Reduced Alcohol Drinking Following Patterned Feeding: Role of Palatability and Acute Contingent Availability

Abstract

Recent studies from our lab have demonstrated that intermittent high-fat diet access reduces alcohol drinking in rats. However, it was unclear if caloric overload, palatability, or diet itself triggered reduced alcohol drinking. It is also unknown if a similar paradigm could reduce relapse-like alcohol drinking. The presented study tested the hypothesis that acute intermittent PD access would rescue relapse-like drinking and palatability, but not diet itself contributes to reduced drinking. Male Long Evans rats received six-weeks intermittent or chronic chow (controls) or palatable diets (PDs; high-fat diet, high-sugar diet) exposure, and alcohol testing occurred following PDs suspension. Alcohol intake was not significantly different among groups in either condition, suggesting that diet itself did not impact alcohol drinking. A subset of these rats received two-weeks intermittent PDs (Int-PDs) exposure and alcohol testing reinitiated while Int-PDs access continued. Alcohol intake significantly escalated (~137% compared to baseline; alcohol deprivation effect) in the chow controls, whereas it remained unchanged in PD groups. These data demonstrate the critical importance of acute intermittent PDs availability and its protective effect in relapse-like drinking. To assess the contribution of palatability in reduced alcohol drinking, a separate group of rats received two-weeks intermittent high-sugar diet (Int-HSD) or saccharin (Int-SAC) access and tested for alcohol drinking while Int-HSD/SAC continued. Alcohol drinking significantly decreased (~30%) in both HSD and SAC groups compared to the controls. These data identify the critical parameters by which acute intermittent PD access reduces alcohol drinking and could have important therapeutic implications in the management of alcoholism.

1. INTRODUCTION

Alcohol use disorder (AUD) is the second most prevalent psychiatric illness after major depressive disorder in patients with dysregulated eating [1]. Unlike other drugs of abuse, alcohol represents a significant source of calories (7 kcal/g), and problematic alcohol drinking could impact energy balance, reduce food intake and nutrient absorption [2]. In this context, reduction in food/calorie intake before, during, or after alcohol drinking episodes has been reported, which could be a compensatory behavior and/or an attempt to maximize alcohol intoxication [3,4]. Therefore, chronic alcohol abuse could result in impaired body composition and nutritional status [5,6], which may contribute to the pathology of AUD [7–9]. Interestingly, alcoholics who stayed sober longer displayed increased intake of highly palatable food during early recovery [10,11], suggesting a protective role of nutrition-related interventions in the management of AUD. However, the manner by which altered nutritional status interacts to promote the initiation or maintenance of AUD is poorly understood.

Several studies have examined alcohol drinking behavior following palatable diets exposure and have obtained mixed results. For example, studies exist supporting a positive [12–14], negative [15–18], and no [19] association between palatable diets (PDs) exposure and alcohol drinking. A detailed account of these findings is recently reviewed [20]. Briefly, earlier studies reporting escalated alcohol drinking in rats evaluated alcohol drinking following acute, one week, or four weeks of high-fat diet (HFD) exposure [12,13]. Whereas, Krahn and Gosnell selected rats based on their inherent preference for high-fat diet and reported increased alcohol intake in this group [14]. However, it has been suggested that macronutrient preference may not reliably predict alcohol drinking behavior [19]. On the other hand, a relatively recent study reported blunted ethanol preference following five weeks of ad libitum HFD exposure in mice [15]. Another study evaluating sex-differences in alcohol intake in mice documented reduced alcohol drinking following three weeks of HFD exposure [16]. Reduced alcohol drinking has also been reported in golden hamsters maintained on HFD [17] and in male Wistar rats following four weeks of cafeteria diet composed of high-calorie junk food [18].

While the palatable HFD in these studies was provided in an acute or chronic manner, studies have also evaluated the impact of binge-like intake of HFD on alcohol intake. For example, studies from our lab have demonstrated that patterned feeding of a nutritionally complete HFD reduces alcohol drinking in rats [21–23]. In these studies, diet and alcohol were never provided on the same day, and alcohol testing occurred under acute withdrawal, while the rats were still maintained on intermittent HFD cycling. This was not the case with the previously mentioned studies in which palatable diets were provided in a chronic manner, and alcohol testing occurred once dietary manipulations were released. Therefore, it was unclear if the HFD itself or its intermittent availability triggered reduced alcohol drinking in our studies. It was also unknown if the diet-induced effects on alcohol drinking are macronutrient specific, and if a diet high in sugar would produce similar effects on alcohol drinking. While the reduction in alcohol drinking behavior following intermittent palatable diet exposure has been repeatedly demonstrated, the functional significance of these findings remains unexplored. In the present study, a series of experiments were designed to address these questions.

First, we assessed if the effects of a palatable diet on alcohol drinking are nutrient specific and if this effect would differ when PDs are provided in an intermittent or chronic manner. The impact of six weeks intermittent and chronic HFD and HSD access on alcohol drinking was evaluated once the palatable diet manipulations were released, and rats were maintained on chow. We have recently shown that acute intermittent availability of a palatable diet is critical to observe reduced alcohol drinking behavior [23]. Therefore, it was predicted that chronic or intermittent PDs exposure would have no effect on alcohol drinking.

While acute intermittent availability of a palatable diet has been shown to reduce the acquisition and maintenance of alcohol drinking behavior [22,23], the functional significance of these findings in the management of alcoholism remains unexplored. For example, it is unknown whether a similar approach would rescue relapse-like drinking. We evaluated the alcohol deprivation effect, which has been successfully utilized in the laboratory to assess relapse-like alcohol drinking [24]. It was predicted that acute intermittent availability of a palatable diet would have a protective effect when tested under alcohol deprivation condition (relapse-like drinking).

Acute intermittent access to a palatable diet is necessary to observe a reduction in alcohol drinking behavior [23]. Since the intermittent palatable diet exposure produces a caloric overconsumption/underconsumption pattern, it could be that reduction in alcohol intake is triggered by caloric overload. However, in the same study, we demonstrated that energy homeostasis pathways are probably less likely involved in this phenomenon. To clarify the role of caloric overload vs. palatability, we assessed if intermittent exposure to a non-caloric palatable substance would impact alcohol drinking in a similar manner. Rats received saccharin (a non-caloric palatable substance) in a similar two-weeks intermittent access paradigm, and alcohol drinking was evaluated to examine the contribution of palatability in reduced alcohol drinking. It was predicted that intermittent saccharin access would reduce alcohol drinking.

Overall, it was hypothesized that acute intermittent PD access would rescue relapse-like drinking and palatability, but not diet itself triggers reduced alcohol drinking following our intermittent PD access paradigm.

2. METHODS

2.1. Animals

Adult male Long-Evans rats (~ 300 g, nine weeks old) were used and were supplied by Envigo RMS, Inc (Indianapolis, IN). Upon arrival, the rats were quarantined for ten days before being individually caged in the experimental rooms with appropriate conditions: 60%−70% humidity, 65–75° F, and 12h/12h reverse light/dark cycle. During the first week, animals were gently handled, and baseline data (body weight and food/water intake) were collected. Food and water were available ad libitum to all the rats during the entire study. All experimental procedures were approved by the Institutional Animal Care and Use Committee at the Xavier University of Louisiana.

2.2. Diets

All rats had ad libitum access to standard rodent chow (Tekland-Envigo Diets #2020X). The experimental groups received a high-fat diet (HFD; Research Diets #D03082706), high sugar diet (HSD; Research Diets #D10001), or saccharin (SAC; Necta Sweet, NSI sweeteners INC). Fig 1A presents detailed dietary compositions of the palatable conditions used in the present study.

Figure 1. Schematic representation of palatable conditions and experimental protocol.

A) Caloric compositions of the palatable diets/solutions used in the present study: Standard Chow, a high-fat diet (HFD), high-sugar diet (HSD), and Saccharin (SAC; 0.2% w/v). B) A flow chart/timeline with brief descriptions of various experiments conducted in the present study. All rats had ad libitum access to chow and water.

2.3. Experimental Procedure

Rats matched for their body weight, food intake, and water intake were randomly divided into respective control and experimental groups. Three sets of experiments were conducted as described below, also outlined in Fig 1B.

Experiment 1:

To assess the impact of diet and exposure conditions on alcohol drinking, groups of rats received intermittent (n=5/group) access (24h, Tue and Thu) or chronic (n=10/group) access (24 hr every day) to palatable diets (HFD or HSD) for six-weeks. Each experimental condition had its own control that received chow under identical conditions. Standard rodent chow and water were available ad libitum throughout the experiment to all groups of rats. Food intake was measured every 24 hrs from Monday to Friday only. While food intake data are unavailable during weekends and two other separate occasions, all animals received appropriate diets during these periods, as outlined in Fig 1B. Following six weeks of intermittent or chronic PD exposure, access to palatable diets was suspended, and all animals were maintained on chow for the duration of alcohol testing. All rats were provided with alcohol bottles daily in their home-cage in a two-bottle choice paradigm (explained in detail in section 2.5 below). Alcohol intake was measured every 24 hr in 8 testing sessions, in which 8% (v/v) alcohol was provided in the first five, and 20% (v/v) alcohol was provided in the last three alcohol drinking sessions. All animals, except a subset used for experiment 2, were euthanized at the end of this experiment.

Experiment 2:

Next, we examined if acute intermittent PD access would be protective against relapse-like alcohol drinking. The alcohol deprivation effect, which is a temporary increase in alcohol drinking behavior compared to baseline following renewed access to ethanol after an alcohol deprivation period, has been successfully utilized in the laboratory to assess relapse-like alcohol drinking [24]. Since rats in the experiments 1 had already established free-choice alcohol drinking behavior, a subset of rats (n=4–5/group) with no between-group differences in alcohol intake was used in this experiment. At the conclusion of experiment 1, their alcohol access was terminated, and they received the same diet they received during experiment 1 (chow or PDs) in an intermittent access manner for two-weeks. Following two weeks of pre-exposure to the respective diets, alcohol consumption (20% v/v) was examined on chow-only days (Mon, Wed and Fri) while the intermittent access cycling (Tue and Thu) continued, which is consistent with the previous studies utilizing this intermittent PD access paradigm in our lab. Please refer to section 2.5 for a detailed alcohol testing procedure. All animals were euthanized at the end of this experiment.

Experiment 3:

A new set of rats were obtained and randomly divided into three groups (n=6-10/group) matched for their baseline weight, food, and water intakes. Group of rats received intermittent access to high-sugar diet (HSD) or saccharin (SAC) for two-weeks. The control group of rats received standard rodent chow during these times. SAC was dissolved (0.2%, w/v) in drinking water and was provided in addition to the normal drinking water bottle in the home cages. Again, standard rodent chow and water were available ad libitum throughout the experiment to all groups of rats. Food intake was measured every 24 hrs from Monday to Friday only. While food intake data are unavailable during weekends and two separate occasions, all animals received appropriate diets during these periods, as outlined in Fig 1B. Following two-weeks of intermittent HSD/SAC pre-exposure, the rats were tested for alcohol consumption (20% v/v) on chow-only days (Mon, Wed, Fri) while the intermittent access cycling (Tue and Thu) continued. All animals were euthanized at the end of this experiment.

2.4. Intermittent Access to Palatable Diet Feeding Paradigm:

Bodyweight, food, and water intake matched rats were randomly divided into the palatable diets (PDs) groups or chow controls. The experimental group of rats received 24 hr access to respective palatable diets (HFD, HSD, SAC) on Tuesday and Thursday for an initial exposure duration, as specified above (see section 2.3). Following initial PD pre-exposure, alcohol testing occurred on the chow access days (Mon, Wed, and Fri), and intermittent PD cycling continued during these alcohol testing sessions for Experiments 2 and 3, whereas it was suspended for Experiment 1. PDs and alcohol were never presented on the same day.

2.5. Alcohol testing

190 proof alcohol was purchased from Greenfield Global, CT, and the desired concentrations were freshly prepared every week at least one day in advance of testing. Since patterned feeding could promote future bouts of palatable food intake [25], alcohol was not mixed with any sweetener in the present study and was presented unsweetened. On the days of alcohol testing, rats were provided with unsweetened alcohol (8 or 20% v/v) and water bottle in their home-cages under a two-bottle choice paradigm. Bodyweight, water intake, and alcohol intake (every 24hr) were measured these days. Water and alcohol bottles were alternated every other alcohol testing day to account for conditioning effects on alcohol intake. All bottles were manually weighed, and alcohol bottles (50 ml tubes) were refreshed on each testing session. Two alcohol and two water bottles were handled similarly in an empty cage and these values were subtracted from respective experimental bottles measurements to account for any drip or evaporation. Alcohol intake is expressed as g/kg, and alcohol preference was calculated as alcohol intake/water intake.

2.6. Statistical Analysis

One-way ANOVA analyzed baseline body weight, food, and water intake data. A mixed model two-way ANOVA analyzed body weight, food intake, alcohol intake, total fluid intake, and alcohol preference across exposure/testing duration, followed by appropriate post-hoc (LSD) testing. The between-group variables were the different feeding conditions (i.e., chow, HFD, HSD, or SAC), and the within-subject variable was the time interval between different sessions. A univariate analysis of variance analyzed 24 hr food intake on the palatable diet days or chow only days. All statistical comparisons were conducted at 0.05 α level and power (1-β, β=0.2).

3. RESULTS

3.1. Six weeks of intermittent and chronic palatable diets exposure: body weight

While there was a main effect of time (F_{1.259, 15.111} = 645.093, p<0.000), no significant diet × time interaction or between-group body weight differences existed in rats receiving six-weeks intermittent access to HFD or HSD compared to the chow controls (Fig 2A). However, in case of six weeks chronic HFD or HSD exposure a main effect of time (F_{1.569, 42.359} = 1538.868, p<0.001), a significant diet × time interaction (F_{3.138, 42.359} = 11.591, p<0.001) and a main effect of diet (F_{2, 27} = 4.987, p<0.014, power=0.77) existed. Post hoc analysis further revealed that body weight (g) was significantly (p<0.01) elevated in the chronic HFD group (g; 528.3±13.2) of rats compared to the chow (g; 479.3±4.5) controls (Fig 2B). No other statistically significant differences existed.

Figure 2. Bodyweight changes during intermittent and chronic palatable diets feeding.

Mean (± SEM) body weight gain during six-weeks of intermittent and chronic palatable diets (PDs) exposure, including baseline (day 1). A) Intermittent PD access feeding paradigm did not change body weight in either HFD or HSD group of rats compared to controls. B) HFD group of rats in the chronic PD access-group gained significantly higher body weight over the exposure duration compared to the chow controls. *p<0.05 main effect of PD exposure.

3.2. Six weeks of intermittent and chronic palatable diets exposure: food intake

Rats receiving intermittent HFD or HSD diet exposure displayed a pattern of overconsumption/ underconsumption (Fig 3 A, B). A mixed model Two-Way ANOVA identified a main effect of time (F_{5.428, 65.133} = 103.41, p<0.001), a significant diet × time interaction (F_{10.855,65.133} = 31.88, p<0.001) and a main effect of diet (F_{2, 12} = 12.26, p<0.01, power=0.98). Post-hoc analysis further revealed that both HFD and HSD group of rats significantly (p<0.001) overconsumed (Kcal) on the palatable diet access days (kcal; Chow=67.5±2.4, HFD=121.5±4.4, HSD= 104.4±3.5) and underconsumed (p<0.01; HFD group, p<0.05; HSD group) on the chow access days (kcal; Chow=69.4±1.7, HFD=59.8±0.9, HSD= 63.2±2.5), respectively. Furthermore, overall mean caloric intake during six weeks was also significantly higher (p<0.01) in the HFD group of rats compared to the HSD group of rats (Fig 3B) on PD access days. Cumulative food intake data were also analyzed using a mixed model Two-Way ANOVA, which revealed similar relationship as a main effect of time (F_27,324 = 3245, p<0.001), a significant diet × time interaction (F_54,324 = 10.78, p<0.0001) and a main effect of diet (F_{2, 12} = 13.02, p=0.001). Cumulative food intake (kcal; Chow= 1920.6±54.9, HFD=2350.8±57.9, HSD= 2220.8±74.1) was significantly elevated in the HFD (p<0.01) and HSD (p<0.05) group of rats compared to the chow controls (Fig 3 C).

Figure 3. Food intake during six-weeks of intermittent or chronic palatable diets exposure.

Data compares mean (±SEM) food intake (kcal) among chow and palatable diet groups. A) Intermittent HFD and HSD group of rats displayed a pattern of overconsumption/underconsumption compared to the chow controls. **p<0.01 main effect of PD exposure. B) Overall, average caloric intake during six weeks of intermittent HFD and HSD group of rats was significantly elevated on palatable diet access days, and a significantly under consumption was observed on chow-only access days. *p<0.05, **p<0.01, ****p<0.001 compared to chow. ^δδp<0.01 compared to HFD. C) Mean (±SEM) cumulative food intake. ***p<0.001 main effect of PD exposure. D) Daily caloric consumption of rats in the chronic HFD and HSD groups compared to the chow controls. ^λλλλp<0.0001 main effect of PD exposure. E) Overall, average food intake during six weeks in the chronic HFD and HSD groups of rats was significantly elevated compared to chow controls. ****p<0.001 compared to chow. ^δδp<0.01 compared to HFD. F) Mean (±SEM) cumulative food intake. ^λλλλp<0.0001 main effect of PD exposure.

Chronic HFD or HSD access group of rats significantly overconsumed daily during the six weeks exposure period compared to the chow controls (Fig 3 D). A mixed model Two- Way ANOVA revealed a main effect of time (F_{9.087, 245.346} = 3.957, p<0.001), a significant diet × time interaction (F_{18.174, 245.346} = 2.454, p<0.01) and a main effect of diet (F_{2, 27} = 29.758, p<0.001, power=1.0). The post-hoc analysis further revealed that overall mean caloric intake (kcal) during six weeks was significantly higher (kcal; Chow=69.5±0.7, HFD=96.5±4.1, HSD= 82.4±2.5) (p<0.01) in HFD group of rats compared to the HSD group of rats (Fig 3 E). Cumulative food intake data were also analyzed using a mixed model Two-Way ANOVA, which revealed similar relationship as a main effect of time (F_26,702 = 2933, p<0.0001), a significant diet × time interaction (F_52,702 = 23.72, p<0.0001) and a main effect of diet (F_{2, 27} = 30.91, p<0.0001). Cumulative food intake (kcal; Chow= 1876.9±18.4, HFD=2604.7±110.4, HSD= 2310.6±31.3) was significantly elevated in the HFD (p<0.0001) and HSD (p<0.001) group of rats compared to the chow controls. Furthermore, caloric intake was also significantly higher (p<0.05) in the HFD group of rats compared to the HSD group of rats (Fig 3 F).

3.3. Six weeks of intermittent and chronic palatable diets exposure: alcohol drinking

A mixed model Two- Way ANOVA analyzed alcohol drinking data in rats following six-weeks of intermittent or chronic exposure to HFD and HSD. While there was a significant diet × time interaction (F_{14, 84} = 2.043, p<0.05) in the case of intermittent PD group of rats, alcohol drinking was not significantly different between groups (Fig 4 A). There was no significant diet × time interaction or between diet groups difference in alcohol drinking in chronic palatable exposure conditions (Fig 4 B).

Figure 4. Alcohol drinking following six-weeks of intermittent and chronic PD pre-exposure.

Data presents daily mean (±SEM) alcohol consumption (8% or 20% v/v) following six-weeks of (A) intermittent and (B) chronic PD exposure compared to the respective chow controls. The palatable diet exposure was suspended during the alcohol testing period, and all rats were maintained on chow. Alcohol (8.0% v/v for the first five sessions and 20.0% last three sessions) was provided in the home cage daily, and 24 hr alcohol intake was assessed by manually weighing the bottles. Alcohol drinking was not significantly different among any groups or exposure conditions.

3.4. Intermittent palatable diets availability: relapse-like drinking

Alcohol drinking was further evaluated on Mon, Wed, and Fri after two-weeks intermittent palatable diets exposure while the animals remained on intermittent palatable-diet cycling. A mixed model Two-Way ANOVA further revealed a significant between-group effect (F _2,11 = 6.594, p<0.05, power=0.81). In this case, alcohol intake (g/kg) was significantly (p<0.05) escalated in the chow group (g/kg; 2.47±0.4) of rats compared to their baseline (g/kg; 1.13±0.17), whereas no such effect occurred in the HFD or HSD group of rats (Fig 5). The post-hoc analysis further confirmed that alcohol intake in both HFD (p<0.05) and HSD (p<0.01) group of rats was significantly decreased compared to the chow controls.

Figure 5: Acute intermittent palatable diet access and relapse-like alcohol drinking.

Following alcohol testing, a subset of animals from experiment 1, was placed on an intermittent feeding pattern for 2 weeks during which alcohol was unavailable. Following two-weeks, alcohol (20% v/v) was re-introduced, and alcohol intake was recorded on chow-only access days (Mon, Wed, Fri), while the intermittent PD access continued during alcohol testing. Data present mean (±SEM) alcohol consumption (20% v/v) as g/kg during these alcohol drinking sessions. Rats in the chow control significantly escalated their alcohol consumption from baseline (alcohol deprivation effect/relapse-like drinking), whereas alcohol consumption remained unchanged in the case of intermittent PD group of rats. *p<0.05, **p<0.01 compared to the baseline.

3.5. Two weeks of intermittent Saccharin or HSD exposure: food intake

Average daily food intake data from animals receiving two-weeks intermittent access to saccharin (SAC) or HSD were analyzed by mixed model Two-Way ANOVA, which revealed a main effect of time (F_{2.778, 61.119} = 31.709, p<0.001), a significant exposure condition × time interaction (F_{5.556, 61.119} = 28.905, p<0.001) and a main effect of exposure condition (F_{2, 22} = 6.350, p<0.01, power=0.85). The post-hoc analysis further identified that food intake was significantly (p<0.01) higher in the HSD group of rats, whereas food intake in the SAC group of rats was not significantly different from chow controls (Fig 6A). Further analysis revealed that HSD group of rats significantly (p<0.001) overconsumed (kcal) on the palatable diet access days (kcal; Chow=59.7±1.7, HSD=93.4±5.5, SAC= 62.0±1.4) and underconsumed (p = 0.07) on the chow access days (kcal; Chow=60.3±1.7, HSD=54.7±2.2, SAC= 61.8±2.1), whereas no such pattern existed in case of SAC group of rats compared to the chow controls (Fig 6B).

Figure 6. Food intake during intermittent saccharin and HSD exposure.

A separate set of animals received two-weeks intermittent access to HSD or saccharin (SAC). Data compare mean (±SEM) average daily food intake among chow controls, intermittent-HSD, and intermittent-SAC. A) Intermittent-HSD rats, but not intermittent-SAC rats, displayed caloric overconsumption/underconsumption pattern. **p<0.01 main effect of PD exposure B) Overall, caloric intake in the intermittent HSD group of rats was significantly higher on palatable diet access days and lower (a trend p=0.07) on the chow access days compared to the chow controls. On the other hand, caloric intake in the intermittent-SAC rats was similar to the chow controls under these conditions. ****p<0.001 compared to chow.

3.6. Two weeks of intermittent Saccharin or HSD exposure: alcohol drinking

A mixed model Two-Way ANOVA identified a main effect of time (F_{3.17, 69.748} = 3.076, p<0.05), a significant exposure condition × time interaction (F_{6.341, 69.748} = 2.515, p<0.05) and a main effect of exposure condition (F_{2, 22} = 4.548, p<0.05 power=0.71) (Fig 7A). We further analyzed weekly alcohol intake and found that alcohol drinking (g/kg) was not significantly different between groups (g/kg; Chow=2.66±0.7, HSD=2.87±0.24, SAC= 1.52±0.39) in the first week. However, a mixed model Two-Way ANOVA identified a significant main effect of exposure conditions during second (F_{2, 22} = 5.556, p<0.05, power=0.80) and third (F_{2, 22} = 6.614, p<0.05, power=0.87) weeks of alcohol testing. Furthermore, a significant decrease in alcohol intake was evident in the third week (g/kg; Chow=3.93±0.5, HSD=2.31±0.16, SAC= 1.68±0.52) in both HSD (p<0.05) and SAC (p<0.01) group of rats compared to the chow controls (Fig 7B). Under these conditions, the total between-group fluid intake on alcohol drinking day was not significantly different (Fig 7C). Alcohol preference was also reduced in experimental groups compared to the chow control as a mixed-model ANOVA also revealed a strong trend towards significance (P = 0.05) in between-group effects (Fig 7D). Similar to the alcohol drinking data, alcohol preference was significantly lower in the HSD (p<0.05) and SAC (p<0.01) group of rats at later time points only (Fig 7D).

Figure 7. Impact of intermittent HSD and SAC access on alcohol drinking.

Data present mean (±SEM) 24 hr alcohol consumption (20% v/v) on chow-only days (Mon, Wed, Fri) after an initial two-weeks of PD pre-exposure. Intermittent access paradigm was continued during the alcohol testing period. A) Reduced alcohol drinking was observed in palatable diet/solution groups compared to the chow controls. *p<0.05 main effect of PD exposure. B) While the alcohol intake was not different among groups for the first week, a significant reduction in alcohol consumption was observed gradually in both intermittent HSD and SAC groups of rats. *p<0.05, **p<0.01 compared to chow controls. C) Total fluid intake was not significantly different among groups during alcohol testing days. D) A non-significant (p=0.05) decrease in the alcohol preference (ratio of alcohol to water) was also observed in the intermittent HSD and SAC group of rats compared to chow controls, an effect statistically significant during later time points.

4. DISCUSSION

The present study evaluated the impact of intermittent access of palatable diets on relapse-like alcohol drinking and whether palatability or diet itself contributes to the reduced alcohol drinking following intermittent palatable diet exposure. We found that neither intermittent nor chronic six-weeks of HFD/HSD exposure had any impact on alcohol drinking by itself. However, when alcohol was re-introduced following two-weeks of intermittent HFD/HSD exposure, rats exposed to chow significantly escalated their alcohol intake (alcohol-deprivation effect), whereas no such effect was present in groups of rats receiving intermittent PDs exposure. Finally, acute intermittent access to both HSD and SAC could reduce alcohol drinking, but only HSD induced caloric overconsumption/underconsumption pattern, whereas caloric intake in the SAC group of rats was identical to chow controls. These data collectively support the hypothesis and suggest that acute intermittent availability of a palatable diet is critical to trigger reduced alcohol drinking and identifies the contributing factors resulting in reduced alcohol drinking following intermittent palatable diets exposure.

We recently reported that six-weeks intermittent pre-exposure of a nutritionally complete high-fat diet significantly reduced alcohol drinking in rats [21,22]. However, it remained to be identified if a similar paradigm could also rescue relapse-like drinking behavior. It was also unclear if the diet, its intermittent availability, palatability, or caloric overload triggered reduced alcohol drinking. The present study was designed to address these questions. To assess whether a palatable diet itself or the intermittent pattern could trigger a reduced alcohol drinking phenotype, a group of rats received intermittent or chronic access to HFD or HSD for six-weeks. Rats in the intermittent PD group displayed a pattern of overconsumption/underconsumption, whereas rats in the chronic PD group consistently overate every day (Fig 3). When tested following the suspension of PDs, alcohol drinking was not significantly different among groups in either condition (Fig 4), suggesting that neither the diet nor the patterned feeding was capable of impacting alcohol drinking.

It is important to note that several studies have examined the impact of chronic HFD exposure on alcohol drinking and have obtained mixed results, and data exist supporting both increased [13,14] and decreased [15–18] alcohol drinking in the HFD group of rats compared to controls. Only a handful of studies have examined the impact of high carbohydrate and sugar exposure on alcohol drinking and found a negative relationship between carbohydrate intake and alcohol drinking [13,26]. However, when rats received intermittent access to sugar (10% sucrose) solution, an increase in voluntary ethanol intake has been reported [27]. Interestingly, intermittent access group in this study were food-deprived (12hr) followed by 12 hr access to chow (Inter-C) or chow and sugar (Inter-C+S) and was compared to the corresponding ad libitum chow (Adl-C) and sugar and chow (Adl-C+S) conditions. Alcohol intake in both intermittent conditions was significantly escalated compared to the ad libitum groups, but alcohol intake was not significantly different between Inter-C and Inter-C+S. These data suggest that intermittent food availability/deprivation, instead of sugar, could have triggered increased alcohol drinking. In fact, data exists showing that irrespective of alcohol concentration, food deprivation significantly increases alcohol drinking compared to the sated state in rats [28]. Furthermore, length and exposure conditions of palatable food can produce fundamentally different behavioral states [29–31]. Therefore, several procedural/experimental variations among these studies could be responsible for these inconstancies and have been reviewed recently [20].

An important difference between the above-mentioned studies and recent studies from our lab demonstrating reduced alcohol drinking in the intermittent HFD access model is that the above-mentioned studies assessed alcohol drinking while the dietary manipulations were released, similar to what was used in (Fig 4). Whereas in our intermittent HFD access paradigm, intermittent diet cycling (Tue and Thr) continues during alcohol testing sessions (Mon, Wed, and Fri) [22,23] following an initial pre-exposure duration. Using the latter strategy, we have recently shown that reduction in alcohol drinking correlated with the extent of palatable diet pre-exposure duration and two-weeks intermittent HFD pre-exposure was sufficient to reduce alcohol drinking significantly. Furthermore, acute intermittent availability of the palatable diet was critical to observe reduced alcohol drinking behavior as this effect gradually disappeared following the suspension of the PD [23].

While intermittent availability of a palatable diet has been repeatedly shown to reduce acquisition and maintenance of alcohol drinking behavior [22,23], whether a similar approach would rescue relapse-like drinking in alcohol-experienced rats remains to be explored. Therefore, following alcohol testing, a subset of these rats, with no significant differences in alcohol drinking, received two-weeks of intermittent HFD/HSD exposure and alcohol testing was re-initiated. Interestingly, alcohol intake in the chow group of rats was significantly escalated compared to their baseline (alcohol-deprivation effect), whereas alcohol drinking remained un-affected in both HFD/HSD group of rats (Fig 5). Alcohol deprivation effect, with face and predictive validity, is considered an excellent laboratory model to study relapse-like alcohol drinking [32]. The present study is first to demonstrate that a feeding paradigm could rescue relapse-like alcohol drinking in rodents. It is also worth noting that alcohol drinking was significantly reduced in both HFD and HSD group of rats compared to chow controls (Fig 5), suggesting that this effect is not macronutrient specific.

While the reduction in the acquisition, maintenance, and relapse-like alcohol drinking by a feeding paradigm is interesting and may have clinical relevance, the underlying mechanisms are unknown. We have recently shown that homeostatic mechanisms are less likely to be engaged in this process as the peripheral feeding peptides and hypothalamic neurotransmitters gene expression was not impacted under these conditions. In contrast, significant alterations in several neurotransmitters gene expression were detected in the brain reward circuity in rats with reduced alcohol drinking following this intermittent palatable food paradigm [23]. Based on this, we predicted that palatability but not the caloric overload as a contributing factor in reduced alcohol drinking following our paradigm. To test this hypothesis, a group of rats received non-caloric palatable saccharin (SAC) solution in a similar intermittent access paradigm, and alcohol drinking was evaluated following two-weeks. Consistent with the hypothesis, intermittent exposure to SAC significantly reduced alcohol drinking, similar to a high-calorie palatable diet (HSD). It is important to note that caloric intake in the SAC and Chow group of rats was identical, whereas only rats in the HSD group displayed caloric overconsumption and underconsumption pattern. While the palatability of the HSD and SAC may or may not be the same, these data at least suggest that caloric overload in our intermittent access paradigm may not be a contributing factor in reducing alcohol drinking. These data also highlight that both intermittent feeding and drinking of a palatable substance could effectively reduce alcohol drinking. While intermittent access to a palatable diet could have an exposure-dependent reduction in alcohol drinking [23], whether the same is true in the case of SAC drinking is not clear and further studies are needed to explore this possibility. Future studies are also needed to explore possible sex differences, if any, in the effects of intermittent palatable diets access on alcohol drinking.

Malnutrition, a consequence of decreased intake and absorption of essential nutrients from food, [33–35], is frequently reported in alcoholics along with impaired physiological and emotional states [36–40]. An impaired nutritional status could contribute to escalated alcohol consumption and behavioral impairments commonly observed in alcohol dependence [2,33–35,41]. Interestingly, alcoholics display increased intake of highly palatable food during recovery, which could be an attempt to restore caloric deficits or alleviate the negative consequences of alcohol withdrawal or both [10,11]. Notably, increased intake of palatable food has been shown to promote greater abstinence in alcohol-dependent outpatients undergoing detoxification [10,11]. Results from the present study may help to understand this as it is possible that palatable food intake may have protective effects against excessive alcohol consumption and could help sustain sobriety. Since alcohol and drugs of abuse interact with the similar brain reward circuitry as of palatable food and common neurobiological mechanisms are involved in food and drug reward, it could be that a palatable food may compete with the propensity to drink alcohol and thereby reducing alcohol drinking. In this context, it is clear that contingency management (providing a tangible reward for drug abstinence) is an effective method for the management of substance use and alcohol use disorders [42]. A very recent study reported that an alternative palatable food reward could maintain prolonged abstinence from methamphetamine [43]. Therefore, it is possible that improving nutritional status during abstinence would not only compensate for general malnutrition in alcoholics but also could serve to ameliorate some of the adverse symptoms observed in alcohol withdrawal, thereby enhancing prospects of recovery, a contention needs further evaluation.

In this regard, a clinically appropriate dietary recommendation must account for the fact that there are known health concerns associated with prolonged consumption of high calorie/palatable foods. In this context, studies exist demonstrating the effectiveness of two-weeks pre-exposure and even 2 hr of palatable diets every third day in reducing alcohol drinking without impacting body weight or composition [21,23]. These data at least raise the possibility of testing such strategies in early detoxification stages with a gradual reduction in intake to minimize PDs-related negative health consequences. As we are beginning to understand the link between palatable diets exposure and alcohol intake, these data provide critical insight into this relationship and a step forward in this direction.

In conclusion, the present study demonstrates that intermittent access to a palatable diet, irrespective of its macronutrient composition, can reduce acquisition and relapse-like alcohol drinking in rats, possibly as a result of hyper palatability. Since impaired nutritional status is frequently demonstrated in alcoholic patients, which may contribute to the never-ending cycle of alcoholism, these data have important clinical implications in the management of AUD. A palatable diet could not only rectify nutritional deficiencies but also could be protective against excessive alcohol drinking and warrants further investigation.

HIGHLIGHTS.

• Neither high-fat nor -sugar diet (PDs) had long-lasting impact on alcohol drinking.

• Acute intermittent (A-Int) PDs access reduced relapse-like alcohol drinking.

• Palatability may trigger reduced alcohol intake in A-Int PDs access paradigm.