- Email: [email protected]
Sugar withdrawal and differential reinforcement of low rate (DRL) performance in rats
Physiology & Behavior 139 (2015) 468–473
Contents lists available at ScienceDirect
Physiology & Behavior journal homepage: www.elsevier.com/locate/phb
Sugar withdrawal and differential reinforcement of low rate (DRL) performance in rats Victor Mangabeira a,⁎, Miriam Garcia-Mijares b, M. Teresa A. Silva c,1 a b c
Experimental Psychology Department, Institute of Psychology, University of São Paulo, Av. Prof. Mello Moaraes 1721, São Paulo, SP CEP 05508-900, Brazil Experimental Psychology Department, Institute of Psychology, University of São Paulo, Av. Prof. Mello Moraes 1721, São Paulo, SP CEP 05508-900, Brazil Experimental Psychology Department, Institute of Psychology, University of São Paulo, Av. Prof. Mello Moaraes 1721, São Paulo, SP CEP 05508-900, Brazil
H I G H L I G H T S • • • •
Sugar withdrawal resulted in the impairment of DRL performance. Sugar abstinent rats showed shorter IRTs and higher response rates per reinforcer. The estimated relative treatment effect was greater after abstinence. Observed effects parallel to those of addictive drug abstinence
a r t i c l e
i n f o
Article history: Received 20 February 2014 Received in revised form 3 July 2014 Accepted 29 September 2014 Available online 5 December 2014 Keywords: Sugar addiction DRL model Impulsivity Sugar abstinence
a b s t r a c t Sugar consumption is assumed to induce a behavioral state that is similar to the one provoked by addictive substances. Drug withdrawal increases impulsivity, assessed by differential reinforcement of low rate (DRL) performance. The present study investigated the effect of withdrawal from a prolonged period of sugar consumption on DRL performance. Water-deprived rats were trained under a DRL 20 s (DRL 20) schedule. The animals were allowed to choose between plain water and a sucrose solution (E group) or water only (C group) for 30 days. The sucrose solution was then removed, and measures of DRL 20 performance were obtained on 3 consecutive days. Results showed that DRL performance in the C group improved after sugar withdrawal, whereas performance in the E group led to the loss of reinforcers. An analysis of variance-type analysis showed that the E group had higher response rates per reinforcer, lower IRTs, and greater differences between baseline and abstinence than the C group after 3 days of sugar withdrawal. Thus, sugar abstinence after a relatively long consumption period resulted in impairment of DRL performance, confirming the parallel effects of addictive drugs and sugar and suggesting an increase in impulsivity as a consequence of sugar deprivation. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Extended sugar consumption may produce neurochemical changes in brain loci related to feeding and reinforcement [40]. Although these sites evolved to respond to natural reinforcers, they are also affected by drugs of abuse, particularly those related to the opioidergic and dopaminergic systems [2,17,18,25,36,39]. In rats, dopamine receptor antagonists attenuate the hedonic value of sweet-tasting nutrients [40]. Addictive drugs and sugar lead to increased dopamine and opioid release [6,24,34]. Glucose consumption for 30 days has been shown to alter dopamine and opioid receptor binding and provoke withdrawal
⁎ Corresponding author at: Av. Prof. Mello Moraes 1721, São Paulo 05508-000, Brazil. E-mail addresses: [email protected] (V. Mangabeira), [email protected] (M. Garcia-Mijares), [email protected] (M.T.A. Silva). 1 CNPq grant 301747/2005-9.
http://dx.doi.org/10.1016/j.physbeh.2014.09.017 0031-9384/© 2014 Elsevier Inc. All rights reserved.
symptoms after naloxone injection [12,13]. Sugar withdrawal causes alterations in the mRNA levels of dopamine and opioid receptors, similar to those seen in morphine-dependent rats [38]. Neuroimaging studies support the hypothesis that palatable food and addictive substances share powerful reinforcing properties [41]. The consumption of hyperpalatable substances, such as sugar, has been suggested to induce a behavioral state that is similar to the one induced by addictive substances [23], an effect that is evident even during the perinatal period in rats [8] and is reflected by similar patterns of neural activation in addictive-like eating behavior and drug dependence [21]. The obesity epidemic makes this relationship even more relevant [1,2,4]. Excessive intermittent sugar intake can lead to addiction [5] and induce symptoms of abstinence, demonstrated by behavioral observations, enhanced responding for sugar, ultrasonic vocalizations, anxiety-like behavior in the elevated plus maze, in vivo microdialysis, and increased sucrose-seeking behavior [3,12,19,38]. Withdrawal from highly palatable food can lead to depressive-like behavior that may be
V. Mangabeira et al. / Physiology & Behavior 139 (2015) 468–473
relieved by compulsive eating in rats [26]. Even withdrawal from the non-caloric sugar substitute saccharin has been shown to produce behavioral effects, such as increases in aggressive and sexual behavior in rats [7]. The reinforcement of low rate (DRL) schedule has been used to assess the effects of drugs, such as the psychomotor stimulant D amphetamine, the anxiolytic compound chlordiazepoxide, and the antidepressant desipramine. In this schedule, a minimum interval is required between lever presses, and the difficulty an animal has in keeping up with the delay in reinforcement demand has been considered an index of impulsivity. Impulsivity is a multifactor behavioral trait that is thought to contribute to various psychiatric disorders, including drug abuse and eating disorders [14,16,29,30,32]. Many clinical questionnaires have been used to measure impulsivity in humans [14]. Human studies have related impulsivity to food addiction using the Yale Food Addiction Scale (YFAS), which was developed to assess signs of substance dependence in eating behavior [20]. For example, the impulsivity dimension on an emotion dysregulation scale had the strongest specific relationship with “food addiction” on the YFAS [22]. Although the concept of impulsivity involves multiple subjacent elements, a factor analysis revealed at least two contributing factors that can be related to substance abuse: lack of inhibitory control and heightened reward sensitivity [14]. The DRL procedure models these two factors and has been used to assess impulsive action in animals [14,16,27, 30,32,35]. Withdrawal from amphetamine disrupted DRL performance, with increased responding and a reduced number of reinforcers earned [33]. Sugar-dependent rats exhibited enhanced responding for sugar after abstinence [3]. Considering the neurochemical and behavioral similarities between the effects of addictive drugs and sugar, the objective of the present study was to determine whether sugar abstinence after a relatively long consumption period impairs DRL performance. This would indicate a sugar deprivation effect that is analogous to drug withdrawal, suggesting increased impulsivity.
2. Materials and methods 2.1. Animals Naive male Wistar rats (n = 27), approximately 60 days old (range, 58–63 days) and weighing an average of 180 g (SE = 16.8 g), were used as subjects. They were singly housed under a 12 h/12 h light/dark cycle (dark at 7:00 PM) and controlled temperature (22 ± 1 °C) in semitransparent 25 × 40 × 20 cm plastic home-cages that contained a feeder on top. Depending on the experimental phase, one or two bottles were positioned above the cage lid, with their nozzle protruding inside so it could be easily reached. The regular food ration in the laboratory, Nuvilab food pellets, was freely available. Food pellets were made of whole corn, soybean meal, wheat bran, limestone, dicalcium phosphate, sodium chloride, and vitamin mineral premix. Water was restricted to 2 h daily.
2.2. Apparatus An operant 27.5 × 22.5 × 28.0 cm conditioning chamber was used that was encased in a sound-attenuating box (Med Associates, St. Albans, VT, USA). The front and back walls and ceiling were made of Plexiglas, and the lateral walls and grid rod floor in inox. A ventilation fan provided background noise. A standard lever was located on the right wall. Water was delivered for 2 s by a 0.1 ml rising dipper located at floor level in the center of that wall. After this time of water delivery, the dipper was lowered so it became inaccessible to the subject. A computer was programmed to control stimulus presentation and data recording.
469
2.3. Procedure The experiment consisted of three phases.
2.3.1. Phase 1: Baseline (BL) A bottle that contained water was withdrawn for 22 h after each session. All of the rats were shaped for bar-pressing for water under a continuous reinforcement schedule for 2 days. Afterward, DRL training began. Pressing the lever after the required interval elapsed resulted in water delivery. The initial reinforcement requirement was a DRL 5 s schedule, in which the animal had to wait for 5 s until the next lever press to receive water. The requirement was gradually increased in steps of 5 s until an interval of 20 s between each response (DRL 20) was reached. The subjects' performance thus progressed from DRL 10 s (14 experimental sessions) to DRL 15 s (25 sessions), to DRL 20 s (50 sessions). The sessions were run Monday through Friday, but the water and food regimen was maintained on the weekends. Performance on the DRL 20 was the dependent variable and used as an index of impulsive action. When the mean responses per reinforcement rate were stable between 4 and 5, training was interrupted, and the second phase began. The data from the last three sessions were used as baseline (BL).
2.3.2. Phase 2: Sugar consumption (SUG) This phase sought to generate sugar dependence. The rats were randomly assigned to two groups: Experimental (E; n = 14) and Control (C; n = 13). The animals in both groups received food and water ad libitum in their home cages, but a second bottle that contained either 10% commercial sugar solution in the E group or plain water in the C group was added. The position of the bottle in Phase 1 was maintained, and the second bottle was placed to its left. The duration of Phase 2 was 30 days.
2.3.3. Phase 3: Abstinence (ABS) The conditions of Phase 1 were reinstated. Ad libitum-fed rats were subjected to water restriction, with water available for only 2 h daily. The bottles that were previously added in Phase 2 were withdrawn so only the water bottle remained in the cage. Three DRL 20 sessions were conducted under this condition. Additional sessions were not conducted because the animals had begun to suffer deleterious effects of the long period of water restriction.
2.4. Statistical analysis A nonparametric longitudinal analysis of variance (ANOVA) that is equivalent to an ANOVA (ANOVA-type) was performed [9]. Nonparametric techniques that are based on the concept of the relative effect of treatments were proposed by Brunner et al. [10]. The ANOVAtype statistic yields results analogous to those obtained by repeatedmeasures ANOVA [28] and may be used even when the response distribution is unknown. The mean responses per reinforcement and the in inter-response times (IRTs) in the last three BL sessions and three ABS sessions were calculated. The ANOVA-type analysis was performed to test the effects of independent groups E and C (between factors) and longitudinal factors LB and ABS (within factors) and interactions between these two factors. The estimated relative treatment effect (ERE) for longitudinal measures was also computed. The ERE is a nonparametric analysis that determines the probability of a randomly selected observation from a given group at a given moment (e.g., the E group in ABS) have a greater value than another observation selected from the whole set [9,37]. An Excel macro was used for data computation (www.ime.usp.br/~jmsinger; accessed June 28, 2014). The accepted level of significance was 0.05 for all of the tests.
470
V. Mangabeira et al. / Physiology & Behavior 139 (2015) 468–473
3. Results Fig. 1 presents mean ± SE water intake (ml) in the C and E groups and mean sucrose solution intake in the E group during the 30 SUG sessions (Phase 2). Comparisons between sugar solution consumption by the E group and water consumption by the C group in the last day of phase 2 showed significant differences between groups (t(26) = 9.75, p b 0.001). There were also differences between groups on water consumption (t(26) = 1366, p b 0.001) and between sugar solution and water consumption by the E group (t(13) = 14.81, p b 0.001) in the last day of phase 2. So, sugar consumption in the E group was higher than water intake by both groups and the C group consumed more water than the E group. Daily sugar solution consumption was maintained between 25% and 40% of body weight. The mean increase in body weight was 64 g (15.6%) in the E group during the sugar intake period and 46 g (9.8%) in the C group. Median responses per reinforcer in each of the last three BL sessions and in each ABS session are presented in Fig. 2. Both groups exhibited relatively stable performance at BL. On the last day of ABS, an increase in the median number of responses per reinforcer was observed in the E group, and a decrease in the median number of responses per reinforcer was observed in the C group. An ANOVA-type analysis of the BL sessions and the first, second, and third ABS sessions (within-factors) was performed, with the E and C groups as the between-factors. Fig. 3 presents the ERE probability (see Section 2.4) in the C and E groups, comparing (a) the last three BL sessions with the three ABS sessions, (b) the last three BL sessions with the first ABS session, (c) the last three BL sessions with the second ABS session, and (d) the last three BL sessions with the third ABS session. As shown in Fig. 3a, the overall ERE in C group remained nearly constant from BL to the ABS condition (from 0.48 to 0.49) but increased from 0.43 to 0.61 in the E group. However, those differences were not significant (QA: between = 0.13, df = 1, p = 0.72; within = 3.72, df = 1, p = 0.053; interaction = 2.9, df = 1, p = 0.08). The analysis of the difference between the last three BL sessions and first ABS session (Fig. 3b) showed similar results, in which no significant changes in ERE values were observed between BL and ABS, with no interaction between these factors (QA: between = 0.03, df = 1, p = 0.87; within = 2.53, df = 1, p = 0.11; interaction = 0.49, df = 1, p = 0.48). A significant
difference was found between BL and the second ABS session (QA: between = 0.03, df = 1, p = 0.87; within = 4.25, df = 1, p = 0.04; interaction = 1.93, df = 1, p = 0.17). Differences between groups were statistically evident in the third ABS session (Fig. 3d), in which an interaction between the within and between factors was observed and only the E group responded more for the sucrose solution (QA: between = 0.84, df = 1, p = 0.36; within = 0.00, df = 1, p = 0.98; interaction = 6.36, df = 1, p = 0.01). An increase in the ERE difference was observed in the E group between BL and ABS and between the first and third ABS sessions (from 0.47 to 0.61), but a decrease in this effect was observed in the C group (from 0.53 to 0.38). Thus, the obtained daily ERE values in the E group indicate that this effect gradually increased from the first to third session after withdrawal. Median IRTs in each of the last three BL sessions and in each ABS session are presented in Fig. 4. Both groups showed relatively stable performance at BL on the last two days of BL, but the median of the C group decreased on the last day. On the other hand, IRTs of the E group showed a slight decrease on the three first days of ABS when compared to BL, whereas IRTs of the C group showed a slight increase on the first two days of ABS and a higher increase in the last day. As indicated in Fig. 5, when the overall ERE of IRTs was computed (last three BL sessions compared to three first ABS sessions; Fig. 5a), no significant differences in EREs between groups (QA = 0.03, df = 1, p = 0.850), nor between BL and ABS (QA = 0.72, df = 1, p = 0.395) or interaction (QA = 1.56, df = 1, p = 0.212) were found. Also, there were no differences between these parameters when BL was compared to the first ABS session (QA: between = 0.13, df = 1, p = 0.718; within = 1.75 df = 1, p = 0.186; interaction = 0.51, df = 1, p = 0.476; Fig. 5b) or the second ABS session (QA: between = 0.329, df = 1, p = 0.566; within = 1.916 df = 1, p = 0.166; interaction = 0.42, df = 1, p = 0.516; Fig. 5c). However, in the third session of abstinence, groups displayed different EREs (QA: between = 0.125, df = 1, p = 0.724; within = 0.448, df = 1, p = 0.485; interaction = 5.127, df = 1, p = 0.024; Fig. 5c), indicating shorter IRTs for the E group as shown in Fig. 4. 4. Discussion The present study investigated the possible signs of abstinence in DRL performance after a prolonged period of sugar consumption.
Fig. 1. Mean (±SE) water intake in the C and E groups and mean (±SE) sugar intake in the E group during the 30 SUG sessions (Phase 2). **p b 0.001 for comparisons between E (sugar) and C (water). ##p b 0.001 for comparisons between E (sugar) and C (water). ++p b 0.001 between E (sugar) and E (water).
V. Mangabeira et al. / Physiology & Behavior 139 (2015) 468–473
471
Fig. 2. Median responses per reinforcer in the C and E groups in the three last BL sessions (48, 49, 50) and three ABS sessions (1, 2, 3). The 25th and 75th percentiles are presented for each group in each session. Δ represents outliers of the C distribution, and ○ represents outliers of the E distribution.
Sucrose intake in the animals that had a choice between plain water and a sucrose solution indicated a clear preference for sugar. Abstinence affected DRL performance, leading to higher response rates per reinforcer and shorter IRTs, compared with baseline after 3 days of withdrawal. Individual intrasubject differences and the analysis of the ERE showed a similar deterioration in performance, reflected by an increase in the loss of reinforcers. Only animals that were subjected to sugar abstinence presented impaired DRL performance. The observed impairment of DRL performance suggests an increase in impulsive behavior following sugar deprivation, similar to the effect produced by withdrawal from drugs of abuse, such as amphetamine and phencyclidine [11,33]. After sugar abstinence, the control of responses 0.8
a
0.7
0.7
0.5
0.5
0.4
0.4
0.3
0.3
0.1
0.1
0
0
0.8
c
0.7
Probability
E
0.2
0.2
0.5
0.5
0.4
0.4
0.3
0.3
0.2
0.2
0.1
0.1 ABS
Phase
**
0.7
0.6
BL
C
0.8
0.6
0
b
0.6
0.6
Probability
by long-term reinforcement, a sensitive indicator of impulsivity, was attenuated in the E group. This result may be relevant for studies of obesity and several behaviors that are affected by deficiencies in self-control. The DRL procedure was used because it is considered a typical animal model of impulsivity [30]. However, in the present study the DRL procedure was confounded by the time needed to reach stable performance. The animals had to be subjected to an extended period of water restriction to attain the stipulated criterion, which should perhaps be revised. Water restriction imposed physiological alterations, but this constraint was unavoidable because the metabolic effects of food restriction would have outweighed sensorial factors. New procedures for DRL training might be implemented, and other procedures,
0
BL
Phase
d
ABS
Fig. 3. Estimated relative effect (ERE) of responses per reinforcer and its 95% confidence intervals in the C and E groups in comparisons between (a) the last three BL sessions and three ABS sessions, (b) last three BL sessions and first ABS session, (c) last three BL sessions and second ABS session, and (d) last three BL sessions and third ABS session. **p b 0.01 for differences between groups.
472
V. Mangabeira et al. / Physiology & Behavior 139 (2015) 468–473
Fig. 4. Median inter-response time (seconds) in the C and E groups in the three last BL sessions (48, 49, 50) and three ABS sessions (1, 2, 3). 25th and 75th percentiles are presented for each group in every session. Δ represents outliers of C distribution and ○ outliers of E distribution.
such as the paced fixed consecutive number test [15], could be used as alternative models. The DRL schedule confirmed the similar effects of addictive drugs and sucrose. The present data corroborate the behavioral changes observed after excessive sugar consumption, such as signs of withdrawal [12,26,38], and similarities revealed by brain imaging techniques in craving in response to palatable food and drugs of abuse [6,41]. Given
the typical effect of antidepressants on DRL performance in reducing response rates and increasing reinforcement rates [31], a pertinent issue would be to investigate whether these drugs would reverse the effect of sugar abstinence reported herein, suggesting relief from depression [26]. Further research could also address the relationship between behavioral and neural changes elicited by sugar withdrawal in the DRL procedure.
Fig. 5. Estimated relative effect (ERE) of IRTs and its 95% confidence intervals in the C and E groups in comparisons of (a) the last three BL sessions and three ABS sessions, (b) last three BL sessions and first ABS session, (c) last three BL sessions and second ABS session, and (d) last three BL sessions and third ABS session. *p b .005 for differences between groups.
V. Mangabeira et al. / Physiology & Behavior 139 (2015) 468–473
5. Conclusion The animal procedure utilized in the present study contributes to our understanding of sugar addiction, demonstrating similarities with drug addiction in a model of impulsivity. This conclusion supports the hypothesis that the excessive consumption of highly palatable food and drugs of abuse can generate similar behavioral consequences. Such consequences are likely paralleled by brain changes caused by hedonic eating and various types of addiction. As stated by Hoebel in 2009 (http://www.foodaddictionsummit.org/presenters-hoebel.htm; accessed June 15, 2014), “The brain is getting addicted to its own opioids as it would to morphine or heroin. Drugs give a bigger effect, but it is essentially the same process”. References [1] N.M. Avena, M.E. Bocarsly, B.G. Hoebel, M.S. Gold, Overlaps in the nosology of substance abuse and overeating: the translational implications of “food addiction.”, Curr. Drug Abus. Rev. 4 (2011) 133–139, http://dx.doi.org/10.2174/1874473711104030133. [2] N.M. Avena, J.A. Gold, C. Kroll, M.S. Gold, Further developments in the neurobiology of food and addiction: update on the state of the science, Nutrition 28 (2012) 341–343, http://dx.doi.org/10.1016/j.nut.2011.11.002. [3] N.M. Avena, K.A. Long, B.G. Hoebel, Sugar-dependent rats show enhanced responding for sugar after abstinence: evidence of a sugar deprivation effect, Physiol. Behav. 84 (2005) 359–362, http://dx.doi.org/10.1016/j.physbeh.2004.12.016. [4] N.M. Avena, S. Murray, M.S. Gold, Comparing the effects of food restriction and overeating on brain reward systems, Exp. Gerontol. 48 (2013) 1062–1067, http://dx.doi. org/10.1016/j.exger.2013.03.006. [5] N.M. Avena, P. Rada, B.G. Hoebel, Evidence of sugar addiction: behavioral and neurochemical effects of intermittent, excessive sugar intake, Neurosci. Biobehav. Rev. 32 (2008) 20–39, http://dx.doi.org/10.1016/j.neubiorev.2007.04.019. [6] N.M. Avena, P. Rada, B.G. Hoebel, Sugar and fat bingeing have notable differences in addictive-like behavior, J Nutr. 139 (2009) 623–628, http://dx.doi.org/10.3945/jn. 108.097584. [7] I.V. Belozertseva, I.A. Sukhotina, J.M. Vossen, A.Y. Bespalov, Facilitation of aggressive and sexual behaviors by saccharin deprivation in rats, Physiol. Behav. 80 (2004) 531–539, http://dx.doi.org/10.1016/j.physbeh.2003.10.012. [8] M.E. Bocarsly, J.R. Barson, J.M. Hauca, B.G. Hoebel, S.F. Leibowitz, N.M. Avena, Effects of perinatal exposure to palatable diets on body weight and sensitivity to drugs of abuse in rats, Physiol. Behav. 107 (2012) 568–575, http://dx.doi.org/10.1016/j. physbeh.2012.04.024. [9] E. Brunner, F. Langer, Nonparametric analysis of ordered categorical data in designs with longitudinal observations and small sample sizes, Biom. J. 42 (2000) 663–675, http://dx.doi.org/10.1002/1521-4036(200010)42:6b663::AID-BIMJ663N3.0.CO;2-7. [10] E. Brunner, U. Munzel, M.L. Puri, Rank-score tests in factorial designs with repeated measures, J. Multivar. Anal. 70 (1999) 286–317, http://dx.doi.org/10.1006/jmva. 1999.1821. [11] M.E. Carroll, J.L. Mach, R.M. La Nasa, J.L. Newman, Impulsivity as a behavioral measure of withdrawal of orally delivered PCP and nondrug rewards in male and female monkeys, Psychopharmacology 207 (2009) 85–98, http://dx.doi.org/10.1007/ s00213-009-1636-y. [12] C. Colantuoni, P. Rada, J. McCarthy, C. Patten, N.M. Avena, A. Chadeayne, B.G. Hoebel, Evidence that intermittent, excessive sugar intake causes endogenous opioid dependence, Obes. Res. 10 (2002) 478–488, http://dx.doi.org/10.1038/oby.2002.66. [13] C. Colantuoni, J. Schwenker, J. McCarthy, P. Rada, B. Ladenheim, J.L. Cadet, G.J. Schwartz, T.H. Moran, B.G. Hoebel, Excessive sugar intake alters binding to dopamine and mu-opioid receptors in the brain, Neuroreport 12 (2001) 3549–3552, http://dx.doi.org/10.1097/00001756-200111160-00035. [14] S. Dawe, N.J. Loxton, The role of impulsivity in the development of substance use and eating disorders, Neurosci. Biobehav. Rev. 28 (2004) 343–351, http://dx.doi. org/10.1016/j.neubiorev.2004.03.007. [15] J.L. Evenden, The pharmacology of impulsive behavior in rats: II. The effects of amphetamine, haloperidol, imipramine, chlordiazepoxide and other drugs on fixed consecutive number schedules (FCN 8 and FCN 32), Psychopharmacology 138 (1998) 283–294, http://dx.doi.org/10.1007/s002130050673. [16] J.L. Evenden, Varieties of impulsivity, Psychopharmacology 146 (1999) 348–361, http://dx.doi.org/10.1007/PL00005481. [17] J.D. Fernstrom, S.D. Munger, A. Sclafani, I.E. de Araujo, A. Roberts, S. Molinary, Mechanisms for sweetness, J. Nutr. 142 (2012) 1134S–1141S, http://dx.doi.org/10. 3945/jn.111.149567.
473
[18] B.M. Geiger, M. Haburcak, N.M. Avena, M.C. Moyer, B.G. Hoebel, E.N. Pothos, Deficits of mesolimbic dopamine neurotransmission in rat dietary obesity, Neuroscience 159 (2009) 1193–1199, http://dx.doi.org/10.1016/j.neuroscience.2009.02.007. [19] J.W. Grimm, A.M. Fyall, D.P. Osincup, Incubation of sucrose craving: effects of reduced training and sucrose pre-loading, Physiol. Behav. 84 (2005) 73–79, http:// dx.doi.org/10.1016/j.physbeh.2004.10.011. [20] A.N. Gearhardt, W.R. Corbin, K.D. Brownell, The preliminary validation of the Yale Food Addiction Scale, Appetite 52 (2009) 430–436, http://dx.doi.org/10.1016/j. appet.2008.12.003. [21] A.N. Gearhardt, C.M. Grilo, R.J. DiLeone, K.D. Brownell, M.N. Potenza, Can food be addictive? Public health and policy implications, Addiction 106 (2011) 1208–1212, http://dx.doi.org/10.1111/j.1360-0443.2010.03301.x. [22] A.N. Gearhardt, M.A. White, R.M. Masheb, P.T. Morgan, R.D. Crosby, C.M. Grilo, An examination of the food addiction construct in obese patients with binge eating disorder, Int. J. Eat. Disord. 45 (2012) 657–663, http://dx.doi.org/10.1002/eat.20957. [23] A.N. Gearhardt, S. Yokum, P.T. Orr, E. Stice, W.R. Corbin, K.D. Brownell, Neural correlates of food addiction, Arch. Gen. Psychiatr. 68 (2011) 808–816, http://dx.doi.org/ 10.1001/archgenpsychiatry.2011.32. [24] A. Hajnal, R. Norgren, Accumbens dopamine mechanisms in sucrose intake, Brain Res. 904 (2001) 76–84, http://dx.doi.org/10.1016/S0006-8993(01)02451-9. [25] B.G. Hoebel, N.M. Avena, M.E. Bocarsly, P. Rada, Natural addiction: a behavioral and circuit model based on sugar addiction in rats, J. Addict. Med. 3 (2009) 33–41, http://dx.doi.org/10.1097/ADM.0b013e31819aa621. [26] A. Iemolo, M. Valenza, L. Tozier, C.M. Knapp, C. Kornetsky, L. Steardo, V. Sabino, P. Cottone, Withdrawal from chronic, intermittent access to a highly palatable food induces depressive-like behavior in compulsive eating rats, Behav. Pharmacol. 23 (2012) 593–602, http://dx.doi.org/10.1097/FBP.0b013e328357697f. [27] P. Jayarajan, R. Nirogi, A. Shinde, Effect of olanzapine on scopolamine induced deficits in differential reinforcement of low rate 72 s (DRL-72 s) schedule in rats: involvement of the serotonergic receptors in restoring the deficits, Eur. J. Pharmacol. 720 (2013) 344–354, http://dx.doi.org/10.1016/j.ejphar.2013.10.005. [28] M.H. Kutner, C. Nachtsheim, J. Neter, W. Li, Applied Linear Statistical Models, 5th ed. McGraw-Hill, Boston, 2004. [29] B.A. McMillen, L.W. Means, J. Matthews, Comparison of the alcohol-preferring P rat to the Wistar rat in behavioral tests of impulsivity and anxiety, Physiol. Behav. 63 (1998) 371–375 (PII S0031–9384(97)00442–3). [30] J. Monterosso, G. Ainslie, Beyond discounting: possible experimental models of impulse control, Psychopharmacology 146 (1999) 339–347, http://dx.doi.org/10.1007/ PL00005480. [31] J.M. O'Donnell, G.J. Marek, L.S. Seiden, Antidepressant effects assessed using behavior maintained under a differential-reinforcement-of-low-rate (DRL) operant schedule, Neurosci. Biobehav. Rev. 29 (2005) 785–798, http://dx.doi.org/10.1016/j.neubiorev. 2005.03.018. [32] J.L. Perry, M.E. Carroll, The role of impulsive behavior in drug abuse, Psychopharmacology 200 (2008) 1–26, http://dx.doi.org/10.1007/s00213-008-1173-0. [33] J.D. Peterson, M.E. Wolf, F.J. White, Impaired DRL 30 performance during amphetamine withdrawal, Behav. Brain Res. 143 (2003) 101–108, http://dx.doi.org/10. 1016/S0166-4328(03)00035-4. [34] J.D. Pomonis, D.C. Jewett, C.M. Kotz, J.E. Briggs, C.J. Billington, A.S. Levine, Sucrose consumption increases naloxone-induced c-Fos immunoreactivity in limbic forebrain, Am. J. Physiol. Regul. Integr. Comp. Physiol. 278 (2000) R712–R719 (PubMed 10712293). [35] J.B. Richards, K.E. Sabol, L.S. Seiden, DRL interresponse-time distributions: quantification by peak deviation analysis, J. Exp. Anal. Behav. 60 (1993) 361–385, http:// dx.doi.org/10.1901/jeab. 1993.60-361. [36] M.T.A. Silva, F.L. Gonçalves, M. Garcia-Mijares, Neural events in the reinforcement contingency, Behav. Anal. 30 (2007) 17–30 ([PubMed] 22478485). [37] J.M. Singer, Z. Poleto, P. Rosa, Parametric and nonparametric analyses of repeated ordinal categorical data, Biom. J. 46 (2004) 460–473, http://dx.doi.org/10.1002/bimj. 200310045. [38] R. Spangler, K.M. Wittkowski, N.L. Goddard, N.M. Avena, B.G. Hoebel, S.F. Leibowitz, Opiate-like effects of sugar on gene expression in reward areas of the rat brain, Mol. Brain Res. 124 (2004) 134–142, http://dx.doi.org/10.1016/j.molbrainres.2004.02.013. [39] N.D. Volkow, G.J. Wang, D. Tomasi, R.D. Baler, Obesity and addiction: neurobiological overlaps, Obes. Rev. 14 (2013) 2–18, http://dx.doi.org/10.1111/j.1467-789X.2012. 01031.x. [40] R.A. Wise, Role of brain dopamine in food reward and reinforcement, Philos. Trans. R. Soc. Lond. B Biol. Sci. 361 (2006) 1149–1158, http://dx.doi.org/10.1098/rstb.2006.1854. [41] Y. Zhang, K.M. von Deneen, J. Tian, M.S. Gold, Y. Liu, Food addiction and neuroimaging, Curr. Pharm. Des. 17 (2011) 1149–1157 (BSP/CPD/E-Pub/000438 [pii]).
Contents lists available at ScienceDirect
Physiology & Behavior journal homepage: www.elsevier.com/locate/phb
Sugar withdrawal and differential reinforcement of low rate (DRL) performance in rats Victor Mangabeira a,⁎, Miriam Garcia-Mijares b, M. Teresa A. Silva c,1 a b c
Experimental Psychology Department, Institute of Psychology, University of São Paulo, Av. Prof. Mello Moaraes 1721, São Paulo, SP CEP 05508-900, Brazil Experimental Psychology Department, Institute of Psychology, University of São Paulo, Av. Prof. Mello Moraes 1721, São Paulo, SP CEP 05508-900, Brazil Experimental Psychology Department, Institute of Psychology, University of São Paulo, Av. Prof. Mello Moaraes 1721, São Paulo, SP CEP 05508-900, Brazil
H I G H L I G H T S • • • •
Sugar withdrawal resulted in the impairment of DRL performance. Sugar abstinent rats showed shorter IRTs and higher response rates per reinforcer. The estimated relative treatment effect was greater after abstinence. Observed effects parallel to those of addictive drug abstinence
a r t i c l e
i n f o
Article history: Received 20 February 2014 Received in revised form 3 July 2014 Accepted 29 September 2014 Available online 5 December 2014 Keywords: Sugar addiction DRL model Impulsivity Sugar abstinence
a b s t r a c t Sugar consumption is assumed to induce a behavioral state that is similar to the one provoked by addictive substances. Drug withdrawal increases impulsivity, assessed by differential reinforcement of low rate (DRL) performance. The present study investigated the effect of withdrawal from a prolonged period of sugar consumption on DRL performance. Water-deprived rats were trained under a DRL 20 s (DRL 20) schedule. The animals were allowed to choose between plain water and a sucrose solution (E group) or water only (C group) for 30 days. The sucrose solution was then removed, and measures of DRL 20 performance were obtained on 3 consecutive days. Results showed that DRL performance in the C group improved after sugar withdrawal, whereas performance in the E group led to the loss of reinforcers. An analysis of variance-type analysis showed that the E group had higher response rates per reinforcer, lower IRTs, and greater differences between baseline and abstinence than the C group after 3 days of sugar withdrawal. Thus, sugar abstinence after a relatively long consumption period resulted in impairment of DRL performance, confirming the parallel effects of addictive drugs and sugar and suggesting an increase in impulsivity as a consequence of sugar deprivation. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Extended sugar consumption may produce neurochemical changes in brain loci related to feeding and reinforcement [40]. Although these sites evolved to respond to natural reinforcers, they are also affected by drugs of abuse, particularly those related to the opioidergic and dopaminergic systems [2,17,18,25,36,39]. In rats, dopamine receptor antagonists attenuate the hedonic value of sweet-tasting nutrients [40]. Addictive drugs and sugar lead to increased dopamine and opioid release [6,24,34]. Glucose consumption for 30 days has been shown to alter dopamine and opioid receptor binding and provoke withdrawal
⁎ Corresponding author at: Av. Prof. Mello Moraes 1721, São Paulo 05508-000, Brazil. E-mail addresses: [email protected] (V. Mangabeira), [email protected] (M. Garcia-Mijares), [email protected] (M.T.A. Silva). 1 CNPq grant 301747/2005-9.
http://dx.doi.org/10.1016/j.physbeh.2014.09.017 0031-9384/© 2014 Elsevier Inc. All rights reserved.
symptoms after naloxone injection [12,13]. Sugar withdrawal causes alterations in the mRNA levels of dopamine and opioid receptors, similar to those seen in morphine-dependent rats [38]. Neuroimaging studies support the hypothesis that palatable food and addictive substances share powerful reinforcing properties [41]. The consumption of hyperpalatable substances, such as sugar, has been suggested to induce a behavioral state that is similar to the one induced by addictive substances [23], an effect that is evident even during the perinatal period in rats [8] and is reflected by similar patterns of neural activation in addictive-like eating behavior and drug dependence [21]. The obesity epidemic makes this relationship even more relevant [1,2,4]. Excessive intermittent sugar intake can lead to addiction [5] and induce symptoms of abstinence, demonstrated by behavioral observations, enhanced responding for sugar, ultrasonic vocalizations, anxiety-like behavior in the elevated plus maze, in vivo microdialysis, and increased sucrose-seeking behavior [3,12,19,38]. Withdrawal from highly palatable food can lead to depressive-like behavior that may be
V. Mangabeira et al. / Physiology & Behavior 139 (2015) 468–473
relieved by compulsive eating in rats [26]. Even withdrawal from the non-caloric sugar substitute saccharin has been shown to produce behavioral effects, such as increases in aggressive and sexual behavior in rats [7]. The reinforcement of low rate (DRL) schedule has been used to assess the effects of drugs, such as the psychomotor stimulant D amphetamine, the anxiolytic compound chlordiazepoxide, and the antidepressant desipramine. In this schedule, a minimum interval is required between lever presses, and the difficulty an animal has in keeping up with the delay in reinforcement demand has been considered an index of impulsivity. Impulsivity is a multifactor behavioral trait that is thought to contribute to various psychiatric disorders, including drug abuse and eating disorders [14,16,29,30,32]. Many clinical questionnaires have been used to measure impulsivity in humans [14]. Human studies have related impulsivity to food addiction using the Yale Food Addiction Scale (YFAS), which was developed to assess signs of substance dependence in eating behavior [20]. For example, the impulsivity dimension on an emotion dysregulation scale had the strongest specific relationship with “food addiction” on the YFAS [22]. Although the concept of impulsivity involves multiple subjacent elements, a factor analysis revealed at least two contributing factors that can be related to substance abuse: lack of inhibitory control and heightened reward sensitivity [14]. The DRL procedure models these two factors and has been used to assess impulsive action in animals [14,16,27, 30,32,35]. Withdrawal from amphetamine disrupted DRL performance, with increased responding and a reduced number of reinforcers earned [33]. Sugar-dependent rats exhibited enhanced responding for sugar after abstinence [3]. Considering the neurochemical and behavioral similarities between the effects of addictive drugs and sugar, the objective of the present study was to determine whether sugar abstinence after a relatively long consumption period impairs DRL performance. This would indicate a sugar deprivation effect that is analogous to drug withdrawal, suggesting increased impulsivity.
2. Materials and methods 2.1. Animals Naive male Wistar rats (n = 27), approximately 60 days old (range, 58–63 days) and weighing an average of 180 g (SE = 16.8 g), were used as subjects. They were singly housed under a 12 h/12 h light/dark cycle (dark at 7:00 PM) and controlled temperature (22 ± 1 °C) in semitransparent 25 × 40 × 20 cm plastic home-cages that contained a feeder on top. Depending on the experimental phase, one or two bottles were positioned above the cage lid, with their nozzle protruding inside so it could be easily reached. The regular food ration in the laboratory, Nuvilab food pellets, was freely available. Food pellets were made of whole corn, soybean meal, wheat bran, limestone, dicalcium phosphate, sodium chloride, and vitamin mineral premix. Water was restricted to 2 h daily.
2.2. Apparatus An operant 27.5 × 22.5 × 28.0 cm conditioning chamber was used that was encased in a sound-attenuating box (Med Associates, St. Albans, VT, USA). The front and back walls and ceiling were made of Plexiglas, and the lateral walls and grid rod floor in inox. A ventilation fan provided background noise. A standard lever was located on the right wall. Water was delivered for 2 s by a 0.1 ml rising dipper located at floor level in the center of that wall. After this time of water delivery, the dipper was lowered so it became inaccessible to the subject. A computer was programmed to control stimulus presentation and data recording.
469
2.3. Procedure The experiment consisted of three phases.
2.3.1. Phase 1: Baseline (BL) A bottle that contained water was withdrawn for 22 h after each session. All of the rats were shaped for bar-pressing for water under a continuous reinforcement schedule for 2 days. Afterward, DRL training began. Pressing the lever after the required interval elapsed resulted in water delivery. The initial reinforcement requirement was a DRL 5 s schedule, in which the animal had to wait for 5 s until the next lever press to receive water. The requirement was gradually increased in steps of 5 s until an interval of 20 s between each response (DRL 20) was reached. The subjects' performance thus progressed from DRL 10 s (14 experimental sessions) to DRL 15 s (25 sessions), to DRL 20 s (50 sessions). The sessions were run Monday through Friday, but the water and food regimen was maintained on the weekends. Performance on the DRL 20 was the dependent variable and used as an index of impulsive action. When the mean responses per reinforcement rate were stable between 4 and 5, training was interrupted, and the second phase began. The data from the last three sessions were used as baseline (BL).
2.3.2. Phase 2: Sugar consumption (SUG) This phase sought to generate sugar dependence. The rats were randomly assigned to two groups: Experimental (E; n = 14) and Control (C; n = 13). The animals in both groups received food and water ad libitum in their home cages, but a second bottle that contained either 10% commercial sugar solution in the E group or plain water in the C group was added. The position of the bottle in Phase 1 was maintained, and the second bottle was placed to its left. The duration of Phase 2 was 30 days.
2.3.3. Phase 3: Abstinence (ABS) The conditions of Phase 1 were reinstated. Ad libitum-fed rats were subjected to water restriction, with water available for only 2 h daily. The bottles that were previously added in Phase 2 were withdrawn so only the water bottle remained in the cage. Three DRL 20 sessions were conducted under this condition. Additional sessions were not conducted because the animals had begun to suffer deleterious effects of the long period of water restriction.
2.4. Statistical analysis A nonparametric longitudinal analysis of variance (ANOVA) that is equivalent to an ANOVA (ANOVA-type) was performed [9]. Nonparametric techniques that are based on the concept of the relative effect of treatments were proposed by Brunner et al. [10]. The ANOVAtype statistic yields results analogous to those obtained by repeatedmeasures ANOVA [28] and may be used even when the response distribution is unknown. The mean responses per reinforcement and the in inter-response times (IRTs) in the last three BL sessions and three ABS sessions were calculated. The ANOVA-type analysis was performed to test the effects of independent groups E and C (between factors) and longitudinal factors LB and ABS (within factors) and interactions between these two factors. The estimated relative treatment effect (ERE) for longitudinal measures was also computed. The ERE is a nonparametric analysis that determines the probability of a randomly selected observation from a given group at a given moment (e.g., the E group in ABS) have a greater value than another observation selected from the whole set [9,37]. An Excel macro was used for data computation (www.ime.usp.br/~jmsinger; accessed June 28, 2014). The accepted level of significance was 0.05 for all of the tests.
470
V. Mangabeira et al. / Physiology & Behavior 139 (2015) 468–473
3. Results Fig. 1 presents mean ± SE water intake (ml) in the C and E groups and mean sucrose solution intake in the E group during the 30 SUG sessions (Phase 2). Comparisons between sugar solution consumption by the E group and water consumption by the C group in the last day of phase 2 showed significant differences between groups (t(26) = 9.75, p b 0.001). There were also differences between groups on water consumption (t(26) = 1366, p b 0.001) and between sugar solution and water consumption by the E group (t(13) = 14.81, p b 0.001) in the last day of phase 2. So, sugar consumption in the E group was higher than water intake by both groups and the C group consumed more water than the E group. Daily sugar solution consumption was maintained between 25% and 40% of body weight. The mean increase in body weight was 64 g (15.6%) in the E group during the sugar intake period and 46 g (9.8%) in the C group. Median responses per reinforcer in each of the last three BL sessions and in each ABS session are presented in Fig. 2. Both groups exhibited relatively stable performance at BL. On the last day of ABS, an increase in the median number of responses per reinforcer was observed in the E group, and a decrease in the median number of responses per reinforcer was observed in the C group. An ANOVA-type analysis of the BL sessions and the first, second, and third ABS sessions (within-factors) was performed, with the E and C groups as the between-factors. Fig. 3 presents the ERE probability (see Section 2.4) in the C and E groups, comparing (a) the last three BL sessions with the three ABS sessions, (b) the last three BL sessions with the first ABS session, (c) the last three BL sessions with the second ABS session, and (d) the last three BL sessions with the third ABS session. As shown in Fig. 3a, the overall ERE in C group remained nearly constant from BL to the ABS condition (from 0.48 to 0.49) but increased from 0.43 to 0.61 in the E group. However, those differences were not significant (QA: between = 0.13, df = 1, p = 0.72; within = 3.72, df = 1, p = 0.053; interaction = 2.9, df = 1, p = 0.08). The analysis of the difference between the last three BL sessions and first ABS session (Fig. 3b) showed similar results, in which no significant changes in ERE values were observed between BL and ABS, with no interaction between these factors (QA: between = 0.03, df = 1, p = 0.87; within = 2.53, df = 1, p = 0.11; interaction = 0.49, df = 1, p = 0.48). A significant
difference was found between BL and the second ABS session (QA: between = 0.03, df = 1, p = 0.87; within = 4.25, df = 1, p = 0.04; interaction = 1.93, df = 1, p = 0.17). Differences between groups were statistically evident in the third ABS session (Fig. 3d), in which an interaction between the within and between factors was observed and only the E group responded more for the sucrose solution (QA: between = 0.84, df = 1, p = 0.36; within = 0.00, df = 1, p = 0.98; interaction = 6.36, df = 1, p = 0.01). An increase in the ERE difference was observed in the E group between BL and ABS and between the first and third ABS sessions (from 0.47 to 0.61), but a decrease in this effect was observed in the C group (from 0.53 to 0.38). Thus, the obtained daily ERE values in the E group indicate that this effect gradually increased from the first to third session after withdrawal. Median IRTs in each of the last three BL sessions and in each ABS session are presented in Fig. 4. Both groups showed relatively stable performance at BL on the last two days of BL, but the median of the C group decreased on the last day. On the other hand, IRTs of the E group showed a slight decrease on the three first days of ABS when compared to BL, whereas IRTs of the C group showed a slight increase on the first two days of ABS and a higher increase in the last day. As indicated in Fig. 5, when the overall ERE of IRTs was computed (last three BL sessions compared to three first ABS sessions; Fig. 5a), no significant differences in EREs between groups (QA = 0.03, df = 1, p = 0.850), nor between BL and ABS (QA = 0.72, df = 1, p = 0.395) or interaction (QA = 1.56, df = 1, p = 0.212) were found. Also, there were no differences between these parameters when BL was compared to the first ABS session (QA: between = 0.13, df = 1, p = 0.718; within = 1.75 df = 1, p = 0.186; interaction = 0.51, df = 1, p = 0.476; Fig. 5b) or the second ABS session (QA: between = 0.329, df = 1, p = 0.566; within = 1.916 df = 1, p = 0.166; interaction = 0.42, df = 1, p = 0.516; Fig. 5c). However, in the third session of abstinence, groups displayed different EREs (QA: between = 0.125, df = 1, p = 0.724; within = 0.448, df = 1, p = 0.485; interaction = 5.127, df = 1, p = 0.024; Fig. 5c), indicating shorter IRTs for the E group as shown in Fig. 4. 4. Discussion The present study investigated the possible signs of abstinence in DRL performance after a prolonged period of sugar consumption.
Fig. 1. Mean (±SE) water intake in the C and E groups and mean (±SE) sugar intake in the E group during the 30 SUG sessions (Phase 2). **p b 0.001 for comparisons between E (sugar) and C (water). ##p b 0.001 for comparisons between E (sugar) and C (water). ++p b 0.001 between E (sugar) and E (water).
V. Mangabeira et al. / Physiology & Behavior 139 (2015) 468–473
471
Fig. 2. Median responses per reinforcer in the C and E groups in the three last BL sessions (48, 49, 50) and three ABS sessions (1, 2, 3). The 25th and 75th percentiles are presented for each group in each session. Δ represents outliers of the C distribution, and ○ represents outliers of the E distribution.
Sucrose intake in the animals that had a choice between plain water and a sucrose solution indicated a clear preference for sugar. Abstinence affected DRL performance, leading to higher response rates per reinforcer and shorter IRTs, compared with baseline after 3 days of withdrawal. Individual intrasubject differences and the analysis of the ERE showed a similar deterioration in performance, reflected by an increase in the loss of reinforcers. Only animals that were subjected to sugar abstinence presented impaired DRL performance. The observed impairment of DRL performance suggests an increase in impulsive behavior following sugar deprivation, similar to the effect produced by withdrawal from drugs of abuse, such as amphetamine and phencyclidine [11,33]. After sugar abstinence, the control of responses 0.8
a
0.7
0.7
0.5
0.5
0.4
0.4
0.3
0.3
0.1
0.1
0
0
0.8
c
0.7
Probability
E
0.2
0.2
0.5
0.5
0.4
0.4
0.3
0.3
0.2
0.2
0.1
0.1 ABS
Phase
**
0.7
0.6
BL
C
0.8
0.6
0
b
0.6
0.6
Probability
by long-term reinforcement, a sensitive indicator of impulsivity, was attenuated in the E group. This result may be relevant for studies of obesity and several behaviors that are affected by deficiencies in self-control. The DRL procedure was used because it is considered a typical animal model of impulsivity [30]. However, in the present study the DRL procedure was confounded by the time needed to reach stable performance. The animals had to be subjected to an extended period of water restriction to attain the stipulated criterion, which should perhaps be revised. Water restriction imposed physiological alterations, but this constraint was unavoidable because the metabolic effects of food restriction would have outweighed sensorial factors. New procedures for DRL training might be implemented, and other procedures,
0
BL
Phase
d
ABS
Fig. 3. Estimated relative effect (ERE) of responses per reinforcer and its 95% confidence intervals in the C and E groups in comparisons between (a) the last three BL sessions and three ABS sessions, (b) last three BL sessions and first ABS session, (c) last three BL sessions and second ABS session, and (d) last three BL sessions and third ABS session. **p b 0.01 for differences between groups.
472
V. Mangabeira et al. / Physiology & Behavior 139 (2015) 468–473
Fig. 4. Median inter-response time (seconds) in the C and E groups in the three last BL sessions (48, 49, 50) and three ABS sessions (1, 2, 3). 25th and 75th percentiles are presented for each group in every session. Δ represents outliers of C distribution and ○ outliers of E distribution.
such as the paced fixed consecutive number test [15], could be used as alternative models. The DRL schedule confirmed the similar effects of addictive drugs and sucrose. The present data corroborate the behavioral changes observed after excessive sugar consumption, such as signs of withdrawal [12,26,38], and similarities revealed by brain imaging techniques in craving in response to palatable food and drugs of abuse [6,41]. Given
the typical effect of antidepressants on DRL performance in reducing response rates and increasing reinforcement rates [31], a pertinent issue would be to investigate whether these drugs would reverse the effect of sugar abstinence reported herein, suggesting relief from depression [26]. Further research could also address the relationship between behavioral and neural changes elicited by sugar withdrawal in the DRL procedure.
Fig. 5. Estimated relative effect (ERE) of IRTs and its 95% confidence intervals in the C and E groups in comparisons of (a) the last three BL sessions and three ABS sessions, (b) last three BL sessions and first ABS session, (c) last three BL sessions and second ABS session, and (d) last three BL sessions and third ABS session. *p b .005 for differences between groups.
V. Mangabeira et al. / Physiology & Behavior 139 (2015) 468–473
5. Conclusion The animal procedure utilized in the present study contributes to our understanding of sugar addiction, demonstrating similarities with drug addiction in a model of impulsivity. This conclusion supports the hypothesis that the excessive consumption of highly palatable food and drugs of abuse can generate similar behavioral consequences. Such consequences are likely paralleled by brain changes caused by hedonic eating and various types of addiction. As stated by Hoebel in 2009 (http://www.foodaddictionsummit.org/presenters-hoebel.htm; accessed June 15, 2014), “The brain is getting addicted to its own opioids as it would to morphine or heroin. Drugs give a bigger effect, but it is essentially the same process”. References [1] N.M. Avena, M.E. Bocarsly, B.G. Hoebel, M.S. Gold, Overlaps in the nosology of substance abuse and overeating: the translational implications of “food addiction.”, Curr. Drug Abus. Rev. 4 (2011) 133–139, http://dx.doi.org/10.2174/1874473711104030133. [2] N.M. Avena, J.A. Gold, C. Kroll, M.S. Gold, Further developments in the neurobiology of food and addiction: update on the state of the science, Nutrition 28 (2012) 341–343, http://dx.doi.org/10.1016/j.nut.2011.11.002. [3] N.M. Avena, K.A. Long, B.G. Hoebel, Sugar-dependent rats show enhanced responding for sugar after abstinence: evidence of a sugar deprivation effect, Physiol. Behav. 84 (2005) 359–362, http://dx.doi.org/10.1016/j.physbeh.2004.12.016. [4] N.M. Avena, S. Murray, M.S. Gold, Comparing the effects of food restriction and overeating on brain reward systems, Exp. Gerontol. 48 (2013) 1062–1067, http://dx.doi. org/10.1016/j.exger.2013.03.006. [5] N.M. Avena, P. Rada, B.G. Hoebel, Evidence of sugar addiction: behavioral and neurochemical effects of intermittent, excessive sugar intake, Neurosci. Biobehav. Rev. 32 (2008) 20–39, http://dx.doi.org/10.1016/j.neubiorev.2007.04.019. [6] N.M. Avena, P. Rada, B.G. Hoebel, Sugar and fat bingeing have notable differences in addictive-like behavior, J Nutr. 139 (2009) 623–628, http://dx.doi.org/10.3945/jn. 108.097584. [7] I.V. Belozertseva, I.A. Sukhotina, J.M. Vossen, A.Y. Bespalov, Facilitation of aggressive and sexual behaviors by saccharin deprivation in rats, Physiol. Behav. 80 (2004) 531–539, http://dx.doi.org/10.1016/j.physbeh.2003.10.012. [8] M.E. Bocarsly, J.R. Barson, J.M. Hauca, B.G. Hoebel, S.F. Leibowitz, N.M. Avena, Effects of perinatal exposure to palatable diets on body weight and sensitivity to drugs of abuse in rats, Physiol. Behav. 107 (2012) 568–575, http://dx.doi.org/10.1016/j. physbeh.2012.04.024. [9] E. Brunner, F. Langer, Nonparametric analysis of ordered categorical data in designs with longitudinal observations and small sample sizes, Biom. J. 42 (2000) 663–675, http://dx.doi.org/10.1002/1521-4036(200010)42:6b663::AID-BIMJ663N3.0.CO;2-7. [10] E. Brunner, U. Munzel, M.L. Puri, Rank-score tests in factorial designs with repeated measures, J. Multivar. Anal. 70 (1999) 286–317, http://dx.doi.org/10.1006/jmva. 1999.1821. [11] M.E. Carroll, J.L. Mach, R.M. La Nasa, J.L. Newman, Impulsivity as a behavioral measure of withdrawal of orally delivered PCP and nondrug rewards in male and female monkeys, Psychopharmacology 207 (2009) 85–98, http://dx.doi.org/10.1007/ s00213-009-1636-y. [12] C. Colantuoni, P. Rada, J. McCarthy, C. Patten, N.M. Avena, A. Chadeayne, B.G. Hoebel, Evidence that intermittent, excessive sugar intake causes endogenous opioid dependence, Obes. Res. 10 (2002) 478–488, http://dx.doi.org/10.1038/oby.2002.66. [13] C. Colantuoni, J. Schwenker, J. McCarthy, P. Rada, B. Ladenheim, J.L. Cadet, G.J. Schwartz, T.H. Moran, B.G. Hoebel, Excessive sugar intake alters binding to dopamine and mu-opioid receptors in the brain, Neuroreport 12 (2001) 3549–3552, http://dx.doi.org/10.1097/00001756-200111160-00035. [14] S. Dawe, N.J. Loxton, The role of impulsivity in the development of substance use and eating disorders, Neurosci. Biobehav. Rev. 28 (2004) 343–351, http://dx.doi. org/10.1016/j.neubiorev.2004.03.007. [15] J.L. Evenden, The pharmacology of impulsive behavior in rats: II. The effects of amphetamine, haloperidol, imipramine, chlordiazepoxide and other drugs on fixed consecutive number schedules (FCN 8 and FCN 32), Psychopharmacology 138 (1998) 283–294, http://dx.doi.org/10.1007/s002130050673. [16] J.L. Evenden, Varieties of impulsivity, Psychopharmacology 146 (1999) 348–361, http://dx.doi.org/10.1007/PL00005481. [17] J.D. Fernstrom, S.D. Munger, A. Sclafani, I.E. de Araujo, A. Roberts, S. Molinary, Mechanisms for sweetness, J. Nutr. 142 (2012) 1134S–1141S, http://dx.doi.org/10. 3945/jn.111.149567.
473
[18] B.M. Geiger, M. Haburcak, N.M. Avena, M.C. Moyer, B.G. Hoebel, E.N. Pothos, Deficits of mesolimbic dopamine neurotransmission in rat dietary obesity, Neuroscience 159 (2009) 1193–1199, http://dx.doi.org/10.1016/j.neuroscience.2009.02.007. [19] J.W. Grimm, A.M. Fyall, D.P. Osincup, Incubation of sucrose craving: effects of reduced training and sucrose pre-loading, Physiol. Behav. 84 (2005) 73–79, http:// dx.doi.org/10.1016/j.physbeh.2004.10.011. [20] A.N. Gearhardt, W.R. Corbin, K.D. Brownell, The preliminary validation of the Yale Food Addiction Scale, Appetite 52 (2009) 430–436, http://dx.doi.org/10.1016/j. appet.2008.12.003. [21] A.N. Gearhardt, C.M. Grilo, R.J. DiLeone, K.D. Brownell, M.N. Potenza, Can food be addictive? Public health and policy implications, Addiction 106 (2011) 1208–1212, http://dx.doi.org/10.1111/j.1360-0443.2010.03301.x. [22] A.N. Gearhardt, M.A. White, R.M. Masheb, P.T. Morgan, R.D. Crosby, C.M. Grilo, An examination of the food addiction construct in obese patients with binge eating disorder, Int. J. Eat. Disord. 45 (2012) 657–663, http://dx.doi.org/10.1002/eat.20957. [23] A.N. Gearhardt, S. Yokum, P.T. Orr, E. Stice, W.R. Corbin, K.D. Brownell, Neural correlates of food addiction, Arch. Gen. Psychiatr. 68 (2011) 808–816, http://dx.doi.org/ 10.1001/archgenpsychiatry.2011.32. [24] A. Hajnal, R. Norgren, Accumbens dopamine mechanisms in sucrose intake, Brain Res. 904 (2001) 76–84, http://dx.doi.org/10.1016/S0006-8993(01)02451-9. [25] B.G. Hoebel, N.M. Avena, M.E. Bocarsly, P. Rada, Natural addiction: a behavioral and circuit model based on sugar addiction in rats, J. Addict. Med. 3 (2009) 33–41, http://dx.doi.org/10.1097/ADM.0b013e31819aa621. [26] A. Iemolo, M. Valenza, L. Tozier, C.M. Knapp, C. Kornetsky, L. Steardo, V. Sabino, P. Cottone, Withdrawal from chronic, intermittent access to a highly palatable food induces depressive-like behavior in compulsive eating rats, Behav. Pharmacol. 23 (2012) 593–602, http://dx.doi.org/10.1097/FBP.0b013e328357697f. [27] P. Jayarajan, R. Nirogi, A. Shinde, Effect of olanzapine on scopolamine induced deficits in differential reinforcement of low rate 72 s (DRL-72 s) schedule in rats: involvement of the serotonergic receptors in restoring the deficits, Eur. J. Pharmacol. 720 (2013) 344–354, http://dx.doi.org/10.1016/j.ejphar.2013.10.005. [28] M.H. Kutner, C. Nachtsheim, J. Neter, W. Li, Applied Linear Statistical Models, 5th ed. McGraw-Hill, Boston, 2004. [29] B.A. McMillen, L.W. Means, J. Matthews, Comparison of the alcohol-preferring P rat to the Wistar rat in behavioral tests of impulsivity and anxiety, Physiol. Behav. 63 (1998) 371–375 (PII S0031–9384(97)00442–3). [30] J. Monterosso, G. Ainslie, Beyond discounting: possible experimental models of impulse control, Psychopharmacology 146 (1999) 339–347, http://dx.doi.org/10.1007/ PL00005480. [31] J.M. O'Donnell, G.J. Marek, L.S. Seiden, Antidepressant effects assessed using behavior maintained under a differential-reinforcement-of-low-rate (DRL) operant schedule, Neurosci. Biobehav. Rev. 29 (2005) 785–798, http://dx.doi.org/10.1016/j.neubiorev. 2005.03.018. [32] J.L. Perry, M.E. Carroll, The role of impulsive behavior in drug abuse, Psychopharmacology 200 (2008) 1–26, http://dx.doi.org/10.1007/s00213-008-1173-0. [33] J.D. Peterson, M.E. Wolf, F.J. White, Impaired DRL 30 performance during amphetamine withdrawal, Behav. Brain Res. 143 (2003) 101–108, http://dx.doi.org/10. 1016/S0166-4328(03)00035-4. [34] J.D. Pomonis, D.C. Jewett, C.M. Kotz, J.E. Briggs, C.J. Billington, A.S. Levine, Sucrose consumption increases naloxone-induced c-Fos immunoreactivity in limbic forebrain, Am. J. Physiol. Regul. Integr. Comp. Physiol. 278 (2000) R712–R719 (PubMed 10712293). [35] J.B. Richards, K.E. Sabol, L.S. Seiden, DRL interresponse-time distributions: quantification by peak deviation analysis, J. Exp. Anal. Behav. 60 (1993) 361–385, http:// dx.doi.org/10.1901/jeab. 1993.60-361. [36] M.T.A. Silva, F.L. Gonçalves, M. Garcia-Mijares, Neural events in the reinforcement contingency, Behav. Anal. 30 (2007) 17–30 ([PubMed] 22478485). [37] J.M. Singer, Z. Poleto, P. Rosa, Parametric and nonparametric analyses of repeated ordinal categorical data, Biom. J. 46 (2004) 460–473, http://dx.doi.org/10.1002/bimj. 200310045. [38] R. Spangler, K.M. Wittkowski, N.L. Goddard, N.M. Avena, B.G. Hoebel, S.F. Leibowitz, Opiate-like effects of sugar on gene expression in reward areas of the rat brain, Mol. Brain Res. 124 (2004) 134–142, http://dx.doi.org/10.1016/j.molbrainres.2004.02.013. [39] N.D. Volkow, G.J. Wang, D. Tomasi, R.D. Baler, Obesity and addiction: neurobiological overlaps, Obes. Rev. 14 (2013) 2–18, http://dx.doi.org/10.1111/j.1467-789X.2012. 01031.x. [40] R.A. Wise, Role of brain dopamine in food reward and reinforcement, Philos. Trans. R. Soc. Lond. B Biol. Sci. 361 (2006) 1149–1158, http://dx.doi.org/10.1098/rstb.2006.1854. [41] Y. Zhang, K.M. von Deneen, J. Tian, M.S. Gold, Y. Liu, Food addiction and neuroimaging, Curr. Pharm. Des. 17 (2011) 1149–1157 (BSP/CPD/E-Pub/000438 [pii]).