Objective: Prior research has shown that fasting alternated with a diet of standard rodent chow and a 10% sucrose solution produces bingeing on the sucrose, but animals remain at normal body weight. The present study investigated whether restricted access to a highly palatable combination of sugar and fat, without food deprivation, would instigate binge eating and also increase body weight.
Methods and Procedures: Male rats were maintained for 25 days on one of four diets: (i) sweet-fat chow for 2 h/day followed by ad libitum standard chow, (ii) 2-h sweet-fat chow only 3 days/week and access to standard chow the rest of the time, (iii) ad libitum sweet-fat chow, or (iv) ad libitum standard chow.
Results: Both groups with 2-h access to the sweet-fat chow exhibited bingeing behavior, as defined by excessively large meals. The body weight of these animals increased due to large meals and then decreased between binges as a result of self-restricted intake of standard chow following binges. However, despite these fluctuations in body weight, the group with 2-h access to sweet-fat chow every day gained significantly more weight than the control group with standard chow available ad libitum.
Discussion: These findings may have implications for the body weight fluctuations associated with binge-eating disorder, as well as the relationship between binge eating and the obesity epidemic.
Binge eating affects almost 6% of the general population, making it the most common of all eating disorders (1,2). Binge eating is defined as repeated, discrete, and intermittent bouts of consuming unusually large amounts of food (3). Patients with bulimia nervosa or binge eating disorder have binges that are calorically dominated by sugar and fats and often consist of foods such as ice cream, cake, and potato chips (4,5). Many animal models of binge eating use limited access to palatable foods such as sugar solutions (6) or fat (7,8). The present study investigated limited access to a sugar-fat combination (sweet-fat chow) as a clinically relevant model of binge eating in humans. Previous studies using animal models have found that bingeing on fat or sugar does not result in body weight dysregulation (6,7,8). Since binge eating can be a behavioral component of obesity, the goal of the present study was to use a combination of sweet and fat to test whether binge eating would cause weight fluctuations or weight gain, or both.
Methods and Procedures
Male Sprague-Dawley rats (N = 40; Taconic Farms, Germantown, NY), weighing on average 315–325 g (9–10 weeks in age) at the onset of the experiment, were housed individually on a 12-h reversed light/dark cycle. All rats had access to water ad libitum.
Rats were divided into four groups (n =10/group) matched for body weight. The 2-h daily sweet-fat group received access to sweet-fat chow that was nutritionally complete (4.7 kcal/g) and available from 6 to 8 h after the onset of the dark cycle (Research Diets, New Brunswick, NJ, #12451; 45% fat, 20% protein, and 35% carbohydrate, which was 17% sucrose, simple carbohydrate, and 17% maltodextrin-10 and corn starch, complex carbohydrate). For the other 22 h of the day, these rats were given ad libitum access to standard lab chow (LabDiet #5001, PMI Nutrition International, Richmond, IN; 10% fat, 20% protein, 70% carbohydrate, 3.01 kcal/g). The 2-h MWF (Monday, Wednesday, Friday) sweet-fat group was maintained on the same schedule, but only given access to the sweet-fat chow on Mondays, Wednesdays, and Fridays, with standard chow available in the interim. This experimental design is a modification of previously published models (7,8,9). For both of these limited-access groups, the present study made the sweet-fat chow available 6 h into the dark phase. Food was offered at this time to ensure that the animals would not be hungry, because they had presumably been feeding on standard chow since the onset of their normal feeding cycle that ensues at the onset of the dark.
The ad libitum sweet-fat group had access to only the sweet-fat chow, and the ad libitum chow group had only standard chow available. All rats were maintained on their respective diets for 25 days. Food intake was measured daily at 6 and 8 h after the onset of the dark cycle (i.e., before and after the 2-h access to the sweet-fat chow) for all groups. Body weights were also measured during these times on days 1–7 (week 1) and days 18–24 (week 3).
Statistical analyses were carried out for both prebinge body weights (measured 6 h after the onset of the dark cycle) and postbinge body weights (measured 8 h after the start of the dark cycle). Chow intake in grams was converted into kcal and divided into bins of 5-day averages. Body weight and food intake comparisons were made using a two-way ANOVA (time X group) and post hoc Tukey or least significant difference tests when justified. When comparisons were made for all four groups, binge-day data were used unless otherwise specified. For the 2-h MWF sweet-fat group, analyses were carried out comparing binge-day data with data from corresponding days from other groups. Similarly, non-binge-day data for the 2-h MWF sweet-fat group was compared with Tuesday, Thursday, Saturday, and Sunday data from the other groups. Body weight analysis was conducted with all data points for each week included. ANOVA results represent main effect of group unless otherwise specified.
Rats with limited access to sweet-fat chow binge eat
Compared with the control groups, rats in the 2-h daily sweet-fat group consumed excessively large amounts of the palatable chow in the 2 h of access (12.4 ± 1 g, 57% of their daily calories) indicating bingeing behavior (kcal: effect of group: F (3,36) = 98, P < 0.001; post hoc compared with ad libitum chow and ad libitum sweet fat, P < 0.001; Figure 1). The rats with 2-h access to sweet-fat chow on Mondays, Wednesdays, and Fridays also binged (P < 0.001 compared with ad libitum chow and ad libitum sweet-fat groups). They consumed 58% of their daily calories (11.7 ± 0.7 g) in the 2-h access period. There was no significant difference between the binge size for rats given 2-h sweet-fat daily vs. MWF. For comparison, the ad libitum chow and ad libitum sweet-fat chow rats consumed 3.7 ± 1 g and 2.7 ± 1 g of food, respectively, during this 2-h time.
Rats with limited access to sweet-fat chow self-restrict their standard chow intake
During the 22 h that the sweet-fat chow was not available, the 2-h daily sweet-fat group decreased their standard chow intake to levels significantly below those of all other groups (F (3,36) = 81.9, P < 0.0001; compared with ad libitum control groups, P < 0.001, and 2-h MWF sweet-fat group, P < 0.03; Figure 2). This suggests that they self-restricted chow intake between binges. Rats in the MWF group also adjusted their 22-h intakes. On binge days (Mondays, Wednesdays, and Fridays), these rats significantly reduced their calorie intake between binges (F (3,36) = 80.5, P < 0.0001; compared with the ad libitum control groups = P < 0.001). They restricted their 22-h intake more on binge days than on nonbinge days (F (1,9) = 92.2, P < 0.001).
Rats with 2-h daily sweet-fat chow fail to regulate their total daily caloric intake
Despite the self-restriction of standard chow intake, the total daily caloric intake was dysregulated in the 2-h daily sweet-fat group (F (3,36) = 6.1, P < 0.01; Figure 3). Post hoc tests revealed that this group consumed more calories in 24 h than the ad libitum chow group (P < 0.05). The daily caloric intake of the MWF sweet-fat group was also dysregulated. On binge days, they ate significantly more than the ad libitum chow group (P < 0.01). Conversely, on nonbinge days their 24-h caloric intake was significantly lower than the other groups (F (3,36) = 6.9, P < 0.01; compared with 2-h daily sweet fat, P < 0.005 and ad libitum sweet fat, P < 0.01). Within group, the 2-h MWF sweet-fat group consumed significantly more calories on the days they binged compared with the days they did not binge (F (1,9) = 238.23, P < 0.0001).
Rats with 2-h daily sweet-fat chow have body weight fluctuations between binges but their body weights increase overall
By week 3, there was a significant difference between groups in terms of body weight both before the 2-h binge (F (18,216) = 7.5, P < 0.001) and afterward (F (18,216) = 135.3, P < 0.0001). Prior to bingeing each day, the 2-h daily sweet-fat group weighed significantly more than the ad libitum standard chow group (P < 0.05; Figure 4a). A similar effect was observed at the postbinge measurement time, again with the 2-h daily sweet-fat rats weighing significantly more than the ad libitum chow rats (P < 0.05; Figure 4a). The 2-h MWF sweet-fat group did not weigh significantly more than any of the groups either pre- or postbinge, nor were there any other significant group differences in pre- or postbinge body weights.
As shown in Figure 4a, and in greater detail in Figure 4b, c, despite the overall increase in body weight for the 2-h daily sweet-fat group, these rats had fluctuations in body weight between binges (week 1: F (3,39) = 4.76, P < 0.01; week 3: F (3,39) = 77.6, P < 0.001). During week 1, the body weights of the ad libitum sweet-fat group increased, whereas there was no change in body weight for the 2-h daily sweet-fat group (P < 0.01). This effect was also evident and more pronounced during week 3, when the 2-h daily sweet-fat rats lost significantly more weight between binges than both of the ad libitum control groups (P < 0.001). While the body weights of the rats with 2-h daily access decreased between binges, the body weights of the control rats increased. The body weights of the 2-h MWF sweet-fat group also decreased between binges (P < 0.001 compared with the two ad libitum-fed control groups, week 3). Thus, it was only the daily binge eating that led to a significant rise in body weight.
The present results suggest an animal model of binge eating in which nonfood-deprived rats binge on a palatable sugar- and fat-rich food when it is available intermittently. These animals self-restrict their standard chow intake, which results in a between-binge decrease in body weight followed by an increase in body weight after binge eating. Overall, the body weight of these animals increases if they engage in this behavior every day. These findings are supported collectively by other animal models that have been used to describe similar phenomenon using different diets, palatable-food access periods, and periods of food deprivation. The present results are novel in showing that binge eating of a palatable food can increase body weight significantly, while ad libitum access to the same food was less effective.
Defining binge eating
Binge eating is difficult to define in clinical populations perhaps because there is a fine line between what one might consider simple overeating and abnormal binge eating. We used a modification of the definition provided by the DSM-IV-TR (10) and operationally defined binge eating as consuming more of a given food in a discrete (2 h) period of time than would normally be consumed under similar circumstances. Our findings suggest that both the 2-h daily sweet-fat and MWF sweet-fat groups exemplify binge eating. The control group given ad libitum access to the sweet, high-fat diet was included to allow for the differentiation of bingeing from palatability-induced intake. These control animals had access to the same palatable food, but because they were provided access to it continuously, they did not consume it in a binge-like manner. On the other hand, the rats with daily 2-h access to the palatable diet, as well as the MWF-access group, exhibit what we consider to be binge intake. The limited-access schedule is not meant to fully model all aspects of disorders involving binge eating but does model the intermittent and excessive intake that defines these disorders.
Animal models of bingeing
Several other animal models of binge eating have been proposed (9). Many of these models used rebound hyperphagia paradigms that generally consist of experimenter-imposed limited food access (11,12,13). Like the results of the present study, previous studies have shown that bingeing on palatable food can be induced without food deprivation (7,8,9,14). This lack of food deprivation increases the clinical validity of animal models of binge eating associated with eating disorders such as bulimia, because bulimics typically continue to eat between binges but restrict themselves to healthier foods during this time (4).
Corwin et al. (8) have reported that rats maintained on a diet with highly limited access to vegetable fat (e.g., 2-h MWF access) consumed 51% of their daily calories in a 2-h binge period, but rats with access to the vegetable fat for 2 h every day only consumed 32% of their daily caloric intake during that time. In this case, highly restricted access is superior to daily access in eliciting binge-eating behavior. The present study found that both the 2-h daily sweet-fat and the 2-h MWF sweet-fat rats consumed by week 3 a similar number of calories in the 2-h access period (58 and 57% of daily intake, respectively). This suggests that when the palatable diet is comprised of sugar and fat, both daily access and MWF access can increase bingeing behavior equivalently.
Self-restriction of intake prior to bingeing
Rats self-restricted their intake of standard chow between periods of access to sweet-fat chow (Figure 2). Rats, like humans, have evolved “opportunistic” feeding abilities, or the capacity to change feeding and activity patterns based on the availability and type of food in the environment (15). One may interpret this finding as mere compensation for the calories obtained from the palatable chow during the previous binge. However, it is also a possibility that the 2-h daily access group was hyperphagic on the sweet-fat pellets because they learned to wait for the sweet-fat chow as suggested by their restriction of standard chow intake prior to sweet-fat access. Regardless of the cause, the pattern of food intake that emerges is similar to the bingeing and self-restriction observed in patients who binge eat. Anticipatory negative contrast to the standard chow diet that precedes the availability of the palatable food is also a possibility, as suggested by the work of Cottone et al. (16).
Daily bingeing on sweet-fat chow increases total caloric intake and body weight
Two-hour daily sweet-fat rats did not adjust their intakes enough to keep their overall body weights at the same levels as control animals. The self-imposed chow restriction of the 2-h daily sweet-fat and 2-h MWF sweet-fat rats caused their weight to decrease between binges, whereas rats maintained on ad libitum standard lab diet or ad libitum sweet-fat chow consistently gained weight in these 22 h. Thus a “saw-tooth” body weight pattern is evident in the 2-h daily sweet-fat rats (Figure 4b) but not in the ad libitum-fed control groups. The 2-h daily sweet-fat group gained significantly more weight than chow controls.
In terms of overall body weight, some studies have found that bingeing on fat or sugar does not result in weight dysregulation (6,7,8,13), whereas others have shown an increase in body weight (16,17). In the present study, the body weights of rats in the ad libitum sweet-fat group appeared to increase more than those of ad libitum chow rats but this difference was not statistically significant. Weight dysregulation is significantly more pronounced and accelerated when access is daily and limited such that a binge-fast cycle is created. Daily bingeing rats weighed significantly more than the chow-controls both prior to and after the binge. This may be a consequence of their increased cumulative caloric intake as suggested by Figure 3. The observed differences in body weight might be even more pronounced if one were to carry out the experiment for a longer period of time.
It is possible that the daily changes in body weight may be due to the amount of food being consumed in a short time and subsequently digested. Even if the daily fluctuations in body weight are largely a result of the food ingested and digested and partially excreted, these meals are still having residual effects on body weight as shown by the increase in overall body weight (Figure 4a), which is not seen in the other groups. Thus, digestion and excretion fail to compensate for binge eating as shown by gradually increasing body weight.
Clinical implications: an animal model of binge eating with restriction
The animal model presented in this study shares many behavioral aspects of some eating disorders. Patients with binge eating disorder or bulimia often restrict their food intake when they are not bingeing, deliberately fighting physiological signals and urges to eat (18), but ultimately succumbing to binge eating (19). Such restrained eaters are much more sensitive to visual and olfactory cues of palatable foods than nonrestrained eaters and report more intense cravings and better liking of palatable foods (18). In the present study, the 2-h daily sweet-fat rats did not restrict enough to maintain a normal body weight, but instead increased their overall body weight, suggesting a model of similar behavior and outcome as in humans.
This study's use of the limited-access model allows for insights into the relationship between dieting and the obesity epidemic. In humans, bingeing often alternates with fasting and some have hypothesized that obesity might involve a predilection for sweet and fatty foods that dysregulates opioid systems and increases the tendency to binge, leading to a cycle of bingeing and dieting (20). Historically, the reward signals associated with palatable foods were adaptive as they indicated calorically dense sugar and fat-rich foods that were high-energy sources good for survival. Now, however, these sugary, high-fat foods are ubiquitous in many societies, and attempts to avoid them only exacerbate bingeing behaviors (21). Thus, the intermittent intake of highly palatable meals in large quantities that is frequent in some cultures is likely to precipitate bingeing and may contribute to increased body weight and possibly obesity in some individuals.
This study presents a new animal model of weight and total caloric intake dysregulation in rats that binge on a sweet-fat diet. Just as previous models using other diets and periods of limited access to sugar- or fat-rich diets have done (6,7,8,16), this model of limited sweet-fat access could facilitate further characterization of binge eating in relation to physiological and neurological changes. Access to sweet-fat chow for 2 h per day produces a more pronounced increase in body weight compared with standard chow available ad libitum. Between binges on a sweet-fat chow, rats self-restrict standard lab chow intake to an extent that their body weights decrease between binges, but if the access schedule to the sweet-fat chow is limited and daily, overall caloric intake and body weight significantly increase despite daily fluctuations. In summary, 2 h of daily access to a sweet-fat chow is a model of voluntary, non-experimenter-restricted binge eating that results in daily body weight fluctuations and overall increase in body weight.
This work was supported by USPHS grants MH-65024 (to B.T. Walsh, B.G.H. et al.), DA-10608 (to B.G.H.) and DA-16458 and DK-79793 (fellowships to N.M.A.).