Ghrelin enhances cue-induced bar pressing for high fat food

Highlights

• Ghrelin into the VTA increases the number of operant responses to food-predicting cues in a reinstatement paradigm.

• We confirm that cues are more powerful in reinstating operant responses than food pellets themselves or an overnight fast.

• Intra-VTA ghrelin in satiated rats produces responses to cues that are about 60–70% of those produced when they were hungry.

Abstract

Ghrelin is an orexigenic hormone produced by the stomach that acts on growth hormone secretagogue receptors (GHSRs) both peripherally and centrally. The presence of GHSRs in the ventral tegmental area (VTA) suggests that ghrelin signaling at this level may increase the incentive value of palatable foods as well as other natural and artificial rewards. The present investigation sought to determine if ghrelin plays a role in relapse to such foods following a period of abstinence. To achieve this, thirty-six male Long Evans rats were trained to press a lever to obtain a high fat chocolate food reward on a fixed ratio schedule of 1. Following an extinction period during which lever presses were not reinforced, rats were implanted with a cannula connected to a minipump that continuously delivered ghrelin, a GHSR antagonist ([d-Lys-3]-GHRP-6), or saline in the VTA for 14 days. One week later, food reward-associated cues, food reward priming, and an overnight fast were used to induce reinstatement of the lever pressing response. Our results indicate that intra-VTA ghrelin enhances cue-induced reinstatement of responses for palatable food pellets. To the extent that the reinstatement paradigm is considered a valid model of relapse in humans, this suggests that ghrelin signaling facilitates relapse to preferred foods in response to food cues through GHSR signaling in the VTA.

Introduction

Relapse to preferred foods and unhealthy eating habits is a primary concern for people undergoing weight reduction diets (Elfhag and Rossner, 2005, Kramer et al., 1989). This phenomenon is similar to drug relapse in that it can be triggered by the same environmental stimuli: re-exposure to the substance and/or cues associated with it, and exposure to stressful situations (Elfhag and Rossner, 2005, Grilo et al., 1989, Kayman et al., 1990, Gorin et al., 2004). Interestingly, some have demonstrated that cues associated with reinforcing stimuli can elicit feeding responses in satiated animals that similar to those seen when these test animals are hungry (Weingarten, 1983). Experimentally, relapse to food seeking is most frequently studied using an operant conditioning model that is also used in drug addiction research. The paradigm is characterized by 3 successive phases: self-administration training, extinction, and reinstatement (Epstein et al., 2006, Nair et al., 2009a). Over the last decade, the use of this model has facilitated the identification of a number of peptides and neurotransmitters that act centrally to modulate the reinstatement of food seeking, many of which also impact reinstatement of drug seeking (reviewed in Nair et al., 2009a). These include, but are not limited to, orexigenic peptides such as orexin and melanin-concentrating hormone (Cippitelli et al., 2010, Richards et al., 2008, James et al., 2011, Nair et al., 2008, Nair et al., 2009b, Boutrel et al., 2005, Wang et al., 2009). In this context, the role of ghrelin, the only circulating orexigenic hormone currently known, is relatively under-studied.

Ghrelin is a 28 amino acid-long peptide that is produced predominantly in the oxyntic glands of the stomach in times of negative energy balance (reviewed in Castaneda et al., 2010). It is passively transported across the blood brain barrier and acts on growth hormone secretagogue receptors (GHSRs) both inside and outside of the central nervous system (Banks et al., 2002, Diano et al., 2006, Ghigo et al., 2005). Peripheral, intra-cranial and hypothalamic ghrelin administration produces a robust dose-dependent feeding response that is not seen in animals lacking a functional GHSR gene (Lawrence et al., 2002, Wren et al., 2001, Tschop et al., 2000). Ghrelin administration also results in a number of outcomes that are typically observed in the presence of both natural (ex: food) and artificial (ex: drug) rewards, such as increased dopamine (DA) levels in the nucleus accumbens (NAc), increased locomotion and development of a conditioned place preference, suggesting a general role of ghrelin signaling in reward processing (Abizaid et al., 2006, Abizaid, 2009, Jerlhag et al., 2006a/Jerlhag et al., 2006b, Jerlhag, 2008). Likewise, disruption of ghrelin or ghrelin signaling via genetic or pharmacological manipulations can alter the behavioral and physiological response to reinforcers, including operant responding for palatable foods and drugs of abuse (Jerlhag and Engel, 2011, Landgren et al., 2011a, Skibicka et al., 2011a, Skibicka et al., 2011b, King et al., 2011, Landgren et al., 2011b, Clifford et al., 2011, Wellman et al., 2011, Perello et al., 2010, Jerlhag et al., 2010, Jerlhag et al., 2009, Egecioglu et al., 2010, Abizaid et al., 2011). Ghrelin's reward-related effects likely involve the dopaminergic mesolimbic system. Over 60% of DA cells within the ventral tegmental area (VTA) express the GHSR, and ghrelin binding in this area causes synaptic re-organization of local DA cells, ultimately increasing the ratio of excitatory (glutaminergic) to inhibitory (gabaergic) afferents to these cells (Abizaid et al., 2006). The resulting reduction in DA neurons' firing threshold and increased accumbal DA turnover might underlie ghrelin's orexigenic effects at this level (Abizaid et al., 2006). Ghrelin's action in the mesolimbic system may be specifically related to adding incentive value to food stimuli (Skibicka et al., 2011a, Egecioglu et al., 2010).

A role for ghrelin in the reinstatement of food seeking is suggested by the fact that ghrelin activates the HPA stress response system, and that stress is intimately linked with food seeking relapse in dieting individuals and rats tested in the reinstatement model (Ghitza et al., 2006, Asakawa et al., 2001, Greeno and Wing, 1994, Grilo et al., 1989, Tassone et al., 2003). Indeed, mice with targeted deletion of the GHSR gene or mice treated with a ghrelin receptor antagonists show lower reinstatement of operant responses to cues previously associated with food (Walker et al., 2012). In addition, caloric restriction, which increases plasma ghrelin levels, facilitates reinstatement of both food and drug seeking (Nair et al., 2009a). Elevated ghrelin levels are also associated with alcohol cravings in people who try to abstain from drinking, and such cravings often precede relapse episodes (Koopmann et al., 2012, Koob and Volkow, 2010). Finally, endogenous ghrelin levels are positively correlated to the reinstatement of cocaine seeking in rats (Tessari et al., 2007).

In the current study, we sought to determine if ghrelin administration within the VTA could restore previously extinguished bar pressing for pellets, a rodent model of reinstatement or relapse in rats. This was examined in response to stimuli that have been associated with the reinstatement of reward seeking behaviors in the past. Specifically, we evaluated the reinstatement of these responses after exposure to a cue previously associated with the reward, the reward itself (i.e. re-exposure to the high fat pellet), or an overnight fast. We chose chocolate-flavored high-fat food pellets as the reward because preferred foods are often sweet and rich in fat and ghrelin preferentially increases intake of sweet and high-fat foods (Shimbara et al., 2004, Disse et al., 2010). The choice of the administration route was based on the presence of GHSRs within this node of the reward system and evidence suggesting the VTA plays a role in the reinstatement of both cocaine and heroin seeking (Abizaid et al., 2006, Guan et al., 1997, Zigman et al., 2006, Wang et al., 2009, Wang et al., 2007, Bossert et al., 2004, Stewart, 1984). In order to limit the effects of endogenous ghrelin levels, which naturally rise in times of negative energy balance at GHSRs located outside of the mesolimbic system, reinstatement was tested in animals that were sated (i.e. not chronically food restricted) and therefore not in negative energy state.