Rats exposed to chronic alcohol display protracted insomnia and daytime sleepiness-like behavior during alcohol withdrawal

Abstract

Introduction:

Alcohol use disorder (AUD), a chronic brain disorder, is characterized by a multitude of symptoms, including insomnia, during withdrawal. Previously, we have shown that rats exposed to chronic alcohol displayed insomnia-like symptoms during acute withdrawal. Since insomnia lasts for several years and is a major risk factor of relapse to alcoholism, the present study is designed to investigate the long-term effects of alcohol withdrawal on sleep-wakefulness.

Methods:

Adult male Sprague-Dawley rats, instrumented with sleep recording electrodes, were divided into two groups: Alcohol and Control. Rats were either administered alcohol (35% v/v), mixed with infant formula (Alcohol group) or control mixture containing water and infant formula (Controls; 10 mL/kg) every 8 h for 4 days using Majchrowicz’s chronic binge drinking protocol. Electrographic recordings of sleep-wakefulness were performed until withdrawal day 7, however, the data was analyzed for withdrawal days 3, 5 and 7 in both Control and Alcohol groups.

Results:

As compared to the controls, alcohol-exposed rats displayed insomnia-like symptoms as revealed by a) significant reduction in the quantity and quality of sleep during the light (inactive) period and b) a significant increase in NREM sleep with a concomitant reduction in the amount of time spent in the wakefulness during the dark (active) period of alcohol withdrawal.

Conclusion:

Our results suggest that the chronic binge model of alcohol dependence mimics clinical symptoms of AUD especially protracted insomnia and is suitable for understanding the mechanisms associated with alcohol withdrawal-induced behaviors.

INTRODUCTION

Alcohol use disorder (AUD) is a chronic relapsing brain disorder and accounts for 2.6% of all deaths in the United States [1, 2]. People with AUD display multitude of symptoms however, insomnia (difficulty in initiating and maintaining sleep) is the most common, severe, and protracted symptom associated with alcohol withdrawal [3, 4]. In fact, insomnia is one criterion for the diagnosis of AUD [5]. Moreover, insomnia lasts for up to 3 years during alcohol withdrawal and if left untreated, may interfere with recovery from the AUD and contribute to relapse [4, 6, 7]. Interestingly, people who relapse to alcohol use display more sleep disturbances as compared to abstainers [8]. Moreover, in general, there is a strong correlation between insomnia and impaired quality of life [9]. Overall, sleep disturbances including insomnia during alcohol withdrawal are considered as risk factors for relapse in recovering alcoholics and may thus be targeted to improve treatment outcomes for the AUD patients [10]. In addition, there is evidence that suggests that people with AUD display excessive daytime sleepiness during withdrawal, a condition linked to serious life-threatening complications [11, 12]. However, the exact relationship between AUD and insomnia and associated sleep disturbances during withdrawal is not yet completely understood.

There has been a lack of preclinical evidence in support of human studies. For example, in animal studies, the effects of long-term alcohol exposure on sleep-wake behavior in rats [13, 14] and mice [15] have been investigated. These studies either did not find any substantial effects on sleep-wakefulness or the long-term effects of alcohol withdrawal were not investigated on alcohol dependent animals.

Previously, we have shown that alcohol-exposed animals display severe insomnia-like symptoms during acute withdrawal (day 1 of alcohol withdrawal) using Majchrowicz’s chronic binge model [16]. Now we ask if this chronic binge model is able to mimic protracted symptoms of alcohol withdrawal as observed in patients with AUD. Hence, in the present study, using a similar protocol, we have investigated the chronic (for 7 days) effects of alcohol withdrawal on sleep-wakefulness. We hypothesized that alcohol-exposed rats will display protracted insomnia and daytime sleepiness-like behavior during alcohol withdrawal.

Materials and methods

Animals

Male Sprague-Dawley rats (200–300g; Charles River, Wilmington, MA, USA) were housed in a standard light-dark cycle with light onset (ZT0) at 10:00 am and dark onset (ZT12) at 10:00 pm. The animals were kept at ambient room temperature (25°C) with ad libitum access to water and standard laboratory rat chow. We made every effort to minimize the number of animals used in this study. All animal experiments were performed according to the ARRIVE guidelines, American Association for Accreditation of Laboratory Animal Care’s policy and Guide for the Care and Use of Laboratory Animals issued by the Institute for Laboratory Animal Research. The experimental protocol was approved by the Animal’s Committee of the Harry S.Truman Memorial Veterans Hospital.

Surgery

Under isoflurane anesthesia and aseptic conditions, sleep recording electrodes were surgically implanted in the rat brain as previously described [17–19]. Briefly, the anesthetized rat was mounted on stereotaxic apparatus and sleep recording electrodes were screwed into the skull surface at the level of epidural space for electroencephalogram (EEG) recording and sutured close to the nuchal muscle for electromyogram (EMG) recording [20]. These EEG and EMG electrodes were connected to a multichannel electrode pedestal (MS363, Plastics One Inc, Roanoke, VA). The whole assembly was then fixed on the cranium by using dental cement.

Post-operative recovery and habituation

After surgery, the animals were individually kept in a sleep recording cage, connected to the amplifier, and were allowed to habituate with the sleep recording cables at-least for 7 days or until they showed stable sleep-wake behavior [17]. Once a stable sleep-wake behavior was established, a 24 h baseline sleep-wakefulness recording was performed.

Binge alcohol administration

a). Alcohol mixture

The alcohol mixture was prepared fresh by mixing alcohol (200 proof; 35 ml; Fisher Scientific, Pittsburgh, PA) with water (80 ml) and 1 scoop (~8.5 gm) of infant formula (PBM Nutritionals. Georgia, VT). Once the infant formula was thoroughly mixed, the final volume was made up to 100 ml by the addition of water to achieve 35 % (v/v) alcohol concentration.

b). Administration of alcohol

After baseline recording of sleep-wakefulness, the animals were randomly divided into two groups: Alcohol exposed (Experimental) and Non-alcohol exposed (Control). We used Majchrowicz’s protocol to induce alcohol dependence [21]. This method has been extensively used to induce alcohol dependency in rats [16, 22]. In this method, three doses of alcohol were administered across 24 hours (after every 8 hours) for four consecutive days mimicking frequent binge drinking episodes in human alcoholics. The experiment was initiated at 8.00 AM (at the end of the dark period; Light onset = 10:00 AM) with the administration of a priming dose of alcohol (5 g/Kg) mimicking human alcohol consumption during late evening/early night. After administration of the priming dose, subsequent doses (every 8 hours) of alcohol were administered depending on the degree of intoxication, exhibited by the animal, prior to each dose. This way, two doses were administered in the dark (active) period and one in the light (inactive) period on each of the four days. The level of intoxication was determined by observing the animal’s behavior for 5 min before each administration [16, 21, 22]. The control mixture (10 ml/Kg; prepared fresh by mixing one scoop of infant formula in 100 ml of water) was administered to the controls. Administration of alcohol (Alcohol group) or control (Control group) mixture was performed by gastric intubation using a 10 ml syringe connected to a thin long and slightly curved metal injector tube with a round bulb at its end [18]. All animals had free access to rat chow and water throughout the experiment. With this procedure, the rats received a daily dose of ~10 g/kg/day of ethanol. No sleep/wake recording was performed during the alcohol administration.

Sleep/wake recording

The EEG and EMG signals, obtained from freely behaving rats, were amplified and acquired as previously described [16]. The duration of each sleep/wake recording, which was performed on baseline and withdrawal days, was of 24 h duration starting from light-onset (ZT0). The baseline (24 h) was recorded before the beginning of alcohol administration. After four days of alcohol exposure, the sleep/wake recording resumed again on withdrawal day 3 starting at light onset and continued until withdrawal day 7.

Data analysis.

The data captured on baseline day and withdrawal days 3, 5 and 7, was separated into light and dark period and analyzed using SleepSign software (version 3; Kissei Comtec Co. Ltd., Nagano, Japan). Later, the data was visually scored as wakefulness, NREM, and REM sleep in 10-sec epochs by the experimenter who was blinded to the experiment as described previously [16]. To determine the behavioral state transitions, we also analyzed the number of bouts of each state viz wakefulness, NREM, and REM sleep, and their average duration (in seconds). Sleep onset latency (amount of time in minutes between light onset and first un-interrupted 60 s bout of NREM sleep) was examined. We also examined the delta (1–4 Hz) and beta (12–20 Hz) activities during the light period of alcohol withdrawal as previously described [16, 18]. The reduced delta and increased beta activities are considered as markers of insomnia [23]. The relative contribution of each of the frequency ranges was determined by dividing the power in these frequency ranges with the total power obtained from all the frequencies (1 – 20 Hz).

Statistics:

To achieve scientific rigor and reproducibility, all groups (Controls and Alcohol group) were run in parallel and repeated at least twice. The animals were randomly assigned to each group using the Online GraphPad randomization calculator, just before the beginning of the experiment. The effect and sample size were calculated by performing a priori power analysis (α =0.05; power ≥ 0.9; N = 3/group; G*Power [24]). Two-Way Repeated Measure ANOVA (TW-RM-ANOVA) was performed separately for each behavioral state viz. wakefulness, NREM and REM to determine the effect of alcohol withdrawal on sleep-wakefulness. Two independent variables were used: Alcohol (2 levels: Alcohol Dependent and Control) as between-the groups and Days (4 levels: Baseline, Day 3, Day 5, and Day 7) as within the group variable. The variables used for analysis were a) NREM sleep latency, b) NREM delta activity, c) the percentage of time spent in wakefulness, NREM and REM sleep along with d) number of bouts and e) average duration of bouts of each of these states. The significance level of p<0.05 was accepted. Once the significance was achieved by TW-RM-ANOVA, a post-hoc test (Bonferroni’s) was performed to identify the source of significance. Correlation analyses was performed to determine the relationship between percent time spent in NREM sleep and, NREM delta (1 to 4 Hz) or Beta (12–20 Hz) power. GraphPad Prism Software (La Jolla, CA) was used to perform all the statistical tests.

RESULTS

G*Power analysis (F test; α = 0.05; power = 0.95; N = 3/group) conducted after preliminary experiment suggested a total of 10 rats (5/Group) with an effect size of 0.51.

Light Period

A. Wakefulness:

Two-Way RM ANOVA suggested that there was a significant main effect of treatment (F_{1, 24} = 19.70, p<0.01), time (F_{3, 24} = 13.23, p < 0.001) and interaction (F_{3, 24} = 5.18, p < 0.01) on time spent in wakefulness. Post-hoc analysis confirmed that as compared to the rats in the control group, alcohol exposed rats displayed a significant increase in the amount of time spent in wakefulness on Days 3 (p < 0.001) and 5 (p < 0.01) (Figure 1A). No significant difference was observed on baseline day and Day 7 of alcohol withdrawal.

Figure 1:

Protracted changes in sleep-wakefulness of alcohol exposed rats during the light period of alcohol withdrawal. As compared to the rats in the control group, alcohol exposed rats displayed a significant increase in the amount of time spent in wakefulness (Panel A) with a concomitant reduction in NREM sleep (Panel B) on Days 3 and 5. No significant changes in the REM sleep was observed (Panel C). The percent time spent in NREM sleep was positively correlated with NREM delta activity (Panel D) and negatively correlated with NREM beta activity (Panel E). **p<0.01 and ***p<0.001 vs control group.

There was a no significant main effect of treatment (F_{1, 24} = 1.36, p > 0.05) on frequency of wake bouts however, a significant main effect of treatment (F_{1, 24} = 17.57, p < 0.01), time (F_{3, 24} = 6.17, p < 0.01) and interaction (F_{3, 24} = 3.08, p < 0.05) on duration of wake bouts, was observed. Bonferroni’s post-hoc analysis confirmed that as compared to the rats in the control group, alcohol exposed rats displayed a significant increase in the duration of wake bouts on Days 3 (p < 0.01) and 5 (p < 0.01) (see Table 1). The values on baseline day and Day 7 of alcohol withdrawal were comparable between the groups.

Table 1:

Frequency and duration of bouts of wakefulness, NREM and REM sleep during light (inactive) period of alcohol withdrawal days 3, 5 and 7.

Number of Bouts
	Wakefulness		NREM		REM
	Control	Alcohol	Control	Alcohol	Control	Alcohol
Baseline	100.7±3.5	102.2±3.7	100.9±3.5	102.4±3.8	40.5±2.4	43.5±7.3
Day 3	102.9±3.2	107.7±5.3	103.1±3.2	107.5±5.2	37.3±2.2	39.6±4.1
Day 5	102.7±3.5	99.4±8.4	103.1±3.5	99.7±8.5	41.8±3.3	42.7±2.2
Day 7	96.1±1.5	102.6±5.3	100.3±2.1	102.9±5.3	40.6±2.9	49.3±4.5
Duration per Bouts (s)
	Wakefulness		NREM		REM
	Control	Alcohol	Control	Alcohol	Control	Alcohol
Baseline	134.8±4.5	137.0±3.7	264.8±13.9	256.6±13.4	77.4±2.0	76.0±8.0
Day 3	142±2.7	192.6±16.7**	249.2±13.7	188.0±12.7*	82.6±3.2	69.6±8.9
Day 5	154.0±10	204.8±17.6**	236.2±3.8	209.8±16.5	77.8±3.9	69.4±3.4
Day 7	148.0±5.1	154.6±10.7	258.2±7.4	236.6±20.9	78.6±4.5	70.8±7.2

Alcohol = Alcohol exposed;

^*p<0.05 and

^**p<0.01 vs Controls

B. NREM sleep

Quantity:

There was a significant main effect of treatment (F_{1, 24} = 21.22, p<0.01), time (F_{3, 24} = 13.34, p < 0.001) and interaction (F_{3, 24} = 5.08, p < 0.01) on time spent in NREM sleep. As compared to the rats in the control group, alcohol exposed rats displayed a significant reduction in the amount of time spent in NREM sleep on Days 3 (p < 0.001) and 5 (p < 0.01) (Figure 1B). No significant difference was observed in NREM sleep between the groups on baseline day and Day 7 of alcohol withdrawal.

Two-Way RM ANOVA suggested that there was a no significant main effect of treatment (F_{1, 24} = 0.36, p > 0.05) on frequency of NREM sleep bouts however, there was a significant main effect of treatment (F_{1, 24} = 9.74, p < 0.05) and time (F_{3, 24} = 4.19, p < 0.05) on duration of NREM sleep bouts. Post-hoc analysis confirmed that as compared to the rats in the control group, alcohol exposed rats displayed a significant reduction in the duration of NREM bouts on Day 3 (p < 0.05) of alcohol withdrawal (see Table 1). The values on baseline day, Day 5 and Day 7 of alcohol withdrawal were comparable.

C. REM sleep:

Two-Way RM ANOVA suggested that there was no significant main effect of treatment found on time spent in REM sleep (Figure 1C). Frequency and duration of REM sleep bouts were also comparable between the groups (see Table 1).

Quality:

a). Sleep Latency:

A significant main effect of treatment (F_{1, 24} = 27.64, p < 0.001), time (F_{3, 24} = 18.43, p < 0.001) and interaction (F_{3, 24} = 23.01, p < 0.001) was observed on NREM sleep latency. Bonferroni’s Post-hoc analysis confirmed that rats exposed to chronic binge alcohol displayed a significant increase in the latency to NREM sleep on Days 3 (p < 0.001) and 5 (p < 0.01) as compared to the rats in the control group. However, values for NREM sleep latency were comparable between the groups on baseline day and Day 7 of alcohol withdrawal (see Table 2).

Table 2:

NREM sleep latency, delta and beta activity during light (inactive) period of alcohol withdrawal days 3, 5 and 7.

	NREM sleep latency (Min.)		NREM delta activity (μV2)		NREM beta activity (μV2)
	Control	Alcohol	Control	Alcohol	Control	Alcohol
Baseline	14.0±1.0	12.0±1.26	0.37±0.02	0.38±0.04	0.17±0.03	0.18±0.02
Day 3	12.4±2.5	40.20±2.5***	0.39±0.01	0.25±0.02***	0.14±0.01	0.28±0.02***
Day 5	13.0±2.9	26.4±2.9**	0.39±0.02	0.28±0.02**	0.15±0.01	0.23±0.01**
Day 7	14.0±2.0	9.8±2.0	0.40±0.04	0.36±0.04	0.14±0.02	0.17±0.02

Alcohol = Alcohol exposed;

^**p<0.01 and

^***p<0.001 vs Controls

b). NREM delta activity:

There was a significant main effect of treatment (F_{1, 24} = 12.04, p<0.01) and interaction (F_{3, 24} = 4.07, p < 0.05) on NREM delta activity. Post-hoc analysis confirmed that rats in the Alcohol group displayed a significant reduction in NREM delta activity on Days 3 (p < 0.001) and 5 (p < 0.01) as compared to the rats in the control group. However, NREM delta values were comparable between the groups on baseline day and Day 7 of alcohol withdrawal (see Table 2). Correlation analysis suggested that there was a significant (R² = 0.29, p < 0.001) positive correlation between the percent time spent in NREM sleep and NREM delta activity (Figure 1D).

c). NREM beta activity:

Two-Way RM ANOVA suggested that there was a significant main effect of treatment (F_{1, 24} = 16.02, p<0.01) and interaction (F_{3, 24} = 5.09, p < 0.01) on NREM beta activity. Post-hoc analysis confirmed that as compared to the rats in the control group, rats in the Alcohol group displayed a significant increase in NREM beta activity on Days 3 (p < 0.001) and 5 (p < 0.01). However, NREM beta values were comparable between the groups on baseline day and Day 7 of alcohol withdrawal (see Table 2). Correlation analysis suggested that there was a significant (R² = 0.18, p < 0.01) negative correlation between the percent time spent in NREM sleep and NREM beta activity (Figure 1E).

Dark Period

A. Wakefulness:

Two-Way RM ANOVA suggested that there was a significant main effect of treatment (F_{1, 24} = 13.23, p<0.01), time (F_{3, 24} = 4.73, p < 0.01) and interaction (F_{3, 24} = 12.81, p < 0.001) on time spent in the wakefulness. Post-hoc analysis confirmed that as compared to the rats in the control group, alcohol exposed rats displayed a significant reduction in the wakefulness on Days 3 (p < 0.001) and 5 (p < 0.01) (Figure 2A). The time spent in wakefulness on baseline day and Day 7 were comparable between the groups.

Figure 2:

Protracted changes in sleep-wakefulness of alcohol exposed rats during the dark period of alcohol withdrawal. As compared to the rats in the control group, alcohol exposed rats displayed a significant reduction in the wakefulness (Panel A) with a concomitant increase in NREM sleep (Panel B) on Days 3 and 5. The time spent in wakefulness on baseline day and Day 7 was comparable between the groups. No significant changes in the REM sleep were observed (Panel C). **p<0.01 and ***p<0.001 vs control group.

Two-Way RM ANOVA suggested that there was a significant main effect of treatment (F_{1, 24} = 13.60, p < 0.01), time (F_{3, 24} = 15.18, p < 0.001) and interaction (F_{3, 24} = 11.21, p < 0.001) on number of wake bouts. Post-hoc analysis confirmed that as compared to the rats in the control group, alcohol exposed rats displayed a significant increase in the frequency of wake bouts on Day 3 (p < 0.001) and Day 5 (p < 0.01) of alcohol withdrawal. No significant difference was observed in the number of wake bouts between the groups on baseline day and Day 7 of alcohol withdrawal (Table 3).

Table 3:

Frequency and duration (s) of bouts of wakefulness, NREM and REM sleep during dark (active) period of alcohol withdrawal days 3, 5 and 7.

Number of Bouts
	Wakefulness		NREM		REM
	Control	Alcohol	Control	Alcohol	Control	Alcohol
Baseline	73.8±2.5	75.5±3.1	74.0±2.5	75.9±3.2	40.6±0.7	29.8±2.5
Day 3	70.2±4.2	111.4±6.5***	70.4±4.2	110.4±6.3***	33.2±1.6	43.1±5.4
Day 5	82.0±7.6	108.3±4.8**	75.5±3.5	115.2±4.5***	34.8±3.5	31.8±5.2
Day 7	71.8±5.4	71.6±2.8	79.5±5.2	71.6±2.8	33.7±3.2	30.3±3.2
Duration per Bouts
	Wakefulness		NREM		REM
	Control	Alcohol	Control	Alcohol	Control	Alcohol
Baseline	430.8±15.3	421.4±16.3	256.6±13.4	132.6±15.8	44.0±2.1	68.2±10.0
Day 3	458.4±28.8	232.0±16.1***	188.0±12.7	139.2±3.8	53.0±2.9	60.6±6.0
Day 5	401.4±27.4	263.4±15.2***	209.8±16.5	116.0±4.0	51.4±4.4	55.2±4.3
Day 7	431.0±32.0	437.0±9.4	236.6±20.9	148.2±14.8	49.0±2.8	53.4±3.9

Alcohol = Alcohol exposed;

^**p<0.01 and

^***p<0.001 vs Controls

Two-Way RM ANOVA suggested that there was a significant main effect of treatment (F_{1, 24} = 24.89, p < 0.01), time (F_{3, 24} = 14.48, p < 0.001) and interaction (F_{3, 24} = 15.70, p < 0.001) on duration of wake bouts. Post-hoc analysis confirmed that as compared to the rats in the control group, alcohol exposed rats displayed a significant reduction in the duration of wake bouts on Day 3 (p < 0.001) and Day 5 (p < 0.001) of alcohol withdrawal. No significant difference was observed in the duration of wake bouts between the groups on baseline day and Day 7 of alcohol withdrawal (Table 3).

B. NREM sleep:

Two-Way RM ANOVA suggested that there was a significant main effect of treatment (F_{1, 24} = 14.17, p<0.01), time (F_{3, 24} = 7.55, p < 0.001) and interaction (F_{3, 24} = 15.29, p < 0.001) on time spent in NREM sleep. Post-hoc analysis confirmed that as compared to the rats in the control group, alcohol exposed rats displayed a significant increase in NREM sleep on Days 3 (p < 0.001) and 5 (p < 0.01) (Figure 2B).

Two-Way RM ANOVA suggested that there was no significant treatment (F_{1, 24} = 0.036, p > 0.05) main effect of treatment on duration of NREM bouts however, a significant main effect of treatment (F_{1, 24} = 33.51, p < 0.01), time (F_{3, 24} = 12.94, p < 0.001) and interaction (F_{3, 24} = 18.86, p < 0.001) on number of NREM bouts. Post-hoc analysis confirmed that as compared to the rats in the control group, alcohol exposed rats displayed a significant increase in the frequency of NREM bouts on Day 3 (p < 0.001) and Day 5 (p < 0.001) of alcohol withdrawal. No significant difference was observed in the number of NREM bouts between the groups on baseline day and Day 7 of alcohol withdrawal (Table 3). During the dark (active) period, analysis for NREM delta (F_{1, 24} = 0.10, p > 0.01) and beta (F_{1, 24} = 0.02, p > 0.01) activities during NREM sleep did not show any significant difference between control and Alcohol-exposed groups (Data not shown).

C. REM sleep:

There was no significant difference found in the REM sleep between control and alcohol exposed groups during the dark period of alcohol withdrawal (Figure 2C).

DISCUSSION

In the present study, we have shown that alcohol exposed rats displayed protracted insomnia-like symptoms during alcohol withdrawal. The main findings of the study are that alcohol exposed rats displayed a) insomnia-like symptoms as revealed by a significant reduction in the quantity and quality of sleep during the light (inactive) period and b) excessive daytime sleepiness-like behavior as revealed by a significant increase in NREM sleep with a concomitant reduction in the amount of time spent in the wakefulness during the dark (active) period of alcohol withdrawal.

We used Majchrowicz’s protocol to induce alcohol dependency [21]. This method involves multiple binge alcohol episodes over the course of four days in order to maintain uninterrupted BAC to induce alcohol dependence in rats mimicking conditions as observed in human alcoholics such as high blood alcohol levels (>200mg/dL), heavy bingeing and withdrawal symptoms [25, 26]. This chronic alcohol method has been extensively used to understand the effects of alcohol dependency in experimental animals including sleep disruptions during acute withdrawal [16, 22].

In the present study, we have investigated long term (7 days) effects of alcohol withdrawal on sleep-wakefulness in alcohol exposed rats. We did not measure blood alcohol concentration (BAC) to verify alcohol withdrawal, however, on day 3 of alcohol withdrawal, animals should be free of alcohol as we have observed in our previous study that the BAC was negligible after 12 h of last dose of alcohol [16]. Our results suggest that alcohol exposed rats displayed severe and protracted sleep disruptions during withdrawal.

During the light period of alcohol withdrawal (Day 3 and Day 5), the alcohol exposed rats displayed a significant reduction in the quantity (reduction in the amount of time spent in NREM sleep) and quality (increased sleep latency, reduced NREM delta activity and increased NREM beta activity) of sleep. NREM delta and beta activities are considered as markers for insomnia [23]. Corroborating our findings, evidence from clinical studies suggests that alcoholics with symptoms of insomnia display a reduction in the delta and an increase in beta activity during alcohol abstinence (reviewed in [27]). In addition, behavioral state transition analysis suggests that the increased wakefulness in alcohol exposed rats is caused mainly due to a significant increase in the duration of wake bouts during the light period. We did not find an overall significant change in REM sleep across days however, we did find a trend in reduction in REM sleep on Day 3. Although we did not thoroughly investigate the underlying cause of such long, uninterrupted bouts of wakefulness, based on our previous study it is very likely that homeostatic sleep mechanism is impaired in these animals [16]. Such impairments in sleep homeostatic mechanisms include a reduction in adenosinergic mechanisms such as reduced adenosine levels along with a reduction in the expression of equilibrative nucleoside transporter and adenosine 1 receptor in the basal forebrain [16].

Our results are in line with previous clinical and preclinical studies. For example, several polysomnographic studies investigating sleep-wake behavior in alcoholics suggested that acute and sustained withdrawal results in sleep disturbances including difficulty in falling and staying asleep, reduced sleep efficiency or percentage time spent in NREM sleep and increased wakefulness during the night that lasted for several months during abstinence (reviewed in [8, 28]). In the preclinical studies, Mendelson and his coworkers showed insomnia-like symptoms in alcohol exposed rats. However, they examined sleep-wakefulness during acute withdrawal (4 days) and only during 8 h of the light period of each withdrawal day [29]. Similarly, Veatch (2006) examined the effect of chronic alcohol exposure in mice during acute withdrawal (4 days) and reported insomnia-like symptoms in mice however, these animals were not alcohol exposed.

During the dark period, alcohol exposed rats displayed an increased amount of NREM sleep with a concomitant reduction in wakefulness during withdrawal days 3 and 5. Moreover, further analysis suggests that there was a significant increase in the number of NREM bouts during the dark period suggesting that the maintenance of wakefulness was affected. Corroborating our findings, there is some evidence that suggests that, during alcohol withdrawal, patients with AUD display narcolepsy-like symptoms such as uncontrollable excessive daytime sleepiness and disrupted nocturnal sleep, a disease caused by loss of the orexinergic system [11, 30, 31]. Recently, we have shown that alcohol exposed rats displayed a significant downregulation in the expression of orexin in the lateral hypothalamus implicating that downregulation of the orexin gene may be a possible link between excessive daytime sleepiness-like behavior and alcohol withdrawal [32]. We did not find an overall significant change in REM sleep across days during the dark period however, we did find an increasing trend in REM sleep on Day 3. This may be due to a rebound effect of REM sleep reduction, although not significant, which we observed during the light period of Day 3. Such a rebound in REM sleep has also been observed in human alcoholics during alcohol withdrawal [33].

In human alcoholics, sleep disturbances reduce to some extent with sobriety in the first six months. However, this is only observed in patients who remained sober but not in the patients who relapse to heavy drinking [34]. Similarly, in our study, we have observed that the alcohol exposure-related alterations in sleep-wakefulness (i.e., changes in wakefulness, NREM sleep, and delta and beta activity) returned to baseline by withdrawal day 7 which is equivalent to 9–12 human months. This is important as it has provided us with a clinically relevant time window to understand the mechanism of alcohol-associated sleep disturbances in mice.

One limitation of this study is that it was conducted in males only. We are in the process of conducting a separate study to examine the effects of alcohol withdrawal on sleep-wakefulness in females.

In summary, our results suggest that alcohol withdrawal leads to severe and protracted sleep disturbances in the alcohol exposed rats as revealed by a reduction in the quantity and quality of sleep during the light (inactive) period and an increase in NREM sleep during the dark (active) period mimicking the clinical condition of insomnia associated sleep disturbances in AUD patients. In conclusion, the chronic binge model of alcohol dependence mimics clinical symptoms of AUD especially protracted insomnia, and is suitable for understanding the mechanisms associated with alcohol withdrawal-induced behaviors.

Highlights.

During withdrawal, alcohol dependent rats:

• displayed protracted sleep disturbances.

• showed insomnia-like symptoms during light (inactive) period.

• showed a significant increase in the NREM sleep during dark (active) period.

• did not display any significant changes in the REM sleep.

Abbreviations:

AUD

Alcohol use disorder