Palatable food access impacts expression of amylin receptor components in the mesocorticolimbic system

Abstract

Amylin is a pancreatic and brain-derived peptide that acts within the central nervous system (CNS) to promote negative energy balance. However, our understanding of the CNS sites of action for amylin remains incomplete. Here, we investigate the effect of amylin receptor (AmyR) activation in the nucleus accumbens core (NAcC) on the intake of bland and palatable foods. Intra-NAcC injection of the AmyR agonist salmon calcitonin (sCT) or amylin itself in male chow-fed rats had no effect on food intake, meal size, or meal number. However, in chow-fed rats with access to fat solution, although fat intake was not affected by intra-NAcC AmyR activation, subsequent chow intake was suppressed. Given that mesolimbic AmyR activation suppresses energy intake in rats with access to fat solution, we tested whether fat access changes AmyR expression in key mesocorticolimbic nuclei. Whereas fat exposure did not affect NAcC AmyR expression, in the accumbens shell, expression of receptor activity modifying protein (RAMP)-3 was significantly reduced in fat-consuming rats. We show that all components of AmyRs are expressed in medial prefrontal cortex (mPFC) and central nucleus of the amygdala (CeA); fat access significantly reduced expression of calcitonin receptor (CTR)-A in the CeA and RAMP-2 in the mPFC. Taken together, these results indicate that intra-NAcC AmyR activation can suppress energy intake, and furthermore, suggest that AmyR signaling in a broader range of mesocorticolimbic sites may have a role in mediating the effects of amylin on food intake and body weight.

Introduction

Obesity is one of the most common ailments in modern society. Although its prevalence is increasing rapidly in many parts of the world (Meldrum, Morris, & Gambone, 2017; Ng et al., 2014; Swinburn et al., 2019; Wolfenden, Ezzati, Larijani, & Dietz, 2019), treatment options are insufficient. This is, at least in part, due to the fact that obesity is a disease of a very complex nature involving multiple physiological systems and caused by a combination of factors that contribute to its etiology, including genetic, psychological, and environmental factors (Hasselbalch, 2010; Pereira-Lancha, Campos-Ferraz, & Lancha, 2012; Roberto et al., 2015; Swinburn et al., 2019). The interaction of these factors can promote overconsumption of food, especially highly palatable foods that are typically energy-dense (Roberto et al., 2015; Swinburn et al., 2019). Understanding the neurohormonal mechanisms influencing food intake and food reward is critical to the identification of novel pharmacotherapies to treat obesity.

The mesocorticolimbic system is central to the regulation of motivated behavior, including palatable food intake. It consists of several nuclei interconnected through dopaminergic neurons, including the ventral tegmental area (VTA), the nucleus accumbens (NAc), the prefrontal cortex (PFC), and the amygdala, among others (Cai, Haubensak, Anthony, & Anderson, 2014; Douglass, Kucukdereli, Ponserre, Markovic, Gründemann, et al., 2017; Kim, Zhang, Muralidhar, LeBlanc, & Tonegawa, 2017; Land et al., 2014; Petrovich, Ross, Holland, & Gallagher, 2007; Petrovich, Ross, Mody, Holland, & Gallagher, 2009). The NAc, comprising the core (NAcC) and shell (NAcSh) subregions, has been examined for its role in feeding and other motivated behaviors (Baldo & Kelley, 2007; Kelley, 1999; Pierce & Wolf, 2013; Schall, Wright, & Dong, 2020). In general, although the NAcC and NAcSh share neuroanatomical connections (van Dongen et al., 2005), these subnuclei have distinct roles in energy balance control. For example, there are differences in the food intake responses driven by glutamate or GABA between the subnuclei (Floresco, McLaughlin, & Haluk, 2008; Kelley, 1999; Kelley & Swanson, 1997; Maldonado-Irizarry & Kelley, 1994; Maldonado-Irizarry, Swanson, & Kelley, 1995), and although both areas are involved in motivated feeding, the NAcSh appears to be responsive to novelty of food, whereas the NAcC plays a role in learning and conditioning related to food (Bassareo, De Luca, & Di Chiara, 2002; Bassareo & Di Chiara, 1999). The PFC encodes information on motivational value (Bassareo et al., 2002) and its effects on feeding include roles in cue-induced feeding (Petrovich, 2013; Petrovich et al., 2007), inhibitory control over food intake (Selleck & Baldo, 2017; Sinclair, Klump, & Sisk, 2019), and impulsivity in food intake (Anastasio et al., 2019). Roles of the CeA in food intake and energy balance control include encoding information on the reinforcing value and positive or negative valence of food and other stimuli (Douglass, Kucukdereli, Ponserre, Markovic, Grundemann, et al., 2017; Hardaway et al., 2019), as well as changes in feeding related to stress and aversive stimuli (Herzog, 2020; Petrovich et al., 2009). It is important to note that these areas of the brain share neuroanatomical connections, directly and/or indirectly (Del Arco & Mora, 2009; Keistler et al., 2017; McGlinchey, James, Mahler, Pantazis, & Aston-Jones, 2016; Roura-Martinez et al., 2020; van Dongen et al., 2005), and many neuropeptides involved in the regulation of food intake have receptors in mesocorticolimbic sites where they alter dopaminergic neurotransmission and therefore influence reward-motivated feeding (Howell et al., 2019; Mebel, Wong, Dong, & Borgland, 2012; Risco & Mediavilla, 2018; Schele, Bake, Rabasa, & Dickson, 2016).

Amylin is a peptide hormone secreted by pancreatic β-cells along with insulin that helps to control blood glucose levels (Pullman, Darsow, & Frias, 2006; Reidelberger, Kelsey, & Heimann, 2002; A. Young, 2005; A. A. Young, Gedulin, Vine, Percy, & Rink, 1995). Amylin is also produced in the brain (Dobolyi, 2009; Li, Kelly, Heiman, Greengard, & Friedman, 2015; Szabo, Cservenak, & Dobolyi, 2012) and binds in numerous sites throughout the neuraxis, including mesocorticolimbic areas such as the VTA and NAc (Beaumont, Kenney, Young, & Rink, 1993; Paxinos et al., 2004; Sexton, Paxinos, Kenney, Wookey, & Beaumont, 1994). In addition to well-established hindbrain sites of action for amylin including brainstem nuclei such as the area postrema and the nucleus of the solitary tract (Lutz, Mollet, Rushing, Riediger, & Scharrer, 2001; Lutz et al., 1998), amylin acts directly in reward-processing mesocorticolimbic nuclei such as the VTA (Mietlicki-Baase et al., 2017; Mietlicki-Baase et al., 2015; Mietlicki-Baase et al., 2013) and the NAc (Baisley & Baldo, 2014; Baldo & Kelley, 2001) to produce weight loss and reduce palatable food intake. These findings strongly indicate the relevance of amylin-mediated signaling in the mesocorticolimbic system for its effect on energy balance. They also provide evidence in favor of utilizing the amylin system as a promising potential therapeutic strategy for the treatment of obesity (Boyle, Lutz, & Le Foll, 2018; Hay, Chen, Lutz, Parkes, & Roth, 2015; Mietlicki-Baase & Hayes, 2014). Yet, for that end goal to be achieved, a more thorough understanding of the central sites at which amylin acts to suppress feeding is needed.

The VTA and the NAc express all components of amylin receptors (AmyR) (Mietlicki-Baase et al., 2013), and direct pharmacological activation of VTA AmyRs suppresses both chow intake (Mietlicki-Baase et al., 2013) and palatable food intake (Mietlicki-Baase et al., 2017; Mietlicki-Baase et al., 2015). Interestingly, intra-VTA administration of an AmyR agonist may have a more potent anorectic effect on fat intake versus sucrose intake (Mietlicki-Baase et al., 2017). Amylin administration into the NAc, however, is an understudied area of research. It has been shown that intra-NAcSh amylin reduces feeding and drinking (Baldo & Kelley, 2001), but little is known about the effects of direct NAc core (NAcC) AmyR activation on energy balance control. One study showed that intra-NAcC amylin had no significant effect on feeding; however, the experiment was conducted in water-deprived rats (Baldo & Kelley, 2001). Given the tight association between food and water intake (Kissileff, 1969), the water-deprived state of the animals could have influenced the results. It is important to evaluate whether intra-NAcC AmyR activation in ad libitum-fed rats impacts energy intake. Furthermore, as AmyR activation in other reward-related sites such as the VTA potently reduces palatable food intake (Mietlicki-Baase et al., 2017; Mietlicki-Baase et al., 2015), this highlights the importance of evaluating the effects of AmyR signaling in the NAcC on both bland (e.g., chow) as well as palatable food intake. We hypothesized that AmyR signaling in the NAcC would reduce chow intake, but would have more potent suppressive effects on palatable food intake.

Although previously published studies collectively suggest a role for amylin in controlling food intake via mesocorticolimbic nuclei, the involvement of other mesocorticolimbic sites besides the VTA and NAc, such as the prefrontal cortex (PFC) and the central nucleus of the amygdala (CeA), remains largely unexplored. This is surprising because amylin binds in these sites (Beaumont et al., 1993; Paxinos et al., 2004; Sexton et al., 1994) and these nuclei are known to have roles in controlling ingestive behavior (Cai et al., 2014; Douglass, Kucukdereli, Ponserre, Markovic, Grundemann, et al., 2017; Kim et al., 2017; Land et al., 2014; Petrovich et al., 2007). The AmyR is a heterodimer composed of a calcitonin G-protein coupled receptor (CTR) and receptor activity modifying protein (RAMP). The CTR exists in two isoforms (CTR-A and CTR-B), while RAMPs exist in three (RAMP-1, RAMP-2, and RAMP-3). CTRs and RAMPs can dimerize in different combinations to form AmyRs of varying degrees of sensitivity and selectivity to amylin (Christopoulos et al., 1999; Hay, Christopoulos, Christopoulos, & Sexton, 2004; Muff, Buhlmann, Fischer, & Born, 1999). One paper showed that some components of the AmyR are expressed in PFC and CeA (Kalafateli et al., 2019), but did not fully characterize expression of all potential AmyR components within these nuclei. Thus, the particular components of the AmyR that are expressed in these sites are not completely resolved. Furthermore, it is unclear whether palatable food access / intake may alter expression of AmyRs in the brain. Given that mesolimbic AmyR activation has particularly potent effects on fat intake (Mietlicki-Baase et al., 2017), we hypothesized that fat exposure may alter AmyR expression in the mesocorticolimbic system. Therefore, we characterized expression of AmyR components and examined the effect of fat exposure on their expression in several mesocorticolimbic sites. In addition, since AmyR activation in the VTA has been found to alter dopaminergic neurotransmission in the NAcC to suppress food intake (Mietlicki-Baase et al., 2015), we examined whether fat access changed expression of dopamine receptors within the NAc. Collectively, the studies here support a role for NAcC AmyR activation in energy balance control and also show that fat access suppresses expression of particular AmyR components in several mesocorticolimbic nuclei.

Methods

Ethical Approval

All procedures were carried out according to the institutional guidelines for animal experimentation and were approved by the Institutional Animal Care and Use Committee of the University at Buffalo, State University of New York (IACUC Project #ENS01086Y).

Animals

Adult male Sprague Dawley rats were purchased (Charles River) and housed individually in hanging-wire cages. Rats were maintained on a 12-h light/dark cycle in a temperature- and humidity-controlled environment. Except where stated, rats had ad libitum access to food (2018 Teklad global 18% protein rodent diet; Envigo Teklad, Madison, WI, USA) and tap water. Mean body weight of the rats was 384.7±52.4g at beginning of experiments (start of training in burette studies and start of testing in feedometer study).

Drugs

Salmon calcitonin (sCT; Bachem) and amylin (Bachem) were dissolved in artificial cerebrospinal fluid (aCSF; Harvard Apparatus) for intraparenchymal injections. Doses used were based on the literature (Mietlicki-Baase et al., 2013).

Surgery

For behavioral pharmacology studies, rats were surgically implanted with a chronic indwelling cannula (26ga; Plastics One) aimed at the nucleus accumbens core (NAcC). After acclimation to the animal facility, rats were anesthetized via an intramuscular injection of a cocktail containing ketamine, acepromazine, and xylazine (KAX). KAX cocktail consisted of either 40 mg/kg ketamine, 5 mg/kg xylazine, and 1 mg/kg acepromazine; or, 90 mg/kg ketamine, 2.7 mg/kg xylazine, and 0.64 mg/kg acepromazine. Rats were placed in a stereotaxic apparatus and bupivacaine (0.5%) was applied topically to the scalp. A unilateral guide cannula was implanted aimed at the NAcC (guide cannula coordinates: 2.0mm anterior to bregma, 1.7mm lateral to midline, 4.8mm ventral to skull; internal cannula extended 2.0mm beyond guide cannula). Preoperative (5 mg/kg meloxicam or carprofen, SC; 0.5% bupivacaine, topical) and postoperative analgesic (5 mg/kg meloxicam or carprofen, SC, for 2 days post-surgery) was provided. Animals were allowed at least one week to recover from surgery before behavioral testing began.

Nucleus accumbens – behavioral pharmacology

Experiments were conducted using a counterbalanced within-subjects design, with treatments separated by at least 48h. Separate rats were used for each behavioral experiment. After each behavioral experiment was complete, rats were euthanized via CO₂ and pontamine sky blue ink (100nl) was injected through the NAcC cannula. Brains were removed from the crania and drop-fixed in 10% formalin. For each brain, coronal sections containing the NAcC (40μm) were cut using a cryostat (Minotome Plus; Triangle Biomedical Sciences, Durham, NC, USA) and injection placement was verified by examining ink placement under a microscope. Only rats with correct cannula placements were included in analyses. A representative image of cannula placement is shown in Figure 1C.

Figure 1. Activation of AmyRs in the NAcC reduces chow intake but has no significant effects on body weight or meal patterns.

Rats were given intra-NAcC administration of amylin (0.2μg) or the AmyR agonist sCT (0.04μg). No significant effects of intra-NAcC AmyR activation were observed for cumulative chow intake (A) or body weight gain (B). A representative image taken under 2x objective showing injection placement in a 40μm coronal section containing NAcC is shown in panel (C); aca, anterior part of the anterior commissure. When meal patterns were analyzed, no significant differences were detected in cumulative meal size (D) or cumulative meal number (E). Data are shown as mean + SD. Key in (A) applies to all bar graph panels.

Chow intake:

To evaluate the effect of intra-NAcC AmyR activation on chow intake, rats (n=13) were housed in a commercially available automated feedometer system (BioDAQ; Research Diets, Inc.) to measure feeding and meal patterns after pharmacological activation of NAcC AmyRs. On test days, rats received a unilateral intra-NAcC injection of sCT (0.04μg), amylin (0.2μg), or vehicle (100nl aCSF) beginning approximately 30 min before lights off. Doses of sCT and amylin were selected from our previous work, and this dose of sCT is subthreshold for prolonged effects on feeding when delivered directly into the cerebroventricular system but reduces feeding when administered directly into the parenchyma (Mietlicki-Baase et al., 2013). Food intake was continuously monitored in the feedometer system for the 24h post-injection. Crumb spillage was collected and accounted for in food intake measurements. Body weight change over the 24h post-injection was also measured. For each treatment, the percentage of food intake occurring as meal-related intake was calculated for each rat. Three rats had meal-related intake values exceeding 100%, suggesting technical errors in measurement, and were excluded from further analysis. Statistical outliers were defined as rats with 24h chow intake or body weight change values that fell as outliers using Tukey’s method; specifically, rats with 24h chow intake or body weight values 1.5 times the interquartile range above the third quartile or below the first quartile were removed from the data set and excluded from statistical analyses. One rat met these criteria and was excluded from statistical analyses. This resulted in a final n=9 for this experiment.

Palatable macronutrient intake:

Rats were trained to consume a macronutrient solution [either 10% Intralipid^® (Baxter Healthcare, Deerfield, IL, USA) or 25% sucrose] in a 60-minute access period. We have used these concentrations of fat and sucrose solutions in our previous work on the effects of mesolimbic AmyR signaling (Mietlicki-Baase et al., 2017). Other rodent preference studies also suggest that fat and sucrose solutions at similar concentrations are palatable (Nissenbaum & Sclafani, 1987; Sakamoto et al., 2015). Food and water were removed from the cage for 1h, prior to the 60-minute macronutrient access period beginning 30 minutes after lights on. No other food or water were available during the period in which the macronutrient solution was available. Rats were trained to drink their assigned macronutrient solution for 5 days prior to testing. Rats consistently consuming ≥5mL of the macronutrient solution for 3 of the training days were included (n=6 for fat, n=10 for sucrose). On test days, rats received a unilateral intra-NAcC injection of either sCT (0.04μg), amylin (0.2μg), or vehicle (100nl aCSF) beginning approximately 30 min before macronutrient access. Macronutrient solution intake was measured to the nearest 0.1mL every 10 minutes for the 60-minute access period. After the 1h access period, the burette containing the macronutrient solution was removed, and chow and water were returned to the cage. The effects of NAcC AmyR activation on subsequent chow intake were monitored by measuring chow intake for ~21.5h following the 1h burette access period. Crumb spillage was collected and was accounted for in chow intake measurements. Body weight was also measured at 0 and 24h. Due to the within-subjects design of the experiments, at least one training day (e.g., burette access but no injection) always preceded the next test day. Statistical outliers for these studies were defined as rats with 60min fluid intake, subsequent chow intake, or body weight change values that fell as outliers using Tukey’s method as described above. No rats from the fat intake experiment met these criteria and all rats were included in the final statistical analyses. For the sucrose experiment, two rats had outlying values using this method and were excluded from subsequent statistical analyses, resulting in a final n=8 for this study.

Amylin receptor and dopamine receptor expression studies

Sixteen rats were randomly divided into 2 groups: a control group (n=7) and a fat intake group (n=9). Rats were deprived of food and water approximately 1h before the beginning of the light phase. Thirty minutes into the light phase, they were provided with access to one solution in a graduated glass burette for 60 minutes: either tap water or a 10% Intralipid^® fat solution (1 kcal/mL). After this period, each burette was removed and food and water were returned to the cages. This procedure was repeated at the same time for ~1 month (total 21 training days with no more than 2 consecutive days with no fat access). Body weights of the rats were recorded on each training day and just before sacrifice. In addition, total fluid intake during the 60-minute burette access period over the last 3 training days was averaged and compared between the 2 groups.

Upon completion of the training period, rats were deeply anesthetized with KAX cocktail and decapitated. The brains were quickly removed, flash-frozen in isopentane, and stored at −80°C until sectioning. Brains were later sliced in a caudal to rostral direction with a cryostat (Minotome Plus) at −20°C into 40μm-coronal sections until the caudal end of sites of interest were reached. A rat stereotaxic atlas (Paxinos & Watson, 2014) was used to identify anatomical locations of sites of interest, namely the mPFC, NAcC, NAcSh, and CeA. Bilateral tissue punches (~1mm³ per hemisphere) enriched for each site of interest were collected and stored at −80°C until further processing.

Total RNA was extracted from each tissue sample using TRIzol reagent (Invitrogen, Grand Island, NY, USA) and the RNeasy Mini Kit (Qiagen, Valencia, CA, USA) according to the manufacturer’s instructions. Commercially available cDNA synthesis kits were used to synthesize cDNA from 390 ng of total RNA. Advantage RT-for-PCR Kit (Clontech, Mountain View, CA, USA) was used for RNA samples from the CeA and mPFC and iScript Advanced cDNA Synthesis Kit (BioRad, Hercules, CA, USA) was used for RNA samples from the NAcC and NAcSh according to the respective manufacturers’ instructions. Relative mRNA levels of each component of the AmyR complex was measured using quantitative real-time PCR (qRT-PCR): calcitonin receptor (CTR)-A and CTR-B, as well as all three subtypes of receptor activity modifying protein (RAMP-1, RAMP-2, and RAMP-3), with rat GAPDH as an internal control for each sample. We also analyzed expression of the dopamine D₁ receptor (D₁R) and D₂ receptor (D₂R) in the NAcC and NAcSh. PCR reactions for genes of interest were completed with the following TaqMan gene expression kits (Thermo Fisher Scientific, Waltham, MA, USA; CTR-A: Rn01526770_m1; CTR-B: Rn01526768_m1; RAMP-1: Rn01427056_m1, RAMP-2: Rn00824652_g1, RAMP-3: Rn00571815_m1; D₁R: Rn03062203_s1, D₂R: Rn00561126_m1; GAPDH: Rn01775763_g1) and PCR reagents (Applied Biosystems, Foster City, CA, USA). Analysis of samples was conducted using a CFX-96 Connect multiplex real time PCR thermocycler (Bio-Rad, Hercules, CA, USA) and the comparative threshold cycle method was used for relative quantification of mRNA expression. From these results, CTR expression data were normalized to CTR-A levels of control rats, and RAMP expression data were normalized to RAMP-1 levels of control rats. D₁R expression data were normalized to D₁R levels of control rats, and similarly, D₂ receptor expression data were normalized to D₂R levels of control rats. Outliers were defined using Tukey’s method, as described above, and were removed from the data set and excluded from statistical analyses. This resulted in a final n=5-9 per group.

Data analyses

For all studies, statistical significance was set at p<0.05. Statistical analyses were performed using Statistica (TIBCO Software Inc., Palo Alto, CA, USA) , Prism (GraphPad, San Diego, CA, USA), and Microsoft Excel 2016. For the experiment testing effects of intra-NAcC AmyR activation on chow intake and meal patterns, a meal was defined as at least 0.25g of food intake and a minimum of 10min between feeding bouts (Mietlicki-Baase et al., 2015; Mietlicki-Baase et al., 2013). All chow intake measurements (e.g., meal-related chow intake, meal size, and meal number at each time point in feedometer study, and chow intake in palatable macronutrient studies), as well as body weight change in all experiments, were analyzed by separate repeated measures analysis of variance (ANOVA) accounting for the within-subject effect of drug. Cumulative meal size was calculated by dividing the cumulative meal-related intake (greater than 98% of chow intake in this experiment) at each time point by the cumulative number of meals that had been taken at that time point. For studies of palatable macronutrient intake, intake of the palatable macronutrient solution was analyzed by repeated measures ANOVA, accounting for within-subject effects of time and drug. For the gene expression studies, the effect of fat exposure on the expression levels of each AmyR component or dopamine receptor subtype, as well as average fluid intake and terminal body weights, were tested by one-way ANOVA. When a significant effect was observed in an overall ANOVA, Student-Newman-Keuls post-hoc analysis was used to reveal differences between conditions. Unless otherwise stated, all values are presented as mean ± SD.

Results

3.1. Effect of intra-NAcC AmyR activation on chow intake and meal patterns

To test the hypothesis that AmyR activation in the NAcC reduces food intake, we first evaluated the effect of direct NAcC injection of amylin (0.2μg) or the AmyR agonist sCT (0.04μg) on chow intake in ad libitum-fed rats. We detected no statistically significant effects of AmyR activation in the NAcC on chow intake at any time during the 24h post-injection (Figure 1A; all F_2,16≤2.24, all p>0.05). No significant change in 24h body weight gain occurred in these rats (Figure 1B; F_2,16=3.18, p>0.05). Meal patterns were also examined in this study, including latency to onset of feeding, meal size, and meal number. There were no significant changes in latency to the first meal after amylin or sCT at the doses examined (vehicle: 1463.33±1560.94 s; amylin: 2343.11±2960.40 s; sCT: 1817.11±2209.23 s; F_2,16=1.36, p>0.05). Additionally, no significant changes in cumulative meal size (Figure 1D; all F_2,16≤1.97, all p>0.05) or cumulative meal number (Figure 1E; all F_2,16≤1.93, all p>0.05) were observed after intra-AmyR activation with amylin or sCT at these doses.

3.2. Effect of NAcC AmyR activation on intake of palatable fat and sucrose solutions

Given the important role of the NAcC in intake of palatable foods (Brown, McCutcheon, Cone, Ragozzino, & Roitman, 2011; Vucetic & Reyes, 2010), we hypothesized that AmyR activation in this site may have more potent intake-suppressive effects for palatable foods compared to a bland food like chow. In addition, because previous research has demonstrated that AmyR activation in the VTA may differentially impact intake of different palatable macronutrient solutions (Mietlicki-Baase et al., 2017), we evaluated intake of fat and, in a separate group of rats, of sucrose after intra-NAcC amylin or sCT. In these studies, rats were trained to consume either a 10% fat solution (Intralipid^®; 1 kcal/ml) or 25% sucrose solution (1 kcal/ml, isocaloric to fat solution) for a 1h access period, and received the intra-NAcC injection of the AmyR agonist just before the access period began; intake of the palatable solution and subsequent chow intake were measured. In rats given fat solution during the 1h access period, no significant effect of NAcC AmyR activation was observed on fat intake (Figure 2A; no main effect of drug, F_2,10=0.17, p>0.05; no interaction between time and drug, F_10,50=1.31, p>0.05). Surprisingly, however, when chow was returned after the 1h access period, intake of chow over the next ~21.5h was suppressed by AmyR activation (Figure 2B; F_2,10=5.05, p<0.05; post-hoc tests, vehicle versus amylin, p<0.08; vehicle versus sCT, p<0.05). Body weight gain over the test period was not significantly different between groups (Figure 2C; F_2,10=1.40, p>0.05). In contrast, in the rats that received a sucrose solution, we saw no significant effects of NAcC amylin or sCT on sucrose intake (Figure 3A; no main effect of drug, F_2,14=0.20, p>0.05; no interaction between time and drug, F_10,70=0.57, p>0.05), subsequent chow intake (Figure 3B; F_2,14=3.13, p<0.08), or body weight gain (Figure 3C; F_2,14=0.47, p>0.05). These findings suggest that although fat intake was not changed by NAcC AmyR activation, access to and/or intake of fat changed the effects of an AmyR agonist on chow intake later in the post-injection period.

Figure 2. AmyR activation in the NAcC reduces chow intake after exposure to a fat solution.

Although intra-NAcC administration of the AmyR agonist sCT (0.04μg) did not affect the intake of 10% fat solution when made available to the rats for 1h after drug injection (A), it significantly reduced the intake of chow over the ~21.5h period following fat exposure (B). Intra-NAcC amylin (0.2μg) injection also reduced, although not significantly, chow intake after fat exposure (B). Neither sCT (0.04μg) or amylin (0.2μg) injection affected body weight gain over the test period (C). Data are shown as mean + SD. * indicates p<0.05 and # indicates p<0.1 by repeated measures ANOVA. Key applies to all panels.

Figure 3. AmyR activation in the NAcC has no effect on sucrose intake.

Administration of amylin (0.2μg) or the AmyR agonist sCT (0.04μg) into the NAcC did not affect the intake of 25% sucrose solution made available to the rats for 1h after drug injection (A). The treatments had no effect on chow intake (B) or on body weight gain (C) during the test period. Data are shown as mean + SD. Key applies to all panels.

3.3. Expression of AmyR components in mesocorticolimbic nuclei and impact of dietary fat exposure

Given the results from our behavioral studies suggesting that fat access / intake may influence later feeding effects of NAcC AmyR activation, we hypothesized that fat exposure may change the expression of AmyR components in the NAcC. To test this hypothesis, rats were exposed to a fat solution (10% Intralipid^®) for 1h a day for ~1 month with no more than 2 days at a time of no fat access. Rats exposed to tap water instead of fat served as a control. The two groups significantly differed in the average fluid intake during the last three days of training (Figure 4A; F_1,14=27.38; p<0.05). However, before sacrificing the rats at the end of the training period, fat-exposed and water-exposed rats had comparable body weights (Figure 4B; F_1,14=0.25; p>0.05). RT-qPCR was used to quantify the expression of the components of the AmyR (both subtypes of CTRs and the three subtypes of RAMPs) in NAcC-enriched micropunches as well as micropunches taken from other key mesocorticolimbic nuclei, specifically, the NAcSh, CeA, and mPFC.

Figure 4. One-hour daily fat exposure did not cause a significant change in the body weight of rats.

Although the average fluid intake of fat-exposed rats (10% fat solution) was significantly higher than water-exposed controls during the one-hour daily test period (A), the two groups did not differ in their average weight at the end of the experiment (B). Data are shown as mean + SD.* indicates p<0.05 by one-way ANOVA.

Our results in this study recapitulated previous findings (Mietlicki-Baase et al., 2013) demonstrating that all AmyR components are expressed in the NAcC. However, fat exposure was not associated with any significant changes in mRNA expression of AmyR components in this nucleus (Figures 5A and 5B; CTR-A, F_1,13=2.47; CTR-B, F_1,13=0.81; RAMP-1, F_1,13=0.08; RAMP-2, F_1,13=1.28; RAMP-3, F_1,13=0.89; all p>0.05). The present results also confirm the previous finding that AmyR components are present in the NAcSh (Mietlicki-Baase et al., 2013), but unlike our findings in the NAcC, we observed a significant reduction in mRNA levels of RAMP-3 in the NAcSh in fat-exposed rats compared to control (Figure 5D; F_1,11=7.80; p<0.05). No other differences in AmyR component expression were detected in the NAcSh (Figures 5C and 5D; CTR-A, F_1,12=0.08; CTR-B, F_1,12=0.07; RAMP-1, F_1,12=0.0003; RAMP-2, F_1,12=0.12; all p>0.05).

Figure 5. All components of the amylin receptor are expressed in the NAcC and NAcSh.

While all components of the AmyR are expressed in the NAcC, daily exposure to fat solution did not cause significant changes in the expression of any AmyR components in the NAcC (A and B). In the NAcSh, although fat exposure did not significantly affect the expression of CTR-A or CTR-B (C), it did cause a significant reduction in the expression of RAMP-3 compared to water-exposed control rats (D). The expression levels of RAMP-1 and RAMP-2 in the NAcSh were unaffected (D). Because expression of RAMP-2 and RAMP-3 were low in both NAcC and NAcSh, RAMP-1 expression is indicated by the left y-axis in panels (B) and (D), whereas RAMP-2 and RAMP-3 expression are indicated by the right y-axis in these panels. Data are shown as mean + SD.* indicates p<0.05 by one-way ANOVA. Key applies to all panels.

Given that dopamine receptors within the NAcC appear to mediate the intake-suppressive effects of VTA AmyR activation (Mietlicki-Baase et al., 2015), we also tested whether fat access influenced dopamine D₁ receptor (D₁R) or D₂ receptor (D₂R) expression in the NAcC or NAcSh. We quantified the expression of D₁R and D₂R in both the NAcC and NAcSh in fat-exposed rats and water-exposed controls. Interestingly, we have found no significant differences in D₁R or D₂R expression between the two groups: NAcC (Figure 6A; D₁R, F_1,12=0.64, p>0.05; D₂R, F_1,12=0.49, p>0.05), and NAcSh (Figure 6B; D₁R, F_1,12=0.03, p>0.05; D₂R, F_1,11=0.11; p>0.05).

Figure 6. Dopamine D₁ and D₂ receptor expression in the NAcC and NAcSh are not changed by fat access.

Dopamine D₁ receptor (D₁R) and D₂ receptor (D₂R) mRNA was detected within the NAcC (A) and NAcSh (B). Fat access did not influence expression of either receptor subtype in either site. Data are shown as mean + SD. Key applies to all panels.

Finally, we tested the hypothesis that all AmyR components are expressed in key mesocorticolimbic sites where amylin binding has been observed in vitro, specifically the mPFC and the CeA, and examined the effect of fat exposure on AmyR expression in these nuclei. qPCR data revealed that all AmyR components are expressed in the CeA and mPFC (Figure 7). Significant changes in AmyR expression after fat exposure were detected in the CeA and mPFC. In CeA from rats that had access to fat solution, CTR-A levels were significantly lower than those in rats that were not exposed to fat (Figure 7A; F_1,9=8.37; p<0.05) while no detectable differences in the levels of CTR-B (Figure 7A; F_1,9=0.11; p>0.05) or RAMPs (Figure 7B; RAMP-1, F_1,10=0.49; RAMP-2, F_1,10=1.36; RAMP-3, F_1,10=0.47; all p>0.05) were found between the two groups of rats. In contrast, in the mPFC, RAMP-2 expression was significantly lower in rats exposed to fat (Figure 7D; F_1,12=8.39; p<0.05) but no differences were detected in the levels of CTRs (Figure 7C; CTR-A, F_1,14=0.004; CTR-B, F_1,10=0.23; all p>0.05) or other RAMPs (Figure 7D; RAMP-1, F_1,14=0.85; RAMP-3, F_1,14=0.26; all p>0.05) in this site.

Figure 7. All components of the amylin receptor are expressed in the CeA and mPFC.

In the CeA, exposure to 10% fat solution for 1 h/day resulted in a significant decrease in the expression levels of CTR-A but not CTR-B (A). The expression levels of RAMP-1, -;2, and -3 in the CeA remained unaffected as a result of this treatment (B).On the other hand, in the mPFC, fat exposure did not affect the expression of CTR-A or CTR-B (C), but resulted in a significant decrease in RAMP-2 levels compared to rats exposed to water (D) No changes in the expression of RAMP-1 or RAMP-3 were observed (D). Data are shown as mean + SD. * indicates p<0.05 by one-way ANOVA. Key applies to all panels.

Discussion

Amylin is a pancreas and brain-derived peptide that promotes negative energy balance and weight loss. Since obesity is linked to the consumption of highly palatable foods (Ryan, Woods, & Seeley, 2012), investigating the actions of amylin in mesocorticolimbic nuclei and its potential role in suppressing reward-motivated food consumption could be of great translational value. Here, we investigated the effect of direct NAcC AmyR activation on the intake of bland versus palatable foods and found that, in rats with access to fat solution, there was no change in intake of the fat solution but later chow intake was suppressed after intra-NAcC AmyR activation. We also fully characterized the expression of AmyR components in other mesocorticolimbic sites and showed that all CTR and RAMP subtypes are expressed in mPFC and CeA. Finally, we evaluated the effect of dietary exposure to a fat solution on the expression of different AmyR components in key mesocorticolimbic sites. Our results indicated that fat access reduced CTR-A expression in the CeA, RAMP2 expression in the mPFC, and RAMP-3 expression in the NAcSh.

Previous research has shown that AmyR activation in the NAcSh can suppress food and fluid intake (Baisley & Baldo, 2014; Baldo & Kelley, 2001). However, a role of AmyRs in the NAcC for food intake control in ad libitum-fed rats has not been reported. This is surprising as AmyRs are expressed in the NAcC (Mietlicki-Baase et al., 2013) and the NAcC is well-established to play an important role in the control of food intake (Bassareo et al., 2002; Floresco et al., 2008; Mietlicki-Baase et al., 2015). It is particularly important to examine this as an animal’s intake after food restriction is different than that under ad libitum-fed conditions (Del Prete, Balkowski, & Scharrer, 1994; Larue-Achagiotis & Le Magnen, 1980). By examining the effects of AmyR activation in ad libitum-fed rats, we are able to determine the impact of AmyR signaling on natural feeding patterns when food is freely available. We first examined the effects of intra-NAcC AmyR activation on food intake in rats fed chow only and found no significant suppression of intake, meal size, or meal number in amylin- or sCT-treated rats.

Next, we examined whether NAcC AmyR activation suppressed intake of palatable fluids (sucrose or fat solution) or subsequent chow intake in rats given palatable fluid access. This was of interest because previous research has demonstrated that activation of AmyRs in the VTA, another key mesolimbic nucleus with direct projections to NAc, suppresses intake of fat solution and of sucrose solution (Mietlicki-Baase et al., 2017). However, suppression of fat intake in response to VTA AmyR activation may be more robust than suppression of sucrose intake (Mietlicki-Baase et al., 2017), making it important to examine both types of fluid. In the present study, we observed no significant suppression of intake of either palatable fluid after intra-NAcC AmyR activation. In contrast, we saw suppressive effects of intra-NAcC AmyR activation on food intake when chow was returned after the palatable fluid access period. Specifically, in chow-fed rats given 1h access to fat solution, activation of AmyR in the NAcC had no effect on fat intake but did suppress subsequent chow intake. However, there was no significant effect on chow intake in rats with sucrose access. These current results stand in contrast to effects of intra-VTA AmyR activation on palatable fluid intake, as direct administration of sCT into the VTA potently suppresses either fat or sucrose intake in one-bottle tests and selectively suppresses fat intake under some two-bottle conditions (Mietlicki-Baase et al., 2017). However, as particularly robust intake-suppressive effects are observed for fat intake after intra-VTA AmyR activation (Mietlicki-Baase et al., 2017) and for later chow intake after intra-NAcC AmyR activation in rats with fat access in the present studies, this may suggest that VTA and NAcC AmyRs play separate but complementary roles in suppressing food intake in rats with fat access.

In these studies, we observed differences in the effects of amylin itself versus the AmyR agonist sCT on intake measures. For example, in our findings in rats with fat access followed by chow access, intra-NAcC AmyR activation with sCT significantly suppressed later chow intake, whereas NAcC amylin injection produced a trend for chow intake reduction. Although sCT is a well-established AmyR agonist (Lutz, Tschudy, Rushing, & Scharrer, 2000) and shares binding sites with amylin (Beaumont et al., 1993), there are key differences between sCT and amylin that might explain the difference in significance observed in our results. Binding affinity of AmyRs for sCT is greater than that of amylin itself (Christopoulos et al., 1999) and the effects of sCT are generally more durable than those of amylin, perhaps due to its longer-lasting effects at the receptor (Hilton, Dowton, Houssami, & Sexton, 2000; Lutz et al., 2000). The longer action of sCT at its receptors may have resulted in the significant reduction in food intake in comparison to the trend produced by amylin in the present studies. Nevertheless, it is important to note that the direction of the chow intake-suppressive effects produced by intra-NAcC sCT and amylin in the present studies were the same, suggesting that the effects of both sCT and amylin are likely mediated by NAcC AmyRs.

The delayed effect of agonist administration into the NAcC after fat access in our behavioral pharmacology studies was surprising, as no longer-term (e.g., 24h) effects on chow intake were observed in our rats fed chow only. Furthermore, the effect was specific to rats given access to fat and was not observed in rats with sucrose access. This suggested that perhaps fat access was changing later effects of AmyR activation in the NAcC to suppress food intake. We therefore pursued the hypothesis that fat exposure changes AmyR expression in key mesocorticolimbic nuclei. First, we replicated our previous finding that all AmyR components are expressed in the NAcC and NAcSh (Mietlicki-Baase et al., 2013). Since AmyR activation in the mesolimbic system produces robust suppression of fat intake (Mietlicki-Baase et al., 2017), AmyR expression in rats exposed to fat solution was compared to that of fat solution-naïve (control) rats. Our experimental paradigm, in which the fat solution was made available to the rats for only 60 mins per day, was designed such that we could test the impact of fat exposure on AmyR expression without producing major weight gain in the test group of rats. This was done to rule out the possibility that any differences observed in AmyR expression levels between the two groups of rats would not be due to differences in inflammatory processes or other physiological changes known to be associated with overweight and obesity. Surprisingly, fat exposure was not associated with any changes in the expression of AmyR components in the NAcC, despite the fact that our behavioral data suggested later reductions in food (chow) intake in rats trained to drink fat solution. Nevertheless, this does not entirely rule out the possibility that fat exposure affects amylin signaling in the NAcC, as there could be other explanations for this observation that are worth exploring in future work. For example, fat exposure may induce molecular changes in the NAcC beyond the level of AmyR, such as at the level of second messenger systems. In contrast, in the NAcSh, fat exposure was associated with a reduction in RAMP-3 expression but not other components of the AmyR. The physiological relevance of this reduction in RAMP-3 is not yet clear, particularly in light of the fact that NAcSh RAMP-3 expression is quite low even in control animals. On one hand this could suggest reduced sensitivity to amylin in this nucleus (Christopoulos et al., 1999; McLatchie et al., 1998; Muff et al., 1999), but it is also possible that given the much higher levels of RAMP-1 and RAMP-2 expression in this site, the role of RAMP-3 in the response to amylin in this nucleus may be of lesser importance. Nevertheless, given that AmyR activation in the NAcSh has been associated with changes in ingestive behavior (Baisley & Baldo, 2014; Baldo & Kelley, 2001), this possibility will be addressed in future studies.

Previous research has demonstrated that the VTA is a physiologically and pharmacologically relevant site of action for amylin signaling in the control of food intake (Mietlicki-Baase et al., 2015; Mietlicki-Baase et al., 2013), and that dopaminergic projections from VTA to NAcC play an important role in the intake-suppressive effects of VTA AmyR activation. Specifically, intra-NAcC administration of dopamine D₁ and D₂ receptor agonists attenuated the intake- and body weight-suppressive effects of intra-VTA sCT (Mietlicki-Baase et al., 2015). Given the role of NAcC dopamine receptors in the intake-suppressive effects of mesolimbic amylin signaling, and that VTA dopamine neurons project monosynaptically to the NAc (Beier et al., 2015; Fallon & Moore, 1978), we also investigated whether fat intake altered dopamine receptor expression in the NAcC or NAcSh. We found no differences in D₁R or D₂R expression in either area between rats with fat access and control animals. The previous research examining the effect of NAcC D₁R / D₂R activation on the hypophagic effects of intra-VTA AmyR activation showed similar effects whether rats were maintained on chow or on a high-fat diet (Mietlicki-Baase et al., 2015). The present finding that expression of D₁R and D₂R was not changed in the NAc by a history of fat access appears to be consistent with these prior findings, although here we only examined NAc receptor expression at the mRNA level.

We also fully characterized all components of the AmyR in the mPFC and CeA. Previous research has demonstrated expression of a subset of AmyR components in these nuclei (Kalafateli et al., 2019) but here we expand upon these prior data by investigating both subtypes of CTR and all three types of RAMP, as well as the impact of fat access on expression of these components. Our data indicate that all CTR and RAMP subtypes are expressed in the mPFC and CeA. Together with previous data showing amylin binding in these sites (Beaumont et al., 1993; Paxinos et al., 2004; Sexton et al., 1994), this supports the likelihood that functional AmyR complexes exist in these nuclei, although again it must be noted that our results only examined expression at the mRNA level. Indeed, further research will need to investigate expression of and heterodimerization of the AmyR components at the protein level to confirm whether functional AmyRs are expressed within these sites. Fat exposure was associated with a reduction in the expression of RAMP-2 in the mPFC and with lower CTR-A expression in the CeA. The potential ability of AmyR activation in these sites to suppress food intake has not been investigated in the literature, and it is unknown whether these changes in expression of AmyR components may indicate a change in amylin sensitivity in these nuclei after exposure to dietary fat solution. Thus, these data lay foundational work warranting further behavioral studies that probe into the effect of AmyR agonists in these sites on food intake and body weight gain.

Conclusions

In summary, this study shows that although AmyR activation in the NAcC produces minimal effects on feeding in chow-fed rats, a delayed suppression of chow intake is observed in rats with access to a fat solution, suggesting an interesting interaction between diet and AmyR effects on energy balance control. Further, our work demonstrates that all AmyR components are expressed in several mesocorticolimbic sites relevant to food intake and energy balance control. Fat exposure is associated with downregulation of some of these receptor components in the CeA, mPFC, and NAcSh, which may modulate the ligand binding affinity of AmyR complexes and the potency by which amylin activates these receptor complexes in these sites. This highlights the possibility that the consequences of AmyR activation may differ depending on diet of the animal. Collectively, these findings encourage further examination of the role of the mesocorticolimbic system in amylinergic regulation intake of food intake and subsequently the potential of amylin signaling in novel sites such as the CeA and mPFC as possible pharmacological targets for the treatment of obesity.

New Findings.

What is the central question of this study?

We tested whether intra-nucleus accumbens core (NAcC) amylin receptor (AmyR) activation suppresses feeding, and evaluated whether palatable food intake influences mesocorticolimbic AmyR expression.

What is the main finding and its importance?

Intra-NAcC AmyR activation reduces food intake in some dietary conditions. We showed that all components of the AmyR are expressed in prefrontal cortex and central nucleus of the amygdala, and demonstrated that fat access impacts AmyR expression in these and other mesocorticolimbic nuclei. These results suggest that palatable food intake may alter amylin signaling in the brain and shed further light onto potential sites of action for amylin.