Sleep disturbance after cessation of cannabis administration in mice
Asano et al.
Department of Pharmaceutical Therapy and Neuropharmacology, Faculty of Pharmaceutical Sciences Graduate School of Medicine and Pharmaceutical Sciences, University of Toyama Toyama Japan
Department of Biology Graduate School of Science & Engineering, University of Toyama Toyama Japan
* CorrespondenceAtsumi Nitta, Department of Pharmaceutical Therapy & Neuropharmacology, Faculty of Pharmaceutical Sciences, Graduate School of Medicine and Pharmaceutical Sciences, University of Toyama, Toyama, Japan.
Email: nitta@pha.u-toyama.ac.jp
Abstract
Cannabis withdrawal syndrome (CWS) in humans is characterized by various somatic symptoms, including sleep disturbances. In the present study, we investigated sleep alterations in mice after the cessation of arachidonylcyclopropylamide (ACPA), a cannabinoid type 1 receptor agonist, administration. ACPA‐administered mice (ACPA mice) displayed an increased number of rearings after the cessation of ACPA administration compared to saline‐administered mice (Saline mice). Moreover, the number of rubbings was also decreased in ACPA mice compared with those of the control mice. Electroencephalography (EEG) and electromyography (EMG) were measured for 3 days after the cessation of ACPA administration. During ACPA administration, there was no difference in the relative amounts of total sleep and wake time between ACPA and Saline mice. However, ACPA‐induced withdrawal decreased total sleep time during the light period in ACPA mice after ACPA cessation. These results suggest that ACPA cessation induces sleep disturbances in the mouse model of CWS.
Toc
Cessation of arachidonylcyclopropylamide (ACPA)induces somatic symptoms and sleep disturbance, a CB1 receptor agonist, to mice, Although ACPA itself does not affect them.
Boxed Text
Article notes
Asano T , Takemoto H , Horita T , Tokutake T , Izuo N , Mochizuki T , et al. Sleep disturbance after cessation of cannabis administration in mice. Neuropsychopharmacol Rep. 2023;43:505–512. 10.1002/npr2.12329 36905178 PMC10739061
1INTRODUCTION
Cannabis use is increasing worldwide with a concomitant increase in the annual number of cannabis users. 1 Habitual cannabis users have difficulty limiting their cannabis intake, which results in various social and economic problems. 2 The acute effects of cannabis use include anxiety, suicidal ideations or tendencies, and psychotic symptoms. 3 , 4 By contrast, chronic cannabis use induces mood disorders and neurocognitive impairments. 4 Cannabis has also been reported to contribute to the development of psychosis and worsen its prognosis. 5 , 6 Thus, cannabis users attempt to cease or decrease their intake; however, most are unsuccessful in their cessation attempts due to the development of cannabis withdrawal syndrome (CWS). 7 , 8 , 9 CWS typically presents with symptoms such as cognitive impairment, psychosis, anxiety, and sleep disturbances. 1 , 8 , 9 , 10 , 11 , 12 , 13
Sleep disturbances are a prominent feature of CWS, which cause cannabis use relapse in approximately 65% of patients who attempt cannabis cessation. 9 , 14 Sleep disturbances have also been reported to persist after cannabis withdrawal for up to 45 days of cessation in humans. 15 However, there are no effective treatments for CWS sleep disorders due to a lack of appropriate animal models to study CWS sleep disorders. Thus, the mechanisms underlying CWS‐associated sleep disorders have not been elucidated. In this study, we aimed to establish a better, practical mouse model of sleep disturbance in CWS.
The primary psychoactive component in cannabis is Δ‐9‐tetrahydrocannabinol (THC), which binds to the cannabinoid type 1 receptor (CB1R). THC exhibits various psychoactive effects, including pleasant sensations, hallucinations, and sedation via stimulation of CB1R in the brain. THC causes only mild withdrawal symptoms in animal models due to its lipophilic nature and long half‐life. 16 Therefore, it is inappropriate to administer THC to establish a cannabis withdrawal model in mice. Arachidonylcyclopropylamide (ACPA), a CB1 agonist that selectively binds to CB1 receptors in the brain, has a higher affinity for CB1 but not CB2 receptors compared to THC. Previous studies using rodent models of CWS establish either precipitated or spontaneous withdrawal models. 17 The precipitated withdrawal model is established by administering a CB1R antagonist, typically rimonabant, which immediately blocks the effects of chronic administration of a CB1R agonist. 17 , 18 However, the precipitated withdrawal model does not reflect CWS symptoms in humans because of CB1R antagonist administration after CB1R agonist or THC withdrawal. On the other hand, the spontaneous withdrawal model is established by chronically administering the CB1 agonist, then abruptly ceasing treatment and observing the natural behavioral response to drug cessation. 17 , 19 , 20 , 21 The spontaneous withdrawal model induces somatic symptoms, including decreased rubbing and grooming during drug withdrawal in mice. 22 , 23 , 24
In this study, we examined physical symptoms and changes in sleep after cannabis cessation in mice by assessing sleep disturbances using a more clinically relevant model of spontaneous withdrawal following cessation of ACPA administration, a CB1R agonist.
2METHODS
2.1Animals and treatments
Twenty‐three male adult C57BL/6J mice (7–8 weeks old) weighing 21–28 g were purchased from Japan SLC, Inc. (Shizuoka, Japan) and housed under a temperature (23 ± 2°C) and humidity (50 ± 5%) controlled environment with a 12/12‐h light–dark cycle (07:00–19:00). The mice were provided with standard rodent chow (CE‐2, CLEA Japan, Inc., Tokyo, Japan) and water ad libitum. All mice were acclimated to the environment for 1 week. All animal experiments were performed according to the National Institute of Health Guidelines for the Care and Use of Laboratory Animals. Animal experimental protocols were approved by the Animal Care and Use Committee of the University of Toyama (Approval Number: A2018PHA‐4) and conducted in accordance with Institutional Animal Experiment Handling Rules of the University of Toyama. The number of animals used was carefully estimated and kept to the minimum necessary for meaningful interpretation of the data. The minimum number of animals required was calculated using G*power with a significance level of 0.05 and a power of 0.80, based on the behavioral test values of previous studies. 25 ACPA was purchased from Funakoshi Co., Ltd. (Tokyo, Japan) and dissolved in saline, prepared at the time of use. For experiment 1 (Behavioral tests), the control group was administered saline intraperitoneally for 1 week at 08:00 (Saline group [n = 6]). The ACPA group (n = 6) was administered ACPA intraperitoneally (0.02 mg/kg) for 1 week at 08:00. We used a previously established ACPA dosage (0.02 mg/kg, ip). 25 After drug administration for 1 week, behavioral experiments were conducted for 3 days of withdrawal (Figure 1A). For experiment 2 (Sleep recordings), the control group was administered saline intraperitoneally for 1 week at 08:00 (Saline group [n = 6]). The ACPA group (n = 5) was administered ACPA (0.02 mg/kg) intraperitoneally for 1 week at 08:00. Before the administration of ACPA or saline, the mice were acclimated by intraperitoneal saline administration for 3 days. Sleep recordings were performed before and during ACPA administration, as well as for 3 days after drug withdrawal (Figure 2A).
2.2Behavioral tests
We followed previously established behavioral protocols with minor modifications. 19 , 22 , 23 , 24 Mice were placed in an acrylic chamber (30 × 30 × 35 cm) different from their home cage for 20 min, for 3 days (Days 8–10) after the cessation of ACPA or saline. The locomotor activity and number of rearing were recorded using a Scanet MV‐40 (Melquest Ltd., Toyama, Japan). As a preliminary experiment, the locomotor activity and the number of rearing in 20 min were measured in mice treated with ACPA (0.02 mg/kg, ip) or saline for 1 week. In addition, the numbers of rubbing and grooming were counted from the recorded video of the freely moving mice in the acrylic chamber (Figure 1A).
2.3Surgery for sleep recording and assessment of sleep disturbances
Surgery for polysomnograph recordings and assessment of sleep–wake status were performed as described previously. 26 , 27 Mice (Saline group, n = 6; ACPA group, n = 5) were intraperitoneally injected with a combination of anesthetics (medetomidine [0.3 mg/kg, Nippon Zenyaku Kogyo Co., Ltd., Fukushima, Japan], midazolam [4.0 mg/kg, Sandoz K.K., Tokyo, Japan], and butorphanol [5.0 mg/kg, Meiji Seika Pharma Co., Ltd., Tokyo, Japan]), and were placed in a stereotaxic frame (SR‐5M, Narishige, Tokyo, Japan). The skull was implanted with a pair of electroencephalography (EEG) electrodes (±1.0 mm rostrocaudal and 1.5 mm lateral from bregma) and the implanted EEG electrode was fixed using Super‐Bond C & B (Sun Medical Co., Ltd., Shiga, Japan) and UNIFASTIII (Geniee, Inc., Tokyo, Japan). The trapezius muscle of mice was implanted with a pair of electromyography (EMG) wires. Mice were transferred to individual soundproof boxes after a recovery period and habituated to recoding cables for 3 days under a 12/12‐h light–dark cycle (07:00–19:00). After habituation, mice were administered ACPA (0.02 mg/kg) or saline intraperitoneally once a day for 1 week at 08:00 (Figure 2A). The EEG and EMG signals were recorded 1 day before drug administration (Day 0), on the last day of ACPA administration (Day 7), and 3 days of ACPA withdrawal (Day 8–10). Data were taken from 07:00 to 19:00 for the light period and from 19:00 to 07:00 for the dark period. The EEG and EMG signals were acquired using digital amplifiers (Biotex, Kyoto, Japan) and stored in PC at a sampling rate of 128 Hz using SleepSign software version 3.0 (Kissei Comtec Co., Ltd., Nagano, Japan). Video of the cage‐top view, as well as spontaneous locomotor activity counts using an infrared sensor (Biotex), were also captured simultaneously and incorporated with the SleepSign recordings. The EEG and EMG signals were, first automatically scored as wake, rapid eye movement (REM) sleep, or non‐REM (NREM) sleep by 10 s epochs, then the scored data were visually corrected according to standard criteria. 26 , 27
2.4Statistical analysis
The statistical differences between the two groups were determined using Student's t‐tests. All data are expressed as mean ± standard error of the mean. Statistical significance was set at *p < 0.05.
3RESULTS
3.1Somatic symptoms induced by withdrawal after ACPA administration ACPA
A previous study reported that spontaneous withdrawal of another CB1R agonist resulted in increased locomotor activity and rearing in mice. 22 CB1R agonist withdrawal in mice also caused somatic symptoms including decreased rubbing and grooming. 17 , 19 , 20 , 21 , 22 First, to assess whether ACPA affects behavior in mice, we observed the behavior of mice immediately after 1 week administration with ACPA (0.02 mg kg). As shown in Figure S1 (Row data of Figure S1: Table S1), the locomotor activity in the ACPA group (10299.2 ± 462.6 counts/20 min) did not change compared to the saline group (10831.2 ± 457.9 counts/20 min). The number of rearing in the ACPA group (156.5 ± 6.5 counts/20 min) did not change compared to the saline group (152.2 ± 16.0 counts/20 min). Therefore, it is suggested that ACPA by itself does not affect the behavior, including locomotor activity and the number of rearing. Here, we assessed somatic symptoms in mice by measuring locomotor activity, rearing, rubbing, and grooming for 3 days (Days 8–10) after the cessation of ACPA administration (0.02 mg/kg, ip; Figure 1B–D). We did not observe a difference in locomotor activity between the ACPA (Day 8: 10831.2 ± 457.9 counts/20 min, Day 9: 9793.5 ± 508.3 counts/20 min, Day 10: 8355.5 ± 508.7 counts/20 min) and the Saline groups (Day 8: 10299.2 ± 462.6 counts/20 min, Day 9: 8321.5 ± 588.9 counts/20 min, Day 10: 8408.3 ± 581.4 counts/20 min; Figure 1B–D). However, the number of rearing was significantly increased in the ACPA group (Day 8: 227.2 ± 9.7 counts/20 min, Day 9: 221.5 ± 16.8 counts/20 min, Day 10: 172.0 ± 13.1 counts/20 min), compared with that in the Saline group (Day 8: 180.0 ± 9.5 counts/20 min, Day 9: 159.8 ± 10.2 counts/20 min, Day 10: 158.7 ± 9.7 counts/20 min) for 3 days after drug cessation (Day 8: p = 0.0059, Day 9: p = 0.0046, Day 10: p = 0.0390, **p < 0.01, *p < 0.05, Figure 1B–D). Moreover, the number of rubbing was significantly decreased in the ACPA group (Day 9: 11.5 ± 2.8 counts) compared to that in the Saline group (Day 9: 24.7 ± 4.8 counts; p = 0.0392, *p < 0.05, Figure 1C). Finally, the number of grooming was also mildly decreased in the ACPA group (Day 9: 43.7 ± 5.6 counts, Day 10: 46.8 ± 8.4 counts) compared with that in the Saline group (Day 9: 100.8 ± 30.5 counts, Day 10: 70.0 ± 10.1 counts) (Day 9: p = 0.0949, Day 10: p = 0.1083, Figure 1C,D). These results suggest that ACPA cessation induced somatic symptoms in mice. All row data of Figure 1 are shown in Table S1.
3.2Sleep disturbances induced by ACPA cessation
We assessed sleep alterations after ACPA cessation by conducting long‐term measurements of EEG/EMG recordings: before ACPA administration (Day 0), during ACPA administration (Day 7), and after ACPA cessation (Days 8–10) (Figure 2A). Data are expressed as % of sleep/wake amount on Day 0. First, we evaluated sleep after 1 week of ACPA or saline administration (Figure 2B,C). The wakefulness and sleep during ACPA or saline administration showed no significant differences between the ACPA and the Saline groups during the light (07:00–19:00) and the dark period (19:00–07:00) (Figure 2B,C). These results suggest that chronic administration of ACPA does not affect wakefulness and sleep in mice. EEG and EMG recordings were continuously measured for 3 days (Days 8–10) after ACPA cessation (Figure 2D–I). During the light period on Day 8, wakefulness was significantly increased in the ACPA group (93.9 ± 3.9%) than in the Saline group (83.1 ± 1.4%) (p = 0.0216, *p < 0.05, Figure 2D). In contrast, total sleep time, including both NREM and REM sleep, during the light period was significantly decreased in the ACPA group (104.6 ± 3.0%) than in the Saline group (112.7 ± 0.9%; p = 0.0181, *p < 0.05, Figure 2D). Next, we assessed the NREM and REM sleep time during the light period separately, and observed a mild decrease in NREM sleep in the ACPA group (106.7 ± 3.1%) compared with the Saline group (112.8 ± 0.8%; p = 0.0670, Figure 2D). Similarly, REM sleep was also mildly decreased in the ACPA group (92.1 ± 8.9%) compared with the Saline group (112.6 ± 5.3%; p = 0.0704, Figure 2D). There was no difference in sleep and wakefulness during the dark period between the ACPA and the Saline groups (Figure 2E), suggesting that ACPA administration did not cause sleep disturbances during the active phase. In contrast, total sleep time was significantly decreased after ACPA cessation during the light period, suggesting that the sleep disturbances were induced by cessation after ACPA administration, specifically during the rest phase. During the light period on Day 9, wakefulness tended to increase in the ACPA group (93.6 ± 2.8%) compared with the Saline group (83.4 ± 3.7%; p = 0.0527, Figure 2F), and total sleep time was significantly decreased in the ACPA group (104.7 ± 2.6%) compared with the Saline group (112.4 ± 1.9%; p = 0.0387, *p < 0.05, Figure 2F). There were no differences in NREM sleep, however REM sleep was significantly decreased in the ACPA group (88.7 ± 6.1%) compared to the Saline group (116.0 ± 7.1%; p = 0.0189, *p < 0.05, Figure 2F). During the dark period, there was no significant difference in wakefulness and sleep between the ACPA and the Saline groups (Figure 2G). These results suggest that the sleep disturbances after ACPA cessation persisted for 2 days. On Day 10, there was no significant difference in wakefulness and sleep between the ACPA and the Saline groups during the light and dark periods (Figure 2H,I). These results suggest that ACPA cessation resulted in sleep disturbances in mice. All row data of Figure 2 are shown in Table S2.
4DISCUSSION
In the present study, we demonstrated that ACPA cessation after administration in mice induced somatic symptoms including increased rearing and decreased rubbing with no effects on locomotor activity. In addition, we showed that chronic administration of ACPA had no effect on wakefulness and sleep in mice. However, during the light period, ACPA cessation resulted in significantly decreased total sleep time, including NREM and REM sleep.
A previous study reported that administration of SR141716A, a potent and selective antagonist of the brain CB1R, increased locomotor activity in THC‐administrated mice. 28 Additionally, a CWS model in mice showed increased locomotion, severe tremors, and wet‐dog shakes. 18 , 29 , 30 However, these earlier reports used CB1R antagonists, typically rimonabant, to model CWS and did not reflect clinically‐relevant withdrawal symptoms. 17 These results could be explained by the lipophilic nature and a long half‐life in the blood of THC, which is metabolized slowly and is less likely to cause dependence. As a result, animal models have reported no or mild withdrawal symptoms. 16 ACPA does not bind to CB2R and is potently selective for CB1R. 31 In this study, we assessed somatic symptoms in a more clinically relevant spontaneous withdrawal mouse model without administering a CB1R antagonist (Figure 1A). Previous studies have assessed somatic symptoms, including locomotor activity, rearing, rubbing, and grooming, in spontaneous withdrawal after 1–3 days of CP‐55940, a CB1R agonist, administration. 23 , 24 Here, this study was conducted for 3 days after the ACPA administration to examine somatic symptoms during spontaneous withdrawal of ACPA. As shown in Figure 1B–D, ACPA‐induced withdrawal did not affect locomotor activity. In contrast, as shown in Figure 1B–D, ACPA‐induced withdrawal resulted in persistently increased rearing for 3 days (Days 8–10). It has been reported that spontaneous withdrawal of CP‐55940, a CB1R selective agonist, increases locomotor activity and the number of rearing. 24 In this experiment, ACPA withdrawal did not affect the amount of locomotor activity, rubbing, and grooming, but the increase in rearing was consistent with previous studies. There have been no previous reports evaluating the behavior of ACPA withdrawal, and in addition, this study is the first finding for the sleep disturbances by ACPA withdrawal. Our results may reflect a relatively mild effect of ACPA because it models spontaneous withdrawal without using CB1 antagonists. We previously reported that the administration of ACPA (0.02 mg/kg, ip) had no effect on locomotor activity. 25 Here, we confirmed that ACPA did not affect locomotor activity both during administration and after withdrawal. These observations are consistent with the spontaneous withdrawal of another CB1R agonist, which also resulted in increased rearing. 22 , 23 , 24
In addition, the ACPA‐induced withdrawal significantly decreased the number of rubbing (Figure 1C) and showed a tendency to decrease the number of grooming. These somatic symptoms were considered to be depression‐like symptoms. In a light/dark box test of anxiety‐like behaviors, the withdrawal after CP‐55940 administration resulted in decreased time in the light box. 22 Moreover, these mice also displayed decreased number of rubbing and grooming. 22 Consistent with this, a similar increase in anxiety‐like behavior was reported in precipitated THC withdrawal in mice, and increased extracellular levels of corticotropin‐releasing factor were reported in sedentary HU‐210 withdrawal in rats. 32 , 33 Thus, our ACPA‐induced withdrawal symptoms, although weak, were consistent with previous studies.
In a human study, cannabis decreased sleep onset latency, 34 and reduced wakefulness after sleep onset. 35 Thus, cannabinoid use has a positive effect on sleep. Contrarily, cannabis‐induced withdrawal decreased sleep duration and reduced REM sleep in humans. 14 , 36 , 37 Recently, the number of people using cannabis has increased, with a concomitant increase in people with disabilities from cannabinoid withdrawal. 1 Therefore, understanding the mechanisms underlying CWS‐induced sleep disorders is critical to develop CWS treatments. However, only a few studies have been reported on sleep disturbances during spontaneous cannabis withdrawal in animal models. Thus, we evaluated wakefulness and sleep in mice during ACPA administration (0.02 mg/kg, ip) for 1 week, followed by ACPA cessation. As shown in Figure 2, chronic administration of ACPA showed no effect on wakefulness and sleep during light and dark periods. Contrary to our results, Missing et al. reported that repeated administration of the high‐efficacy CB1 receptor agonist AM2389 (0.03 mg/kg, s.c.) in mice for 5 days increased wakefulness and decreased NREM and REM sleep. 38 This could be because the Ki value of AM2389 (Ki = 0.16 nmol L−1) is much lower than that of ACPA (Ki = 2.2 nmol L−1), and AM2389 also has a very high affinity for the CB1 receptor. 31 , 39 As shown in Figure 2D,F, ACPA‐induced withdrawal decreased total sleep time during the light period. Kesner et al. recently reported that withdrawal after chronic high‐dose THC administration decreased NREM sleep in mice. 40 However, our results provide new findings that cessation after chronic low‐dose CB1 selective agonist ACPA induced sleep disturbances. In an animal study, precipitated withdrawal by CB1R antagonist SR141716A after chronic administration of HU‐210, a potent CB1R agonist, increased corticotropin‐releasing hormone (CRH) mRNA in the central amygdala. 41 Moreover, intracerebroventricular injection of CRH induced decrease in slow wave NREM sleep in rats. 42 , 43 Therefore, our observations of decreased NREM sleep after ACPA‐induced withdrawal could be explained by increased levels of CRH in the brain. In this study, cessation of ACPA administration to mice resulted in withdrawal symptoms such as increased rearing and decreased rubbing, consistent with previous studies. 22 , 23 , 24 , 32 , 33 CWS has been reported to cause sleep disturbances in humans. 1 , 8 , 14 In the present study, ACPA‐induced withdrawal caused sleep disturbances in mice. Thus, the face validity of the CWS model mouse is more certain, as ACPA withdrawal induced somatic symptoms and sleep disturbances. These results suggest that spontaneous withdrawal of ACPA may be a useful model for a more clinical CWS.
It has already been reported that in humans, withdrawal from various drugs, such as methamphetamine 44 and cocaine, 45 is associated with sleep disturbances. However, clinical trials with illicit drugs are very difficult to conduct, therefore, clinical trials have not yielded consistent results on sleep disturbances. In this study, we identified sleep disturbances during cannabis withdrawal in mice. In the future, a more detailed analysis of sleep disturbances during intense withdrawal after tolerance to cannabis administration may be conducted. REM sleep duration, frequency, and theta power may be used as new indicators of withdrawal. Therefore, our study opens new avenue for developing treatments for sleep disturbances seen in cannabis withdrawal syndrome.
FUNDING INFORMATION
This work was supported by the grant‐in‐aid for Scientific Research (KAKENHI) (Grant number 21K15294, 20K22702 to TA, 21H02635 to AN), JP22H04922, JP16H06276 (AdAMS) from the Japan Society for the Promotion of Science, Kobayashi Foundation, and SRF Grant for Biomedical Research and Foundation. The funders had no role in the study design, data collection and analysis, decision to publish, or manuscript preparation.
CONFLICT OF INTEREST STATEMENT
The authors declare no conflict of interest.
ETHICS STATEMENT
Approval of the Research Protocol by an Institutional Reviewer Board: None.
Informed Consent: None.
Registry and the Registration No. of the Study/Trial: None.
Animal Studies: Animal experimental protocols were approved by the Animal Care and Use Committee of the University of Toyama (Approval number: A2018‐PHA‐4) and conducted in accordance with the Institutional Animal Experiment Handling Rules of the University of Toyama.
Supporting information
ACKNOWLEDGMENTS
We thank Prof Satoshi Morimoto, Kyusyu University, for supplying THC for pre‐study.