et al., 1998). In contrast, contingent operant self-administration experiment. All animal procedures were approved by the local models allow to directly evaluate a drug’s reinforcing property with a ethical committee (Comitè Ètic d’Experimentació Animal-Parc de high predictive value to model voluntary drug consumption in Recerca Biomèdica de Barcelona, CEEA-PRBB, agreement N°9687) humans. The current model provides for the first time an animal and conducted in strict conformity with the guidelines of the model of cannabinoid addiction that recapitulates the diagnostic European Communities Council Directive (2010/63/EU) criteria used in DSM-5 to define this human disorder (Deroche- regulating animal experimentation, in the animal facility at Gamonet et al., 2004; Belin and Everitt, 2008; Maldonado et al., Universitat Pompeu Fabra-Barcelona Biomedical Research Park 2021). Indeed, our animal model has been generated based on the (UPF-PRBB; Barcelona, Spain). All the experiments were three main behavioral hallmarks of addiction that englobe DSM-5 performed under blind and randomized conditions. The male sex addiction criteria (Supplementary Figure S1): 1) persistence of was chosen accordingly with the previous literature that has response (criteria 6 and 7 of DSM-5), 2) motivation for the drug validated the operant WIN 55,212–2 self-administration model (criteria 9 and 10 of DSM-5), 3) compulsive-like behavior defined as only in males (Martellotta et al., 1998; Mendizábal et al., 2006; a disruption of inhibitory control despite negative consequences Mancino et al., 2015; Domingo-Rodriguez et al., 2020; Martín- (criterion 11 of DSM-5). Our model also measures extinction García et al., 2020; García-Blanco et al., 2022). reactiveness and reinstatement, parameters closely related with craving, another addiction diagnostic criteria of the DSM-5, as well as other phenotypic traits of predictive value for the 2.2 Drugs development of cannabinoid addiction. In this study, we used the synthetic cannabinoid agonist WIN WIN 55,212–2 [(R)-(+)-WIN 55,212–2 mesylate salt, Sigma- 55,212–2, which is a potent full agonist of cannabinoid receptor 1 Aldrich, U.S.A.] was dissolved in one drop of Tween 80 (TWEEN 80, (CB1R) (Compton et al., 1992), to generate a model of cannabinoid Sigma-Aldrich, U.S.A.) and then diluted in heparinized (1%) sterile addiction (Maldonado et al., 2011). Previous studies tried to obtain saline solution and made available at two different doses: 0.1 mg/kg an operant intravenous (iv) self-administration of delta9- for intraperitoneal (ip) injection 24 h before the first operant session tetrahydrocannabinol (THC), which is the primary psychoactive and 12.5 μg/kg/infusion for the self-administered iv infusions. The component of the cannabis sativa plant. However, few studies were preparation was covered from the light and stored at room able to maintain a model with persistent, dose-related behavior temperature. After each self-administration session, 0.05 mL of regarding THC iv self-administration (Justinova et al., 2003; Spencer sodic heparin (Hospira 5%, Hospira, Pfizer) was applied through et al., 2018), as well as vapor self-administration (Freels et al., 2020), the iv catheter to avoid coagulation and obstruction of the latter. showing the necessity of alternative methods. In contrast to THC, Thiopental sodium (5 mg/mL, Braun Medical S.A.) was dissolved in mice achieved a reliable operant iv self-administration with the distilled water and injected in a volume of 0.05 mL through the iv synthetic cannabinoid agonist WIN 55,212–2 (Martellotta et al., catheter to evaluate catheter patency. 1998; Mendizábal et al., 2006), although this cannabinoid has not yet been used to validate a model of cannabinoid addiction. This study aimed to validate for the first time a reliable mouse model of 2.3 Operant self-administration apparatus cannabinoid addiction by using WIN 55,212–2 iv operant self- administration. Experiments were performed in mouse operant chambers This behavioral protocol can be combined with multiple (model ENV-307A-CT, Med Associates Inc., Georgia, VT, neurochemical, electrophysiological, optogenetic, and U.S.A.) equipped with two nose-pokes, one randomly selected as Frontiers in Pharmacology 02 frontiersin.org Cajiao-Manrique et al. 10.3389/fphar.2023.1143365 the active hole and the other as the inactive hole. A house light was 3.1.2 WIN 55,212–2 self-administration training located on the chamber’s ceiling, and two stimuli lights (cues) were The operant model was applied accordingly to previous drug placed one inside the active hole and the other above it. Nose-poking self-administration paradigms (Mendizábal et al., 2006; Vallée et al., on the active hole resulted in the delivery of one WIN 2014; Martín-García et al., 2016). To avoid the aversive effects of the 55,212–2 infusion (under the associated schedule) paired with the drug’s first administration, mice received an ip injection of WIN activation of the stimulus light located above the active hole, while 55,212–2 (0.1 mg/kg) only 24 h before the first self-administration nose-poking on the inactive hole had no consequences. The session (Valjent and Maldonado, 2000; Mendizábal et al., 2006; chambers were made of aluminum and acrylic and placed inside Vallée et al., 2014). Subsequently, mice (n = 29) were trained to sound- and light-attenuated boxes equipped with fans providing acquire an operant self-administration conditioning maintained by ventilation and white noise. The chamber’s floor was a grid made iv infusions of WIN 55,212–2. The schedule was a fixed ratio (FR) with metal bars that could conduct electrical current when 1 schedule of reinforcement during 5 consecutive sessions, followed performing the shock test. WIN 55,212–2 (12.5 μL/kg/infusion) by a progression to FR2 for another 5 sessions. All sessions were was delivered in a volume of 23.5 μL over 2 s via a syringe firmly performed at the same time and scheduled every day. Each daily self- attached to a micro infusion pump (PHM-100A, Med-Associates, administration session was started with a priming injection of the Georgia, VT, U.S.A.) and connected with flexible polymer tubing drug (0.0125 mg/kg/infusion) automatically delivered iv through the (0.96 mm outer diameter, Portex Fine Bore Polythene Tubing, catheter when the session was initiated (Mendizábal et al., 2006; Portex Limited, Kent, England) to a single channel liquid swivel Martín-García et al., 2016; Flores et al., 2020), followed by two (375/25, Instech Laboratories, Plymouth Meeting, PA, U.S.A.) and 55 min active periods separated by a 15 min drug-free period for a the mouse iv catheter. total duration of 125 min. The initiation of each session was signaled by turning on the house light only during the first 3 s. The cue lights, together with the noise of the infusion pump, acted as 3 Methods environmental cues signaling the drug infusion. A 10 s time-out period was fixed after each drug delivery, during which the cue light 3.1 WIN 55,212–2 self-administration was off, and no reward was provided after responding to the active nose-poke. Responses to the active and inactive holes and all 3.1.1 Jugular vein catheterization responses executed during the time-out were recorded. During Mice (n = 30) were anesthetized by ip injection (0.2 mL/10 g of the drug-free period, no reinforcer nor cue was delivered, body weight) of ketamine hydrochloride (75 mg/kg of body weight, signaled by the activation of the house light. The session was Ketamidor, Richterpharma ag, Austria) and medetomidine concluded after 50 reinforcers were delivered or after 125 min, hydrochloride (1 mg/kg of body weight, Domtor, Esteve, Spain) whichever occurred first. The acquisition of the self-administration dissolved in 0.9% sterile physiological saline and then implanted behavior was achieved when the three following conditions were with indwelling iv silastic catheters in the right jugular vein, as met: 1) mice maintained 80% of stability in three consecutive previously described (Martín-García et al., 2009). Briefly, a 6 cm training sessions, meaning that the variance during these 3 days long silicone tubing (0.3 mm inner diameter, 0.6 mm outer was 20% or less, 2) at least 75% responding on the active hole, and 3) diameter; Silastic, Dow Corning, Houdeng-Goegnies, Belgium) a minimum of five reinforcers per session. After each session, mice was adapted to a 22-gauge steel cannula (Semat, Herts, were brought back to their home cages (Figure 1). England) curved at a right angle and embedded in a dental cement disk (Dentalon Plus, Heraeus Kulzer, Germany) with a 3.1.3 Three addiction criteria nylon mesh underneath. The catheter tubing was inserted 1.1 cm The development of addictive-like behaviors was evaluated into the right jugular vein and attached with a suture. The at the end of the training sessions based on three addiction-like remaining tubing was inserted subcutaneously (sc) to the criteria that summarize the addiction hallmarks according to cannula, exiting at the midscapular region. All incisions were the DSM-5 (Deroche-Gamonet et al., 2004; Piazza and Deroche- sutured and coated with a local analgesic (Blastoestimulina, Gamonet, 2013; Domingo-Rodríguez et al., 2022). The Almirall, Spain). Post-surgery procedure consisting of an ip addiction score developed was then attributed based on the injection of antibiotic (1 mg/kg of body weight, Gentamicine, results of these three criteria, each determined by the respective Genta-Gobens, Laboratorios Normon, Spain), a sc injection of behavioral test: analgesic (mixture of glucose serum (GlucosaVet, B. Braun Vet Care, Spain) and meloxicam (2 mg/kg of body weight, Metacam, 3.1.3.1 Persistence to response Boehringer Ingelheim, Rhein) and a sc injection of an anesthesia The number of non-reinforced active responses during the reversor, atipamezole hydrochloride (2.5 mg/kg of body weight, 15 min drug-free period was measured as persistence of drug- Revertor, Virbac, Spain), was applied all dissolved in 0.9% sterile seeking behavior. Mice were scored on the three consecutive days physiological saline. Mice were allowed to recover for 3 days, with before the progressive ratio (PR). follow-up analgesics, prior to the initiation of the self- administration sessions. The patency of iv catheters was 3.1.3.2 Motivation assessed by a thiopental sodium test at the end of the self- The PR schedule of reinforcement evaluated the motivation administration experimental sequence. The mouse was removed towards the reinforcer. The responses required to receive one drug from the experiment if prominent signs of anesthesia were not infusion escalated following this series: 1, 5, 12, 21, 33, 51, 75, 90, observed immediately after injection (n = 1 in this study). 120, 155, 180, 225, 260, 300, 350, 410, 465, 540, 630, 730, 850, 1000, Frontiers in Pharmacology 03 frontiersin.org Cajiao-Manrique et al. 10.3389/fphar.2023.1143365 FIGURE 1 Complete experimental protocol of our mouse model of operant WIN 55,212–2 intravenous self-administration. 1200, 1500, 1800, 2100, 2400, 2700, 3000, 3400, 3800, 4200, 4600, during which each drug delivered was associated with a foot-shock- 5000, and 5500. The breaking point, the maximal number of induced punishment. This shock session was performed after a responses mice perform to obtain one infusion defined as the stabilizing FR2 self-administration session following the PR test. motivation value, corresponds to the last ratio completed. The Mice were placed in a different operant box than the one regularly duration of the PR session was maximum 4 h or until mice used for the operant sessions. Then mice underwent an FR2 self- stopped responding to any nose-poke within 1 h. administration schedule of reinforcement for 50 min with two scheduled changes: after one active response, mice received an 3.1.3.3 Compulsivity electric foot-shock (0.18 mA, 2 s), while after the second response, Resistance to punishment, now defined as compulsive-like the electric foot-shock was paired with the drug delivery and the behavior, corresponded to the maintenance of active responding associated cue light. In parallel, if the second response was not behavior despite its association with a negative consequence. It was completed within a min after completing the first response, the measured by the total number of shocks obtained in a 50 min shock test, sequence was reinitiated. Frontiers in Pharmacology 04 frontiersin.org Cajiao-Manrique et al. 10.3389/fphar.2023.1143365 3.1.4 Establishment of mice subpopulations 3.1.6.1 Impulsivity After the three behavioral tests were performed, mice were The number of non-reinforced active responses during the time- categorized into addicted and non-addicted animals based on the out periods (10 s) after each WIN 55,212-2 delivery was measured as number of positive criteria achieved. A mouse was considered to be impulsivity-like behavior, which indicated the inability to stop a positive for an addiction-like criterion when the score of the response once it is initiated. The three consecutive days before the behavioral test was equal to or beyond the 75th percentile of the PR test were considered for this criterion. normal distribution of the saline group. Mice that achieved 2 or 3 criteria were considered addicted and categorized as vulnerable, 3.1.6.2 Sensitivity to reward whereas those reaching 0 or 1 criterion were considered non- The number of reinforcers obtained in 2-h sessions during the addicted and categorized as resilient. last three consecutive FR2 operant conditioning sessions maintained by WIN 55,212–2. Animals with higher levels of sensitivity to reward 3.1.5 Extinction and parameters related with will obtain a higher number of reinforcers. craving Only mice with patent catheters that reached all acquisition criteria continued to the extinction phase. After thiopental testing, mice were 3.2 DREADD approach: Surgery and viral allowed to rest for 1 day, during which they underwent a 2-h locomotion vector microinjection test in individual locomotor activity boxes (10.8 × 20.3 × 18.6 cm, Imetronic, Pessac, France) equipped with infrared sensors to detect The adeno-associated viral (AAV) vectors used were: AAV- locomotor activity and an infrared plane to detect rearings. hM4Di-DREADD (AAV8-hSyn-DIO-hM4D(Gi)-mCherry, During the extinction period, neither WIN 55,212–2 infusions, 1.21E+13 gc/mL) and AAV-retrograde-Cre-EBFP (AAV pmSyn1- priming infusions, nor the associated environmental cues were EBFP-Cre; 8.2E+12 gc/mL) (Viral Vector Production Unit, delivered after nose-poking on the active hole. Mice were exposed to Universitat Autònoma de Barcelona). 2-h daily sessions for 10 consecutive days in the same operant chamber Mice were anesthetized by ip injection (0.2 mL/10 g of body as the self-administration sessions. During this period, mice reached the weight) of ketamine hydrochloride (75 mg/kg of body weight, extinction criterion when responses to the active nose-poke were < 35% Ketamidor, Richterpharma ag, Austria) and medetomidine of the mean responses obtained during the last 3 days of WIN hydrochloride (1 mg/kg of body weight, Domtor, Esteve, Spain) 55,212–2 self-administration across three consecutive extinction dissolved in 0.9% sterile physiological saline and located into a sessions. Only mice that achieved the extinction criterion were stereotaxic apparatus to receive the intracranial AAV injections. All evaluated for the following. Two parameters related with craving injections were performed through a bilateral injection cannula (33- were evaluated before and after this extinction period: gauge internal cannula, Plastics One, United Kingdom) connected via a polyethylene tubing (PE-20, Plastics One, United Kingdom) to 3.1.5.1 Resistance to extinction a 10 μL microsyringe (Model 1701 N SYR, Cemented NDL, 26 s GA, Number of active responses in 2 h during the first extinction 2 in point style 3, Hamilton company, NV). The displacement of an session. Animals with significant sensitivity to drug withdrawal will air bubble along the tubing connecting the syringe to the injection increase their resistance to extinction by increasing the number of needle was utilized to monitor the microinjections. For the precise active nose-pokes to seek the drug when access is prevented. inhibition of the PL-NAc pathway, two bilateral injections were performed, one targeting the PL and the other the NAc core. Mice 3.1.5.2 Drug-seeking behavior were injected with 0.2 μL per site of the AAV-hM4Di-DREADD into The day after achieving the extinction criterion, we performed a PL (rate infusion of 0.05 μL/min) and 0.4 μL per site of the AAV- single cue-induced reinstatement session in the same operant retrograde-Cre-EBFP into the NAc core (rate infusion of 0.10 μL/ chamber, in order to test reinstatement of drug-seeking behavior min). After infusion, the injection cannula was left untouched for an upon exposure to the environmental stimuli after a period of additional 10 min to permit the fluid to diffuse and prevent reflux, abstinence. The cue test was conducted under the same and then slowly withdrawn. A heating pad was used to preserve the conditions used in the acquisition phase, except that active body temperature at 35°C. The coordinates used followed the responding was not reinforced by the drug. This meant that mice Paxinos and Franklin atlas (Franklin and Paxinos, 1997): (PL) were subject to a 90-min FR2 session, where the first 60 min were AP +2.10 mm, L ±0.3 mm, DV -2.3 mm; (NAc core) AP similar to an extinction session but in the last 30 min, nose-poking +1.94 mm, L ±1 mm, DV -4.6 mm. on the active hole resulted in the presentation of all the Clozapine-N-oxide (CNO, Enzo Life Sciences, NY), a environmental cues associated (cue light, pump noise, and behaviorally inert drug, was administered via Alzet osmotic priming injection light) but not the delivery of WIN 55,212–2 minipumps (Model 2004; Alzet, Cupertino, CA) previously filled (Martín-García et al., 2016; García-Blanco et al., 2022). with either CNO (diluted in 0.9% sterile saline; 5 mg/mL) or physiological saline solution. Minipumps were sc implanted in 3.1.6 Behavioral tests to evaluate addiction-like the lower back of each mouse during the jugular vein phenotypic traits catheterization surgery. The osmotic minipumps delivered CNO Two additional phenotypic traits were also evaluated as factors by osmosis at a constant sc flow rate of 0.25 μL/h for 15 days of vulnerability to addiction: (Domingo-Rodríguez et al., 2020; Martín-García et al., 2020). Frontiers in Pharmacology 05 frontiersin.org Cajiao-Manrique et al. 10.3389/fphar.2023.1143365 3.3 Statistical analysis 0.01 in FR2, p < 0.001, DMS Actives vs. Inactives: p < .001]). No significant differences were found in active and inactive nose- 3.3.1 Statistical analysis of behavioral data pokes between CNO- and saline-treated mice in both FR1 and The number of animals (n) in each experimental condition is FR2 schedules, suggesting similar levels of operant conditioning indicated in the figure legends. All statistical comparisons were maintained by WIN 55,212–2 (DMS Actives/Inactives CNO vs. performed with SPSS (IBM, version 25). Comparisons between two Actives/Inactives Saline: n.s.). groups were performed by Student’s t-test or U Mann-Whitney test The addiction score was calculated after the operant training according to the distribution defined by the Kolmogorov-Smirnov using the three addiction-like criteria, as explained above. normality test. ANOVA with repeated measures was used when Extreme subpopulations of mice that present a high necessary to test the evolution over time, followed by post hoc persistence of response, motivation and compulsivity, were analysis (Fisher LSD) for multiple group comparison. The revealed in both saline and CNO groups. Specifically, 26.1% Pearson correlation coefficient was performed to analyze the (23.08% saline, 18.75% CNO), 51.7% (61.54% saline, 43.75% relationships between values in each addiction-like criterion and CNO) and 24.1% (30.77% saline, 18.75% CNO) of mice the final criteria achieved. The chi-square analysis were used to surpassed each criterion’s threshold, suggesting the potential compare the percentage of addicted and non-addicted mice. Results development of addictive-like behaviors after the chronic were expressed as individual values with the median and the operant training. No significant differences were found interquartile range or with the mean ± S.E.M, which is specified between CNO- and saline-treated mice in persistence of in the figure legend. A p-value <0.05 was applied to determine response, motivation or compulsive-like behavior (Figures statistical significance. 2B–D). In the saline group, 23.08% (3/13) were considered The sample size was calculated based on the power analysis. The addicted whereas 12.50% (2/16) were considered addicted in significance criterion (alpha) was set at 0.050, and the statistical test the CNO group (chi-square = 3.77, n.s., Figure 2E). Addicted utilized was a two-sample t-test. With the sample size of 13–16 mice per mice showed a strong tendency for higher persistence of response group, our studies achieved a power superior to 80%. Supplementary compared to non-addicted in both saline and CNO groups and tables (Supplementary Tables S1–S2) provide a complete report of the only for the saline-treated animals in the motivation (Figures statistical results for the data described in the figures. 2F,G), whereas a significantly higher compulsive-like behavior was observed for addicted mice compared to non-addicted mice 3.3.2 Principal component analysis regardless of the treatment (U Mann-Whitney test, U = 3,000 for The principal component analysis (PCA) was performed to NA vs. A saline and U = 1,500 for NA vs. A CNO, p < 0.05, evaluate the multidimensional behavioral data by reducing it to Figure 2H), Moreover, positive correlations were found between fewer dimensions in order to observe trends, clusters, and outliers. the number of criteria achieved and the severity of each criterion PCA and varimax rotation were conducted using the three in both CNO- and saline-treated mice for all addictive-like addiction-like criteria, the two parameters related to craving and criteria except for motivation in CNO-treated mice (Pearson the two phenotypic traits considered as vulnerability factors of correlations, p < 0.05, Figures 2I–K). addiction to dimensionality reduce them to the minimum After FR1 and FR2 training, mice underwent 10 sessions of number of components that best explain and maximize the extinction (Figure 3A). Both groups extinguished the self- variance present in the data set. An eigenvalue greater than administration behavior similarly (Repeated Measures 1 was set as selecting components criterion (Field, 2018). ANOVA, [Active lever presses: F(1,27) = 0.15, n.s., Inactive lever presses: F(1,27) = 0.38, n.s.]), despite a higher number of reinforcers obtained during the first session in saline-treated 4 Results mice compared to the CNO group (U Mann-Whitney test, U = 42,000, p < 0.01, Figure 3B). Responses to the active 4.1 WIN 55,212–2 self-administration led to nose-poke declined over time until reaching 58% and 43.7% the development of an addictive-like decrease of the active nose-poke in the last session compared phenotype in mice to the last operant session for the saline and CNO groups, respectively. We have developed for the first time a mouse model of Animals that responded < 35% of the mean responses cannabinoid addiction by using WIN 55,212–2 self- performed during the last 3 days of WIN 55,212–2 self- administration. In addition, we have evaluated in this model administration across three consecutive extinction sessions the possible involvement of the PL-NAc pathway in the acquired the extinction criteria (30.77% of saline and 12.50% of development of this addictive behavior (Figure 1). Saline and CNO mice, chi-square = 1.92, n.s.). Resistance to extinction, CNO-treated mice were trained to acquire an operant self- measured on the first day of extinction, was significantly lower in administration sustained by iv infusions of WIN 55,212–2 the CNO group compared to saline-treated mice (U Mann-Whitney (Figure 2A). The percentage of animals that achieved test, U = 42.000, p < 0.01, Figure 3B). In contrast, no significant acquisition criteria of stability, discrimination, and number of differences were obtained between groups for the cue-induced reinforcers was 44.83% in the saline group and 55.17% in the reinstatement of drug-seeking behavior (Figure 3C). Non- CNO group (chi-square, C-S = 0.69, n.s.), with a progressive addicted mice in the saline group showed higher levels of increase in the number of active nose-pokes across sessions response in the first extinction session compared to non-addicted (Repeated Measures ANOVA, [F(1,54) = 27.76 in FR1, F(1,54) = mice in the CNO group (U Mann-Whitney test, U = 32.000, p < 0.05, Frontiers in Pharmacology 06 frontiersin.org Cajiao-Manrique et al. 10.3389/fphar.2023.1143365 FIGURE 2 WIN 55,212–2 operant self-administration led to the development of an addictive-like phenotype in mice. (A) Similar number of total active and inactive nose-pokes performed by CNO- and saline-treated groups during 2 h of operant self-administration maintained by intravenous infusions of WIN 55,212–2 in both FR1 and FR2 schedules of reinforcement (mean ± S.E.M., repeated measures ANOVA). (B–D) Mice present similar responses in the three addiction-like criteria tests (individual data with median and interquartile range): (B) Persistence to response: Number of responses to the active nose-poke during the 15 min drug-free period (Student’s t-test). (C) Motivation: Breaking point determined during a 4 h progressive schedule of reinforcement represents the maximal number of responses that an animal is able to do to obtain one drug infusion (U Mann-Whitney). (D) Compulsivity: Number of shocks received following the schedule described in the Materials and Methods section, reflecting the compulsivity level of each group (Student’s t-test). The dashed horizontal line indicates the 75th percentile of the distribution of the group, used as the threshold to consider a mouse positive for one criterion. Addicted mice are represented in grey- and red-filled circles. (E) Percentage of mice categorized as addicted (Chi square). (F–H) Behavioral tests of the three addiction-like criteria showing increased compulsivity in addicted mice compared to non-addicted but similar persistence to response and motivation (individual data with mean ± S.E.M., U Mann-Whitney, *p < 0.05). (I–K) Pearson correlations between individual values of addiction-like criteria and (I) non-reinforced active responses in 15 min, (J) breaking point in 4 h and (K) number of shocks in 50 min (Saline-treated mice: n = 13; CNO-treated mice: n = 16; statistical details are included in Supplementary Table S1). Frontiers in Pharmacology 07 frontiersin.org Cajiao-Manrique et al. 10.3389/fphar.2023.1143365 FIGURE 3 WIN 55,212–2 operant self-administration effects on two parameters related with craving and two phenotypic vulnerability traits to addiction-like behavior in mice. (A) Extinction pattern of the WIN 55,212–2 operant self-administration behavior (mean ± S.E.M., repeated measures ANOVA). (B–C) Behavioral tests of the two parameters related with craving (individual data with median and interquartile range): (B) Resistance to extinction: Number of responses to the active nose-poke during the first 2-h extinction session is significantly higher in saline-compared to CNO-treated mice (U Mann- Whitney, **p < 0.01). (C) Drug-seeking behavior measured by the cue-induced reinstatement after abstinence: Number of active responses performed during the 90 min cue-induced drug-seeking test performed after extinction (Student’s t-test). The dashed horizontal line indicates the 75th percentile of distribution of the group, used as the threshold to consider a mouse positive for one criterion. Addicted mice are represented in grey- and red-filled circles. (D–E) Behavioral tests of the parameters related with craving showing similar responses in cue-induced reinstatement between addicted and non-addicted mice (U Mann-Whitney), whereas a difference is observed between non-addicted mice in the 1st day of extinction (U Mann-Whitney, *p < 0.05) (individual data with mean ± S.E.M.). (F–G) Behavioral tests used to evaluate the two phenotypic traits considered to be factors of vulnerability to (Continued ) Frontiers in Pharmacology 08 frontiersin.org Cajiao-Manrique et al. 10.3389/fphar.2023.1143365 FIGURE 3 (Continued) addiction-like behavior (individual data with median and interquartile range): (F) Impulsivity: Number of responses to the active nose-poke during the 10 s time-out period (U Mann-Whitney) (G) Sensitivity to reward: Number of reinforcers performed to the active nose-poke during the 2 h of the last three sessions of self-administration (Student’s t-test). The dashed horizontal line indicates the 75th percentile of distribution of the group, used as the threshold to consider a mouse positive for one criterion. Addicted mice are represented in grey- and red-filled circles. (H–I) Behavioral tests of the phenotypic traits showing similar responses in impulsivity and behavior during the last 3 days of operant training between addicted and non-addicted mice (U Mann-Whitney) (individual data with mean ± S.E.M.). (J) Body weight: Body weight was measured every week during the self-administration protocol (mean ± S.E.M.; Student’s t-test). (K) Locomotor activity: Activity was measured by the number of beam breaks during a 2 h test (mean ± S.E.M.; Student’s t-test). (Saline-treated mice: n = 13; CNO-treated mice: n = 16; statistical details are included in Supplementary Table S2). FIGURE 4 Principal component analysis of cannabinoid addiction through WIN 55,212–2 operant self-administration. (A) Factor loadings of the principal component 1 (PC1) and principal component 2 (PC2) in all variables studied. (B–C) Order of factor loading of the different variables in PC1 and PC2. The dashed horizontal line marks loading greater than 0.7, mainly contributing to the component. In regards to the addiction criteria, a dissociation between motivation and persistence of response, mainly contributing to PC1, and compulsivity, mainly contributing to PC2, can be observed. Regarding parameters related to craving, drug-seeking behavior weighted more in the PC1, while resistance to extinction weighted more in the PC2, even though its influence is also present in the PC1. For the phenotypic traits, both weighted more in the PC1. Figure 3D), while no significant differences were obtained for the were also evaluated. No significant differences were found drug-seeking behavior (Figure 3E). between CNO (mean ± S.E.M.: 10.85 ± 2.85; 18.08 ± 3.11) and saline-treated mice (mean ± S.E.M.: 8.56 ± 2.66; 19.90 ± 3.42), neither in impulsivity nor reward sensitivity respectively (Figures 4.2 WIN 55,212–2 self-administration effects 3F,G). No significant differences between groups were neither on phenotypic vulnerability traits to observed when the population of saline and CNO groups was addiction-like behavior in mice divided into addicted and non-addicted (Figures 3H,I). To confirm that CNO treatment did not produce any effect that Two phenotypic traits considered as vulnerability factors to could bias our self-administration results (Roth, 2016), the body addiction-like behavior, impulsivity and sensitivity to reward, weight, locomotor activity, and food intake of mice were monitored Frontiers in Pharmacology 09 frontiersin.org Cajiao-Manrique et al. 10.3389/fphar.2023.1143365 FIGURE 5 Correlation heatmaps of the variables of cannabinoid addiction criteria, parameters related to craving, and phenotypic vulnerability traits. (A–B) Pearson correlations between the three addiction-like criteria, the two parameters related with craving and the two phenotypic traits in both (A) non- addicted and (B) addicted groups. (*p < 0.05, **p < 0.01, ***p < 0.001). throughout the experiment. No significant differences were observed parameters related with craving and phenotypic traits were between saline- and CNO-treated mice in terms of body weight taken into account. Principal component 1, which accounts (Figure 3J), and food intake (data not shown) across the entire for 45.0% of the variance (Figure 4A), has strong loadings experiment. Moreover, no significant differences were observed (>0.7) from all behavioral variables except compulsivity. These between groups in the locomotor activity (Figure 3K), sustaining traits are associated to the development of cannabinoid addiction the absence of side effects of CNO treatment. and, therefore, they contribute to this development. The second principal component, which is orthogonal to component 1 and accounts for 20.2% of the variance, is comprised of two variables, 4.3 Principal component analysis of the criteria of compulsivity and resistance to extinction. cannabinoid addiction through WIN Interestingly, impulsivity participates more in the first 55,212–2 self-administration component, while compulsivity is more critical in the second component (Figures 4B,C), resembling the sequential feature of A principal component analysis was used to determine the transition from impulsivity to compulsivity described in whether the behavioral outcomes previously described could addiction. Finally, most of the phenotypic traits of be reduced to fewer dimensions that might display individual vulnerability are in the same component suggesting similar differences in cannabinoid addiction. All addiction criteria, neurological correlates. Frontiers in Pharmacology 10 frontiersin.org Cajiao-Manrique et al. 10.3389/fphar.2023.1143365 4.4 Correlation heatmap of the variables of 1998; Fattore et al., 2001; Mendizábal et al., 2006; Lefever et al., cannabinoid addiction criteria, parameters 2014). However, none of these self-administration studies has been related with craving and vulnerability used to generate a model of cannabinoid addiction. phenotypic traits In our study, we have established a model in which vulnerable mice present persistence to self-administer WIN 55,212–2, When representing the addiction-like criteria, the parameters extremely high motivation to obtain WIN 55,212–2 infusions, related with craving and the phenotypic traits into a heat map, non- and compulsivity to WIN 55,212-2-seeking and self-administer addicted animals revealed significant correlations between despite adverse consequences, which confirms the development persistence to response and sensitivity to reward (r = 0.54, p < of a cannabinoid addiction-like model based on iv infusions of 0.01), motivation and sensitivity to reward (r = 0.62, p < 0.001), WIN 55,212–2. These criteria allowed to separate two populations of impulsivity and sensitivity to reward (r = 0.86, p < 0.001), and mice with vulnerable and resilient phenotypes to develop motivation and impulsivity were significant (r = 0.54, p < 0.01) cannabinoid addiction. Therefore, the establishment of this (Figure 5A). In addicted animals, the significant correlations cannabinoid addiction mouse model using WIN 55,512–2 iv self- between motivation and impulsivity (r = 0.99, p < 0.01), administration represents a pivotal tool for future research allowing motivation and sensitivity to reward (r = 0.95, p < 0.05), and to elucidate the neurobiological correlates underlying resilience and impulsivity and sensitivity to reward (r = 0.98, p < 0.01) were vulnerability to develop this disorder. These addiction-like maintained (Figure 5B). Theses results were in agreement with the hallmarks used to establish this cannabinoid addiction model results obtained in the PCA. Interestingly, compulsivity and have been extracted from an established rat model of cocaine impulsivity showed a negative correlation (non-significant) in addiction (Deroche-Gamonet et al., 2004), and have been coherence with the differential load of each variable in the PCA. repeatedly used as a reference to establish mice models of drug and food addiction (Martín-García et al., 2016; Domingo-Rodríguez et al., 2020). 5 Discussion A main problem in the treatment of addiction is the high rates of relapse to drug use after periods of abstinence (Venniro et al., 2016; Cannabis addiction, defined in the 5th edition of the DSM as Fredriksson et al., 2021). An important advantage of our cannabis use disorder, is a major concern worldwide, and its cannabinoid addiction model is the possibility to evaluate two neurobiological substrate is still largely unknown. The absence of parameters closely related with craving and relapse, resistance to animal models that recapitulate the hallmarks of cannabinoid extinction and cue-induced reinstatement. Resistance to extinction addiction has impaired the animal research focused on this measures an ‘extinction burst’ behavior typically seen in rodents disorder. In this study, we developed for the first time an animal during the first day of extinction (Cooper et al., 1987) that has been model of cannabinoid addiction that recapitulates the main revealed with different drugs of abuse (Peltier et al., 2001; Shalev diagnosis criteria of this disorder defined on the DSM-5 based on et al., 2002; Soria et al., 2008). The resistance to extinguish the an operant paradigm of WIN 55,212–2 iv self-administration. operant behavior revealed in the present study suggests that mice Drug self-administration paradigms based on their positive had developed a reliable and persistent operant WIN 55,212–2 self- reinforcement effects can be used to model certain aspects of the administration behavior, and this behavior has been reported to human addictive behavior (Markou et al., 1999). However, previous reflect a ‘craving-like’ state at the beginning of the extinction training studies have demonstrated the complexity in modeling cannabis (Fredriksson et al., 2021). Craving during abstinence has been self-administration compared to other drugs due to the difficulty to suggested to be directly involved with the vulnerability to relapse obtain a reliable self-administration of cannabinoids in animals (Venniro et al., 2016; Fredriksson et al., 2021). Reinstatement of (Panlilio and Justinova, 2018). Operant self-administration of drug seeking is typically assessed by the extinction-reinstatement THC is difficult to be reliably maintained in animals (Tanda and model (Shaham et al., 2003; Weiss, 2010) and non-reinforced Goldberg, 2003). Only a few number of studies in squirrel monkeys responding to the previously learned active nose-poke is the (Justinova et al., 2003) and rats (Spencer et al., 2018; Freels et al., measure of drug seeking (Stewart and de Wit, 1987). In our 2020) have been able to show that THC can maintain self- model, we performed a cue-induced reinstatement procedure administration (intravenous or vapor) without a previous history after extinction, in which re-exposure to conditioned cues when of exposure to other drugs. The failure of THC to act as reinforcer in responding to the active nose-poke, cues that had been contingently animal studies has been related to its delayed onset of pleasurable paired with drug delivery during acquisition, reinstated drug seeking pharmacological effects, the long duration of its pharmacological (Davis and Smith, 1976). Exposure to drug-associated cues can elicit and behavioral effects, and its partial agonist profile (Tanda and drug desire and drug seeking, effects implicated both in the Goldberg, 2003). Due to these difficulties, most of the operant self- maintenance of ongoing drug use and inducing drug seeking administration studies have used synthetic cannabinoids. WIN after abstinence, which shows important resistance to extinction 55,212–2 is a potent synthetic cannabinoid with a shorter half- (Weiss and Volkow, 2022). However, some studies argue that life than THC and a full agonist of the CB1R (Pertwee et al., 2010). human abstinence is often either forced or voluntary (self- These characteristics may explain the difference in reinforcing imposed) (Venniro et al., 2016; Fredriksson et al., 2021). In fact, properties and the improving characteristics for operant self- drug relapse and craving are commonly triggered not only by drug- administration models in comparison to THC (Maldonado et al., associated cues, but also by acute exposure to the self-administered 2011). Several studies have achieved reliable operant paradigms to drug, stress and short-term or protracted withdrawal symptoms self-administer WIN 55,212–2 in mice and rats (Martellotta et al., (Venniro et al., 2016; Fredriksson et al., 2021). Alternative animal Frontiers in Pharmacology 11 frontiersin.org Cajiao-Manrique et al. 10.3389/fphar.2023.1143365 models have been developed in which abstinence is not obtained by previous literature on drug addiction models (Martellotta et al., extinction training but through forced abstinence, mainly assessed 1998; Mendizábal et al., 2006; Flores et al., 2020). In spite of all these by the incubation of drug craving model (Grimm et al., 2001) or by studies previously performed in male rodents, further studies will be voluntary abstinence, achieved either by the introduction of negative necessary to validate these models in female mice and rats. consequences to ongoing drug-administration (Panlilio et al., 2003; We have established for the first time a novel and reliable mouse Cooper et al., 2007), or of alternative non-drug reinforcers (Ahmed model of cannabinoid addiction using WIN 55,212–2 iv operant et al., 2013). self-administration that allows to evaluate three addiction criteria, This model also evaluates two phenotypic traits related to persistence of response, motivation, and compulsivity, based on the addiction vulnerability factors. Impulsivity is a complex construct addiction hallmarks defined in the DSM-5. This model also allows to composed of motor impulsivity and choice impulsivity (Belcher measure two parameters related craving, resistance to extinction and et al., 2014). In our model, we have considered the non-reinforced reinstatement, and two phenotypic traits related to cannabinoid active nose-pokes during the time-out periods to evaluate the motor addiction, impulsivity and sensitivity to reward. This model impulsivity defined as motor disinhibition, as previously described represents a pivotal tool to elucidate the neurobiological (Domingo-Rodriguez et al., 2022), The impulsivity trait has been substrates of cannabinoid addiction and guide future research associated with drug addiction since it predicts the transition to toward therapeutic strategies to address this disorder. compulsive drug intake (Belin and Deroche-Gamonet, 2012; Weafer et al., 2014), and neurobiological correlates underlying this phenotypic trait could reveal potential biomarkers and/or Data availability statement therapeutic targets for cannabinoid addiction. Reward sensitivity is associated with an increased probability of responding with a The original contributions presented in the study are included in positive hedonic component involving pathways that have a crucial the article/Supplementary Materials, further inquiries can be role in the rewarding properties of drugs of abuse (Koob and directed to the corresponding authors. Volkow, 2016). We have used this cannabinoid addiction model to evaluate the possible involvement of the PL- NAc glutamatergic pathway, which Ethics statement plays a crucial role in food addiction (Domingo-Rodríguez et al., 2020). This pathway is modulated by the endocannabinoid system The animal study was reviewed and approved by Comitè Ètic and deletion of the CB1R of these glutamatergic neurons leads to a d’Experimentació Animal-Parc de Recerca Biomèdica de Barcelona, resilient phenotype to develop food addiction (Domingo-Rodríguez CEEA-PRBB, agreement N°9687. et al., 2020). CB1Rs are the main cannabinoid receptors involved in the development of cannabinoid addiction (Maldonado et al., 2011) and the inhibition of the activity of the PL-NAc pathway may Author contributions modify the development of cannabinoid addiction. However, CNO activation of the inhibitory DREADDs expressed in the PL-NAc EM-G and RM created and conceived the animal model and pathway did not alter the addictive-like behavior regarding the behavioral methodologies associated. MMC-M conducted the addictive criteria and phenotypic vulnerability traits. However, behavioral experiments with the supervision of EM-G and RM. the resistance to extinction was decreased in CNO-treated MMC-M and EM-G performed catheterization surgeries. MMC-M animals, suggesting a protective effect of this manipulation on performed the statistical and data analyses, generated the associated the craving-like state. It is important to underly that the patency graphs, and interpretation of the results under the supervision of of the catheters limited the protocol to 2 weeks. Thus, the time EM-G and RM. MMC-M wrote the manuscript and prepared the between when the minipumps filled with CNO were implanted and figures with the supervision of EM-G and RM. EM-G and RM the performance of the addictive-like behavioral tests was merely revised the manuscript. 15 days. Hence, the CNO had only this short period of time to be released from the minipump and act on the DREADD to inhibit the pathway. We hypothesize this to be the reason for the absent effect of Funding this pathway’s inhibition, as the long-term action of the CNO and the hM4Di DREADDs in inhibiting a neuronal pathway has been This work was supported by the Spanish “Ministerio de Ciencia confirmed in many studies (Domingo-Rodríguez et al., 2020; e Innovación (MICIN), Agencia Estatal de Investigación (AEI)” Martín-García et al., 2020). (PID2020- 120029GB-I00/MICIN/AEI/10.13039/501100011033, The influence of the environment is key in the onset of RD21/0009/0019, to MR; the “Generalitat de Catalunya, consumption as well as the maintenance of cannabinoid AGAUR” (2017 SGR-669, to MR; the “ICREA-Acadèmia” (2020) addiction. Indeed, environments with high levels of social to MR; the “European Commission-DG Research” (PainFact, stressors, lack of opportunities, easy accessibility to drugs, and H2020-SC1-2019-2-RTD-848099, QSPain Relief, H2020-SC1- lack of alternative reinforcers, lead to an elevated risk for 2019-2-RTD-848068) to MR; the Spanish ”la Caixa” Foundation addiction development (Volkow et al., 2019). Consequently, the under the project code LCF/PR/HR22/52420017 to MR, the Spanish biggest shortcoming of this model is the absence of the “Instituto de Salud Carlos III, RETICS-RTA” (RD16/0017/0020) to environmental aspect of the disorder, which we cannot mimic in MR; the Spanish “Ministerio de Sanidad, Servicios Sociales e a mouse model. Moreover, the male sex was chosen considering the Igualdad, Plan Nacional Sobre Drogas” (PNSD- 2021I076, to MR; Frontiers in Pharmacology 12 frontiersin.org Cajiao-Manrique et al. 10.3389/fphar.2023.1143365 PNSD- 2019I006, to ME) and Ministerio de Ciencia e Innovación Publisher’s note (ERA-NET) PCI2021-122073-2A to ME. All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated Acknowledgments organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or We thank R. Martín, M. Linares, D. Real, and F. Porrón for their claim that may be made by its manufacturer, is not guaranteed or technical support. endorsed by the publisher. Conflict of interest Supplementary material The authors declare that the research was conducted in the The Supplementary Material for this article can be found online absence of any commercial or financial relationships that could be at: https://www.frontiersin.org/articles/10.3389/fphar.2023.1143365/ construed as a potential conflict of interest. full#supplementary-material References Ahmed, S. H., Lenoir, M., and Guillem, K. (2013). Neurobiology of addiction versus Fredriksson, I., Venniro, M., Reiner, D. J., Chow, J. J., Bossert, J. M., and Shaham, Y. drug use driven by lack of choice. Curr. Opin. Neurobiol. 23 (4), 581–587. doi:10.1016/j. (2021). Animal models of drug relapse and craving after voluntary abstinence: A review. conb.2013.01.028 Pharmacol. Rev. 73 (3), 1050–1083. doi:10.1124/pharmrev.120.000191 Belcher, A. M., Volkow, N. D., Moeller, F. G., and Ferré, S. (2014). Personality traits Freels, T. G., Baxter-Potter, L. N., Lugo, J. M., Glodosky, N. C., Wright, H. R., Baglot, and vulnerability or resilience to substance use disorders. Trends Cogn. Sci. 18 (4), S. L., et al. (2020). Vaporized cannabis extracts have reinforcing properties and support 211–217. doi:10.1016/j.tics.2014.01.010 conditioned drug-seeking behavior in rats. J. Neurosci. 40 (9), 1897–1908. doi:10.1523/ JNEUROSCI.2416-19.2020 Belin, D., and Deroche-Gamonet, V. (2012) Responses to novelty and vulnerability to cocaine addiction: Contribution of a multi-symptomatic animal model. Cold Spring García-Blanco, A., Domingo-Rodriguez, L., Cabana-Domínguez, J., Fernández- Harb. Perspect. Med., 2, a011940, doi:10.1101/cshperspect.a011940 Castillo, N., Pineda-Cirera, L., Mayneris-Perxachs, J., et al. (2022) miRNA signatures associated with vulnerability to food addiction in mice and humans. Belin, D., and Everitt, B. J. (2008). Cocaine seeking habits depend upon dopamine- J. Clin. Invest., 132 e156281, doi:10.1172/JCI156281 dependent serial connectivity linking the ventral with the dorsal striatum. Neuron 3 (57), 432–441. doi:10.1016/j.neuron.2007.12.019 Grimm, J. W., Hope, B. T., Wise, R. A., and Shaham, Y. (2001) Neuroadaptation. Incubation of cocaine craving after withdrawal. Nature, 412, 141–142. doi:10.1038/ Compton, D. R., Gold, L. H., Ward, S. J., Balster, R. L., and Martin, B. R. (1992). 35084134 Aminoalkylindole analogs: Cannabimimetic activity of a class of compounds structurally distinct from delta 9-tetrahydrocannabinol. J. Pharmacol. Exp. Ther. 263 Hamilton, P. J., and Nestler, E. J. (2019). Epigenetics and addiction. Curr. Opin. (3), 1118–1126. PMID: 1335057. Neurobiol. 59, 128–136. doi:10.1016/j.conb.2019.05.005 Compton, W. M., Wargo, E. M., and Volkow, N. D. (2022). Neuropsychiatric model Justinova, Z., Tanda, G., Redhi, G. H., and Goldberg, S. R. (2003). Self-administration of addiction simplified. Psychiatr. Clin. North Am. 45 (3), 321–334. doi:10.1016/j.psc. of delta9-tetrahydrocannabinol (THC) by drug naïve squirrel monkeys. 2022.05.001 Psychopharmacology 169 (2), 135–140. doi:10.1007/s00213-003-1484-0 Cooper, A., Barnea-Ygael, N., Levy, D., Shaham, Y., and Zangen, A. (2007) A conflict Koob, G., and Volkow, N. D. (2016). Neurobiology of addiction: A neurocircuitry rat model of cue-induced relapse to cocaine seeking. Psychopharmacology, 194, analysis. Lancet Psychiatry 3 (8), 760–773. doi:10.1016/S2215-0366(16)00104-8 117–125. doi:10.1007/s00213-007-0827-7 Lefever, T. W., Marusich, J. A., Antonazzo, K. R., and Wiley, J. L. (2014). Evaluation of Cooper, J., Heron, T., and Heward, W. (1987). Applied behavior analysis. New York: WIN 55,212-2 self-administration in rats as a potential cannabinoid abuse liability Macmillan. model. Pharmacol. Biochem. Behav. 118, 30–35. doi:10.1016/j.pbb.2014.01.002 Davis, W. M., and Smith, S. G. (1976). Role of conditioned reinforcers in the Lüscher, C., Robbins, T. W., and Everitt, B. J. (2020). The transition to compulsion in initiation, maintenance and extinction of drug-seeking behavior. Pavlov. J. Biol. Sci. addiction. Nat. Rev. Neurosci. 21 (5), 247–263. doi:10.1038/s41583-020-0289-z 11 (4), 222–236. doi:10.1007/BF03000316 Maldonado, R., Berrendero, F., Ozaita, A., and Robledo, P. (2011). Neurochemical Deroche-Gamonet, V., Belin, D., and Piazza, P. V. (2004) Evidence for addiction-like basis of cannabis addiction. Neuroscience 181, 1–17. doi:10.1016/j.neuroscience.2011. behavior in the rat. Science, 305, 1014–1017. doi:10.1126/science.1099020 02.035 Domingo-Rodriguez, L., Cabana-Domínguez, J., Fernàndez-Castillo, N., Cormand, Maldonado, R., Calvé, P., García-Blanco, A., Domingo-Rodríguez, L., Senabre, E., and B., Martín-García, E., and Maldonado, R. (2022). Differential expression of miR-1249- Martín-García, E. (2021). Vulnerability to addiction. Neuropharmacology 186, 108466. 3p and miR-34b-5p between vulnerable and resilient phenotypes of cocaine addiction. doi:10.1016/j.neuropharm.2021.108466 Addict. Biol. 27 (5), e13201. doi:10.1111/adb.13201 Mancino, S., Burokas, A., Gutiérrez-Cuesta, J., Gutiérrez-Martos, M., Martín-García, Domingo-Rodriguez, L., Ruiz de Azua, I., Dominguez, E., Senabre, E., Serra, I., Kummer, S., E., Pucci, M., et al. (2015). Epigenetic and proteomic expression changes promoted by et al. (2020). A specific prelimbic-nucleus accumbens pathway controls resilience versus eating addictive-like behavior. Neuropsychopharmacology 40 (12), 2788–2800. doi:10. vulnerability to food addiction. Nat. Commun. 11 (1), 782. doi:10.1038/s41467-020-14458-y 1038/npp.2015.129 European Monitoring Centre for Drugs and Drug Addiction (2022). European drug Markou, A., Arroyo, M., and Everitt, B. J. (1999) Effects of contingent and non- report. Trends Dev. 22, 75644. doi:10.2810/75644 contingent cocaine on drug-seeking behavior measured using a second-order schedule of cocaine reinforcement in rats. Neuropsychopharmacology, 20, 542–555. doi:10.1016/ European Monitoring Centre for Drugs and Drug Addiction (2015). European drug S0893-133X(98)00080-3 report; trends and developments. Lisbon: EMCDDA Martellotta, M. C., Cossu, G., Fattore, L., Gessa, G. L., and Fratta, W. (1998). Self- Fattore, L., Cossu, G., Martellotta, C. M., and Fratta, W. (2001). Intravenous self- administration of the cannabinoid receptor agonist WIN 55,212-2 in drug-naïve mice. administration of the cannabinoid CB1 receptor agonist WIN 55,212-2 in rats. Neuroscience 85 (2), 327–330. doi:10.1016/s0306-4522(98)00052-9 Psychopharmacology 156 (4), 410–416. doi:10.1007/s002130100734 Martín-García, E., Barbano, M. F., Galeote, L., and Maldonado, R. (2009). New Field, A. (2018). Discovering statistics using IBM SPSS statistics. Los Angeles: Sage. operant model of nicotine-seeking behaviour in mice. Int. J. Neuropsychopharmacol. 12 Flores, Á., Maldonado, R., and Berrendero, F. (2020). THC exposure during (3), 343–356. doi:10.1017/S1461145708009279 adolescence does not modify nicotine reinforcing effects and relapse in adult male Martín-García, E., Bourgoin, L., Cathala, A., Kasanetz, F., Mondesir, M., Gutiérrez- mice. Psychopharmacology 237 (3), 801–809. doi:10.1007/s00213-019-05416-8 Rodriguez, A., et al. (2016) Differential control of cocaine self-administration by Franklin, K., and Paxinos, G. (1997). The mouse brain in stereotaxic coordinates. GABAergic and glutamatergic CB1 cannabinoid receptors. Academic Press. Massachusetts, United States. Neuropsychopharmacology, 9, 2192–2205. doi:10.1038/npp.2015.351 Frontiers in Pharmacology 13 frontiersin.org Cajiao-Manrique et al. 10.3389/fphar.2023.1143365 Martín-García, E., Domingo-Rodriguez, L., and Maldonado, R. (2020). An operant Soria, G., Barbano, M. F., Maldonado, R., and Valverde, O. (2008) A reliable method conditioning model combined with a chemogenetic approach to study the neurobiology to study cue-priming-and stress-induced reinstatement of cocaine self-administration of food addiction in mice. Bio Protoc. 10, e3777. doi:10.21769/BioProtoc.3777 in mice. Psychopharmacology, 199:4, 593–603. doi:10.1007/s00213-008-1184-x Mendizábal, V., Zimmer, A., and Maldonado, R. (2006) Involvement of kappa/ Spencer, S., Neuhofer, D., Chioma, V. C., Garcia-Keller, C., Schwartz, D. J., Allen, N., dynorphin system in WIN 55,212-2 self-administration in mice. et al. (2018). A model of d9-tetrahydrocannabinol self-administration and Neuropsychopharmacology, 31, 1957–1966. doi doi:10.1038/sj.npp.1300957 reinstatement that alters synaptic plasticity in nucleus accumbens. Biol. Psychiatry 84 (8), 601–610. doi:10.1016/j.biopsych.2018.04.016 Panlilio, L. V., Goldberg, S. R., Gilman, J. P., Jufer, R., Cone, E. J., and Schindler, C. W. (1998). Effects of delivery rate and non-contingent infusion of cocaine on cocaine self- Stewart, J., and de Wit, H. (1987). “Reinstatement of drug-taking behavior as a administration in rhesus monkeys. Psychopharmacology 137 (3), 253–258. doi:10.1007/ method of assessing incentive motivational properties of drugs,” in Methods of assessing s002130050618 the reinforcing properties of abused drugs (New York: Springer-Verlag). Panlilio, L. V., and Justinova, Z. (2018). Preclinical studies of cannabinoid Tanda, G., and Goldberg, S. R. (2003). Cannabinoids: Reward, dependence, and reward, treatments for cannabis use disorder, and addiction-related effects of underlying neurochemical mechanisms - a review of recent preclinical data. cannabinoid exposure. Neuropsychopharmacology 43 (1), 116–141. doi:10.1038/ Psychopharmacology 169 (2), 115–134. doi:10.1007/s00213-003-1485-z npp.2017.193 Valjent, E., and Maldonado, R. (2000). A behavioural model to reveal place preference Panlilio, L. V., Thorndike, E. B., and Schindler, C. W. (2003). Reinstatement of to delta 9-tetrahydrocannabinol in mice. Psychopharmacology 147 (4), 436–438. doi:10. punishment suppressed opioid self-administration in rats: An alternative model of 1007/s002130050013 relapse to drug abuse. Psychopharmacology 168 (1-2), 229–235. doi:10.1007/s00213- Vallée, M., Vitiello, S., Bellocchio, L., Hébert-Chatelain, E., Monlezun, S., Martín- 002-1193-0 García, E., et al. (2014) Pregnenolone can protect the brain from cannabis intoxication. Peltier, R. L., Guerin, G. F., Dorairaj, N., and Goeders, N. E. (2001). Effects of saline Science: 343, 94–98. doi:10.1126/science.1243985 substitution on responding and plasma corticosterone in rats trained to self-administer Venniro, M., Caprioli, D., and Shaham, Y. (2016). Animal models of drug relapse and different doses of cocaine. J. Pharmacol. Exp. Ther. 299 (1), 114–120. craving: From drug-priming induced reinstatement to incubation of craving after Pertwee, R. G., Howlett, A. C., Abood, M. E., Alexander, S. P. H., Di Marzo, V., voluntary abstinence. Prog. Brain Res. 224, 25–52. doi:10.1016/bs.pbr.2015.08.004 Elphick, M. R., et al. (2010). International union of basic and clinical Pharmacology. Volkow, N. D., Michaelides, M., and Baler, R. (2019). The neuroscience of drug LXXIX. Cannabinoid receptors and their ligands: Beyond CB1 and CB2. Pharmacol. Rev. reward and addiction. Physiol. Rev. 99 (4), 2115–2140. doi:10.1152/physrev.00014.2018 62 (4), 588–631. doi:10.1124/pr.110.003004 Weafer, J., Mitchell, S. H., and Wit, H. (2014). Recent translational findings on impulsivity Piazza, P. V., and Deroche-Gamonet, V. (2013). A multistep general theory of in relation to drug abuse. Curr. Addict. Rep. 1 (4), 289–300. doi:10.1007/s40429-014-0035-6 transition to addiction. Psychopharmacology 229 (3), 387–413. doi:10.1007/s00213- 013-3224-4 Weiss, F. (2010). “Advances in animal models of relapse for addiction research,” in Advances in the neuroscience of addiction (Boca Raton (FL: CRC Press). Chapter 1, Roth, B. L. (2016). DREADDs for neuroscientists. Neuron 89 (4), 683–694. doi:10. PMID: 21656976. 1016/j.neuron.2016.01.040 Weiss, S. R. B., and Volkow, N. D. (2022). Coordinating cannabis data collection Shaham, Y., Shalev, U., Lu, L., de Wit, H., and Stewart, J. (2003). The reinstatement globally: Policy implications. Addiction 117 (6), 1520–1521. doi:10.1111/add.15751 model of drug relapse: History, methodology and major findings. Psychopharmacology 168, 3–20. doi:10.1007/s00213-002-1224-x Zehra, A., Burns, J., Liu, C. K., Manza, P., Wiers, C. E., Volkow, N. D., et al. (2018). Cannabis addiction and the brain: A review. J. Neuroimmune Pharmacol. 13 (4), Shalev, U., Grimm, J. W., and Shaham, Y. (2002). Neurobiology of relapse to heroin 438–452. doi:10.1007/s11481-018-9782-9 and cocaine seeking: A review. Pharmacol. Rev. 54 (1), 1–42. doi:10.1124/pr.54.1.1 Frontiers in Pharmacology 14 frontiersin.org