Somatic and anxiety-like behaviors in male and female rats during withdrawal from the non-selective cannabinoid agonist WIN 55,212-2

Abstract

Synthetic cannabinoids are associated with higher risk of dependence and more intense withdrawal symptoms than plant-derived Δ9-tetrahydrocannabinol (THC). Avoidance of withdrawal symptoms, including anxiogenic effects, can contribute to continued cannabinoid use. Adult male and female Long-Evans rats were given escalating doses of WIN 55,212-2 (WIN) via twice daily intrajugular infusions. Precipitated withdrawal was elicited with SR 141716A (rimonabant) 4 hours after the final infusion. Global withdrawal scores (GWS) were compiled by summing z-scores of observed somatic behaviors over a 30-minute period with locomotor activity simultaneously collected via beam breaks. Rimonabant precipitated withdrawal in female and male rats at 3 or 10 mg/kg, respectively, but the individual behaviors contributing to GWS were not identical. 3 mg/kg rimonabant did not impact locomotor behavior in females, but 10 mg/kg decreased locomotion in male controls. Spontaneous withdrawal observed between 6 to 96 hours after the final infusion was quantifiable up to 24 hours following WIN administration. Individual behaviors contributing to GWS varied by sex and time point. Males undergoing spontaneous withdrawal engaged in more locomotion than females undergoing withdrawal. Separate groups of rats were subjected to a battery of anxiety-like behavioral tests (elevated plus maze, open field test, and marble burying test) one or two weeks after WIN or vehicle infusions. At one week abstinence, sex-related effects were noted in marble burying and the open field test but were unrelated to drug treatment. At two weeks abstinence, females undergoing withdrawal spent more time grooming during marble burying and performed more marble manipulations than their male counterparts. WIN infusions did not impact estrous cycling, and GWS scores were not correlated with estrous at withdrawal. Collectively, these results show qualitative sex differences in behaviors contributing to the behavioral experience of cannabinoid withdrawal supporting clinical findings from THC.

1. Introduction:

Cannabinoids are the 3^rd most commonly used drug class in the world behind alcohol and nicotine with over 200 million users (Bonnet and Preuss, 2017; Herrmann et al., 2015). Frequent users of natural cannabis are more likely to try synthetic cannabinoid receptor agonists (SCRAs) (Mathews et al., 2019). The use of SCRAs, compared to phytocannabinoids, is associated with more rapidly developing use disorder, more severe withdrawal, and other negative health outcomes (Cooper, 2016; Grigg et al., 2019; Van Hout and Hearne, 2017). CB1 receptor activation is linked to the psychotropic effects of cannabinoids. WIN 55212-2 (WIN) is a potent aminoalkylindole synthesized for use in cannabinoid research (Compton et al., 1992; Järbe et al., 2011). It has been used for decades as a “prototypical” mixed selectivity full agonist that activates the cannabinoid 1 (CB1) and cannabinoid 2 (CB2) receptors with binding affinities in the low nanomolar range (Dhopeshwarkar and Mackie, 2016; Patel et al., 2005; Pertwee, 2008, 2005, 1997). As a full agonist, WIN is much more efficacious at activating cannabinoid receptors than the partial agonist Δ9-tetrahydrocannabinol (THC) (Kuster et al., 1993; McPartland et al., 2007; Patel et al., 2020). In rats, the binding affinity at CB1 receptors is 10-fold greater for WIN than THC (McPartland et al., 2007; Pertwee, 2008). WIN is directly related to other aminoalkylindoles, such as JWH-018 or AM678, that make up the major psychotropic cannabinoids in herbal concoctions called “Spice” and “K2” (Järbe et al., 2011; Walsh and Andersen, 2020). These chemicals applied to herbal mixtures are used as substitutes for THC to evade drug tests. AM678 and JWH018 have similar binding affinities for the CB1 receptor, and both have higher affinity relative to THC and WIN (Aung et al., 2000; Järbe et al., 2011; Vardakou et al., 2010). Preclinical studies of WIN dependence model human use of aminoalkylindole SCRAs and help understand withdrawal contributions.

Cannabis withdrawal syndrome (CWS) is a clinically significant response that occurs in about half of regular cannabis users (Connor et al., 2022). CWS often manifests as affective changes in mood and anxiety or changes in sleep in addition to physical symptoms (Bonnet and Preuss, 2017; Connor et al., 2022; Levin et al., 2010). Prolonged and heavy cannabis use results in stronger CWS, linking the severity of CWS with cannabinoid dependence (Bonnet and Preuss, 2017; Martin et al., 2023). Men are more likely to use cannabis or SCRAs for recreational and medicinal purposes, develop cannabinoid dependence, and initiate use at a younger age, but the gender gap is narrowing (Cuttler et al., 2016; Farmer et al., 2015; Greaves and Hemsing, 2020; Johnson et al., 2015; Legleye et al., 2014). Yet, like other drugs of abuse, women progress more rapidly from first use to cannabinoid dependence (Khan et al., 2013; Ridenour et al., 2006). Withdrawal from SCRAs is similar to, but more severe, when compared to cannabis withdrawal (Nacca et al., 2013). Women generally present with different cannabinoid withdrawal symptoms where sex differences are reported, although some studies suggest the presentation of more severe withdrawal (Bonnet and Preuss, 2017; Connor et al., 2022; Herrmann et al., 2015; Levin et al., 2010; Sherman et al., 2017).

In rodents, withdrawal can be precipitated with CB1 receptor antagonists like SR 141716 (rimonabant) which produce intense somatic symptoms (Aceto et al., 2001; Kesner and Lovinger, 2021; Lichtman and Martin, 2002). Where direct sex comparisons have been made following precipitated THC withdrawal, more severe withdrawal symptoms occur in female rodents (Marusich et al., 2014). Spontaneous withdrawal is also observed following abrupt cessation of repeated doses of THC or SCRAs (Aceto et al., 2001; Trexler et al., 2019). An increase in wet dog shakes and facial rubs was observed for at least 24 hours following termination of SCRA administration in male rats, but additional behaviors were not systematically quantified (Aceto et al., 2001). Moreover, the time course of withdrawal in female rats remains unknown. Measurements of spontaneous withdrawal undoubtedly have more face validity along with evaluation of behaviors associated with negative affect. Stress-sensitive measures, such as the elevated plus maze, open field, and marble burying, are beginning to be used to test exploratory strategies and anxiety-like behaviors in rodents during cannabinoid withdrawal (Harte-Hargrove and Dow-Edwards, 2012; Trexler et al., 2018). In adolescent rats, spontaneous withdrawal from THC produced greater anxiety-like behavior (e.g., reduced time in open arms) on the elevated plus maze in female rats while males showed the opposite behavior when compared to vehicle-exposed controls (Harte-Hargrove and Dow-Edwards, 2012). Thus, behavioral measures of emotional reactivity in rodent models add insight into the course of cannabinoid withdrawal.

The current studies systematically evaluating precipitated and spontaneous withdrawal from WIN in adult male and female rats were conducted to facilitate future research delving into the contribution of sex and/or withdrawal on cannabinoid-motivated behavior and underlying neural correlates. To understand how cannabinoid withdrawal contributes to the experience of cannabinoid use disorder it is vital to get a complete picture of the withdrawal syndrome in both sexes. To maximize the probability of observing quantifiable withdrawal, we used the SCRA WIN as it produces robust withdrawal symptoms in males in the presence and absence of rimonabant. Studying both spontaneous and rimonabant-precipitated withdrawal is important because the behavioral expression of withdrawal may be divergent. Understanding the presentation and time course of withdrawal will inform the interpretation of future studies.

2. Materials and Methods:

2.1. Subjects

185 adult male and female Long-Evans rats (9-10 weeks) were obtained from Inotiv. Rats were double housed in individually ventilated cages in temperature (70-71°F) and humidity (38-46%) controlled facilities under a 14/10 light-dark cycle. Standard chow and water were available ad libitum. Body weight was recorded throughout drug treatment and withdrawal observations (Figure S2 and S3). Procedures were preapproved by the Institutional Animal Care and Use Committee (IACUC) of the University of Minnesota.

2.2. Intravenous surgery

Rats were anesthetized with isoflurane (2-5%), and self-fabricated catheters were implanted as previously described (Spencer et al., 2018). The catheters consisted of silastic tubing connected to a modified 22-gauge cannula (Plastics One, Roanoke, VA) embedded in dental cement and attached to surgical mesh (Atrium, Merrimack, NH). Post-operative care was provided for 72 hrs. Catheter patency was confirmed by loss of muscle tone with intravenous (i.v.) administration of 0.1 mL xylazine (5 mg/mL).

2.3. Drugs

WIN 55,212-2 (WIN) (MedChem Express, Monmouth Junction, NJ) was suspended in <1% dimethyl sulfoxide (DMSO) and saline (0.9%). Vehicle consisted of 1% DMSO in saline. Rimonabant (SR141716A; APExBIO, Houston, TX) was dissolved in DMSO and diluted to 30% DMSO in saline.

2.4. Somatic withdrawal characterization

Withdrawal observations were made in an arena (45.7 cm x 22.9 cm x 30.5 cm) with mirrored paper attached to the outside of three of the four walls (Med Associates Inc, Fairfax, VT). Sessions were scored in real time by a trained observer. Overt behavioral signs measured included tallied and binned behaviors (Aceto et al., 2001; Moranta et al., 2009). Tallied behaviors included: wet-dog shakes, fore paw tremors, writhes, retropulsion, hind leg scratches, arched back, head/face twitch, and facial rubs. Binned behaviors, marked as present or absent over 5-min bins, included: ptosis, piloerection, genital licks, teeth chattering, and sniffing. The number of bins in which a positive response was marked were summed for a maximum score of 6 for each endpoint. Supplementary appendix 1 provides a full description of how each of these behavioral signs was scored. During somatic withdrawal, locomotor activity was collected using infrared beam tracking (Med Associates, Activity monitor 5). Time spent engaging in ambulatory and stereotyped locomotor behavior was summed to report the total time spent moving. Stereotyped locomotion was defined as movement within a 5 x 5 cm square; movement of 3+ beam breaks outside the square was counted as ambulatory.

2.5. Experiment 1: Precipitated withdrawal

A total of 70 rats (25 males, 45 females) were used in experiment 1. Sample breakdown for each group is as follows: handling control = 8 males and 8 females; vehicle + 10 mg/kg rimonabant = 8 males; vehicle + 3 mg/kg rimonabant = 9 females; WIN + 10 mg/kg rimonabant = 9 males; WIN + 3 mg/kg rimonabant =11 females; vehicle + 10 mg/kg rimonabant = 8 females (Figure S1); WIN + 10 mg/kg rimonabant = 9 females (Figure S1). Five days of escalating doses of WIN via i.v. infusion was given twice per day at 8:30 and 16:30: Day 1 (0.2, 0.2 mg/kg); Day 2 (0.4, 0.4 mg/kg); Day 3 (0.6, 0.6 mg/kg); Day 4 (0.8, 0.8 mg/kg); Day 5 (0.8 mg/kg and withdrawal) following previously published experiments characterizing WIN withdrawal adapted for i.v. administration (Aceto et al., 2001; Gomez et al., 2021). Intravenous WIN more closely models the pharmacokinetics and pharmacodynamics of inhaled synthetic cannabinoids in comparison to i.p or s.c. delivery and is the route of administration used in self-administration experiments (Lucas et al., 2018). Infusion times were chosen to accommodate the light dark cycle of the vivarium and to avoid the development of the light schedule as a cue. Body weight and estrous samples were taken before each infusion. Vehicle controls received i.v. infusions on the same schedule. A visualization of this time course can be seen in Figure 1A. Handling controls were included to isolate the contribution of rimonabant to assessed behaviors. These rats were handled twice daily and received only an IP vehicle injection prior to the observation session.

Figure 1.

A) Time course of WIN or vehicle administration over the course of dependence and withdrawal in precipitated withdrawal experiments. Composite global withdrawal scores in male (B) and female (C) rats following precipitation of withdrawal with rimonabant (Rimo). B) Administration of 10 mg/kg rimonabant significantly increased the global score of male rats undergoing withdrawal from WIN relative to handled rats receiving only vehicle prior to withdrawal observation (***p<0.001) and relative to rats receiving chronic infusions of vehicle (** p<0.01). C) Administration of 3 mg/kg rimonabant increased the global score of female rats undergoing withdrawal from WIN relative to handled rats receiving only vehicle prior to withdrawal observation (***p<0.001) and relative to rats receiving chronic infusions of vehicle (**** p<0.0001). Time spent locomoting in male (D) and female (E) rats during rimonabant precipitated withdrawal. D) Administration of 10 mg/kg rimonabant reduced time spent moving in vehicle (***p<0.01) but not WIN-treated males compared to handled controls. E) Administration of 3 mg/kg of rimonabant had no effect on total time spent locomoting in female rats. F) Proportion of samples in estrous stages proestrus (P), estrus (E), metestrus (M), and diestrus (D) in female rats during twice daily handling or infusions of either vehicle or WIN 55,212; experimental procedures start in the morning. G) Global withdrawal scores in rats during precipitated withdrawal by hormone state (P/E v M/D).

For precipitated withdrawal, rimonabant was administered intraperitoneally (i.p.) four hours after the final infusion at 12:30 (3 mg/kg in females or 10 mg/kg in males; 10 mg/kg data in females shown in figure S1) (Aceto et al., 2001). Immediately prior to the injection of rimonabant or vehicle, body weight and temperature data (measured via rectal probe) were collected to calculate percent weight and temperature change. Somatic withdrawal behavior was scored in a single 30-minute withdrawal observation period immediately following rimonabant injection. 10 mg/kg rimonabant in females increased somatic withdrawal in vehicle and WIN females (Figure S1). Lowering the dose of rimonabant by a half log step in the females mitigated this problem. The 3 mg/kg dose of rimonabant precipitated withdrawal from WIN but did not increase somatic behaviors in vehicle rats similar to 10 mg/kg in males.

2.6. Experiment 2: Spontaneous withdrawal

A total of 61 male and female rats were used in experiment 2. Sample breakdown is as follows: Vehicle = 15 male and 16 females; WIN = 15 male and 15 females. To characterize somatic behavior across the course of spontaneous withdrawal, the schedule of infusions was shifted to start and end in the afternoon (16:30). Infusions occurred at 16:30 and 8:30 every day: Day 1 (0.2, 0.2 mg/kg); Day 2 (0.4, 0.4 mg/kg); Day 3 (0.6, 0.6 mg/kg); Day 4 (0.8, 0.8 mg/kg); Day 5 (0.8 mg/kg). The shift in administration schedule occurred in order to shift the time of withdrawal observations so that the 16-hour observation time point did not occur at 00:30 and to maintain consistency of infusion timing at 8:30 and 16:30. This schedule of drug administration caused the 6-hour observation time point to be in the late evening (22:30), the 16-hour time point to be the next morning (8:30), and the 24-hour time point to be the following afternoon (16:30). Thereafter, all withdrawal observations (48, 72, and 96 hour) were in the afternoon (16:30). For spontaneous withdrawal, 30-minute withdrawal observations occurred at 6, 16, 24, 48, 72 and 96 hours after the final infusion as described in 2.4. A visualization of this time course can be seen in Figure 2A. Body weight and temperature were taken immediately prior to and following withdrawal observations.

Figure 2.

A) Timeline of spontaneous withdrawal experiments. B) Global withdrawal scores in male and female rats over the course of spontaneous withdrawal. WIN administration and subsequent withdrawal significantly increase global scores at 6, 16, and 24 hrs following final administration of drug or vehicle. *p<0.05, **p<0.01 compared to vehicle-treated rats. C) Total time spent in motion in male and female rats undergoing spontaneous withdrawal from either WIN or vehicle. Males undergoing WIN withdrawal moved significantly more than females experiencing WIN withdrawal. *p<0.05 male WIN versus female WIN at indicated time point. D) Proportion of samples in estrous stages proestrus (P), estrus (E), metestrus (M), and diestrus (D) in female rats during twice daily infusions of vehicle or WIN 55,212. E) Proportion of samples in estrous stages P, E, M, and D in female rats undergoing withdrawal from twice daily infusions of vehicle or WIN 55,212.

2.7. Experiment 3: Non somatic characterization of somatic withdrawal

2.7.1. Experimental design

A total of 54 male and female rats were used in experiment 3. Sample breakdown is as follows: 1-week withdrawal: Vehicle = 7 males and 9 females; WIN = 7 males and 7 females; 2-week withdrawal: Vehicle = 6 males and 6 females; WIN = 6 males and 6 females. Daily infusions occurred at 8:00 and 16:00 every day: Day 1 (0.2, 0.2 mg/kg); Day 2 (0.4, 0.4 mg/kg); Day 3 (0.6, 0.6 mg/kg); Day 4 (0.8, 0.8 mg/kg); Day 5 (0.8 mg/kg). Rats underwent spontaneous withdrawal from WIN or vehicle in the home cage following the 4.5-day drug administration. Rats underwent a battery of tests (open field, marble burying test, elevated plus maze) within a single test day in the order described. Behavioral testing took place in the morning between 9:00 and 11:00, either one week (days 5-7) or two weeks (days 14-16) after the final infusion. A visual representation of the timeline can be seen in Figure 3A. Separate subjects were tested at each withdrawal time point. Estrous samples were taken during each administration point of WIN or vehicle, as well as on test day. All rats underwent the same order of behavioral experiments once and did not undergo multiple test days to avoid habituation to the behavioral tests.

Figure 3.

A) Spontaneous withdrawal timeline depicting the 5-day infusions of either escalating doses of WIN 55,212-2 (WIN) or vehicle (Veh) (1% DMSO in saline) (Created with Biorender.com). Rats underwent anxiety-like behavior screening 5-7 days after the final infusion (withdrawal period). Summary data comparing all vehicle-treated rats (white bars) compared to WIN-treated rats (black bars) is depicted alongside data separated by treatment (open versus closed symbols for Veh versus WIN) and sex (blue vs pink bars for male’s vs females). B) There were no treatment effects or sex differences in marble manipulations. C) There were no treatment effects or sex differences for marbles buried 100%. D) There were no treatment effects for center time in the open field. Male rats spent more time in the center of the open field test compared to female rats; **p<0.01. E) There were no treatment effects or sex differences for total distance traveled during the open field test. F, G) There were no treatment effects or sex differences time spent in the open arms (F) and number of open arm entries (G) during the elevated plus maze.

2.7.2. Behavioral battery

Animals habituated to the experiment room for one hour prior to testing. The open field test was performed using an open arena (45.7 cm x 45.7 cm x 30.5 cm) (MedAssociates) under direct light with locomotion tracking using infrared beam breaks and Activity Monitor 5 Software (MedAssociates). The peripheral and central zones each encompassed 50% of the box. The rat was placed in the center of the arena and monitored for a total of 10 minutes (Gomes et al., 2011).

The marble burying test was performed using a rectangular box (20in x 16in x 8in; Tecniplast) with a 3 cm corn cob bedding covered floor in low, indirect light conditions (Gomes et al., 2011). 30 minutes prior to testing, animals were exposed to the empty box for 5 minutes to remove possible confounds of novel box exploration. During the 30-minute test, rats were placed in the center of the box with 20 glass marbles spread evenly across the floor. A trained researcher recorded marble manipulations, defined as a rodent picking up and handling the marble, and counted marbles 100% buried (Gomes et al., 2011). Somatic behaviors tallied during experimentation included: rears, hind leg scratches, fecal boli, wet dog shakes, and grooms (Aceto et al., 2001).

The elevated plus maze apparatus (MedAssociates) consisted of two opposing open arms (each 10.16 × 50.8 cm) with no rims or walls perpendicular to two opposing closed arms with 40 cm walls elevated on a center base 120 cm above the floor. The experiment was conducted for 5 minutes in a brightly overhead lit room (Handley and Mithani, 1984). The rat was placed in the maze center facing an open arm away from the researcher. A camera filmed the experiment from above. Time spent in each arm and the number of arm entrances was manually determined from the recording using a stop timer and tallying entrances.

2.8. Determination of estrous cycle stage

Estrous cycle stage was determined cytologically following vaginal lavage. Vaginal lavage samples were taken prior to each infusion (08:00/8:30 and 16:00/16:30) and following withdrawal time points. A small amount of saline was used to flush the vaginal canal and pipetted onto a clean microscope slide (Ajayi and Akhigbe, 2020; Goldman et al., 2007). Air-dried samples were stained with a differential staining kit (Newcomer Supply, Middleton, WI). A compound microscope (Olympus MVX10) was used to identify estrous cycle phase (Ajayi and Akhigbe, 2020; Goldman et al., 2007). Proestrus was determined by predominant (≥75% or more) nucleated epithelial cells. Estrus was determined by a preponderance of cornified epithelial cells. Metestrus was determined by scattered cells consisting of nucleated or cornified epithelial cells and leukocytes. Diestrus was recognized by a relative lack of cells and/or predominance of leukocytes (Ajayi and Akhigbe, 2020; Marusich et al., 2014).

2.9. Statistical analysis

Graph pad prism (version 9, San Diego, CA) was used to generate figures, and final statistical analysis was conducted using SPSS (Version 28, IBM). Timelines were created with BioRender (Figures 1A, 2A, 3A, and 4A). Data presented are mean ± standard of the mean (SEM). A global withdrawal score (GWS), including all tallied and binned somatic behaviors, was computed by converting variables to z-scores and summing the z-scores to create a composite score. Composite z-scores were analyzed separately for males and female undergoing precipitated withdrawal (experiment 1). One-way analysis of variance (ANOVA) (factors: handled, vehicle, WIN) with Tukey’s post hoc testing were conducted for behavioral endpoints. Where violations of normality occurred, Brown Forsyth correction is used and indicated with F*. Where asymptotic distribution occurred in individual behavioral endpoints Kruskal-Wallis one-way ANOVA on ranks was conducted followed by pairwise comparisons. The effect of WIN withdrawal on locomotion was assessed using one-way ANOVA followed by Tukey’s post hoc testing.

Figure 4.

A) Spontaneous withdrawal timeline depicting the 5-day infusions of either escalating doses of WIN55212 (WIN) or vehicle (Veh) (1% DMSO in saline) (Created with Biorender.com). Rats underwent anxiety-like behavioral screening 14-16 days after the final infusion (withdrawal period) Summary data comparing all vehicle-treated rats (white bars) compared to WIN-treated rats (black bars) is depicted alongside data separated by treatment (open versus closed symbols for Veh versus WIN) and sex (blue vs pink bars for males vs females). B) There was a main effect of both sex and treatment with marble manipulations;* p<0.05 compared to vehicle controls;**p<0.01 compared to the opposite sex. C) There were no WIN treatment effects or sex differences for marbles buried 100%. D) There were no treatment effects for center time in the open field. Male rats spent more time in the center of the open field compared to female rats; **p<0.01 compared to the opposite sex. There were no treatment effects or sex differences for total distance traveled during the open field test. Neither treatment nor sex impacted the amount of time spent in the open arms (F) and the number of entries into the open arms of the elevated plus maze (G).

For spontaneous withdrawal (experiment 2), repeated measures mixed effects ANOVA (factors: sex x drug x time) was conducted for composite global scores, individual behavioral endpoints, and locomotion. Pairwise estimated marginal means with Bonferroni’s correction were used to identify where interactions between time and sex and/or WIN treatment occurred. Where sphericity was violated, Greenhouse-Geisser correction was used.

The effect of WIN withdrawal on non-somatic behavioral endpoints (experiment 3) were collected separately for one- and two-week withdrawal periods. Behavioral endpoints from marble burying test, open field test, and elevated plus maze were assessed using two-way between subjects ANOVA (factors: sex x drug treatment).

Estrous data collected during infusions was converted into proportion of each stage across the total number of samples (total=9 over 4.5 days) and compared across groups using one-way multivariate ANOVA (MANOVA) (experiment 1-3). Proportions of each stage of estrous during spontaneous withdrawal out of total samples (total=5) were also assessed using one-way MANOVA (experiment 2). Partial correlations across treatments were run with global score using the estrous stage at withdrawal and at final infusion. Staging was collapsed into high hormone (proestrus/estrus: P/E) and low hormone (diestrus/metestrus: D/M) to increase power, and correlations were performed between hormone state at withdrawal or final infusion and global score. Distribution of samples in each stage of estrous at 7 days or 14 days after administration was compared using Chi squared test (experiment 3). Body weight was converted into percent weight change from baseline to account for weight differences between adult male and female rats. The effect of WIN withdrawal on body weight and body temperature was investigated using repeated measures mixed-effects ANOVA followed by Bonferroni’s test for estimated marginal means.

3. Results

3.1. Precipitated Withdrawal: Somatic

Rats in precipitated withdrawal experiments received i.v. infusions following the timeline depicted in Figure 1A. Somatic withdrawal behaviors were assessed independently (Table 1) and transformed into z-scores to be summed to report a composite score (GWS) (Figure 1B, 1C). Precipitating withdrawal from WIN with 10 mg/kg rimonabant increased GWS of male rats [F(2, 24)= 17.296 p<0.0001] compared to vehicle control rats (p<0.01) and handled controls not receiving rimonabant (p<0.0001) (Figure 1B). Importantly, rimonabant did not increase GWS in vehicle control rats relative to handling controls (p=0.239). In female rats, administration of 10 mg/kg rimonabant increased GWS in vehicle-treated rats [t(14)=3.574, p<0.01], producing withdrawal-like behaviors in the absence of cannabinoid administration (Supplementary Figure 1) and obscuring the withdrawal phenotype. Therefore, we repeated this characterization with a lower dose of rimonabant (3 mg/kg) in female rats. Rimonabant at 3 mg/kg elevated GWS in female rats treated with WIN [F(2,27)=11.189, p<0.001] (Figure 1C) compared to handled (p<0.001) and vehicle control rats (p<0.01) without increasing GWS in vehicle-treated rats relative to handled controls (p=0.725).

Table 1.

Average and SEM of each of the individually counted (wet dog shakes, tremors, writhes, retropulsion, hind leg scratch, facial rubs, head/ face twitch, and arched back posture) as well as the average and SEM of the mean of positive bins (ptosis, piloerection, genital licks, sniffing, and teeth chatter) for all male and female rats undergoing withdrawal observations. *p<0.05, **p<0.01 compared to vehicle control of the same sex; #p<0.05, ##p<0.01 compared to handled controls of the same sex.

	Males			Females
		Rimonabant 10 mg/kg			Rimonabant 3 mg/kg
	Handled	Vehicle	WIN 55,212-2	Handled	Vehicle	WIN 55,212-2
Tallied	M (SEM)	M (SEM)	M (SEM)	M (SEM)	M (SEM)	M (SEM)
Wet dog shakes	2.7 (0.4)	3.4 (1.2)	5.9 (1.2)	1.6 (0.9)	2.4 (1.1)	4.7 (2.3)
Tremors	2.3 (1.1)	5.6 (1.9)	12.2 (3.7)#	1.3 (0.7)	3.2 (0.9)	7.0 (1.2)* ##
Writhes	0 (0)	2.5 (1.3)	3.1 (1.2)#	0 (0)	0.7 (0.4)	3.6 (1.1)** ##
Retropulsion	0 (0)	0.6 (0.4)	3.2 (0.7)* ##	0.3 (0.3)	1.0 (0.4)	2.2 (0.7)#
Hind leg scratch	2.9 (1.1)	11.2 (6.4)	17.0 (4.8)	6.1 (1.7)	3.4 (0.8)	4.9 (1.6)
Facial rubs	0.4 (0.3)	2.3 (1.6)	3.1 (1.5)	0.8 (0.4)	0.2 (0.2)	2.5 (0.6)** #
Head/face twitch	0 (0)	0.5 (0.3)	1.4 (0.4)##	0.3 (0.2)	0.4 (0.3)	0.6 (0.2)
Arched back posture	0 (0)	0.5 (0.3)	1.4 (0.7)	0.06 (0.06)	0.7 (0.3)	1.8 (0.8)#
Binned
Ptosis	0 (0)	0 (0.0)	2.3 (0.9)* #	0.1 (0.1)	0 (0)	0.4 (0.3)
Piloerection	0.5 (0.4)	0.9 (0.4)	1.9 (0.8)	0 (0)	0.3 (0.3)	2.3 (0.7)* #
Genital licks	0.8 (0.3)	1.0 (0.3)	0.8 (0.4)	0.3 (0.2)	0.7 (0.5)	1.2 (0.3)
Sniffing	1 (0.6)	0 (0)	0.6 (0.3)	1.5 (0.8)	0.9 (0.4)	0.7 (0.2)
Teeth chatter	0.5 (0.3)	0.9 (0.3)	2.6 (0.4)** ##	0.5 (0.3)	1.1 (0.3)	1.3 (0.2)

Precipitating cannabinoid withdrawal with rimonabant did not change instances of wet dog shakes, hind leg scratches, facial rubs, arched back, piloerection, genital licks, or sniffing (Table 1). Fore paw tremors were increased in male rats undergoing withdrawal from WIN [F*(2, 13.117)=4.179, p<0.05] relative to handled controls (p<0.05). Teeth chatter positive bins increased in male rats undergoing withdrawal [F*(2, 18.566)=10.332, p<0.001] compared to handled controls (p<0.01) and vehicle-treated rats (p<0.01). Due to non-gaussian distribution of retropulsion, writhing, head twitches, and ptosis, Kruskal-Wallis independent samples test was used to assess the effect of WIN withdrawal on these behaviors. Retropulsion increased in males experiencing cannabinoid withdrawal from WIN [H(2)=14.553, p<0.001] compared to handled controls (p<0.001) and vehicle controls (P<0.05). Incidences of writhing increased in rats experiencing WIN withdrawal [H(2)=7.436, p<0.05] compared to handled controls (p<0.05). Head twitches increased in male rats undergoing WIN withdrawal [H(2)=8.612, p<0.05] compared to handled controls (p<0.01). Only male rats in the WIN group displayed ptosis [H(2)=8.058, p<0.05].

In female rats, wet dog shakes, hind leg scratches, head twitches, ptosis, gentle licks, sniffing, and teeth chatter were unchanged by withdrawal (Table 1). WIN withdrawal elevated fore paw tremors [F*(2, 21.7772)=10.015, p<0.001] compared to handled (P<0.01) and vehicle controls (p<0.05). Female rats engaged in significantly more facial rubs [F*(2, 17.033)=9.138, p<0.01] compared to vehicle (p<0.01) and handled controls (p<0.05). Due to the non-normal distribution of retropulsion, arched back, writhing, and piloerection in female rats, Kruskal-Wallis test was used to assess the effect of WIN withdrawal on these behaviors. WIN-treated rats displayed more incidences of retropulsion [H(2)=6.361, p<0.05] and arched back posture [H(2)=6.265, p<0.05] compared to handled controls (p<0.05). Writhes were increased in female rats following WIN withdrawal [H(2)=13.113, p<0.001] compared to both vehicle (p<0.05) and handled rats (p<0.01). Cannabinoid withdrawal increased the average number of bins in which piloerection was marked as present [H(2)=10.578, p<0.01] compared to vehicle treated (p<0.05) and handled controls (p<0.01).

Locomotor behavior was simultaneously measured during observation of GWS. 10 mg/kg rimonabant reduced total locomotor time [F(2,24)=4.234, p<0.05] in vehicle-treated male rats (p<0.05), but not WIN-treated male rats (p=.331) relative to handled controls (Figure 1D). Conversely, 3 mg/kg rimonabant did not significantly modify locomotor behavior in female rats [total locomotor time: F(2,27)=0.004, p=0.996] (Figure 1E).

The proportion of samples in each stage of estrous over the course of drug administration was calculated out of the total samples collected prior to each infusion. One-way MANOVA indicated that infusions of vehicle or WIN did not change the proportions of samples in each stage of estrous over the course of four and a half days [F(6,46)=0.344, p=0.910, Wilk’s Lambda=0.916] (Figure 1F). Global score was not correlated with estrous stage at withdrawal [R(23)=0.045, p=0.860] or with hormone status at withdrawal [R(23)=−0.280, p=0.175] when controlling for the effect of treatment (Figure 1E).

3.2. Spontaneous withdrawal: Somatic

Rats in spontaneous withdrawal experiments received i.v. infusions of WIN or vehicle and behaviors resulting from abrupt cessation of WIN infusions were assessed independently (Table 2) and as composite scores (Figure 2B) following the timeline shown in Figure 1A. At the first three timepoints (6, 16, and 24 hours), GWS was higher in male and female rats undergoing withdrawal from WIN [(drug x time) F(5, 75)=2.934 p<0.05] with a main effect of drug across the data set [F(1,15)=12.934, p<0.01] (Figure 2B).

Table 2.

Data presented are average counts and SEM of wet dog shakes, tremors, writhes, retropulsion, hind leg scratch , face rubs, head twitch, and arched back posture as well as the average number of present bins of ptosis, piloerection, genital licks, sniffing and teeth chatter during each withdrawal observations at 6, 14, 24, 48, 72, and 96 hrs following abrupt cessation of infusions of either WIN or vehicle (Veh) in male (M) and female (F) rats. # = main effect of drug, $= main effect of sex, += main effect of time. Symbols indicated at means indicate significant effect of sex ($), drug (#), or responses at 6 hr post-infusion of the same sex and drug condition (+) at p<0.05.

	M		F		M		F		M		F		M		F		M		F		M		F
	Veh	WIN	Veh	WIN	Veh	WIN	Veh	WIN	Veh	WIN	Veh	WIN	Veh	WIN	Veh	WIN	Veh	WIN	Veh	WIN	Veh	WIN	Veh	WIN	ANOVA
	Mean (SEM)		Mean (SEM)		Mean (SEM)		Mean (SEM)		Mean (SEM)		Mean (SEM)		Mean (SEM)		Mean (SEM)		Mean (SEM)		Mean (SEM)		Mean (SEM)		Mean (SEM)
Tallied
Wet dog shakes	2.8 (1.1)	5.5 (1.3)	4 (1.4)	2.5 (0.7)	3.5$ (0.7)	3$ (0.8)	1 (0.5)	2.1 (0.8)	4$ (0.7)	5$ (1)	1.3 (0.3)	2.2 (0.6)	3.1 (1.2)	6.1#$ (0.8)	1.9 (0.6)	1.4 (0.9)	2.6 (0.9)	4.4 (1.1)	1.9 (0.8)	2.1 (1.5)	1.6 (0.7)	6.1 (1.6)	1.5 (0.4)	1.5 (0.9)	$, $x#
Tremors	2.1 (0.7)	5.6 (1.6)	3.7 (0.9)	7.6 (1.5)	2.9 (0.6)	5.9 (1.5)	2.8 (0.6)	6.7 (1.8)	1.6 (0.6)	5.2 (1.5)	2.7 (0.8)	3.1 (0.8)	2 (0.8)	5.4 (1.3)	2.1 (0.6)	3 (1.3)	0.8 (1.2)	3.4 (0.7)	2.6 (1.2)	1.9 (0.4)	1.4 (0.8)	1.7 (0.7)	2 (0.7)	1.6 (0.4)	--
Writhes	0.3 (0.1)	1.3 (0.5)	0.6 (0.2)	2.5 (0.7)	0.2 (0.1)	1.4 (0.8)	0.3 (0.1)	0.8 (0.4)	0.6 (0.2)	1 (0.4)	0.4 (0.1)	1.6 (0.8)	0.9 (0.4)	0.9 (0.3)	0.7 (0.4)	1.3 (0.7)	0.4 (0.3)	0.7 (0.4)	0.3 (0.3)	0.4 (0.2)	0.8 (0.4)	0.8 (0.3)	0.9 (0.5)	0.8 (0.4)	--
Retropulsion	0.6 (0.3)	3.6# (0.6)	0.9+ (0.4)	3.5# (1.2)	0.5 (0.3)	1.4# (0.3)	1 (0.5)	2.4# (0.5)	0.5 (0.1)	1 (0.3)	1.1 (0.4)	0.6 (0.4)	0.3 (0.3)	1.9# (0.7)	0.1 (0.1)	1.3# (0.5)	0.4+ (0.2)	1.6+ (0.8)	0.7+ (0.3)	0.6+ (0.3)	0.2 (0.2)	1.8 (0.9)	0.7 (0.4)	0.4 (0.2)	+,#
Hind leg scratches	1.2 (0.5)	2.6 (1)	2.3 (0.7)	2.5 (1)	2.5 (1.1)	2.4 (1.2)	2 (0.8)	1.9 (0.6)	4 (1.3)	4.7 (1.2)	1.7 (0.5)	4.4 (0.9)	5.6 (2)	5.8 (1.7)	5.3 (2.2)	1.4 (0.3)	4.8 (1.9)	5.4 (1.6)	2.6 (1.6)	8.4 (4.7)	2.9 (2)	6.3 (2.5)	6.2 (2.3)	6.5 (4)	--
Face rubs	1.5 (0.8)	4.4# (1.1)	1.4 (0.5)	3.7# (0.7)	0.3+ (0.2)	1.4 (0.6)	0.3+ (0.2)	1.2 (0.4)	0.5+ (0.2)	1.6#+ (0.5)	0.4+ (0.2)	0.6#+ (0.4)	0.6+ (0.3)	0.9+ (0.5)	0.06+ (0.06)	0.4+ (0.3)	0.1+ (0.1)	0.8+ (0.3)	0.2+ (0.2)	0.2+ (0.1)	0.8+ (0.4)	0.9+ (0.4)	0.1+ (0.1)	1.8+ (1)	#, +x#
Head twitch	0.3 (0.2)	0.8 (0.5)	0.5 (0.2)	1.3 (0.6)	0.1 (0.1)	0.7 (0.5)	0 (0)	1.6 (1.4)	0.3 (0.1)	0.3 (0.2)	0 (0)	0.6 (0.3)	0.1 (0.1)	0.2 (0.2)	0.2 (0.1)	0.1 (0.1)	0.1 (0.1)	0.6 (0.2)	0.4 (0.2)	0.1 (0.1)	0.1 (0.1)	0.6 (0.2)	0.1 (0.1)	0.4 (0.3)	--
Arched back	0.2 (0.1)	1# (0.3)	0.2 (0.2)	1.2# (0.4)	0.1 (0.1)	0.7# (0.3)	0.1 (0.1)	1.1# (0.3)	0 (0)	0.5# (0.2)	0.04 (0.04)	0.6# (0.4)	0.1 (0.1)	0.2 (0.2)	0.1 (0.1)	0.1 (0.1)	0.2 (0.2)	0 (0)	0.3 (0.4)	0.4 (0.3)	0.3 (0.2)	0.1 (0.1)	0.2 (0.2)	0.4 (0.3)	$,#
Binned
Ptosis	0 (0)	0.4 (0.2)	0 (0)	0.2 (0.1)	0.3 (0.1)	0.05 (0.02)	0 (0)	0.8 (0.5)	0.3 (0.1)	0.7 (0.3)	0.4 (0.4)	0.7 (0.2)	0.1 (0.05)	0.04 (0.02)	0.02 (0.02)	0.06 (0.04)	0.1 (0.06)	0.09 (0.05)	0.07 (0.04)	0.08 (0.06)	0.09 (0.05)	0.1 (0.07)	0.04 (0.02)	0.04 (0.03)	--
Piloerection	0.7 (0.4)	0.8 (0.5)	0.8 (0.6)	1.1 (0.6)	1.3 (0.5)	0.2 (0.07)	0.1 (0.1)	1.5 (0.7)	0.5 (0.4)	1.9# (0.6)	0.5 (0.4)	1.6# (0.5)	0.1 (0.1)	0.2 (0.09)	0.09 (0.04)	0.1 (0.09)	0.1 (0.09)	0.2 (0.09)	0.2 (0.1)	0.2 (0.08)	0.2 (0.07)	0.2 (0.06)	0.2 (0.1)	0.3 (0.2)	#
Genital licks	1 (0.3)	2.1# (0.6)	0.8 (0.1)	2.1# (0.4)	1 (0.3)	0.2 (0.05)	1 (0.3)	1.3 (0.3)	1.1 (0.3)	1.3 (0.4)	1.1 (0.3)	2.1 (0.5)	0.3 (0.1)	0.4 (0.06)	0.2 (0.08)	0.3 (0.08)	0.3 (0.07)	0.3# (0.05)	0.1 (0.04)	0.4# (0.07)	0.1 (0.05)	0.4# (0.08)	0.2 (0.09)	0.5# (0.1)	#
Sniffing	0.5 (0.3)	0.6 (0.3)	0.1 (0.1)	1 (0.4)	0.8 (0.3)	0.1 (0.05)	0.2 (0.1)	0.5 (0.3)	0.5 (0.3)	0.8 (0.3)	0.2 (0.2)	0.5 (0.3)	0.1 (0.04)	0.1 (0.04)	0 (0)	0.02 (0.02)	0.04 (0.02)	0.04 (0.02)	0.06 (0.03)	0.06 (0.03)	0.06 (0.04)	0.02 (0.02)	0.04 (0.02)	0.02 (0.02)	--
Teeth chatter	1.2 (0.2)	0.8 (0.2)	1 (0.4)	1.7 (0.4)	2 (0.4)	0.3 (0.06)	1 (0.3)	1.8 (0.4)	1.1 (0.3)	2.1 (0.4)	0.7 (0.3)	1.7 (0.4)	0.2 (0.07)	0.4 (0.09)	0.06 (0.04)	0.2 (0.1)	0.2 (0.08)	0.3 (0.05)	0.3 (0.08)	0.2 (0.06)	0.2 (0.07)	0.3 (0.09)	0.3 (0.1)	0.3 (0.05)	--

The results for individually assessed behaviors are summarized across the spontaneous withdrawal timeline in Table 2. Neither sex nor treatment impacted tremors, writhes, hind leg scratches, head twitches, ptosis, sniffing, nor teeth chatter between groups or across observation time. An interaction between sex and drug was found for raw counts of wet dog shakes [F(1, 15)=5.736, p<0.05] with male rats undergoing withdrawal displaying this behavior more than their female counterparts (p<0.01). The incidence of retropulsion was increased in male and female rats across the course of withdrawal [main effect: drug F(1,15)=7.078, p<0.05]. Facial rubs showed a main effect of drug [F(1,15)=9.539, p<0.01] and a time x drug interaction [F(5, 75.00)=4.732, p<0.01] such that this behavior peaked at 6 hours in WIN-treated compared to vehicle-treated rats (p<0.001). Presentation of arched back posture was higher in rats undergoing WIN withdrawal [main effect: drug, F(1,15)=10.733, p<0.01] and in female rats [Main effect: sex, F(1,15)=7.550, p<0.05]. WIN withdrawal elevated piloerection scores [main effect: drug F(1,15)=5.231, p<0.05] and genital licks [main effect: drug F(1,15)=24.971, p<0.001].

Male rats spent more time moving during withdrawal observation at 6 and 16 hours compared to female rats [F(5,75)=9.116, p<0.001] (Figure 2C). The increased movement of male rats during these first two timepoints contributed to a main effect of sex in time spent moving [F(1,15)=35.068, p<0.0001]. WIN withdrawal decreased time spent moving in females but not in males [sex x treatment: F(1,15)=13.224, p<0.01].

The proportion of samples at each estrous stage over the course of infusions was not affected by treatment with WIN or vehicle [F(3,27)=1.734, p=0.184], Wilk’s lambda=0.838] (Figure 2D) even though, CB1R levels vary across the estrous cycle (Castelli et al., 2014; Riebe et al., 2010). Abrupt cessation of either WIN or vehicle did not impact the proportion of samples in each stage of estrous during withdrawal [F(3,19)=0.789, p=0.515, Wilk’s lambda=0.889] or the proportion of samples in high or low hormone states [F(1,22)=0.131, p=0.721] (Figure 2E). Estrous stage at the final infusion was not correlated with global scores during withdrawal at any time point. Furthermore, global score at 6 hours did not correlate with estrous at final infusion [R(20)=−0.122, p=0.588] or immediately following observation [R(5)=0.259, p=0.576]. Similarly, global scores at 16 and 24 hours were not correlated with estrous stage following withdrawal observation [16 hour R(20)=0.105, p=0.643; 24 hour R(20)=0.237, p=0.289] when separating data into WIN and vehicle groups.

3.3. Spontaneous Withdrawal: Non-somatic

Rats received i.v. infusions of WIN or vehicle followed by a behavioral battery one week (Figure 3) or two weeks (Figure 4) after the final infusion. During the marble burying test, somatic behaviors, including rears, hind leg scratches, fecal boli, wet dog shakes, and grooms, were recorded (Table 3). In the week 1 withdrawal group, neither sex nor treatment impacted wet dog shakes, hind leg scratches, or grooming behavior assessed during the marble burying test However, there was a trend of increased wet dog shakes in male and female rats experiencing withdrawal compared to vehicle controls [F(1,26)=4.2, p=0.0501] (Table 3). Neither marble manipulations nor number of marbles 100% buried was impacted by WIN withdrawal (Figure 3B, 3C). However, there was a tendency for female rats to bury fewer marbles than male rats [main effect of sex: F(1,26)=4.105, p=0.0531 Figure 3C]. In the open field test, there were no drug treatment effects for time in center or distance traveled (Figure 3E), but female rats spent less time in the center compared to male rats [main effect of sex: F(1,26)= 11.497, p<0.01] (Figure 3D). Time spent in the open arms of the elevated plus maze was unchanged by withdrawal (Figure 3F), and withdrawal did not influence the number of entrances into the open arms of the maze (Figure 3G).

In the week 2 withdrawal group, neither sex nor treatment impacted wet dog shakes, hind leg scratches, or grooming behavior assessed during the marble burying test. However, an interaction between sex and treatment was found for grooming [F(1,20)=7.033, p<0.05], with female rats undergoing withdrawal grooming more than their male equivalents (Table 3). WIN withdrawal increased defecation in male and female rats compared to vehicle controls [F(1,20)=6.914, p=0.0161], and female rats reared more than male rats [F(1,20)=7.625, p<0.05] (Supplemental Table 1). There were main effects for sex and treatment for marble manipulations in the marble burying test [main effect: sex, F(1,20)= 6.925, p=0.05; main effect: treatment, F(1,20)=6.925, p=0.05] (Figure 4B). The number of marbles 100% buried was not impacted by withdrawal. Week 2 observations for the open field test mirrored what was observed at the week 1 withdrawal time point. Male rats spent more time in the center of the field compared to the female rats [F(1,20)=13.306, p=0.01] and neither time in center nor distance traveled was impacted by withdrawal. Likewise, withdrawal did not change time spent in the open arms of the elevated plus maze (Figure 4F) or number of open arm entries (Figure 4G).

The proportion of samples in each stage of estrous over the course of drug administration was calculated out of the total samples collected prior to each infusion. One-way MANOVA indicated that infusions of vehicle or WIN did not change the proportions of samples in each stage of estrous over the course of four and a half days [F(3,24)=1.488, p=0.243, Wilk’s Lambda=0.843] (Supplementary Figure 4A). Likewise, the distribution of samples in each estrous cycle stage did not differ between vehicle and WIN treated rats when collected prior to performance of the anxiety-like behavioral battery at one week [X²(3,16)=1.101 p=0.777] or at two weeks [X²(3,12)=1.333, p=0.721] (Supplementary Figure 4B).

4. Discussion

The aim of the present study was to characterize cannabinoid withdrawal in male and female rats using an extensive set of somatic behaviors to understand the course of precipitated and spontaneous withdrawal (Aceto et al., 2001; Lichtman and Martin, 2002; Moranta et al., 2009). Furthermore, anxiety-like behaviors associated with spontaneous cannabinoid withdrawal were examined at one and two weeks. Repeated administration of escalating doses of WIN renders male and female rats dependent upon cannabinoids over a relatively short period of time. Rats treated with WIN exhibited higher GWS following precipitated withdrawal and when undergoing spontaneous withdrawal out to 24 hours when compared to vehicle treated rats. The increase in GWS was accompanied by hypolocomotion in females undergoing spontaneous withdrawal while minimal effects on locomotion were noted in males. Some somatic behaviors were consistently increased by rimonabant in both sexes (e.g., tremors, retropulsion, and writhes), while other behaviors uniquely contributed to the increased GWS in male or female rats. Similar qualitative differences emerged for the individually scored somatic behaviors during spontaneous withdrawal with both sexes engaging in more retropulsion and facial rubs; but females showing more arched back posture. When anxiety-related behaviors were assayed at extended spontaneous withdrawal timepoints, few subtle differences emerged. However, WIN withdrawal increased marble manipulations in the marble burying test ~2 weeks after the final infusion. WIN administration did not disrupt normal estrous cycling of female rats nor was any specific cycle stage correlated with higher GWS. Collectively, these results indicate that subtle, qualitative sex differences in behaviors contribute to overall cannabinoid withdrawal in Long Evans rats.

Rimonabant at 10 mg/kg has previously been used to precipitate and characterize WIN withdrawal (Aceto et al., 2001; Craft et al., 2012; Moranta et al., 2009, 2007), and we initially selected this high dose because we were interested in comparing the peak precipitated response to the more naturalistic expression of spontaneous withdrawal. Precipitating withdrawal from cannabinoids using rimonabant may impact timeframe and intensity of withdrawal as well as the individual behavioral endpoints. To account for this, we used the same dosing paradigm for WIN in both sexes to characterize both precipitated and spontaneous withdrawal. In male rats, 10 mg/kg rimonabant selectively augmented somatic withdrawal signs in WIN dependent rats. Unfortunately, 10 mg/kg rimonabant increased GWS and decreased locomotion in WIN naïve female rats (Supplemental Figure 1). This high dose of rimonabant significantly increased hind leg scratches, facial rubs, and piloerection in vehicle-treated female rats complicating the interpretation of cannabinoid withdrawal related behaviors. Our solution was to decrease the dose in females by a semi-log step to account for this potential increased sensitivity in females. Precipitating withdrawal with 3 mg/kg rimonabant in female rats did not elevate GWS or individual endpoints in vehicle-treated rats relative to handled controls.

While we are unable to make direct quantitative comparisons about the magnitude of precipitated withdrawal between male and female rats given the disparate doses of rimonabant, we can discuss qualitative differences since multiple individual endpoints contribute to increased GWS. The present study identified only retropulsion, ptosis, and teeth chatter as being higher in dependent male rats treated with 10 mg/kg rimonabant. In female rats, tremors, writhes, facial rubs, and piloerection were higher in rats undergoing withdrawal from WIN versus vehicle-treated rats with 3 mg/kg rimonabant. In contrast, when withdrawal from THC was precipitated with 10 mg/kg rimonabant, no sex differences in somatic withdrawal signs were observed, albeit the complement of behaviors examined did not fully overlap with those assessed here (Marusich et al., 2014). In male rats, administration of 10 mg/kg rimonabant decreased locomotor time in vehicle but not in WIN-treated rats. This effect is somewhat consistent with previous findings, showing that precipitation of withdrawal from THC with the same dose of rimonabant blocked the hypolocomotive effects of THC in male rats (Marusich et al., 2014). In contrast, precipitation of withdrawal in female rats with 3 mg/kg rimonabant failed to change locomotor behavior compared to handled controls. Collectively, these results suggest a different behavioral manifestation of precipitated withdrawal in males versus females.

Previous studies have reported higher rimonabant potency in female versus male rats in the antagonism of THC as a discriminative stimulus and the reversal of cannabinoid-induced antinociception (Craft et al., 2012; Wiley et al., 2017). Indeed, female rodents are more sensitive to most pharmacological cannabinoid effects (Craft et al., 2013; Jiang et al., 2022; Riebe et al., 2010; Tseng et al., 2004; Tseng and Craft, 2001; Wiley et al., 2017; Wiley and Burston, 2014). Despite this, there is no sex difference in plasma levels of circulating rimonabant (Craft et al., 2012). Rimonabant has also been demonstrated to bind the CB1 receptor with greater affinity in female versus male rats (Craft et al., 2012). The dose of rimonabant required to block THC antinociception on the paw pressure test was 10 times higher in males than in females (Craft et al., 2012). However, rimonabant (10 mg/kg) increases corticosterone following restraint stress to a greater degree in female versus male mice (Roberts et al., 2014). CB1R density is frequently lower in naïve female rats compared to male rats, although this finding is highly dependent on brain region. Conversely, receptor activation is often higher in female rats suggesting greater signaling efficiency (Castelli et al., 2014; Farquhar et al., 2019; Gonzalez et al., 2000; Llorente-Berzal et al., 2013; Mateos et al., 2011; Reich et al., 2009; Riebe et al., 2010; Rodriguez de Fonseca et al., 1994; Silva et al., 2016). These findings hint at a fascinating sex difference in CB1 receptor expression or activation in the nervous system. Thus, one explanation for dose-related sex differences observed here in the effects of rimonabant on cannabinoid withdrawal is differential CB1 receptor expression or efficiency. Rimonabant can also display intrinsic activity and produce some withdrawal-related behaviors in drug naïve animals, presumably by blocking the CB1R or dose-dependently acting as an inverse agonist (Erdozain et al., 2012; Porcu et al., 2018; Ward and Raffa, 2011). At high concentrations, rimonabant acts independently of a receptor causing non-specific inhibition of Gαi/o activity (Porcu et al., 2018). Therefore, an equally valid alternative interpretation of our data is that female rats experience more non-specific binding or inverse agonism with 10 mg/kg rimonabant compared to male rats. To confirm whether there is a sex difference in rimonabant potency versus an intrinsic or inverse agonist activity, future studies could employ a neutral CB1 antagonist in addition to comparing a full dose-response. Similarly, administration of peripherally restricted antagonists could be utilized to help parse withdrawal-related effects on central versus peripheral CB1 receptors. One such study in mice found that both rimonabant and AM4113, a CB1 neutral antagonist, increased locomotor activity and tremors in mice upon their administration following repeated dosing with the high affinity CB1 receptor agonist AM2389 (Tai et al., 2015). The peripherally restricted CB1 receptor antagonist AM6545 was without effect on these behaviors (Tai et al., 2015). Of note, these neutral antagonists are being pursued for their potential therapeutic utility for indications like obesity and substance use disorders based on the premise that they will be more tolerable and less aversive compared to rimonabant with its inverse agonist activity (Bosquez-Berger et al., 2023; Soler-Cedeno and Xi, 2022), but additional research is needed. Combined CB1 neutral antagonists/CB2 agonists can spare some of the behaviors associated with rimonabant-precipitated withdrawal, particularly scratching and facial rubs (Tai et al., 2018).

Clinical literature is mixed on whether there are sex differences in spontaneous cannabinoid withdrawal with some support for different symptoms associated with withdrawal in each sex (Bonnet and Preuss, 2017; Connor et al., 2022; Cuttler et al., 2016; Herrmann et al., 2015; Levin et al., 2010; Nacca et al., 2013). Smoked cannabis produces larger subjective drug effects and higher cannabinoid metabolite concentrations in women, which may explain increased sensitivity to cannabinoids (Cooper and Haney, 2014; Herrmann et al., 2015; Lake et al., 2023; Sholler et al., 2021). We found no sex differences during spontaneous WIN withdrawal in the GWS or time course of withdrawal, but some individual behaviors differed. Both sexes engaged in more retropulsion, facial rubs, genital licks, and piloerection during spontaneous withdrawal. Male rats undergoing cannabinoid withdrawal showed more wet dog shakes than male vehicle controls or their female counterparts. In females, arched back posture was elevated during cannabinoid withdrawal, particularly at early time points. Thus, it is possible that, even in the absence of a quantitative sex difference in GWS, a qualitative difference may occur in how individual behavior scores impact the GWS. Notably, earlier pre-clinical studies used mostly male animals to develop methods for quantitating withdrawal; thus, individual endpoints may not be as robust in female rats.

In humans, withdrawal often manifests as affective changes in mood and anxiety or changes in sleep (Connor et al., 2022; Nacca et al., 2013). In SCRA dependent male rats, rimonabant produced a rapid increase in corticotropin releasing factor (CRF) levels in the amygdala (Rodriguez de Fonseca et al., 1994), but it remains unknown if a similar stress response is observed in female rats. Here examining anxiety-related behavioral measures during withdrawal resulted in few treatment effects. In week 2, we observed a higher frequency of marble manipulation in WIN- versus vehicle-treated rats. This behavior, defined as a rodent picking up the marble and handling it, is not typically scored in this test, but we propose that it could represent a display of compulsive or anxiety-like behavior. Though previously unmentioned in the literature, it may be a useful addition to the list of scored endpoints in future studies evaluating marble burying. There was also a trend towards a main effect of sex, in which male rats buried more marbles versus female rats consistent with a study examining the effects of artificial hormone treatment in male and female mice (Goel and Bale, 2008).

Clinically, the prevalence of anxiety disorders is much higher in women than in men (McLean et al., 2011). We report a sex difference in the amount of time spent in the center but not distance traveled in the open field test. Past literature studying sex differences using the open field has shown mixed findings for both endpoints (Knight et al., 2021; Scholl et al., 2019). Inconsistencies could be due to differences in strain or the tracking methods/software. However, a recent study using Long Evans rats to study stress was consistent with our data with male rats spending more time in the center compared to female rats (Bishnoi et al., 2021). Conversely, past literature suggests that female rats tend to spend more time in the open arms and make more open arm entries compared to male rats (Knight et al., 2021; Scholl et al., 2019) although here we did not see significance across sex. The lack of convergence between assays to assess anxiety like behavior in rodents is a consistent observation in the literature and is thought to be contingent upon environmental and genetic factors (Figueiredo Cerqueira et al., 2023). Though few WIN withdrawal related effects emerged, future directions should include more screens of negative and positive valence systems. Assessing cannabinoid withdrawal in response to self-administration or other routes of administration is an important future direction to ascertain whether stimulus contingency might impact withdrawal presentation.

5. Conclusion

Intravenous infusions of escalating doses of WIN produce a robust dependence in a relatively short period of time, without disrupting estrous cycling in females, and resulting in higher somatic GWS during precipitated and spontaneous withdrawal out to 24 hours. Although overall GWS was similar, the underlying profile of behaviors differed between males and females depending upon the endpoint scored and whether withdrawal was precipitated or spontaneous. Anxiety-like behaviors were not significantly elevated one week following spontaneous withdrawal, but marble manipulations were increased at two weeks. These findings suggest differential expression of withdrawal in males versus females but could also suggest that individual quantification of cannabinoid withdrawal is more suited to males as opposed to females. Although the use of rimonabant to precipitate withdrawal and the disparate doses used in male and female rats are both limitations of the current study, results clearly point to sex dependent behavioral profiles of withdrawal in males and females. Future studies could be conducted to disentangle the role of the central nervous system CB1 receptors in cannabinoid withdrawal by using neutral and/or peripherally restricted antagonists, particularly considering increased interest in the therapeutic potential of these compounds for both obesity and substance use disorders.

Highlights.

• Intravenous infusions of escalating doses of WIN 55212 produced somatic spontaneous and precipitated cannabinoid withdrawal in male and female rats without interrupting estrous cycling in female rats.

• Total global withdrawal scores were similar between males and females, however, the profile of individual behaviors differed.

• Anxiety-like behavior was not elevated following one week of abstinence from WIN 55212 but marble manipulations were higher at two weeks abstinence.

• Expression of cannabinoid withdrawal is likely different between the two sexes and/or current methods for quantification of withdrawal are more suited to males.