- Email: [email protected]
6-Hydroxydopamine lesions of the ventral tegmental area suppress ghrelin's ability to elicit food-reinforced behavior
Neuroscience Letters 499 (2011) 70–73
Contents lists available at ScienceDirect
Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet
6-Hydroxydopamine lesions of the ventral tegmental area suppress ghrelin’s ability to elicit food-reinforced behavior Zachary Y. Weinberg, Marjorie L. Nicholson, Paul J. Currie ∗ Department of Psychology, Reed College, Portland, OR 97202, United States
a r t i c l e
i n f o
Article history: Received 15 March 2011 Received in revised form 28 April 2011 Accepted 16 May 2011 Keywords: Appetitive motivation Dopamine Food intake Ghrelin Midbrain Reward Ventral tegmentum
a b s t r a c t While past research suggests that ghrelin stimulates appetite through an action on hypothalamic signaling, recent evidence indicates that the peptide acts via mesotelencephalic dopamine neurons to alter appetitive motivation. In the present study, rats were trained to operantly respond for food on a progressive ratio PR5 schedule until stable breakpoints were established. Ghrelin (30–300 pmol) was then injected directly into the ventral tegmental area (VTA) and the 300 pmol dose was observed to increase breakpoint. The dopaminergic neurotoxin 6-hydroxydopamine (6-OHDA, 6 g) was subsequently administered into the VTA resulting in a significant depletion of striatal dopamine. Stable breakpoints were then re-established. When ghrelin’s effects were reassessed, the peptide’s ability to alter operant responding for food was reliably reduced. Our findings demonstrate that ghrelin induces food-reinforced behavior in the mesotelencephalic reward pathway and that this effect is dependent on intact dopaminergic signaling. We conclude that the metabolic peptide ghrelin interacts with dopamine, within reward circuitry, to modulate appetitive behavior. © 2011 Elsevier Ireland Ltd. All rights reserved.
In the central nervous system ghrelin, a 28-amino acid peptide, is the endogenous ligand for the growth hormone secretagogue receptor (GHS-R1A) [20]. Ghrelin’s orexigenic action in the hypothalamus is well documented [9,24]. In addition, other work examining the effects of ghrelin has implicated the peptide in the induction of anxiety-like behaviors [4,5,8]. Recently, ghrelin’s role in the mesotelencephalic dopamine circuit has been investigated [23]. This pathway, implicated in reward and motivated behaviors, projects from the midbrain ventral tegmental area (VTA) to forebrain structures, including the nucleus accumbens in the ventral striatum. Ghrelin’s possible action in the mesotelencephalic dopamine system could provide evidence of a hypothalamic peptide mediating not only caloric intake but also regulating the rewarding aspects of food. There is strong empirical support for a dopamine-driven model of reward in the brain. When studying reinforced or motivated behavior, an experimental paradigm involving manipulation of accumbal dopamine release is commonly utilized [32,33] along with measures of drug- and food-seeking behaviors [27,29]. Hedonic feeding, eating behavior associated with food palatability, is now known to be mediated, at least in part, by mesotelencephalic dopamine neurons. Prior research has shown that decreases in brain concentrations of dopamine receptors in human subjects are correlated with decreased restraint in food intake [30]. Indeed, it
∗ Corresponding author. Tel.: +1 503 777 7267. E-mail address: [email protected] (P.J. Currie). 0304-3940/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2011.05.034
has been suggested that obesity could be modeled as a form of addiction and that in this model the mesotelencephalic dopamine circuit could be a primary therapeutic target in obesity treatment [31]. Additional investigations have examined the effects of eatingrelated peptides such as neuropeptide Y (NPY) and ghrelin on food-related reward. Microinjections of NPY and ghrelin into the VTA elicit a robust eating response [2,22]. Naleid et al. [22] argue these data demonstrate that ghrelin has a significant effect on reward-based food intake. Ghrelin’s reported action within the VTA raises the possibility that the peptide has a broader effect on reward and motivated behaviors beyond eating-related mechanisms. Jerlhag et al. [14] examined ghrelin’s ability to elicit locomotion and accumbal dopamine release. These effects are typically associated with the action of cocaine and other drugs of abuse within the accumbens and are considered biological indicators of reward. Jerlhag et al. [14] found that when injected into the VTA, ghrelin produced locomotion and increases in accumbal extracellular dopamine in a dose-dependent manner. There are at least two possible mechanisms through which ghrelin’s signaling in the VTA could alter accumbal dopamine levels. Ghrelin’s effects could be mediated by downstream signaling, presumably by the already established dopamine pathway. Alternatively, it remains possible that ghrelin’s effect on food-related reward could be mediated by non-dopaminergic circuits. Previous work has characterized the role of NPY, another orexigenic peptide, on food-reinforced behaviors. Like ghrelin, hypothalamic NPY has been reported to elicit alterations in eat-
Z.Y. Weinberg et al. / Neuroscience Letters 499 (2011) 70–73
ing and energy metabolism [6]. Jewett et al. [17] reported that NPY increased food-motivated behavior when injected into the lateral ventricle. Rats were trained to respond on a progressive ratio (PR) reinforcement schedule. The breakpoint, or point at which the rats stopped responding for food, was assessed. Breakpoint was used as a measure of how motivated the rat was to seek food. This work, combined with past research, suggests that, in addition to a role in mediating energy homeostasis, metabolic peptides might also be implicated in underlying neural circuits mediating reward such as food-reinforced behavior. Based on ghrelin’s feeding effect in the VTA and the VTA’s role in mediating reinforced behaviors, the current study investigated ghrelin’s ability to elicit an increase in breakpoint using an operant paradigm. Once we had established that VTA ghrelin microinjection increased breakpoint, we then examined whether this effect was dependent on intact dopamine signaling. Twelve adult male Sprague-Dawley rats (Harlan Laboratories, Kent, WA), weighing 350–400 g at time of surgery, were housed individually with ad libitum access to water and standard chow (LabDiet, PMI Nutrition, Brentwood, MO). The animal colony room was maintained at a temperature of 22 ± 2 ◦ C and on a 12-h light/dark cycle with lights on at 07:00 and off at 19:00 h. All experiments were approved by the Institutional Animal Care and Use Committee of Reed College. Prior to surgery, rats were trained to respond on a PR5 schedule in standard operant chambers. An initial response was rewarded with a banana pellet (Noyes, Lancaster, NH) and five additional responses were required for each successive reward [17]. In order to establish a consistent breakpoint, test sessions were conducted daily over two weeks, with each rat tested over a 2 h period at the onset of the active cycle. Animals were food deprived for the 6 h immediately prior to testing (in the light cycle). During training, rats were initially exposed to a single fixed interval (FI) session where food was dispensed every 30 s. Following one session of FI training, rats were then manually shaped to respond at fixed ratio 1 (FR1) where each response was rewarded with a pellet. Once rats responded reliably at FR1, the schedule was gradually increased to FR10 over the course of three sessions. Finally, rats were placed on a PR5 schedule for seven sessions, at which point the breakpoint of each rat was within a ±10% range of the rat’s last two sessions. Rats were then provided free access to food and water for 48 h after which surgery was conducted. The animals were surgically implanted with stainless steel guide cannulae (30 gauge, Plastics One, Roanoke, VA) aimed at the VTA. Cannulae were implanted 4 mm dorsal to the target with coordinates relative to bregma: −5.3 mm posterior, ±1.0 mm lateral and −4.1 mm ventral [25] with the incisor bar at −3.6 mm. Guide cannulae were counterbalanced with respect to implantation in the left and right cerebral hemispheres. After a 10 day post-surgery recovery period, rats were once again run in PR5 sessions until the breakpoint stabilized (seven session maximum). In order to examine the effects of ghrelin on breakpoint, ghrelin (30 or 300 pmol; Biopeptide Co. Inc., San Diego, CA) or vehicle was injected into the VTA 30 min prior to testing using a 32-gauge injector (Plastics One, Roanoke, VA) extending 4 mm beyond the permanent guide cannula. The total volume of each injection was 0.4 l. Doses were injected over a 1-min period and the injector was left in place for an additional minute to allow diffusion of the peptide. All animals were subjected to all three treatments (vehicle first followed by ghrelin in high-to-low dose order) and all test sessions occurred at least 48 h apart. In a second experiment, ghrelin’s ability to increase breakpoint was confirmed in a separate group of rats (n = 7). That is, rats were injected with vehicle and ghrelin (300 pmol) and changes in breakpoint were assessed. After this replication, rats were unilaterally injected with 6-hydroxydopamine (6-OHDA) in order to deplete central levels of dopamine. The unilateral 6-OHDA lesion
71
Fig. 1. Breakpoint as a function of treatment condition (n = 12). Ghrelin, at a dose of 300 pmol, significantly increased breakpoint compared to vehicle. * p < 0.05.
is a well-established and robust paradigm for assessing dopaminedependent behavior in rats [28]. Each rat was anesthetized with sodium pentobarbitol (50 mg/kg; Ovation Pharmaceuticals) and immediately injected with desipramine (25 mg/kg, i.p.; Sigma). Thirty min later each rat received a unilateral microinjection of 6-OHDA (6 g; Sigma) directly into the VTA. A 14-day recovery period followed. A stable breakpoint was re-established over 7 PR5 sessions. Finally, rats were injected with vehicle and then ghrelin (300 pmol) into the VTA and alterations in breakpoint were measured. All cannulae placements were confirmed via histological examination. Sections were examined by light microscopy and viewed relative to the stereotaxic atlas of Paxinos and Watson [25]. Coronal sections were taken through the entire striatum, including the ventral striatum [13,25]. Samples were homogenized in 0.2 M perchloric acid, sonicated, and centrifuged at 15,000 rpm over 15 min. Dopamine concentrations were measured using high performance liquid chromatography and electrochemical detection [13]. All rats reported here were found to have injector tracks extending into the VTA. In addition, striatal assays confirmed a 77% decrease in ipsilateral dopamine transmitter compared to a control group [t(12) = −13.81, p < 0.0001]. Our reductions in DA levels are consistent with previous work [28]. In experiment 1, data were analyzed using a one-way repeated measures Analysis of Variance (ANOVA) followed by post hoc Tukey tests. The criterion for significance was p < 0.05. As illustrated in Fig. 1, ANOVA confirmed that ghrelin injection into the VTA increased operant responding for food. Specifically, the dose of 300 pmol significantly increased breakpoint (F(2,22) = 7.64, p < 0.003). The 30 pmol dose, however, did not alter breakpoint values. In experiment 2, two-way ANOVA confirmed a significant lesion × ghrelin interaction (F(2,28) = 13.18, p < 0.02). As shown in Fig. 2, again the 300 pmol dose of ghrelin effectively increased breakpoint compared to control. Following treatment with 6OHDA, a significant decrease in breakpoint was observed when rats were injected with ghrelin, compared to vehicle. Breakpoint values for post-lesioned ghrelin-treated rats were also lower in comparison to pre-lesion ghrelin values. Moreover, there was no effect of lesion on the breakpoints of rats treated with vehicle compared to breakpoints in the pre-lesion vehicle condition. The results of experiment 1 are consistent with the findings of Jewett et al. [17] in showing that ghrelin, like NPY, increases foodreinforced responding. In fact, the ghrelin-dependent increase in breakpoint is comparable to that seen by Jewett et al. with NPY. Our work indicates that ghrelin’s effect on food-related reward can be localized to the VTA. Along with work by Jerlhag et al. [14] and
72
Z.Y. Weinberg et al. / Neuroscience Letters 499 (2011) 70–73
Fig. 2. Breakpoint as a function of treatment condition (n = 7). Prior to 6-OHDA lesion, ghrelin (300 pmol) increased breakpoint relative to vehicle. Following the lesion, ghrelin significantly decreased breakpoint. * p < 0.05 compared to vehicle of the same lesion condition. Vehicle breakpoints did not differ significantly between lesion conditions.
Abizaid et al. [2], these findings further implicate a role for ghrelin in the mediation of reinforced behavior via the mesotelencephalic dopamine pathway. Moreover, the findings of experiment 2 further indicate that ghrelin’s action within the VTA is dependent on dopaminergic signaling. When forebrain dopamine was depleted, operant responding for food reward was significantly decreased from vehicle after ghrelin treatment, though post-lesion vehicle did not differ significantly from pre-lesion vehicle. Prior research has implicated ghrelin as a peptide transmitter acting within the hypothalamus and exerting an effect on energy metabolism. When injected into the paraventricular or arcuate nuclei, ghrelin elicits a robust increase in food intake. Microinjections of ghrelin into these same nuclei produce shifts in energy substrate utilization [7,9]. These findings suggest, therefore, that ghrelin alters appetite and energy metabolism, including the promotion of carbohydrate oxidation and the preservation of fat stores, by an action of hypothalamic circuits mediating positive energy balance. Moreover, the findings of the present study further suggest that ghrelin’s ability to alter food intake is not restricted to hypothalamic metabolic circuitry. The findings of our second experiment raise several important questions regarding the mechanism of ghrelin’s action in the VTA. It is clear that ghrelin exerts a positive effect on foodrelated motivated behavior. It is also clear that ghrelin’s effect is dependent upon intact dopaminergic neurons in the VTA since 6OHDA treatment attenuated ghrelin-induced responding for food rewards. However, if ghrelin’s action in the VTA was on dopaminergic neurons exclusively, then depletion of these neurons should reduce ghrelin-induced breakpoints to levels comparable to vehicle. Instead, we observed that when rats were administered ghrelin following dopamine depletion, these animals not only no longer showed an increase in operant responding for food but in fact exhibited a significant reduction in breakpoint in comparison to both pre- and post-lesion vehicle treatment. This suggests that in the VTA ghrelin may also be acting on non-dopaminergic transmitter systems. Clearly, dopamine plays an important role in reward signaling. However, this role is supplemented by several other transmitter systems including the opioids, GABA, serotonin and acetylcholine. Jerlhag et al. [15] showed that ghrelin’s ability to elicit accumbal dopamine overflow and stimulate locomotion when injected into the VTA or laterodrosal tegmental area (LDTg) is dependent on cholinergic signaling in this region. Additionally, Kawahara et al. [18] demonstrated that bicuculline, a GABAA antagonist, injected into the VTA blocked ghrelin’s effect on accumbal dopamine
release. These findings, coupled with Abizaid and Horvath’s [1] data that show GHS-R1As expressed on GABAergic neurons in the VTA, present two possible non-dopaminergic mechanisms for the reduction in breakpoint seen after 6-OHDA treatment. Furthermore, our lab has previously shown that in the hypothalamus, serotonin modulates ghrelin’s effect on food intake [7] so it is not unreasonable to suggest that VTA-proximal serotonergic circuits may also help to modulate a ghrelinergic effect in the VTA. Over the past decade increasing evidence suggests a role for energy-regulatory peptides in food reward. While ghrelin is prominent among these signals, other relevant neurotransmitters include NPY, insulin, agouti-related peptide, leptin and orexin. Previous research has implicated these peptides in the regulation of food intake, food hedonics, food choice and palatability [10–12,23,26]. The current study shows that beyond increasing food intake, ghrelin increases motivation to obtain food. Other metabolic hormones may produce similar effects. Future studies should investigate the ability of other peptide transmitters to elicit motivation within the mesotelencephalic pathway. As described above, ghrelin induces increased locomotion and accumbal dopamine release, two findings commonly observed after the administration of drugs of abuse into the reward circuit [32]. This suggests that ghrelin could potentially impact states of addiction. A relationship between ghrelin and alcohol addiction has been demonstrated in several clinical populations. During the early stages of withdrawal from alcohol, alcohol-dependent individuals express increased levels of endogenous ghrelin [19,21]. In addition, ghrelin levels are correlated with subjective cravings for alcohol in recovering alcoholics [3]. Experiments have also shown that intact ghrelin signaling is necessary for animal subjects to develop alcohol dependence [16]. This evidence clearly supports a role for ghrelin in alcohol reward. Alcohol is a unique substance of abuse as it elicits an addiction while also containing caloric content. Consequently ghrelin, an orexigenic peptide that also impacts dopamine and reward, could provide evidence of an interaction between systems of feeding behavior and addiction. Finally, Volkow and Wise [31] discuss the overlap in biological mechanisms of feeding, obesity and drug addiction. Numerous studies have indicated that antagonism of telencephalic dopaminergic receptors attenuates free-feeding and operant responding for food reward as well the reinforcing effects of psychomotor stimulants. The mesotelencephalic dopamine projection from the VTA to the accumbens is most frequently implicated in reward although clearly other dopamine projections are involved. Given this overlap, it would seem appropriate to investigate ghrelin and other metabolic peptides in the reward circuit in paradigms commonly associated with assessing drugs of abuse such as conditioned place preference and self-administration. In conclusion, ghrelin induces food-reinforced behavior in the mesotelencephalic reward pathway, specifically when injected into the VTA. This effect is dependent on intact dopaminergic neurotransmission. Our results suggest that the metabolic peptide ghrelin interacts with dopamine within the reward system to modulate appetitive behavior. Acknowledgements This research was funded by a Reed College Science Research Fellowship. The authors thank Dr. Tim Hackenberg for programming and consulting on the operant portion of the experiment. References [1] A. Abizaid, T.L. Horvath, Unraveling neuronal circuitry regulating energy homeostasis: plasticity in feeding circuits, Drug Discov. Today: Dis. Models 2 (3) (2005) 191–196.
Z.Y. Weinberg et al. / Neuroscience Letters 499 (2011) 70–73 [2] A. Abizaid, Z.-W. Liu, Z.B. Andrews, M. Shanabrough, E. Borok, J.D. Elsworth, R.H. Roth, M.W. Sleeman, M.R. Picciotto, M.H. Tschop, X.-B. Gao, T.L. Horvath, Ghrelin modulates the activity and synaptic input organization of midbrain dopamine neurons while promoting appetite, J. Clin. Invest. 116 (12) (2006) 3229–3239. [3] G. Addolorato, E. Capristo, L. Leggio, A. Ferrulli, L. Abenavoli, N. Malandrino, S. Farnetti, M. Domenicali, C. D’Angelo, L. Vonghia, A. Mirijello, S. Cardone, G. Gasbarrini, Relationship between ghrelin levels, alcohol craving, and nutritional status in current alcoholic patients, Alcohol. Clin. Exp. Res. 30 (11) (2006) 1933–1937. [4] V.P Carlini, M.E. Monzon, M.M. Varas, A.B. Cragnolini, H.B. Schioth, T.N. Scimonelli, S.R. de Barioglio, Ghrelin increases anxiety-like behavior and memory retention in rats, Biochem. Biophys. Res. Commun. 299 (5) (2002) 739–743. [5] V.P. Carlini, M.M. Varas, A.B. Cragnolini, H.B. Schioth, T.N. Scimonelli, S.R. de Barioglio, Differential role of the hippocampus, amygdala, and dorsal raphe nucleus in regulating feeding, memory, and anxiety-like behavioral responses to ghrelin, Biochem. Biophys. Res. Commun. 313 (3) (2004) 635–641. [6] P Currie, Integration of hypothalamic feeding and metabolic signals: focus on neuropeptide Y, Appetite 41 (3) (2003) 335–337. [7] P.J. Currie, C.S. John, M.L. Nicholson, C.D. Chapman, K.E. Loera, Hypothalamic paraventricular 5-hydroxytryptamine inhibits the effects of ghrelin on eating and energy substrate utilization, Pharmacol. Biochem. Behav. 97 (1) (2010) 152–155. [8] P.J. Currie, R. Khelemsky, C.S. John, S. Higgs, Cb1 receptor antagonism alters the anxiogenic and feeding-stimulant effects of ghrelin, Appetite 52 (2009) 824. [9] P.J. Currie, A. Mirza, R. Fuld, D. Park, J.R. Vasselli, Ghrelin is an orexigenic and metabolic signaling peptide in the arcuate and paraventricular nuclei, Am. J. Physiol. Regulat. Integr. Comp. Physiol. 289 (2) (2005) R353–R358. [10] D.P. Figlewicz, A.J. Sipols, Energy regulatory signals and food reward, Pharmacol. Biochem. Behav. 97 (1) (2010) 15–24. [11] Y. Furudono, C. Ando, C. Yamamoto, M. Kobashi, T. Yamamoto, Involvement of specific orexigenic neuropeptides in sweetener-induced overconsumption in rats, Behav. Brain Res. 175 (2) (2006) 241–248. [12] V.A. Gaysinskaya, O. Karatayev, J. Shuluk, S.F. Leibowitz, Hyperphagia induced by sucrose: Relation to circulating and csf glucose and corticosterone and orexigenic peptides in the arcuate nucleus, Pharmacol. Biochem. Behav. 97 (3) (2011) 521–530. [13] W. Gong, D. Neill, J.B. Justice Jr., 6-hydroxydopamine lesion of ventral pallidum blocks acquisition of place preference conditioning to cocaine, Brain Res. 754 (1–2) (1997) 103–112. [14] E. Jerlhag, E. Egecioglu, S.L. Dickson, A. Douhan, L. Svensson, J.A. Engel, Ghrelin administration into tegmental areas stimulates locomotor activity and increases extracellular concentration of dopamine in the nucleus accumbens, Addict. Biol. 12 (1) (2007) 6–16. [15] E. Jerlhag, E. Egecioglu, S.L. Dickson, L. Svensson, J.A. Engel, Alpha-conotoxin mii-sensitive nicotinic acetylcholine receptors are involved in mediating the ghrelin-induced locomotor stimulation and dopamine overflow in nucleus accumbens, Eur. Neuropsychopharmacol. 18 (7) (2008) 508–518. [16] E. Jerlhag, E. Egecioglu, S. Landgren, N. Salome, M. Heilig, D. Moechars, R. Datta, D. Perrissoud, S.L. Dickson, J.A. Engel, Requirement of central ghrelin signal-
[17]
[18]
[19]
[20]
[21]
[22]
[23] [24]
[25] [26]
[27]
[28]
[29]
[30]
[31] [32] [33]
73
ing for alcohol reward, Proc. Natl. Acad. Sci. U.S.A. 106 (27) (2009) 11318– 11323. D.C. Jewett, J. Cleary, A.S. Levine, D.W. Schaal, T. Thompson, Effects of neuropeptide Y, insulin, 2-deoxyglucose, and food deprivation on food-motivated behavior, Psychopharmacology (Berl) 120 (3) (1995) 267–271. Y. Kawahara, H. Kawahara, F. Kaneko, M. Yamada, Y. Nishi, E. Tanaka, A. Nishi, Peripherally administered ghrelin induces bimodal effects on the mesolimbic dopamine system depending on food-consumptive states, Neuroscience 161 (3) (2009) 855–864. D. Kim, S. Yoon, B. Choi, T. Kim, Y. Woo, W. Kim, H. Myrick, B. Peterson, Y. Bin Choi, Y. Kim, J. Jeong, Increased fasting plasma ghrelin levels during alcohol abstinence, Alcohol Alcohol. 40 (1) (2005) 76–79. M. Kojima, H. Hosoda, Y. Date, M. Nakazato, H. Matsuo, K. Kangawa, Ghrelin is a growth-hormone-releasing acylated peptide from stomach, Nature 402 (1999) 656–660. T. Kraus, A. Schanze, M. Groschl, K. Bayerlein, T. Hillemacher, U. Reulbach, J. Kornhuber, S. Bleich, Ghrelin levels are increased in alcoholism, Alcohol. Clin. Exp. Res. 29 (12) (2005) 2154–2157. A.M. Naleid, M.K. Grace, D.E. Cummings, A.S. Levine, Ghrelin induces feeding in the mesolimbic reward pathway between the ventral tegmental area and the nucleus accumbens, Peptides 26 (11) (2005) 2274–2279. N.S. Narayanan, D.J. Guarnieri, R.J. DiLeone, Metabolic hormones, dopamine circuits, and feeding, Front. Neuroendocrinol. 31 (1) (2010) 104–112. P.K Olszewski, D. Li, M.K. Grace, C.J. Billington, C.M. Kotz, A.S. Levine, Neural basis of orexigenic effects of ghrelin acting within lateral hypothalamus, Peptides 24 (4) (2003) 597–602. G. Paxinos, C. Watson, The Rat Brain in Stereotaxic Coordinates, Academic Press, New York, 2007. M. Perello, I. Sakata, S. Birnbaum, J.C. Chuang, S. Osborne-Lawrence, S.A. Rovinsky, J. Woloszyn, M. Yanagisawa, M. Lutter, J.M. Zigman, Ghrelin increases the rewarding value of high-fat diet in an orexin-dependent manner, Biol. Psychiatry 67 (9) (2010 May 1) 880–886. K.K. Pitchers, M.E. Balfour, M.N. Lehman, N.M. Richtand, L. Yu, L.M. Coolen, Neuroplasticity in the mesolimbic system induced by natural reward and subsequent reward abstinence, Biol. Psychiatry 67 (9) (2010) 872–879. R.K. Schwarting, J.P. Huston, Unilateral 6-hydroxydopamine lesions of mesostriatal dopamine neurons and their physiological sequelae, Prog. Neurobiol. 49 (3) (1996) 215–266. D.W. Self, Regulation of drug-taking and -seeking behaviors by neuroadaptations in the mesolimbic dopamine system, Neuropharmacology 47 (Suppl. 1) (2004) 242–255. N.D. Volkow, G.-J. Wang, L. Maynard, M. Jayne, J.S. Fowler, W. Zhu, J. Logan, S.J. Gatley, Y.-S. Ding, C. Wong, N. Pappas, Brain dopamine is associated with eating behaviors in humans, Int. J. Eat. Disord. 33 (2) (2003) 136–142. N.D. Volkow, R.A. Wise, How can drug addiction help us understand obesity? Nat. Neurosci. 8 (5) (2005) 555–560. R.A. Wise, Brain reward circuitry: insights from unsensed incentives, Neuron 36 (2) (2002) 229–240. R.A. Wise, Forebrain substrates of reward and motivation, J. Comp. Neurol. 493 (1) (2005) 115–121.
Contents lists available at ScienceDirect
Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet
6-Hydroxydopamine lesions of the ventral tegmental area suppress ghrelin’s ability to elicit food-reinforced behavior Zachary Y. Weinberg, Marjorie L. Nicholson, Paul J. Currie ∗ Department of Psychology, Reed College, Portland, OR 97202, United States
a r t i c l e
i n f o
Article history: Received 15 March 2011 Received in revised form 28 April 2011 Accepted 16 May 2011 Keywords: Appetitive motivation Dopamine Food intake Ghrelin Midbrain Reward Ventral tegmentum
a b s t r a c t While past research suggests that ghrelin stimulates appetite through an action on hypothalamic signaling, recent evidence indicates that the peptide acts via mesotelencephalic dopamine neurons to alter appetitive motivation. In the present study, rats were trained to operantly respond for food on a progressive ratio PR5 schedule until stable breakpoints were established. Ghrelin (30–300 pmol) was then injected directly into the ventral tegmental area (VTA) and the 300 pmol dose was observed to increase breakpoint. The dopaminergic neurotoxin 6-hydroxydopamine (6-OHDA, 6 g) was subsequently administered into the VTA resulting in a significant depletion of striatal dopamine. Stable breakpoints were then re-established. When ghrelin’s effects were reassessed, the peptide’s ability to alter operant responding for food was reliably reduced. Our findings demonstrate that ghrelin induces food-reinforced behavior in the mesotelencephalic reward pathway and that this effect is dependent on intact dopaminergic signaling. We conclude that the metabolic peptide ghrelin interacts with dopamine, within reward circuitry, to modulate appetitive behavior. © 2011 Elsevier Ireland Ltd. All rights reserved.
In the central nervous system ghrelin, a 28-amino acid peptide, is the endogenous ligand for the growth hormone secretagogue receptor (GHS-R1A) [20]. Ghrelin’s orexigenic action in the hypothalamus is well documented [9,24]. In addition, other work examining the effects of ghrelin has implicated the peptide in the induction of anxiety-like behaviors [4,5,8]. Recently, ghrelin’s role in the mesotelencephalic dopamine circuit has been investigated [23]. This pathway, implicated in reward and motivated behaviors, projects from the midbrain ventral tegmental area (VTA) to forebrain structures, including the nucleus accumbens in the ventral striatum. Ghrelin’s possible action in the mesotelencephalic dopamine system could provide evidence of a hypothalamic peptide mediating not only caloric intake but also regulating the rewarding aspects of food. There is strong empirical support for a dopamine-driven model of reward in the brain. When studying reinforced or motivated behavior, an experimental paradigm involving manipulation of accumbal dopamine release is commonly utilized [32,33] along with measures of drug- and food-seeking behaviors [27,29]. Hedonic feeding, eating behavior associated with food palatability, is now known to be mediated, at least in part, by mesotelencephalic dopamine neurons. Prior research has shown that decreases in brain concentrations of dopamine receptors in human subjects are correlated with decreased restraint in food intake [30]. Indeed, it
∗ Corresponding author. Tel.: +1 503 777 7267. E-mail address: [email protected] (P.J. Currie). 0304-3940/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2011.05.034
has been suggested that obesity could be modeled as a form of addiction and that in this model the mesotelencephalic dopamine circuit could be a primary therapeutic target in obesity treatment [31]. Additional investigations have examined the effects of eatingrelated peptides such as neuropeptide Y (NPY) and ghrelin on food-related reward. Microinjections of NPY and ghrelin into the VTA elicit a robust eating response [2,22]. Naleid et al. [22] argue these data demonstrate that ghrelin has a significant effect on reward-based food intake. Ghrelin’s reported action within the VTA raises the possibility that the peptide has a broader effect on reward and motivated behaviors beyond eating-related mechanisms. Jerlhag et al. [14] examined ghrelin’s ability to elicit locomotion and accumbal dopamine release. These effects are typically associated with the action of cocaine and other drugs of abuse within the accumbens and are considered biological indicators of reward. Jerlhag et al. [14] found that when injected into the VTA, ghrelin produced locomotion and increases in accumbal extracellular dopamine in a dose-dependent manner. There are at least two possible mechanisms through which ghrelin’s signaling in the VTA could alter accumbal dopamine levels. Ghrelin’s effects could be mediated by downstream signaling, presumably by the already established dopamine pathway. Alternatively, it remains possible that ghrelin’s effect on food-related reward could be mediated by non-dopaminergic circuits. Previous work has characterized the role of NPY, another orexigenic peptide, on food-reinforced behaviors. Like ghrelin, hypothalamic NPY has been reported to elicit alterations in eat-
Z.Y. Weinberg et al. / Neuroscience Letters 499 (2011) 70–73
ing and energy metabolism [6]. Jewett et al. [17] reported that NPY increased food-motivated behavior when injected into the lateral ventricle. Rats were trained to respond on a progressive ratio (PR) reinforcement schedule. The breakpoint, or point at which the rats stopped responding for food, was assessed. Breakpoint was used as a measure of how motivated the rat was to seek food. This work, combined with past research, suggests that, in addition to a role in mediating energy homeostasis, metabolic peptides might also be implicated in underlying neural circuits mediating reward such as food-reinforced behavior. Based on ghrelin’s feeding effect in the VTA and the VTA’s role in mediating reinforced behaviors, the current study investigated ghrelin’s ability to elicit an increase in breakpoint using an operant paradigm. Once we had established that VTA ghrelin microinjection increased breakpoint, we then examined whether this effect was dependent on intact dopamine signaling. Twelve adult male Sprague-Dawley rats (Harlan Laboratories, Kent, WA), weighing 350–400 g at time of surgery, were housed individually with ad libitum access to water and standard chow (LabDiet, PMI Nutrition, Brentwood, MO). The animal colony room was maintained at a temperature of 22 ± 2 ◦ C and on a 12-h light/dark cycle with lights on at 07:00 and off at 19:00 h. All experiments were approved by the Institutional Animal Care and Use Committee of Reed College. Prior to surgery, rats were trained to respond on a PR5 schedule in standard operant chambers. An initial response was rewarded with a banana pellet (Noyes, Lancaster, NH) and five additional responses were required for each successive reward [17]. In order to establish a consistent breakpoint, test sessions were conducted daily over two weeks, with each rat tested over a 2 h period at the onset of the active cycle. Animals were food deprived for the 6 h immediately prior to testing (in the light cycle). During training, rats were initially exposed to a single fixed interval (FI) session where food was dispensed every 30 s. Following one session of FI training, rats were then manually shaped to respond at fixed ratio 1 (FR1) where each response was rewarded with a pellet. Once rats responded reliably at FR1, the schedule was gradually increased to FR10 over the course of three sessions. Finally, rats were placed on a PR5 schedule for seven sessions, at which point the breakpoint of each rat was within a ±10% range of the rat’s last two sessions. Rats were then provided free access to food and water for 48 h after which surgery was conducted. The animals were surgically implanted with stainless steel guide cannulae (30 gauge, Plastics One, Roanoke, VA) aimed at the VTA. Cannulae were implanted 4 mm dorsal to the target with coordinates relative to bregma: −5.3 mm posterior, ±1.0 mm lateral and −4.1 mm ventral [25] with the incisor bar at −3.6 mm. Guide cannulae were counterbalanced with respect to implantation in the left and right cerebral hemispheres. After a 10 day post-surgery recovery period, rats were once again run in PR5 sessions until the breakpoint stabilized (seven session maximum). In order to examine the effects of ghrelin on breakpoint, ghrelin (30 or 300 pmol; Biopeptide Co. Inc., San Diego, CA) or vehicle was injected into the VTA 30 min prior to testing using a 32-gauge injector (Plastics One, Roanoke, VA) extending 4 mm beyond the permanent guide cannula. The total volume of each injection was 0.4 l. Doses were injected over a 1-min period and the injector was left in place for an additional minute to allow diffusion of the peptide. All animals were subjected to all three treatments (vehicle first followed by ghrelin in high-to-low dose order) and all test sessions occurred at least 48 h apart. In a second experiment, ghrelin’s ability to increase breakpoint was confirmed in a separate group of rats (n = 7). That is, rats were injected with vehicle and ghrelin (300 pmol) and changes in breakpoint were assessed. After this replication, rats were unilaterally injected with 6-hydroxydopamine (6-OHDA) in order to deplete central levels of dopamine. The unilateral 6-OHDA lesion
71
Fig. 1. Breakpoint as a function of treatment condition (n = 12). Ghrelin, at a dose of 300 pmol, significantly increased breakpoint compared to vehicle. * p < 0.05.
is a well-established and robust paradigm for assessing dopaminedependent behavior in rats [28]. Each rat was anesthetized with sodium pentobarbitol (50 mg/kg; Ovation Pharmaceuticals) and immediately injected with desipramine (25 mg/kg, i.p.; Sigma). Thirty min later each rat received a unilateral microinjection of 6-OHDA (6 g; Sigma) directly into the VTA. A 14-day recovery period followed. A stable breakpoint was re-established over 7 PR5 sessions. Finally, rats were injected with vehicle and then ghrelin (300 pmol) into the VTA and alterations in breakpoint were measured. All cannulae placements were confirmed via histological examination. Sections were examined by light microscopy and viewed relative to the stereotaxic atlas of Paxinos and Watson [25]. Coronal sections were taken through the entire striatum, including the ventral striatum [13,25]. Samples were homogenized in 0.2 M perchloric acid, sonicated, and centrifuged at 15,000 rpm over 15 min. Dopamine concentrations were measured using high performance liquid chromatography and electrochemical detection [13]. All rats reported here were found to have injector tracks extending into the VTA. In addition, striatal assays confirmed a 77% decrease in ipsilateral dopamine transmitter compared to a control group [t(12) = −13.81, p < 0.0001]. Our reductions in DA levels are consistent with previous work [28]. In experiment 1, data were analyzed using a one-way repeated measures Analysis of Variance (ANOVA) followed by post hoc Tukey tests. The criterion for significance was p < 0.05. As illustrated in Fig. 1, ANOVA confirmed that ghrelin injection into the VTA increased operant responding for food. Specifically, the dose of 300 pmol significantly increased breakpoint (F(2,22) = 7.64, p < 0.003). The 30 pmol dose, however, did not alter breakpoint values. In experiment 2, two-way ANOVA confirmed a significant lesion × ghrelin interaction (F(2,28) = 13.18, p < 0.02). As shown in Fig. 2, again the 300 pmol dose of ghrelin effectively increased breakpoint compared to control. Following treatment with 6OHDA, a significant decrease in breakpoint was observed when rats were injected with ghrelin, compared to vehicle. Breakpoint values for post-lesioned ghrelin-treated rats were also lower in comparison to pre-lesion ghrelin values. Moreover, there was no effect of lesion on the breakpoints of rats treated with vehicle compared to breakpoints in the pre-lesion vehicle condition. The results of experiment 1 are consistent with the findings of Jewett et al. [17] in showing that ghrelin, like NPY, increases foodreinforced responding. In fact, the ghrelin-dependent increase in breakpoint is comparable to that seen by Jewett et al. with NPY. Our work indicates that ghrelin’s effect on food-related reward can be localized to the VTA. Along with work by Jerlhag et al. [14] and
72
Z.Y. Weinberg et al. / Neuroscience Letters 499 (2011) 70–73
Fig. 2. Breakpoint as a function of treatment condition (n = 7). Prior to 6-OHDA lesion, ghrelin (300 pmol) increased breakpoint relative to vehicle. Following the lesion, ghrelin significantly decreased breakpoint. * p < 0.05 compared to vehicle of the same lesion condition. Vehicle breakpoints did not differ significantly between lesion conditions.
Abizaid et al. [2], these findings further implicate a role for ghrelin in the mediation of reinforced behavior via the mesotelencephalic dopamine pathway. Moreover, the findings of experiment 2 further indicate that ghrelin’s action within the VTA is dependent on dopaminergic signaling. When forebrain dopamine was depleted, operant responding for food reward was significantly decreased from vehicle after ghrelin treatment, though post-lesion vehicle did not differ significantly from pre-lesion vehicle. Prior research has implicated ghrelin as a peptide transmitter acting within the hypothalamus and exerting an effect on energy metabolism. When injected into the paraventricular or arcuate nuclei, ghrelin elicits a robust increase in food intake. Microinjections of ghrelin into these same nuclei produce shifts in energy substrate utilization [7,9]. These findings suggest, therefore, that ghrelin alters appetite and energy metabolism, including the promotion of carbohydrate oxidation and the preservation of fat stores, by an action of hypothalamic circuits mediating positive energy balance. Moreover, the findings of the present study further suggest that ghrelin’s ability to alter food intake is not restricted to hypothalamic metabolic circuitry. The findings of our second experiment raise several important questions regarding the mechanism of ghrelin’s action in the VTA. It is clear that ghrelin exerts a positive effect on foodrelated motivated behavior. It is also clear that ghrelin’s effect is dependent upon intact dopaminergic neurons in the VTA since 6OHDA treatment attenuated ghrelin-induced responding for food rewards. However, if ghrelin’s action in the VTA was on dopaminergic neurons exclusively, then depletion of these neurons should reduce ghrelin-induced breakpoints to levels comparable to vehicle. Instead, we observed that when rats were administered ghrelin following dopamine depletion, these animals not only no longer showed an increase in operant responding for food but in fact exhibited a significant reduction in breakpoint in comparison to both pre- and post-lesion vehicle treatment. This suggests that in the VTA ghrelin may also be acting on non-dopaminergic transmitter systems. Clearly, dopamine plays an important role in reward signaling. However, this role is supplemented by several other transmitter systems including the opioids, GABA, serotonin and acetylcholine. Jerlhag et al. [15] showed that ghrelin’s ability to elicit accumbal dopamine overflow and stimulate locomotion when injected into the VTA or laterodrosal tegmental area (LDTg) is dependent on cholinergic signaling in this region. Additionally, Kawahara et al. [18] demonstrated that bicuculline, a GABAA antagonist, injected into the VTA blocked ghrelin’s effect on accumbal dopamine
release. These findings, coupled with Abizaid and Horvath’s [1] data that show GHS-R1As expressed on GABAergic neurons in the VTA, present two possible non-dopaminergic mechanisms for the reduction in breakpoint seen after 6-OHDA treatment. Furthermore, our lab has previously shown that in the hypothalamus, serotonin modulates ghrelin’s effect on food intake [7] so it is not unreasonable to suggest that VTA-proximal serotonergic circuits may also help to modulate a ghrelinergic effect in the VTA. Over the past decade increasing evidence suggests a role for energy-regulatory peptides in food reward. While ghrelin is prominent among these signals, other relevant neurotransmitters include NPY, insulin, agouti-related peptide, leptin and orexin. Previous research has implicated these peptides in the regulation of food intake, food hedonics, food choice and palatability [10–12,23,26]. The current study shows that beyond increasing food intake, ghrelin increases motivation to obtain food. Other metabolic hormones may produce similar effects. Future studies should investigate the ability of other peptide transmitters to elicit motivation within the mesotelencephalic pathway. As described above, ghrelin induces increased locomotion and accumbal dopamine release, two findings commonly observed after the administration of drugs of abuse into the reward circuit [32]. This suggests that ghrelin could potentially impact states of addiction. A relationship between ghrelin and alcohol addiction has been demonstrated in several clinical populations. During the early stages of withdrawal from alcohol, alcohol-dependent individuals express increased levels of endogenous ghrelin [19,21]. In addition, ghrelin levels are correlated with subjective cravings for alcohol in recovering alcoholics [3]. Experiments have also shown that intact ghrelin signaling is necessary for animal subjects to develop alcohol dependence [16]. This evidence clearly supports a role for ghrelin in alcohol reward. Alcohol is a unique substance of abuse as it elicits an addiction while also containing caloric content. Consequently ghrelin, an orexigenic peptide that also impacts dopamine and reward, could provide evidence of an interaction between systems of feeding behavior and addiction. Finally, Volkow and Wise [31] discuss the overlap in biological mechanisms of feeding, obesity and drug addiction. Numerous studies have indicated that antagonism of telencephalic dopaminergic receptors attenuates free-feeding and operant responding for food reward as well the reinforcing effects of psychomotor stimulants. The mesotelencephalic dopamine projection from the VTA to the accumbens is most frequently implicated in reward although clearly other dopamine projections are involved. Given this overlap, it would seem appropriate to investigate ghrelin and other metabolic peptides in the reward circuit in paradigms commonly associated with assessing drugs of abuse such as conditioned place preference and self-administration. In conclusion, ghrelin induces food-reinforced behavior in the mesotelencephalic reward pathway, specifically when injected into the VTA. This effect is dependent on intact dopaminergic neurotransmission. Our results suggest that the metabolic peptide ghrelin interacts with dopamine within the reward system to modulate appetitive behavior. Acknowledgements This research was funded by a Reed College Science Research Fellowship. The authors thank Dr. Tim Hackenberg for programming and consulting on the operant portion of the experiment. References [1] A. Abizaid, T.L. Horvath, Unraveling neuronal circuitry regulating energy homeostasis: plasticity in feeding circuits, Drug Discov. Today: Dis. Models 2 (3) (2005) 191–196.
Z.Y. Weinberg et al. / Neuroscience Letters 499 (2011) 70–73 [2] A. Abizaid, Z.-W. Liu, Z.B. Andrews, M. Shanabrough, E. Borok, J.D. Elsworth, R.H. Roth, M.W. Sleeman, M.R. Picciotto, M.H. Tschop, X.-B. Gao, T.L. Horvath, Ghrelin modulates the activity and synaptic input organization of midbrain dopamine neurons while promoting appetite, J. Clin. Invest. 116 (12) (2006) 3229–3239. [3] G. Addolorato, E. Capristo, L. Leggio, A. Ferrulli, L. Abenavoli, N. Malandrino, S. Farnetti, M. Domenicali, C. D’Angelo, L. Vonghia, A. Mirijello, S. Cardone, G. Gasbarrini, Relationship between ghrelin levels, alcohol craving, and nutritional status in current alcoholic patients, Alcohol. Clin. Exp. Res. 30 (11) (2006) 1933–1937. [4] V.P Carlini, M.E. Monzon, M.M. Varas, A.B. Cragnolini, H.B. Schioth, T.N. Scimonelli, S.R. de Barioglio, Ghrelin increases anxiety-like behavior and memory retention in rats, Biochem. Biophys. Res. Commun. 299 (5) (2002) 739–743. [5] V.P. Carlini, M.M. Varas, A.B. Cragnolini, H.B. Schioth, T.N. Scimonelli, S.R. de Barioglio, Differential role of the hippocampus, amygdala, and dorsal raphe nucleus in regulating feeding, memory, and anxiety-like behavioral responses to ghrelin, Biochem. Biophys. Res. Commun. 313 (3) (2004) 635–641. [6] P Currie, Integration of hypothalamic feeding and metabolic signals: focus on neuropeptide Y, Appetite 41 (3) (2003) 335–337. [7] P.J. Currie, C.S. John, M.L. Nicholson, C.D. Chapman, K.E. Loera, Hypothalamic paraventricular 5-hydroxytryptamine inhibits the effects of ghrelin on eating and energy substrate utilization, Pharmacol. Biochem. Behav. 97 (1) (2010) 152–155. [8] P.J. Currie, R. Khelemsky, C.S. John, S. Higgs, Cb1 receptor antagonism alters the anxiogenic and feeding-stimulant effects of ghrelin, Appetite 52 (2009) 824. [9] P.J. Currie, A. Mirza, R. Fuld, D. Park, J.R. Vasselli, Ghrelin is an orexigenic and metabolic signaling peptide in the arcuate and paraventricular nuclei, Am. J. Physiol. Regulat. Integr. Comp. Physiol. 289 (2) (2005) R353–R358. [10] D.P. Figlewicz, A.J. Sipols, Energy regulatory signals and food reward, Pharmacol. Biochem. Behav. 97 (1) (2010) 15–24. [11] Y. Furudono, C. Ando, C. Yamamoto, M. Kobashi, T. Yamamoto, Involvement of specific orexigenic neuropeptides in sweetener-induced overconsumption in rats, Behav. Brain Res. 175 (2) (2006) 241–248. [12] V.A. Gaysinskaya, O. Karatayev, J. Shuluk, S.F. Leibowitz, Hyperphagia induced by sucrose: Relation to circulating and csf glucose and corticosterone and orexigenic peptides in the arcuate nucleus, Pharmacol. Biochem. Behav. 97 (3) (2011) 521–530. [13] W. Gong, D. Neill, J.B. Justice Jr., 6-hydroxydopamine lesion of ventral pallidum blocks acquisition of place preference conditioning to cocaine, Brain Res. 754 (1–2) (1997) 103–112. [14] E. Jerlhag, E. Egecioglu, S.L. Dickson, A. Douhan, L. Svensson, J.A. Engel, Ghrelin administration into tegmental areas stimulates locomotor activity and increases extracellular concentration of dopamine in the nucleus accumbens, Addict. Biol. 12 (1) (2007) 6–16. [15] E. Jerlhag, E. Egecioglu, S.L. Dickson, L. Svensson, J.A. Engel, Alpha-conotoxin mii-sensitive nicotinic acetylcholine receptors are involved in mediating the ghrelin-induced locomotor stimulation and dopamine overflow in nucleus accumbens, Eur. Neuropsychopharmacol. 18 (7) (2008) 508–518. [16] E. Jerlhag, E. Egecioglu, S. Landgren, N. Salome, M. Heilig, D. Moechars, R. Datta, D. Perrissoud, S.L. Dickson, J.A. Engel, Requirement of central ghrelin signal-
[17]
[18]
[19]
[20]
[21]
[22]
[23] [24]
[25] [26]
[27]
[28]
[29]
[30]
[31] [32] [33]
73
ing for alcohol reward, Proc. Natl. Acad. Sci. U.S.A. 106 (27) (2009) 11318– 11323. D.C. Jewett, J. Cleary, A.S. Levine, D.W. Schaal, T. Thompson, Effects of neuropeptide Y, insulin, 2-deoxyglucose, and food deprivation on food-motivated behavior, Psychopharmacology (Berl) 120 (3) (1995) 267–271. Y. Kawahara, H. Kawahara, F. Kaneko, M. Yamada, Y. Nishi, E. Tanaka, A. Nishi, Peripherally administered ghrelin induces bimodal effects on the mesolimbic dopamine system depending on food-consumptive states, Neuroscience 161 (3) (2009) 855–864. D. Kim, S. Yoon, B. Choi, T. Kim, Y. Woo, W. Kim, H. Myrick, B. Peterson, Y. Bin Choi, Y. Kim, J. Jeong, Increased fasting plasma ghrelin levels during alcohol abstinence, Alcohol Alcohol. 40 (1) (2005) 76–79. M. Kojima, H. Hosoda, Y. Date, M. Nakazato, H. Matsuo, K. Kangawa, Ghrelin is a growth-hormone-releasing acylated peptide from stomach, Nature 402 (1999) 656–660. T. Kraus, A. Schanze, M. Groschl, K. Bayerlein, T. Hillemacher, U. Reulbach, J. Kornhuber, S. Bleich, Ghrelin levels are increased in alcoholism, Alcohol. Clin. Exp. Res. 29 (12) (2005) 2154–2157. A.M. Naleid, M.K. Grace, D.E. Cummings, A.S. Levine, Ghrelin induces feeding in the mesolimbic reward pathway between the ventral tegmental area and the nucleus accumbens, Peptides 26 (11) (2005) 2274–2279. N.S. Narayanan, D.J. Guarnieri, R.J. DiLeone, Metabolic hormones, dopamine circuits, and feeding, Front. Neuroendocrinol. 31 (1) (2010) 104–112. P.K Olszewski, D. Li, M.K. Grace, C.J. Billington, C.M. Kotz, A.S. Levine, Neural basis of orexigenic effects of ghrelin acting within lateral hypothalamus, Peptides 24 (4) (2003) 597–602. G. Paxinos, C. Watson, The Rat Brain in Stereotaxic Coordinates, Academic Press, New York, 2007. M. Perello, I. Sakata, S. Birnbaum, J.C. Chuang, S. Osborne-Lawrence, S.A. Rovinsky, J. Woloszyn, M. Yanagisawa, M. Lutter, J.M. Zigman, Ghrelin increases the rewarding value of high-fat diet in an orexin-dependent manner, Biol. Psychiatry 67 (9) (2010 May 1) 880–886. K.K. Pitchers, M.E. Balfour, M.N. Lehman, N.M. Richtand, L. Yu, L.M. Coolen, Neuroplasticity in the mesolimbic system induced by natural reward and subsequent reward abstinence, Biol. Psychiatry 67 (9) (2010) 872–879. R.K. Schwarting, J.P. Huston, Unilateral 6-hydroxydopamine lesions of mesostriatal dopamine neurons and their physiological sequelae, Prog. Neurobiol. 49 (3) (1996) 215–266. D.W. Self, Regulation of drug-taking and -seeking behaviors by neuroadaptations in the mesolimbic dopamine system, Neuropharmacology 47 (Suppl. 1) (2004) 242–255. N.D. Volkow, G.-J. Wang, L. Maynard, M. Jayne, J.S. Fowler, W. Zhu, J. Logan, S.J. Gatley, Y.-S. Ding, C. Wong, N. Pappas, Brain dopamine is associated with eating behaviors in humans, Int. J. Eat. Disord. 33 (2) (2003) 136–142. N.D. Volkow, R.A. Wise, How can drug addiction help us understand obesity? Nat. Neurosci. 8 (5) (2005) 555–560. R.A. Wise, Brain reward circuitry: insights from unsensed incentives, Neuron 36 (2) (2002) 229–240. R.A. Wise, Forebrain substrates of reward and motivation, J. Comp. Neurol. 493 (1) (2005) 115–121.