The alpha-7 nicotinic acetylcholine receptor agonist PHA-543613 reduces food intake in male rats

Abstract

Obesity is a prevalent disease, but effective treatment options remain limited. Agonists of the alpha-7 nicotinic acetylcholine receptor (α7nAChR) promote negative energy balance in mice, but these effects are not well-studied in rats. We tested the hypothesis that central administration of the α7nAChR agonist PHA-543613 (PHA) would decrease food intake and body weight in adult male Sprague Dawley rats. Intracerebroventricular (ICV) PHA administration in chow-fed rats produced a suppression of energy intake and weight gain over 24h. Next, to evaluate effects of ICV PHA on palatable food intake, rats were maintained on a choice diet of rodent chow and 45% high fat diet (HFD); under these conditions, ICV PHA produced no significant changes in energy intake from either food, or body weight gain, in the 24h post-injection. However, when given a choice of chow or a higher-fat 60% HFD, ICV PHA reduced intake of 60% HFD, but not chow; body weight gain was also suppressed. Further experiments evaluating conditioned taste avoidance (CTA) and pica in response to ICV PHA suggested that the suppressive food intake and body weight effects after ICV injection of PHA were not due to nausea/malaise. Finally, an operant conditioning study showed that responding on a progressive ratio schedule of reinforcement for high-fat food pellets decreased after ICV PHA. Collectively, these studies show that PHA reduces energy intake under some but not all dietary conditions. Importantly, central PHA decreases both food intake as well as motivation for highly palatable, energy dense foods in rats without inducing nausea/malaise, suggesting that the α7nAChR could be a viable target for developing treatments for obesity.

1. Introduction

Obesity is an ongoing epidemic in the United States. Currently, more than 66% of adults in the United States are overweight or have obesity [1, 2]. Since 1980, global obesity rates have more than doubled [3]. Over the years there have been numerous efforts to minimize the rate at which the obesity epidemic is growing. Today, primary interventions used to combat obesity include diet and lifestyle changes, bariatric surgery, and pharmacological treatment [4–8]. It is well known that diet and lifestyle modification is one of the safest ways to promote healthy weight loss [9]. However, studies show low success rates for long-term weight maintenance in individuals achieving initial weight loss through healthy diet changes [10–12]. Bariatric surgery has been shown to have more long-term sustainability as a weight loss intervention than lifestyle changes, but can pose lifelong health risks such as malabsorption, potentially leading to nutrient deficiencies [13]. Lastly and most relevant to the current research, pharmacological agents have been used for decades as a treatment option for obesity. However, over the years some of these drugs have been discontinued due to increased risk of cardiovascular disease and some cancers [14]. More recently, new pharmacotherapies such as the glucagon-like peptide-1 (GLP-1) receptor agonists liraglutide and semaglutide have been approved for the treatment of obesity and have produced high success rates [15, 16]. Originally approved for the treatment and management of type 2 diabetes, GLP-1 receptor agonists are effective to reduce body weight but also can elicit unwanted side effects such as nausea and vomiting in 50% of patients, leading to the discontinuation of these drugs in those individuals [17]. Further research is needed in order to uncover new pharmacological targets for the treatment of obesity that do not elicit these unwanted side effects.

Acetylcholine (ACh) is a neurotransmitter that has numerous effects in the body such as modulating inflammatory processes, movement, and blood pressure [18]. Additionally, ACh plays a vital role in controlling feeding related mechanisms [19–21]. Numerous studies have demonstrated the ability of central cholinergic signaling to suppress food intake [22–25]. In particular, nicotinic acetylcholine receptors (nAChR) have been implicated in the modulation of feeding behaviors [26].

The α7nAChR is a homomeric nAChR that is expressed throughout the brain [27] and has been shown to be involved in the control of food intake [26]. α7nAChRs can be activated by naturally occurring ligands or by synthetic agonists. Peripheral administration of TC-7020, a selective synthetic agonist of the α7nAChR, reduced food intake and body weight in a diabetic mouse model [28]. More recently, systemic delivery of the partial α7nAChR agonist, GTS-21, was shown to increase serum levels of GLP-1 in wild-type mice [29], suggesting the possibility that GTS-21 could suppress food intake through a GLP-1-related mechanism. However, peripheral or central administration of GTS-21 in rats failed to reduce food intake, nor did it impact circulating GLP-1 [30]. The differential effect of systemic GTS-21 on circulating GLP-1 levels in rats versus mice could be due to a species difference, but further consideration of the discrepant results is also warranted because GTS-21 is a partial agonist of the α7nAChR [31] and therefore may cause unexpected or off-target effects. It is possible that using a selective agonist may uncover potential food intake or body weight effects of α7nAChR activation in rats.

In the current study, we therefore examined the ability of a selective α7nAChR agonist, PHA-543613 [32], to impact feeding and body weight in rats. We focused on the effects of central PHA-543613 administration because α7nAChR are widely expressed in the CNS, which as mentioned previously is critical in the control of food intake. Moreover, to our knowledge, there have been no studies that have tested the food intake effects of a selective α7nAChR agonist after central administration despite the established involvement of α7nAChRs in central feeding pathways [26]. We hypothesized that central administration of PHA-543613 would reduce food intake and body weight in rats.

2. Materials and Methods

2.1. Animals

Adult male Sprague Dawley rats (Charles River; ordered from the vendor at ~8 weeks of age) were singly housed in hanging wire cages in a temperature and humidity-controlled environment with a 12h/12h light cycle. Rats undergoing operant conditioning testing were singly housed in tub cages. Rats were given water and rodent chow (Teklad 2018; 3.1kcal/g) ad libitum unless otherwise noted. A 1-week minimum habituation period for acclimation to the animal facilities and for all diets was allotted for all rats prior to the start of testing. All procedures were approved by the Institutional Animal Care and Use Committee of the University at Buffalo.

2.2. Surgeries

Rats were surgically implanted with a unilateral cannula (Plastics One) aimed at the lateral ventricle (LV). Rats were deeply anesthetized by an intramuscular (IM) injection of a rodent surgical cocktail (KAX; 90mg/kg ketamine, 0.64 mg/kg acepromazine, and 2.7 mg/kg xylazine) and placed in a stereotaxic apparatus. The guide cannula coordinates used (0.9 mm posterior to bregma, 1.6 mm lateral to midline, 2.8 mm ventral to skull) were selected from the literature [33]. An internal cannula extended 1.5–2.5 mm beyond the guide cannula, based on results of functional verification described below. Three jewelers’ screws were used to anchor the cannula to the skull and the incision was sealed with dental cement. Topical bupivacaine was applied to the incision site and analgesic (carprofen, 5 mg/kg, subcutaneous), was given 1h prior to surgery and for 2 days post-surgery.

To functionally confirm correct LV cannula placement, an intracerebroventricular (ICV) injection of angiotensin II (Ang II; 10ng in 1μL aCSF) was given and subsequent water intake was measured for 30 min as described previously [34, 35]. Rats were considered to have passed functional verification if they drank at least 5ml of water within 30 min post-AngII injection. Ang II injections were administered beginning with an internal cannula extending 1.5mm past the guide cannula. Once rats received their injection, tap water was provided immediately in a graduated burette. Rats that did not consume at least 5ml of water were retested on subsequent days using a 2.0mm and/or 2.5mm internal cannula. Rats that failed to drink at least 5ml of water in response to AngII with any of these internal cannula lengths were excluded from further testing.

2.3. ICV PHA-543613 dose-response studies: feeding and pica

All dose-response studies were conducted using counterbalanced, within-subjects designs. The order in which each rat received treatments was generated using a Latin square design. Injections were separated by at least 48 hours to considerably exceed the reported half-life for PHA-543613 [36].

For each experiment, initial body weights were taken and food hoppers were removed from cages right before treatment administration, beginning ~1h prior to the time lights turned off. Just prior to dark onset, rats were given back pre-weighed food hoppers. Food hopper weights and crumb spillage were measured at 1, 3, 6 and 24h after injections. Final body weights were also collected at 24h.

First, chow intake and 24h body weight change were evaluated after ICV administration of PHA-543613. Rats (n=13; 427.2 ± 11.8g at start of testing) had access to rodent chow and water only, and received an ICV injection of PHA-543613 (Sigma-Aldrich) at either a high (0.1mg), medium (0.05mg), low dose (0.025mg), or Veh (2μL artificial cerebrospinal fluid [aCSF], Harvard Apparatus). Food intakes and body weight were measured as detailed above.

To evaluate effects of ICV PHA-543613 on palatable food intake, two dose-response studies were conducted. In the first of these experiments, rats (n=12; 652.6 ± 22.5g at start of testing) had access to separate food hoppers containing rodent chow and 45% high fat diet (HFD45; Research Diets D12451, 45% kcal from fat, 4.7kcal/g). Injections were given as described for the chow-only experiment, and food intakes and body weight were measured as detailed above. One rat was excluded from this study due to technical issues with data collection.

In the second dose-response study involving palatable food access, rats (n=15; 501.1 ± 7.6g at start of testing) had access to rodent chow and 60% high fat diet (HFD60; Research Diets D12492, 60% kcal from fat, 5.24kcal/g). Injections and measures again followed the previously described dose-response studies above.

To examine whether PHA-543613 administration caused malaise, we tested for pica in rats using the same animals from the chow-only dose-response study (n=11; 539.8 ± 14.4g at start of testing); however, here, animals were given access to both rodent chow and kaolin (Research Diets). Rats were habituated to kaolin at for at least 1 week prior to testing. Injections and measures were similar to the previously described dose-response studies above, but in addition to food intake, 24h kaolin intake was also evaluated.

2.4. Conditioned Taste Avoidance (CTA) Study

A two-bottle CTA test was performed as described previously [37–39]. Briefly, rats (n=25 total; 615.6 ± 9.9g at start of testing) were habituated to a water restriction schedule for 7 days in which water access was given once daily for 90 min, 2h after the onset of the light phase. Rats were maintained on chow and HFD60 for this experiment, with food available ad libitum during water habituation. After habituation, rats moved to the training phase. On each training day during the normal 90-min fluid access period, rats were given access to two burettes, one on each side of their cage, containing the same flavor of Kool-Aid [either cherry or grape, flavors equally preferred [40]; 0.14 oz Kool-Aid mix, 10g saccharin, 3L water]. Immediately after flavor exposure, one group of rats (n=12) received ICV injection of PHA (CS+, 0.1mg) or aCSF (CS-, 2μL). A second group (n=13) received intraperitoneal (IP) injection of LiCl (CS+, 0.15 M), a substance known to elicit malaise in rodents [38], as a positive control, or saline (CS-, 1ml/kg). Both groups of rats were given two training days to learn the associated flavors. Drug treatments were counterbalanced across training days and for paired flavor. Each training day was followed by a non-injection day where water was available during the 90-min fluid access period. After training, rats were tested for expression of CTA. Animals were presented with both flavors of fluid (one flavor per burette) during the 90-min access period. Fluid intake was recorded after 45 min, at which time the sides of flavor presentation was switched to avoid side preference. Final fluid intake was then recorded at 90 min.

2.5. Operant Testing

Rats (n=15; 388.3 ± 4.53g at start of testing) had ad libitum access to chow and water outside training and testing times. All training and experimental sessions took place during the light phase in standard operant chambers (Med Associates) with a pellet trough and magazine for the delivery of 45 mg high-fat pellets (Custom-F07679; 60% kcal from fat, 15% kcal from protein, 25% kcal from carbohydrates; BioServ), as well as an active lever and an inactive lever. The composition of the high-fat pellets for this study was chosen to closely resemble the HFD60. MED-PC software was used to control the operant chambers and electronically record data.

Rats were initially trained on a fixed ratio 1 (FR1) schedule of reinforcement, during which each active lever press resulted in the delivery of one reinforcer pellet. Responding on the inactive lever did not produce any consequence. In order to progress to the next training FR schedule, rats were required to display stable responding, operationally defined as obtaining at least 20 reinforcers for two consecutive days with less than 20% variation in active lever presses between the two days. Once these stable response criteria were met, rats progressed to a fixed ratio 5 (FR5) training schedule where five active lever presses are required for the delivery of one reinforcer pellet. FR training sessions ended after 1 hour or when a rat received the maximum of 30 reinforcers per session. Rats were trained on the FR5 training schedule for seven consecutive days before moving onto the testing phase.

Rats were tested under a progressive ratio (PR) schedule. Rats completed two PR sessions, separated by one non-injection day of FR5 responding. On each PR day, rats received either an ICV injection of PHA (0.1mg) or vehicle (2μl aCSF) 30 min prior to the start of testing. The order in which rats received treatments were counterbalanced within the group. After injections were given, food and water were unavailable until completion of the PR session. The response ratio schedule during PR increased using the formula R(i)= 5e^0.2i-5 [41–43]. Breakpoint was defined as the last ratio completed resulting in the delivery of one reinforcer pellet. The session is terminated when no reinforcer was earned for more than 30 min.

2.6. Data and statistical analyses

For feeding studies, all food intake was converted from grams to kilocalories (kcal) in order to be able to compare energy intake from the different types of diets that were used. The α-level was set at p<0.05 for all studies. All statistical analyses were conducted using Statistica (TIBCO) or SPSS (IBM). For ad libitum feeding and pica experiments, data from each time point / diet were analyzed by repeated measures ANOVA accounting for the within-subject factor of drug condition. Significant results from an overall ANOVA were further probed using Student Newman Keuls post hoc tests. Similarly, repeated measures ANOVA was used to analyze data from each dependent variable for operant testing. For the CTA experiment, statistical analyses were conducted using two-tailed t-test. All data are shown as mean ± SEM. For ad libitum feeding and pica experiments, any rat that had 24h food intake or body weight gain more than two standard deviations outside the mean were excluded as statistical outliers. For the chow-only study, 4 rats were excluded as statistical outliers (one in vehicle condition; one in 0.025mg PHA condition; one in 0.1mg PHA condition; one for both 0.05mg PHA and 0.1mg PHA conditions; final n=9). For each of the palatable food studies (chow / HFD45 and chow / HFD60), 2 rats per experiment were found to be statistical outliers and thus excluded from further analyses (HFD45: one in 0.05mg PHA condition, one in 0.1mg PHA condition; HFD60: one in vehicle condition, one in 0.05mg PHA condition), leading to final n’s of n=9 and n=13, respectively. For the pica study, 2 rats were excluded as statistical outliers (one in vehicle condition, one in 0.05mg PHA condition; final n=9).

3. Results

3.1. ICV PHA-543613 reduces chow intake

To examine whether central administration of a selective α7nAChR agonist would reduce food intake and body weight in rats, we examined feeding and weight change in response to ICV administration of PHA-543613 (PHA). We first examined the effects of ICV PHA on these outcomes in rats maintained on chow. We hypothesized that PHA would reduce food intake and body weight in rats, similar to the effects of α7nAChR activation in mice [28]. As shown in Figure 1A, ICV injection of PHA affected food intake at 6h (F_3,24=4.96, p<0.05) and 24h postinjection (F_3,24=3.63, p<0.05) but not at earlier times (1h and 3h, both F_3,24≤1.89, all p>0.05). Posthoc analyses revealed that these effects were driven by a suppressive effect of the 0.05mg dose of PHA on food intake at 6h and 24h compared to that of vehicle-treated animals, with the 0.025mg dose also significantly reducing energy intake at 6h (all p<0.05). A significant effect of ICV PHA on 24h body weight change was also observed (Figure 1B; F_3,24=3.32, p<0.05) again due to the body weight-suppressive effect of the 0.05mg dose of PHA (p<0.05 compared to vehicle treatment).

Figure 1:

Energy intake from chow was reduced at later times by ICV administration of PHA (0.025mg, 0.05mg) in rats (A; p<0.05). The 0.05mg dose of ICV PHA also significantly suppressed 24h body weight change (B; p<0.05). * indicates statistical significance for overall ANOVA (p<0.05). Different letters above bars within a time bin represent significant differences between conditions. Data are shown as mean ± SEM. Key applies to both panels.

3.2. ICV PHA-543613 suppresses 60% high-fat diet intake

Next, to assess the effects of ICV PHA on energy intake and body weight when palatable food is available, we measured these outcomes in rats maintained on a choice of chow and HFD45. ICV administration of PHA, using the same doses tested in the chow-only study (0.025mg, 0.05mg, or 0.1mg), did not significantly affect intake of either food at 1, 3, 6, or 24h (Figure 2A–B; all F_3,24≤2.23, all p>0.05). ICV PHA produced a trend for a decrease in 24h body weight change, although this did not reach statistical significance (Figure 2C; F_3,24=2.63, p=0.073).

Figure 2:

Energy intake from chow was not affected by ICV administration of PHA (0.025mg, 0.05mg, or 0.1mg) in rats (A; p>0.05). ICV administration of PHA also did not significantly affect 45% HFD intake (B; p>0.05). A trend for reduced body weight after ICV administration of PHA was detected at 24h (C; p<0.1). Data are shown as mean ± SEM. Key applies to all panels. #, p<0.1.

Although significant reductions of food intake were not observed in this first experiment, PHA produced a slight numerical decrease in 24h HFD45 intake as well as a trend for lower body weight, possibly due to the lower energy intake. α7nAChR expression is widespread throughout the brain, including regions responsible for reward processing [44]. We hypothesized that increasing the palatability of the food may reveal significant food intake-suppressive effects of PHA. Therefore, we next tested the feeding and body weight effects of ICV PHA in rats that were maintained on chow and HFD60. We hypothesized that ICV PHA would reduce intake of HFD60 in rats.

ICV administration of PHA (0.025mg, 0.05mg, or 0.1mg) did not impact chow intake at 1, 3, 6, or 24h (Figure 3A; all F_3,36≤1.49, p>0.05). However, effects on HFD60 intake were seen throughout the test period (Figure 3B), with significant effects at 3h (F_3,36=3.06; p<0.05) and 24h (F_3,36=5.58; p<0.05) post-injection. Post-hoc tests revealed that the high dose of PHA significantly suppressed HFD60 intake compared to vehicle treatment at both of these times (p<0.05), and also was significantly different from the low dose at 24h (p<0.05). A significant suppressive effect of the medium PHA dose also emerged at 24h (vehicle versus medium dose, p<0.05). Although not statistically significant, trends for PHA-induced reductions in HFD60 intake were observed at the other times as well (1h: F_3,36=2.60, p<0.07; 6h: F_3,36=2.39, p<0.09). Body weight change over the 24h test period was also significantly reduced after ICV administration of the highest dose of PHA tested (Figure 3C; F_3,36=7.57, p<0.05; posthoc analysis shows high dose significantly different from all other doses, p<0.05).

Figure 3:

In rats given a choice of chow and HFD60, although ICV PHA did not impact energy intake from chow at any time (A), ICV PHA at the highest dose suppressed intake of HFD60 at 3h and 24hr (B; 0.1mg; p<0.05). The medium PHA dose (0.05mg) also suppressed intake at 24h (p<0.05). Body weight was reduced by the highest dose of PHA tested (C; p<0.05). * indicates statistical significance for overall ANOVA (p<0.05); # indicates a trend (p<0.09). Different letters above bars within a time bin represent significant differences between conditions. Data are shown as mean ± SEM. Key applies to all panels.

3.3. ICV PHA-543613 does not produce gastrointestinal malaise in rats

Rodents, like many other species, will reduce intake of or avoid foods that they associate with gastrointestinal upset [37]. To determine whether ICV PHA elicits malaise in rats, we conducted a two-bottle conditioned taste avoidance (CTA) test (Figure 4); systemic LiCl was used as a positive control, as it is well-established to produce CTA [45–47]. As expected, rats given IP injection of LiCl consumed less of the (CS+)-paired flavor than the (CS-)-paired flavor during the experiment, confirming that LiCl induced a CTA in this group (p<0.05). However, intake of the (CS-)-paired and (CS+)-paired flavors did not significantly differ in the ICV PHA-treated rats, suggesting that ICV PHA did not produce a CTA after a single exposure (p>0.05). The results of this experiment suggest that the hypophagia in ICV PHA-treated rats is most likely independent of CTA.

Figure 4:

ICV PHA did not produce a conditioned taste avoidance (CTA) in rats (p>0.05). In contrast, administration of LiCl as a positive control reduced intake of the drug-paired flavor (p<0.05), indicating CTA. * indicates statistical significance by a two-tailed t-test (p<0.05). Data are shown as mean ± SEM.

Although there was no statistically significant induction of CTA by a single exposure of ICV PHA, there was a numerical decrease in intake of the drug-paired flavor (p=0.17). It is possible that this could reach statistical significance in a different experimental context, such as after repeated exposure of ICV PHA, so we wanted to further solidify the conclusion that malaise was not responsible for the reductions in food intake observed with PHA. To this end, we evaluated pica in ICV PHA-treated rats given access to chow as well as kaolin. In rats, the consumption of kaolin has been shown to be a response to malaise as this species is unable to vomit [48]. Based on the lack of a significant CTA in the previous experiment, we hypothesized that ICV PHA would not increase kaolin intake, or in other words, would not cause pica. In this experiment, minimal early effects on chow intake after ICV PHA were detected. Specifically, chow intake was significantly affected 1h after ICV treatment (Figure 5A; F_3,24=3.85, p<0.05) driven by reduced intake in rats given the 0.05mg dose compared to vehicle and high doses (p<0.05). No changes in chow intake were detected at any other time point tested for any dose of PHA (Figure 5A; all F_3,24≤2.29, p>0.05). Furthermore, kaolin consumption was minimal in all groups, with no significant changes in 24h kaolin intake induced by PHA (Figure 5B; F_3,24=0.81, p>0.05). Thus, we conclude that ICV PHA does not elicit pica at any of the doses tested. Together with our CTA data, these findings suggest that the intake-suppressive effects of ICV PHA are independent of malaise.

Figure 5:

When rats had access to both chow and kaolin clay, chow consumption decreased at 1h after the medium dose of PHA (0.05mg; A; p<0.05). No change in chow consumption was observed at any other time point tested (A; p>0.05). No change in kaolin intake was observed for any dose of PHA (B; p>0.05). Note that vertical axis values differ between panels. * indicates statistical significance for overall ANOVA (p<0.05). Different letters above bars within a time bin represent significant differences between conditions. Data are shown as mean ± SEM. Key applies to all panels.

3.4. ICV PHA-543613 reduces motivation for a high-fat food

Having determined that the reduction of energy intake induced by ICV PHA is independent of malaise, next we further explored possible mechanisms contributing to PHA-induced suppression of feeding by evaluating motivation to obtain a palatable food in an operant paradigm. We hypothesized that the high dose (0.1mg) of ICV PHA which suppressed ad libitum HFD60 intake would also reduce responding for a food reinforcer in a progressive ratio task, indicating reduced motivation to work for that food.

Figure 6 shows responses to obtain high-fat food pellets on the progressive ratio schedule. The number of active lever presses was higher in vehicle-treated rats than PHA-treated rats (Figure 6A; F_1,14=7.24, p<0.05), as was the number of reinforcers earned (Figure 6B; F_1,14=7.34, p<0.05). The breakpoint for PHA-treated rats was lower than that of vehicle-treated rats (Figure 6C; F_1,14=6.43, p<0.05). Finally, inactive lever presses did not differ between PHA-treated rats and vehicle-treated rats (Figure 6D; F_1,14=0.54, p>0.05).

Figure 6:

Active lever pressing was lower in PHA-treated rats (A; p<0.05). A trend was observed for the difference in reinforcers earned between vehicle and PHA-treated rats (B; p<0.1). Breakpoint was lower in PHA-treated rats than in vehicle-treated rats (C; p<0.05). No difference in inactive lever presses was observed between groups (D; p>0.05). * indicates statistical significance in a repeated measures ANOVA (p<0.05). Data are shown as mean ± SEM.

4. Discussion

The involvement of central nAChRs in feeding has been increasingly studied over the years, but few studies have looked specifically at the role of the α7 subtype in food intake and body weight control. Previously, we found that the partial α7nAChR agonist GTS-21 did not reduce food intake or body weight in rats after central or peripheral administration [30]. We also found that GTS-21 did not change circulating levels of the anorectic peptide hormone GLP-1 in rats, as it had been shown to do in mice [29]. GTS-21, being a partial agonist of the α7nAChR, has been shown to have effects via other receptor subtypes [49], so to better elucidate the role of the α7nAChR in energy balance control, we used the selective α7nAChR agonist PHA-543613 and evaluated its effects on food intake and body weight in rats. Results generally supported our hypothesis that centrally administered PHA would reduce food intake and body weight in rats.

The ability of ICV PHA to produce hypophagia appeared to depend at least in part on the diet available to the rats. When chow alone was available, both energy intake and body weight change over the 24h post-injection were reduced after PHA treatment. However, when chow was available simultaneously to another more palatable diet, or alongside kaolin clay, chow intake was either unaffected by PHA (HFD45/HFD60 experiments) or only transiently reduced (pica study). This outcome in the choice studies where rats could eat either chow or a palatable high-fat diet could be explained by a possible floor effect due to the minimal energy intake from chow, but is more difficult to reconcile with the findings from the pica study, where chow is the only food available. The underlying reason for the different effects of PHA on chow intake between the chow-only and pica studies could be due to consequences of having another hopper available in the pica study, even though it contained a non-nutritive substance. Indeed, prior studies have shown that the number of bottles available in fluid intake testing can influence consumption and preference [50], suggesting that the presence of another option of ingesta can affect intake. Nevertheless, both these experiments demonstrate the same suppressive direction of effect of ICV PHA on chow intake, albeit at different timescales.

Turning attention to the outcomes of the chow/HFD choice studies, although ICV injection of PHA did not significantly reduce intake of chow or HFD45 in rats with a choice of these 2 diets, in a separate experiment where rats could choose chow or HFD60, the highest dose of ICV PHA significantly reduced intake of the HFD60 without affecting chow intake. To the best of our knowledge, there have been no previous studies that have specifically examined the effects of α7nAChR with PHA to study feeding, nor has there been research on the effects of central administration of selective α7nAChR agonists on food intake and body weight. However, some studies have shown reduced preference for palatable substances after peripheral administration of other selective α7nAChR agonists. A recent study showed that mice no longer prefer a low concentration of sucrose after chronic administration of the selective α7nAChR agonist, PNU282987, compared to vehicle-treated mice [51]. Similar to our study, this study also showed standard rodent chow consumption was not affected after administration of PNU282987. Both our study and the study by Kawamura and colleagues show changes in the way selective α7nAChR agonists affect the intake of palatable substances, perhaps involving alterations to processes influencing the perception of or motivation for these foods.

After observing significant reductions in palatable food intake induced by ICV PHA, it was important to rule out the possibility that these hypophagic effects were caused by malaise. As previously mentioned, obesity pharmacotherapies such as liraglutide and semaglutide elicit nausea and vomiting in over 50% of patients [17]. Therefore, not only is it important to identify new targets for weight loss drugs but to also identify targets that do not have these side effects. The results of our CTA and pica studies indicated that the mechanisms of ICV PHA to reduce food intake and body weight are independent of malaise. This is consistent with some limited data on the effects of α7nAChR activation in humans. Tropisetron, a drug currently used to treat nausea and vomiting in patients undergoing chemotherapy, serves as both a serotonergic 5-HT3 antagonist as well as a α7nAChR agonist [52]. This suggests that activation of α7nAChRs may in fact have protective effects against nausea/vomiting. Furthermore, clinical trials using α7nAChR agonists to treat psychological conditions such as schizophrenia, have no reported side effects of nausea and vomiting [52]. The lack of reports of α7nAChR agonists eliciting nausea is consistent with our findings here.

Since ICV PHA reduced food intake and body weight independently of malaise, we next examined whether ICV PHA reduces motivation for a palatable food using progressive ratio operant testing. Indeed, ICV PHA significantly suppressed breakpoint for high-fat pellets, indicating that central α7nAChR activation is reducing motivation to work for the palatable food reward. In a previous study in rats, central α7nAChR activation with a selective agonist, PNU-282987, decreased operant responding in a progressive ratio test for self-administration of nicotine [53]. Furthermore, blockade of central α7nAChR via an α7-antagonist increased operant responding for nicotine. The results of this study are consistent with results of our study despite the difference of food reward vs. drug reward. It is important to note that highly palatable food reward and addiction follow very similar mechanisms as drugs of abuse [54]. One caveat of this study is the use of nicotine as the drug reward choice due to nicotine being a ligand for the α7nAChR [55]. The outcomes of this study could be susceptible to differences of interpretations since PNU-282987 and nicotine are both stimulators of α7nAChRs. There are also important methodological differences to consider; for example, in our study we administered PHA into the LV, while Brunzell and colleagues administered the selective α7nAChR agonist site-specifically in to the nucleus accumbens, a mesolimbic nucleus. Since ICV administration allows the agonist to disperse widely throughout the brain, we cannot conclude which brain regions directly mediate the effects observed in our studies. However, Brunzell’s study was able to pinpoint a more specific region at which α7nAChR activation acts to reduce motivation to obtain a reward. Although Brunzell and colleagues identified the nucleus accumbens as a key site involved in the reduction in motivation to obtain a reward, it is possible that other brain regions within the mesolimbic system are also involved. This could be examined in future studies by administering α7nAChR agonists to other specific nuclei.

A few limitations to the present set of studies should be noted. First, only male rats were used for these experiments. This was done to establish whether any effect of ICV PHA on feeding and body weight in rats could be detected. To date, available data regarding the energy balance effects of central α7nAChR agonists in rats are limited to males [30], so studying these effects in males also allowed us to make more direct comparisons with previous literature. Obesity and overweight are major health concerns in women as well as men [2] and thus it will be important in future studies to determine whether the anorectic effects of central PHA observed here extend to female animals. Another limitation is that the rats used in these studies were singly-housed. Although this was done to accurately measure the food intake of each individual animal, social isolation is a stressor. Stress can alter feeding behavior in rats, so it is possible that different outcomes would be observed in group housed animals not experiencing the stressor of social isolation. However, stress often causes hypophagia in rodents [56], suggesting that the ability of ICV PHA to suppress intake even from a potentially reduced baseline level of intake is likely to occur in non-stressed animals as well.

5. Conclusions

Effects of the α7nAChR on energy balance are generally understudied, although some evidence suggests that stimulation of these receptors may reduce food intake and body weight [26, 28]. In particular, we lack information on central α7nAChR activation and the mechanisms by which they influence energy balance. This is vital as this could represent a novel target for obesity. The current studies reveal a novel role for central α7nAChR in the control of food intake, suppressing energy intake as well as motivation for palatable food. Furthermore, PHA appears to elicit these effects in rats without causing malaise. Although further experiments are undoubtedly required to more fully elucidate the role of α7nAChR signaling in feeding and body weight control, the present findings represent an important step in determining the effects of selective α7nAChR agonists on energy balance control.

Highlights.

• Central PHA-543613 (ICV PHA) reduces food intake and weight gain

• ICV PHA does not elicit malaise, measured by pica or conditioned taste avoidance

• ICV PHA suppresses motivation to work for a palatable high-fat food

• The α7 nicotinic acetylcholine receptor may be a potential target to treat obesity