Aerobic Exercise Improves Food Reward Systems in Obese Rats via

Subscriber access provided by Stockholm University Library

Article

Aerobic exercise improves food reward systems in obese rats via insulin signalling regulation of dopamine levels in the nucleus accumbens Wei Chen, Juan Li, Jun Liu, Dalei Wang, and Lijuan Hou ACS Chem. Neurosci., Just Accepted Manuscript • DOI: 10.1021/acschemneuro.9b00022 • Publication Date (Web): 22 Apr 2019 Downloaded from http://pubs.acs.org on April 23, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Neuroscience

Aerobic exercise improves food reward systems in obese rats via insulin signalling regulation of dopamine levels in the nucleus accumbens

Author names Wei Chen†, Juan Li†, Jun Liu§, Dalei Wang♧, Lijuan Hou*

† Key Laboratory of Measurement and Evaluation in Human Movement and Bioinformation, Physical Education College, Hebei Normal University, Shijiazhuang, Hebei 050024, China * College of Physical Education and Sports, Beijing Normal University, Beijing 100875, China § Department of Health Science, Xi’an Sport University, Xi’an 817006, China ♧ Institute of Military Basic Education, National University of Defense Technology, Changsha

* Corresponding author: Lijuan Hou

1

ACS Paragon Plus Environment

ACS Chemical Neuroscience 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 21

Abstract The dopaminergic pathway, comprising projections from the ventral tegmental area to the nucleus accumbens, constitutes the core of the brain reward system. Insufficient food reward caused by dopamine signalling dysfunction in the nucleus accumbens is an important contributor to obesity and may be associated with insulin signalling. Aerobic exercise has a positive effect both on preventing and treating obesity. In addition, physical exercise is important in striatal dopamine homeostasis and improves insulin sensitivity in the peripheral and central nervous system. Therefore, we hypothesized that aerobic exercise may increase dopamine levels in the nucleus accumbens through insulin signalling, thus improving food reward in obesity. In the present study, we used a rat model of obesity, induced by high fat diet. Obese rats exhibited lower basic dopamine concentration in the nucleus accumbens, induced by eating and/or extracellular insulin; attenuated insulin signalling; and increased fat preference. Interestingly, an 8-week aerobic exercise reversed these symptoms. In addition, we noted a significant increase in insulin Akt/GSK3-β signal transduction in the nucleus accumbens. These data demonstrate that aerobic exercise promotes dopamine release in the nucleus accumbens through insulin signal transduction, which may be constitute an important neurobiological mechanism of exercise against obesity.

Keywords: obesity, insulin resistance, aerobic exercise, dopamine release, reward system, high-fat diet

2


Page 3 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Introduction Food rich in fat is a source of natural reward. When the alternative foods were offered, the rats preferred a high-fat diet and became overweight, which suggests that high-fat food may have a higher food reward effect.1, 2 Previous studies have shown that, in terms of dietary structure, the food’s fat content is the only factor that induces obesity in rodents.3 Long-term consumption of a high-fat diet lowers the responses generated by the reward system after feeding. As individuals need to increase their food intake to produce the anticipated food rewards, the resulting overconsumption of food largely leads to obesity.4 The dopaminergic pathway, extending from the ventral tegmental area (VTA) in the midbrain to the nucleus accumbens, is the core of the brain’s reward circuit.5 When stimulated by food and related messages, the VTA-dopamine (DA) neurons display increased bursting firing, which enhances the motivation to eat.6 The ingestion of palatable foods leading to increased dopamine release in nucleus accumbens, which mediates primary reward and further reinforces the motivation to eat.7 As the VTA-DA neurons express insulin receptors, their functions and activities are regulated by insulin.8 On one hand, during feeding, insulin is released to the circulation and then enters the brain by crossing the blood-brain barrier via active transport. It then acts on the VTA to lower the excitation of DA neurons, thus reducing the motivation to eat.9 On the other hand, insulin acts on the nucleus accumbens to promote DA release at axon terminals, thus enhancing the reward effect.10 Insufficient food reward is considered to be the key factor leading to obesity-related overconsumption.11 Improving central insulin sensitivity to enhance feeding-induced DA release at the nucleus accumbens may have a positive impact on reducing obesity-related overconsumption. Consensus has been reached on the fact that physical exercise significantly improves obesity-related insulin resistance.12, 13, 14, 15 In recent years, studies have begun to focus on whether exercise can regulate eating behaviour for obesity prevention through the reward system. As shown in a preliminary study, exercise intervention ameliorates DA release in the striatum of mice with obesity induced by a high-fat diet, while reducing their weight and diminishing their preference for a high-fat diet.16 However, whether this finding correlates with the increase in insulin sensitivity and subsequent regulation of DA release in the nucleus accumbens due to physical exercise is yet to be elucidated. For this purpose, we established a rat model of obesity, using high-fat diet, and observed the impact of treadmill exercise on insulin signalling and DA levels in the nucleus accumbens of rats, as well as their feeding behaviour, in order to further reveal the underlying neurobiological mechanism of how exercise prevents obesity. 3



Page 4 of 21

Results and Discussion Aerobic exercise inhibits weight gain, decreases energy efficiency, and improves body composition in rats consuming high-fat diet To assess the effects of exercise, we induced obesity in rats by feeding them with high-fat diet for 12 weeks and adopted an exercise intervention for 8 weeks that included passive treadmill training of moderate intensity and duration as an aerobic exercise. During training, no sound, light, electricity, or any other stimuli were present, and the rats demonstrated a relatively good level of autonomy. As shown in figure 2A-B, the high-fat diet significantly increased the weight of rats (271.76 ± 17.22 g vs 383.82 ± 18.58 g; P < 0.01). After 8 weeks of exercise intervention, rats in the obese-exercise group (OEG) weighed significantly less than those in the obese group (428.24 ± 19.87 g vs 476.73 ± 22.35 g; P < 0.01). During the first 12 weeks (Figure 2C), the energy efficiency, which is the ratio of a rat's weight gain to its energy intake during the same period, was significantly higher in rats of the high-fat diet group (HG) than in those of the regular group (RG) (11.24 ± 2.43 % vs 6.41 ± 0.69 %; P < 0.01); after 8 weeks, the exercise intervention significantly decreased the energy efficiency of OEG rats compared to that of OG rats (5.15 ± 1.23 % vs 10.08 ± 2.34 %; P < 0.01). Moreover, OG rats showed a significantly higher percent of visceral fat when compared to rats in the control group (14.83 ± 1.87% vs 9.13 ± 1.21 %; P < 0.01). In contrast, the percent of visceral fat was significantly lower in OEG than OG rats (11.25 ± 1.36 % vs 14.83 ± 1.87%; P < 0.05), as shown in Figure 2D. These findings indicate that the exercise intervention strategy used in this study was effective in significantly lowering the energy efficiency in obese rats, controlling weight gain and promoting weight loss, and improving their body composition.

Aerobic exercise reduces fat preference and weakens the impulse to feed in obese rats The sucrose preference test has been widely applied for the assessment of an animal’s consumption of natural rewards.17 Reduced preference for sucrose is usually associated with insufficient food reward, which is often linked to obesity.17 We subjected rats of all groups to a food preference test. Compared to CG rats, OG rats showed significantly reduced preference for the sucrose solution (8.11 ± 1.16 % vs 14.18 ± 1.64 %; P < 0.01) and milk (14.24 ± 1.62 % vs 18.57 ± 2.23 %; P < 0.01), but significantly stronger preference for high-fat food (77.69 ± 4.48 % vs 67.42 ± 3.47 %; P < 0.01). After the 8-week exercise intervention, OEG rats showed significantly stronger preference for the 4


Page 5 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


sucrose solution (13.44 ± 1.25 % vs 8.11 ± 1.16 %; P < 0.01) and milk (18.48 ± 1.21 % vs 14.24 ± 1.62 %; P < 0.01) but significantly reduced preference for high-fat food (67.26 ± 3.27 % vs 77.69 ± 4.48 %; P < 0.01), compared to OG rats, as shown in Figure 3A. After 24 h of fasting, normal feeding was resumed, and the consumption of food was recorded for 24 h. OG rats consumed a higher amount of food during the first 3 h compared to CG rats (47.27 ± 8.43 % vs 34.28 ± 6.37 %； P < 0.05), suggesting that they have a stronger impulse to feed. Exercise intervention significantly lowered food consumption in OEG rats during the first 3 h, when compared to the food consumption in OG rats (38.37 ± 8.42 % vs 47.27 ± 8.43 %；P < 0.05), as shown in Figure 3B. This implies that aerobic exercise enhances the preference of obese rats for sucrose, reduces their preference for fat, and weakens their impulse to feed. Compared to sugar and protein, fat has a higher energy density but also better palatability.3, 18 As a result, a high-fat diet can more significantly activate the brain's reward system, as evidenced by increased dopamine release in the nucleus accumbens. But chronic exposure to a high-fat diet may disrupt the midbrain dopaminergic system. Under certain circumstances, the motivation to feed, arising from the activation of the brain’s reward system by palatable food that is rich in fat, may offset the body’s energy homeostasis signalling, in turn promoting feeding behaviour.18,19 This implies that, after causing obesity, high-fat diets can lower the sensitivity of the reward system and lead to insufficient food reward.20,21 Physical exercise intervention, to a certain extent, can regulate the preference for high-fat food possibly by increasing the sensitivity of the reward system, as aerobic exercise is the most common way to promote the release of neurotransmitters, such as dopamine, in the striatum.22,23 Moody et al. reported that exercise reduces animals’ preference for high-fat diets and lowers their body weights by compensating for the food reward;24 however, the regulatory mechanism involved remains unclear.

Aerobic exercise improves insulin resistance in obese rats Long-term intake of high-fat food can induce obesity and insulin resistance, causing a significant increase in the peripheral and brain insulin levels.25, 26 Moreover, the insulin concentration in the brain is proportional to that in peripheral blood and to body fat content.27 To address this in our animals, we assessed insulin levels in the blood plasma and calculated the insulin resistance index and tolerance. The blood glucose concentration was measured at 10, 30, and 60 min after an intraperitoneal injection of insulin. The plasma glucose concentration was significantly higher at 10 5



Page 6 of 21

(7.56 ± 0.52 mmol/L vs 5.53 ± 0.54 mmol/L; P < 0.01) and 30 min (5.21 ± 0.38 mmol/L vs 3.34 ± 0.47 mmol/L; P < 0.01) after insulin injection in OG than in CG rats. Compared to rats in the OG, those in the OEG had significantly lower plasma glucose concentrations at both time-points (6.52 ± 0.50 mmol/L vs 7.56 ± 0.52 mmol/L at 10 min after injection, P < 0.01; 4.32 ± 0.44 mmol/L vs 5.21 ± 0.38 mmol/L at 30 min after injection, P < 0.01) after insulin injection (Figure 4A). The area under the curve (AUC) of the plasma glucose curve was significantly higher in OG rats than in their CG counterparts (321.35 ± 36.57 vs 227.82 ± 25.39; P < 0.01), but significantly lower in OEG than in OG rats (285.64 ± 27.71 vs 321.35 ± 36.57; P < 0.01), as shown in Figure 4B. By using the homeostasis model assessment of insulin resistance (HOMA-IR), we calculated that the HOMA-IR index was significantly higher in OG than in CG rats (3.87 ± 0.65 vs 2.63 ± 0.48; P < 0.01). In contrast, the HOMA-IR index was significantly lower in OEG than in OG rats (2.81 ± 0.62 vs 3.87 ± 0.65; P < 0.05), as shown in Figure 4C. George et al. showed that a 12-week aerobic exercise program, even without weight loss, improves insulin resistance in obese sedentary adolescents.28 The present study also indicates that aerobic exercise significantly improves insulin resistance in obese rats. Moreover, high fat diet-induced obesity is usually associated with insulin resistance. Thus, the beneficial effects of aerobic exercise in obesity are largely attributed to the improved insulin resistance.

Aerobic exercise effectively increases DA levels in the nucleus accumbens of obese rats and improves insulin-induced changes in these levels during feeding

Since exercise enhanced the sensitivity of obese rats’ reward system and diminished their preference for high-fat diets, we wondered whether this could be related to its effects on improving central nervous system insulin sensitivity. To test this hypothesis, we applied microdialysis and electrochemical high-performance liquid chromatography to track the changes in DA levels in the nucleus accumbens after milk consumption without or with insulin administration. Under baseline conditions, DA levels in the nucleus accumbens were significantly lower in OG than in CG rats (P < 0.01) but significantly higher in OEG than in OG rats (P < 0.01), as shown in Figure 5A. Then, we provided rats with milk containing 5% fat ad libitum for 15 min. The milk consumption in rats of CG, CEG, OG and OEG were 1.75 ± 0.84g, 1.83 ± 0.81g, 1.68 ± 0.66g and 1.78 ± 0.60g over 15 minutes, respectively. But no significant differences were observed between 6


Page 7 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


groups. We found that DA levels significantly increased in the nucleus accumbens of CG (from 165.02 ± 16.88 ng/mL to 204.52 ± 20.05 ng/mL; P < 0.01), CEG (from 172.33 ± 18.06 ng/mL to 214.16 ± 20.91 ng/mL; P < 0.05), and OEG (from 134.13 ± 15.92 ng/mL to 167.66 ± 19.56 ng/mL; P < 0.05) rats, but no significant change was observed in OG rats (from 108.89 ± 10.86 ng/mL to 120.24 ± 15.61 ng/mL; P > 0.05), as shown in figure 5B. The AUC of DA levels was significantly higher in OEG than in OG rats (40723 ± 4011 vs 31905 ± 4023; P < 0.01), as shown in figure 5C. This indicated that consumption of milk containing 5% fat induces the increase of DA in the rat nucleus accumbens. DA levels in the nucleus accumbens of obese rats not only remained low under baseline conditions but also displayed no significant change after milk consumption. This was significantly improved by aerobic exercise in OEG rats, indicating that aerobic exercise significantly elevates DA levels in the nucleus accumbens of obese rats and their response to feeding. Stouffer et al. found that the main effect of insulin on dopamine metabolism in the nucleus accumbens was to enhance its release, but the influence of insulin on dopamine level is diet dependent.10 Therefore, we perfused insulin into the rats’ nucleus accumbens simultaneously with the consumption of milk containing 5% fat. The milk consumption in rats of CG, CEG, OG and OEG were 1.82 ± 0.63g, 2.02 ± 0.76g, 1.87 ± 0.65g and 1.85 ± 0.77g in 15 minutes, respectively. But also no significant differences were observed between groups. This led to a significant elevation of DA levels in the nucleus accumbens of rats in all groups, including the OG (P < 0.01), as shown in figure 6A-B, indicating that insulin promotes DA increase and induces food reward by acting on this brain region during feeding. The ratio of DA level in the nucleus accumbens post- to pre-insulin injection was significantly lower in OG than in CG rats (124.35 ± 8.44 % vs 148.45 ± 12.31 %; P < 0.01). Thus, during the feeding process, the dopamine in nucleus accumbens is increased in response to the intake of palatable food, i.e., milk containing 5% fat or the high fat diet, and induces food reward. In obese rats, the insulin resistance might weaken the function of insulin in inducing DA release during feeding; as a result, the rats might fail to gain the anticipated reward through average feeding.5,19 In addition, The ratio of DA level in the nucleus accumbens post- to pre-insulin injection was significantly higher in OEG than in OG rats (139.62 ± 9.07 % vs 124.35 ± 8.44 %; P < 0.05), as shown in Figure 6C, indicating that the effect of insulin on DA rise in the nucleus accumbens is blunted in obese rats, but alleviated in rats subjected to aerobic exercise intervention. Thus, it might not be unreasonable to infer that aerobic exercise can enhance food 7



Page 8 of 21

reward in obese rats, and that insulin signalling in the nucleus accumbens may be involved in this process. It is noteworthy that, unlike milk alone, insulin injected into the nucleus accumbens promoted milk-induced dopamine increase, but the dopamine levels in the nucleus accumbens dropped rapidly in CG, CEG and OEG, but not in OG after 60 min (P < 0.05), as shown in figure 7. It is possible that there is a time effect of insulin-induced dopamine change in the nucleus accumbens during eating; however, the related regulatory mechanism is still unclear. This phenomenon may be related to insulin-induced changes in dopamine reuptake. Fordahl et al. found that a high-fat diet can disrupt dopamine reuptake by weakening the insulin signalling at the dopamine terminal.29 Taken together, the above results imply that aerobic exercise regulate DA level in the nucleus accumbens during feeding by increasing insulin sensitivity; this ultimately ameliorates the deficiency of food reward and reduces the overconsumption of high-fat diet by obese rats. It should be noted that the dopamine level in nucleus accumbens is not only related to the release of the axon terminals of dopamine neurons, but also regulated by the dopamine transporter.29 Therefore, it is not clear which of the above factors is related to the effect of exercise intervention on dopamine levels in the nucleus accumbens of obese rats. Aerobic exercise promotes insulin receptor (InsR)/protein kinase B (Akt)/glycogen synthase kinase 3 (GSK3) signalling pathways in the nucleus accumbens of obese rats Consistent with our results, existing literature indicates that, after insulin enters the brain, it acts on InsRs in the nucleus accumbens and activates cholinergic relay neurons to promote DA release.10,30 This was shown to depend on the phosphoinositide 3-kinase (PI3K) and Akt pathway.10 Other reports have indicated that the Akt/GSK3 signalling pathway participates in the reward process mediated by DA in the nucleus accumbens.31,32 GSK3 is a serine protein kinase, which has been shown to mediate dopaminergic neurotransmission in the nucleus accumbens; it is divided into two subgroups: GSK3α and GSK3β. After phosphorylation of a serine at the amino terminus of GSK3 (Ser21 in GSK3α or Ser9 in GSK3β), its activity is inhibited; activated Akt can also phosphorylate this site.33, 34 We thus assessed the expression levels of InsR, phosphorylated Akt, and phosphorylated GSK-3β in the nucleus accumbens. We found that these levels decreased by 34.25, 43.28, and 38.45%, respectively, in OG rats compared to the levels in CG rats (P < 0.01 for all), indicating that insulin signalling in the nucleus accumbens of obese rats is weaker. However, the expression levels of these proteins increased by 20.27, 26.33, and 32.54%, respectively, in OEG rats, as compared to 8


Page 9 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


those in OG rats (P < 0.05, P < 0.01, and P < 0.01, respectively), as shown in Figure 8. These results demonstrate that obesity induced by high-fat diet inhibits insulin signal transduction in the nucleus accumbens, thereby weakening the regulatory effects of insulin on DA release. They also indicate that aerobic exercise promotes Akt/GSK-3/insulin signalling transduction in the nucleus accumbens of obese rats. Nicole et al. also discovered that high-fat diet can inhibit the levels of Akt phosphorylation in the rat substantia nigra and striatum, thus affecting the metabolism of DA and increasing feeding behaviour.35 This implies that brain insulin resistance is also induced by high-fat diet in rats and aligns with the obesity-induced peripheral insulin resistance. Moreover, the effects of physical exercise on improving obesity-related insulin resistance have been fully validated.

Conclusions Our study demonstrates that exercise intervention promotes DA release in the nucleus accumbens during feeding in obese rats via the Akt/GSK3-β/insulin signalling pathway, in turn increasing DA levels in the nucleus accumbens, promoting the functional recovery of the reward system, lowering preference for high-fat diets, and decreasing energy efficiency. This helps to control weight gain in these rats. Therefore, the enhancement of insulin signal transduction in the nucleus accumbens is possibly one of the major mechanisms through which exercise prevents obesity. Evidently, increasing physical activities and restricting high-fat diets are both important measures for preventing obesity. In addition to following a healthy diet, individuals who are suffering from obesity should consider improving their insulin resistance and elevating their insulin sensitivity as important training goals during exercise intervention.

Methods Animals and grouping All animal experiments were approved by the Ethical Commission of Hebei Normal University and were performed in accordance with all national or local guidelines and regulations. We used 60 male Sprague Dawley rats of clean grade (age, 5 weeks; weight, 170-200 g) purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. Rats were fed and caged separately on a 12:12 light-dark cycle, with water given ad libitum. The animal room was kept at 25 ± 2 °C, with a relative humidity of 50 ± 5%. The rats were allowed one week to adapt to the animal room before being randomly assigned to two groups: RG (n = 24) and HG (n = 36). After 12 weeks, obese rats 9



Page 10 of 21

from the HG were categorized randomly into two subgroups: OG (n = 12) and OEG (n = 12), the remaining 12 rats in HG did not meet the criteria of the obesity model and were not included in the present study. RG rats were divided into the CG (n = 12) and CEG (n = 12).

Obesity rat model RG rats were fed a regular diet (2.90 kcal/g, containing 13% fat, 60% sugar, and 27% protein); HG rats were fed a high-fat diet (4.20 kcal/g, containing 45% fat, 33% sugar, and 22% protein). After 12 weeks, HG rats that weighed 20% heavier than the average weight of RG rats were categorized as obese. Rats in the CEG and OEG were subjected to an exercise intervention using a treadmill, whereas rats in the CG and OG were placed on an immobile treadmill for the same amount of time but did not exercise.

Food preference test After completing the 12-week period of specific diets, rats in all groups retained their original dietary habits and were provided with an additional solution of 15% sucrose and milk with 5 % fat. The food preference test was administered during the last week of the exercise intervention. The daily average consumption of the sucrose solution, milk, and food, along with the caloric intake from the various energy sources was calculated for one week. Moreover, after 24 hours of fasting, the proportions of caloric intake from the different energy sources were calculated at different intervals within 24 hours upon the resumption of a normal diet.

Exercise intervention plan The exercise intervention was carried out using an electric treadmill with 0 slope. The running speed was set at 5 m/min for 10 min prior to the exercise intervention, 15 m/min for 25 min during exercise, and 18 m/min for the last 10 min of the session. Every training session lasted about 40 min, with an intensity of approximately 50-70 % VO2 max. One training session per day was carried out on weekdays, with rest during the weekend. The training lasted for 8 weeks. All rats underwent 2-3 adaptive training sessions before the official exercise intervention.

Sample collection and handling Six rats were randomly selected from each group to test blood glucose levels, using a blood glucose 10


Page 11 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


meter, and insulin levels using enzyme-linked immunosorbent assay. One week before the end of the experiment, we tested the insulin resistance. The rats first went through 4 h of fasting, and their blood glucose concentrations were measured either immediately, or 10, 30, and 60 min after intraperitoneal injection of 0.75 U/kg insulin. Forty-eight hours after the last exercise session, 10% chloral hydrate was injected intraperitoneally (0.50 mL/100 g) to anesthetize rats, followed by a blood sample collection from their hearts. Serum was separated and stored at 4 °C until use. For tissue collection, the thoracic cavity of each rat was opened before a needle was inserted into the left ventricle for saline perfusion; simultaneously, the right auricle was cut open for drainage until the fluid flowing out was clear. The brain was immediately harvested and stored at -80 °C before testing. The intact visceral fat surrounding the epididymis, kidneys, groin, and omentum was isolated and weighed. After weighing the rats, the percent of visceral fat was calculated according to the following equation: percent of visceral fat = (total fat content from the four areas/body weight) x 100%.

Microdialysis After week 12, 6 rats were randomly selected from each group and anesthetized with 10% chloral hydrate (0.50 mL/100 g) via intraperitoneal injection. They were then stabilized using a stereotaxic apparatus. A guiding catheter was then inserted into the core of the right nucleus accumbens (AP: +1.6 mm, ML: 1.8 mm, DV: -7.2 mm) and fixated by screws and dental cement. The location of microdialysis probe in the rats’ brain as shown figure 1. Normal feeding was resumed 24 h after the rats regained consciousness. During dialysis, a probe needle was inserted into the guiding catheter and connected to the microdialysis pump. Afterwards, artificial cerebral fluid was pumped continuously at a speed of 2 μL/min into the nucleus accumbens. A sample was collected every 30 min. Under baseline conditions, each animal was given 5% milk without or with insulin (0.5 μg/kg) dissolved in 10% DMSO, which was pumped into the nucleus accumbens along with the artificial cerebral fluid. The samples collected were then stored at -80 °C until testing. After the experiment, the recycling rate of the probe needle was analysed. The localization of the microdialysis probe and guiding catheter in the target site was verified after the experiment by histological examination of sections of brain tissue fixed with paraformaldehyde.3 High-performance liquid chromatography and electrochemistry were performed to analyse DA levels in the dialysis solution. The chromatographic conditions are set as follows. The composition of mobile phase includes 11



Page 12 of 21

sodium dihydrogen dihydrate solution of 0.09mol/L, sodium octyl sulfonate of 0.0017mol/L, citric acid monohydrate of 0.05mol/L and ethylenediamine tetraacetic acid of 50 mol/L, after the above-mentioned solution is mixed, the pH is adjusted to 3.00 using phosphoric acid. The mobile phase B is methanol, methanol is filtered by 0.22 um organic membrane, degassed by ultrasonic vibration. The flow rate is set to 0.2 ml/min, the column temperature is 30℃, the working electrode is glass carbon electrode, the reference electrode is Ag/AgCl electrode, and the working voltage is set to 0.52 Volt. The standard solution of dopamine was configured as follows. The standard substance of dopamine was accurately weighted, and the standard dopamine solution was prepared into 10, 1, 0.1, 0.01, 0.001 mol/L with artificial cerebrospinal fluid. The concentration of dopamine in the sample was calculated by plotting the standard curve of concentration-peak area according to the peak area corresponding to the standard substance.

Western blotting The expression levels of InsR, Akt, GSK3-β, and of their phosphorylated forms in the nucleus accumbens were evaluated by western blot. The striatum was rapidly isolated from the frozen brain tissue on ice. An appropriate amount of the tissue was then placed into a pre-cooled homogenizer for thorough and even homogenization. The homogenate was allowed to lyse for 30 min on ice, followed by centrifugation at 12000 rpm/min for 10 min. Afterwards, the supernatant was retrieved and aliquoted. The BCA protein assay was carried out to quantify the protein concentration in the supernatant. Samples were separated by SDS-PAGE gel electrophoresis, proteins were transferred to membranes (PVDF, Abcam, Cambridge, UK), and membranes were incubated with diluted monoclonal antibodies (InsR: 1/200, Akt: 1/500, p-Akt-Thr308: 1/500, GSK3-β: 1/1000, pGSK3-β-Ser9: 1/500; Abcam, Cambridge, UK) overnight at room temperature. Secondary antibodies labelled with horseradish peroxidase (1/5000; ZSJQ, China) were added for 2 h at room temperature. After incubation and exposure, the membrane was developed using chemiluminescent ECL reagent (Abcam, Cambridge, UK). As a loading control, we used β-actin (43 kDa) antibody on the same PVDF membrane. After developing the film, the Quantity One software (Bio-Rad Laboratories, Inc. California, Berkeley, USA) was used to analyse the optical density.

Statistical analysis The SPSS 18.0 statistical software (SPSS Inc, Chicago, IL, USA) was applied for all statistical 12


Page 13 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


analyses, and quantitative data were expressed as mean ± standard deviation. Data were analyzed using independent sample t test or one-way ANOVA with pair wise comparisons (least significant difference test). The microdialysis data were analyzed using two-way ANOVA (group × time) with repeated measures and Fisher post hoc analysis, with P < 0.05 indicating significant differences.

Abbreviations Akt, protein kinase B; CG, control group; CEG, control-exercise group; DA, dopamine; GSK3, glycogen synthase kinase 3; HG, high-fat diet group; HOMA-IR, homeostasis model assessment of insulin resistance; InsR, insulin receptor; OEG, obese-exercise group; OG, obese group; PI3K, phosphoinositide 3-kinase; VTA, ventral tegmental area

Author Information Corresponding Authors: [email protected] Postal address: College of Physical Education and Sports, Beijing Normal University, No. 19, XinJieKouWai St., HaiDian District, Beijing 100875, P. R. China

Author Contributions Wei Chen collected and analysed the experimental data. Wei Chen and Juan Li wrote the paper. Wei Chen, Juan Li, Li Juan Hou and Jun Liu participated in study design. Wei Chen and Da Lei Wang participated in microdialysis experiment.

Funding Sources This work was supported by the Research Initiation Fund for Doctors (Post) of Hebei Normal University (L2018B24). This work was also supported by the National Natural Science Fund (31401018).

Conflict of Interest The authors declare no competing financial interest.

Acknowledgment We would like to thank Dr. Dong Sheng Yang at the Centre for Physical Health Research of Zhe 13



Page 14 of 21

Jiang University of Technology, and Dr. Yan Ru Hu at PE Department of Central South University, for their expert technical assistance and valuable suggestions. References (1) Liang, N. C., Bello,N. T., and Moran, T. H. (2015) Wheel running reduces high-fat diet intake, preference and mu-opioid agonist stimulated intake. Behav. Brain. Res. 284, 1-10. (2) Narayanaswami, V., Thompson, A. C., Cassis, L. A., Bardo, M.T., and Dwoskin, L.P. (2013) Diet-induced obesity: dopamine transporter function, impulsivity and motivation. Int. J. Obes (Lond). 37, 1-20. (3) Sumei, Hu., Lu, W., Dengbao,Y., Li, L., Jacques,Togo., Yingga, Wu., Quansheng, Liu., Baoguo, Li., Min, L., Guanlin, W., Xueying, Z., Chaoqun, N., Jiabao, L., Yanchao, X., Elspeth, C., Andrew, W. D., Mohsen, M., Lijuan, L., Shengnan, W., Alex, D., and John, R.S. (2018) Dietary fat, but not protein or carbohydrate, regulates energy intake and causes adiposity in mice. Cell. Metab. 28, 415-437. (4) Geiger, B.M., Haburcak, M., Avena, N. M., Moyer, M. C., Hoebel, B. G., and Pothos, E. N. (2009) Deficits of mesolimbic dopamine neurotransmission in rat dietary obesity. Neuroscience. 159, 1193-1199. (5) Volkow, N. D., Wang, G. J., and Baler, R. D. (2011) Reward, dopamine and the control of food intake: implications for obesity. Trends. Cogn. Sci. 15, 37-46. (6) Van der Plasse, G., Van Zessen, R., Luijendijk, M. C. M., Erkan, H., Stuber, G. D., Ramakers, G. M. J., and Adan, R. A. H. (2015)Modulation of cue-induced firing of ventral tegmental area dopamine neurons by leptin and ghrelin. Int. J. Obe. 39, 1742–1749. (7) Kelley, A.E. (2004) Ventral striatal control of appetitive motivation: role in ingestive behavior and reward-related learning. Neurosci. Biobehav. Rev. 27, 765-776. (8) Figlewicz, D.P., Evans, S.B., Murphy, J., Hoen, M., and Baskin, D.G. (2003) Expression of receptors for insulin and leptin in the ventral tegmental area/substantia nigra (VTA/SN) of the rat. Brain. Res. 964, 107–115. (9) Mebel, D. M., Wong, J. C. Y., Dong, Y. J., and Borgland S. L. (2012) Insulin in the ventral tegmental area reduces hedonic feeding and suppresses dopamine concentration via increased reuptake. Eur. J. Neurosci. 36, 2336-2346. (10) Stouffer, M. A., Woods, C. A., Patel, J. C., Lee, C.R., Witkovsky, P., Bao, L., Machold, R.P., Jones, K. T., Vaca, S.C.D., Reith, M.E.A., Carr, K.D., and Rice, M. E. (2015) Insulin enhances striatal dopamine release by activating cholinergic interneurons and thereby signals reward. Nat. Commun. 6, 8543. (11) Johnson, P. M., and Paul, J. K. (2010) Addiction-like reward dysfunction and compulsive eating in obese rats: Role for dopamine D2 receptors. Nat. Neurosci. 13, 635–641. (12) Le, S., Mao, L., Lu, D., Yang, Y., Tan, X., Wiklund, P., and Cheng, S. (2016) Effect of aerobic exercise on insulin resistance and central adiposity disappeared after the discontinuation of intervention in overweight women. J. Sport. Health. Sci. 5, 166-170. (13) Zhao, R. R., O’Sullivan, A.J., and Singh, M. A. F. (2018) Exercise or physical activity and cognitive function in adults with type 2 diabetes, insulin resistance or impaired glucose tolerance: a systematic review. Eur. Rev. Aging. Phys. Act. 15, 1. (14) Shih, K.C., and Kwok, C. F. (2018) Exercise reduces body fat and improves insulin sensitivity and pancreatic β-cell function in overweight and obese male Taiwanese adolescents. BMC. Pediatr.18, 80. (15) Swift, D.L., Johannsen, N. M., Lavie, C. J., Earnest, C. P., Blair, S. N., and Church, T. S. 14


Page 15 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(2016) Effects of clinically significant weight loss with exercise training on insulin resistance and cardiometabolic adaptations. Obesity (Silver Spring). 24, 812–819. (16) Chen, W., Wang, H.J., Shang, N.N., Liu, J., Li, J., Tang, D.H., and Li, Q. (2017) Moderate intensity treadmill exercise alters food preference via dopaminergic plasticity of ventral tegmental area-nucleus accumbens in obese mice. Neurosci. Lett. 641, 56-61. (17) Stice, E., Yokum, S., Blum, K., and Bohon, C. (2010) Weight gain is associated with reduced striatal response to palatable food. J. Neurosci. 30,13105-3109. (18) Tellez, L.A., Medina, S., Han, W., Ferreira, J.G., Limón, P.L., Ren, X., Lam, T.T., Schwartz, G. J., and Araujo, I.E. (2013) A gut lipid messenger links excess dietary fat to dopamine deficiency. Science. 341, 800-802. (19) Stice, E., Figlewicz, D.P., Gosnell, B.A., Levine, A. S., and Pratt, W. E. (2013) The contribution of brain reward circuits to the obesity epidemic. Neurosci. Biobehav. Rev. 37, doi:10.1016/j.neubiorev.2012.12.001. (20) Stice, E., Spoor, S., Bohon, C., and Small, D.M. Relation between obesity and blunted striatal response to food is moderated by TaqlA1 gene. Science. 322,449-452. (21) Kenny, P.J. (2011) Reward mechanisms in obesity: new insights and future directions. Neuron. 69, 664–679. (22) Foley, T. E., and Fleshner, M. (2008). Neuroplasticity of dopamine circuits after exercise: implications for central fatigue. Neuromolecular Med. 10, 67-80. (23) Heijnen, S., Hommel, B., Kibele, A., and Colzato, L. S. (2016) Neuromodulation of aerobic exercise-a review. Front. Psychol. 6, 1890. (24) Moody, L., Liang, J., Choj, P. P., Moran, T.H., and Liang, A. C. (2015) Wheel running decreases palatable diet preference in Sprague-Dawley rats. Physiol. Behav. 150, 53-63. (25) Pipatpiboon, N., Pintana, H., Pratchayasakul, W., Chattipakorn, N., and Chattipakorn, S. C. (2013) Dpp4-inhibitor improves neuronal insulin receptor function, brain mitochondrial function and cognitive function in rats with insulin resistance induced by high-fat diet consumption. Eur. J. Neurosci. 37, 839-849. (26) Woods, S.C., Seeley, R.J., Baskin, D.G., and Schwartz, M.W. (2003) Insulin and the blood brain barrier. Curr. Pharm. Des. 9, 795-800. (27) Banks, W. A. (2004) The source of cerebral insulin. Eur. J. Pharmacol. 490, 5–12. (28) Nassis,G.P., Papantakou, K.S., Triandafillopoulou, M., Kavouras, S.A., Yannakoulia, M., Chrousos, G.P., and Sidossis, L.S. (2005) Aerobic exercise training improves insulin sensitivity without changes in body weight, body fat, adiponectin, and inflammatory markers in overweight and obese girls. Metabolism. 54, 1472-1479. (29) Fordahl, S. C., and Jones, S. R. (2017) High-fat-diet-induced deficits in dopamine terminal function are reversed by restoring insulin signalling. ACS. Chem. Neurosci. 8, 290-299. (30) Vogt, M. C., and Bruning, J. C. (2013) CNS insulin signaling in the control of energy homeostasis and glucose metabolism-from embryo to old age. Trends. Endocrinol. Metab. 24, 76–84. (31) Shi, X., Miller, J.S., Harper, L.J., Poole, R.L., Gould, T.J., and Unterwald, E. M. (2014) Reactivation of cocaine reward memory engages the Akt/GSK3/mTOR signaling pathway and can be disrupted by GSK3 inhibition. Psychopharmacology. 231, 3109-3118. (32) Kim, W.Y., Jang, J.K., Lee, J.W., Jang, H., and Kim, J.H. (2013) Decrease of GSK3β phosphorylation in the rat nucleus accumbens core enhances cocaine-induced hyper-locomotor acitivity. J. Neurochem. 125, 642-648. (33) Peineau,S., Bradley, C., Taghibiglou, C., Doherty, A., Bortolotto, Z.A., Wang, Y.T., and 15



Page 16 of 21

Collingridge, G.L. (2008) The role of GSK-3 in synaptic plasticity. Br. J. Pharmacol. 153, S428-S437. (34) Niswender, K.D., Daws, L.C., Avison, M. J., and Galli, A. (2011) Insulin regulation of monoamine signaling: pathway to obesity. Neuropsychopharmacology. Reviews. 36, 359-360. (35) Speed, N., Saunders, C., Davis, A.R., Owens, W.A., Matthies, H.J.G., Saadat, S., Kennedy, J.P., Vaughan, R.A., Neve, R.L., Lindsley, C.W., Russo, S.J., Daws, L.C., Niswender, K.D., and Galli, A. (2011) Impaired striatal Akt signaling disrupts dopamine homeostasis and increases feeding. PLoS. One. 6, e25169.

16


Page 17 of 21

Figure legends

Figure 1. The location of microdialysis probe implantation in the of rats’ nucleus accumbens.

B

††

## Exercise intervention period

** Week

C RG HG CG CEG OG OEG

**

Modeling period

**

** ** ** **

Exercise intervention period

**

*

*

CG CEG OG OEG

D Visceral fat (%)

Body weight (g)

Modeling period

Weight change (%)

A

Energy efficiency (%)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


** **

*

ns

CG CEG OG OEG

Figure 2. An 8-week aerobic exercise intervention inhibits weight gain, decreases energy efficiency, and improves body composition in rats consuming high-fat diet. (A) Comparison of body weight changes in different groups of rats. **Significant differences between HG and RG (P < 0.01). †† Significant differences between OG and CG (P < 0.01). ## Significant differences between OEG and OG (P < 0.01). (B) Comparison of body weight change rates in different groups of rats. * Significant differences (P < 0.05), ** Significant differences (P < 0.01). (C) Comparison of energy efficiency in different groups of rats in different periods. ** Significant differences (P < 0.01). (D) Comparison of percent of visceral fat in different groups of rats. *Significant differences (P < 0.05), ** Significant differences (P < 0.01). Data are represented as mean ± standard deviation. RG: regular diet group and HG: high-fat diet group. CG: control group; CEG: control exercise group; OG: obese group; OEG: obese + exercise group. 17



Page 18 of 21

Figure 3. An 8-week aerobic exercise intervention reduces fat preference and weakens the impulse to feed in obese rats. (A) Comparison of food preference changes in different groups of rats. ** Significant differences (P < 0.01). Food preference = (the food energy intake / total energy intake) × 100%. (B) Comparison of feeding ratios of rats in different time periods after 24 h of fasting, normal feeding was resumed for 24 h. * Significant differences (P < 0.05), ** significant differences (P < 0.01). Data are represented as mean ± standard deviation. CG: control group; CEG: control exercise group; OG: obese group; OEG: obese + exercise group. .

Figure 4. An 8-week aerobic exercise intervention improves insulin resistance in obese rats. (A) Comparison of insulin tolerance in different groups of rats. ** Significant differences between OG and CG (P < 0.01), †† significant differences between OEG and OG (P < 0.01). (B) Comparison of the area under the curve (AUC) of the plasma glucose in different groups of rats. ** Significant differences (P < 0.01). (C) Comparison of homeostasis model assessment of insulin resistance (HOMA-IR) in different groups of rats. * Significant differences (P < 0.05), ** significant differences (P < 0.01). Data are represented as mean ± standard deviation. CG: control group; CEG: control exercise group; OG: obese group; OEG: obese + exercise group. .

18


Page 19 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 5. An 8-week aerobic exercise intervention effectively increases DA levels changes induced by milk containing 5% fat in the nucleus accumbens of obese rats. (A) Comparison of dynamic DA levels changes in the nucleus accumbens induced by milk containing 5% fat in different groups of rats. ** Significant differences between OG and CG (P < 0.01), † significant differences between OEG and OG (P < 0.05), †† significant differences between OEG and OG (P < 0.01). (B) Comparison of DA levels in nucleus accumbens before and after consumes milk containing 5% fat in different groups of rats. * Significant differences between post-injection and pre-injection (P < 0.05), ** significant differences between post-injection and pre-injection (P < 0.01). In figure 5B, each different symbol represents a specific rat, and two identical symbols represent the same rat. (C) Comparison of DA change rate in nucleus accumbens before and after consumes milk containing 5% fat in different groups of rats. ** Significant differences (P < 0.01). The gray rectangle represents giving milk. Data are represented as mean ± standard deviation. The pre and post refers to before and after consumption of milk, respectively. CG: control group; CEG: control exercise group; OG: obese group; OEG: obese + exercise group.

Figure 6. An 8-week aerobic exercise intervention effectively increases DA levels insulin-induced changes in the 19



Page 20 of 21

nucleus accumbens of obese rats during feeding. (A) Comparison of dynamic DA levels changes in the nucleus accumbens induced by insulin injection in different groups of rats during feeding. **Significant differences between OG and CG (P < 0.01). †† Significant differences between OEG and OG (P < 0.01). (B) Comparison of DA levels in nucleus accumbens pre and post insulin injection during feeding in different groups of rats. ** Significant differences between post-injection and pre-injection (P < 0.01). In figure 6B, each different symbol represents a specific rat, and two identical symbols represent the same rat. (C) Comparison of the ratio of DA level in the nucleus accumbens post- to pre-insulin injection in different groups of rats. ** Significant differences (P < 0.01). The gray rectangle represents giving milk, and the arrow represents insulin injection. Data are represented as mean ± standard deviation. The pre and post refers to before and after insulin injection, respectively. CG: control group; CEG: control exercise group; OG: obese group; OEG: obese + exercise group.

Figure 7. An 8-week aerobic exercise intervention effectively improved the regulation effect of insulin on DA level of nucleus accumbens in obese rats. Compared to before receiving milk (0 min), * significant differences (P < 0.05), and ** significant differences (P < 0.01). Compared to before receiving milk + insulin (0 min), # significant differences (P < 0.05). Compared to 30 minutes after receiving milk + insulin (30 min), †† significant differences (P < 0.01). Data are represented as mean ± standard deviation. CG: control group; CEG: control exercise group; OG: obese group; OEG: obese + exercise group.

Figure 8. An 8-week aerobic exercise intervention promotes insulin receptor (InsR)/protein kinase B (Akt)/glycogen synthase kinase 3 (GSK3) signalling pathways in the nucleus accumbens of obese rats. Comparison of proteins expression in the nucleus accumbens in different groups of rats. Compared to CG, ** significant differences (P < 0.01), compared to OG, † significant differences (P < 0.05), †† significant differences (P < 0.01). Data are represented as mean ± standard deviation. CG: control group; CEG: control exercise group; OG: obese group; OEG: obese + exercise group. 20


Page 21 of 21 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


For Table of Contents Use Only Manuscript title: Aerobic exercise improves food reward systems in obese rats via insulin signalling regulation of dopamine levels in the nucleus accumbens Author names: Wei Chen, Juan Li, Jun Liu, Da L. Wang, Li J. Hou

21


Aerobic Exercise Improves Food Reward Systems in Obese Rats via

Recommend Documents