Overweight and obesity are defined as abnormal or excessive fat accumulation that may impair health(1). In 2008, almost a quarter of adults (24 % of men and 25 % of women aged 16 years or over) in England were classified as obese (BMI 30 kg/m2 or over)(2), while in Brazil, 13·9 % of adults are obese(3). The excessive accumulation of adipose tissue that characterises obesity is considered an important risk factor for the development of diseases such as type 2 diabetes mellitus(Reference Chan, Rimm and Colditz4, Reference Colditz, Willett and Rotnitzky5). The rise in the number of cases of obesity appears to be the result of changes in food intake, eating habits and physical activity levels. Recent data indicate that the food service industry has been gradually increasing the sizes of current marketplace foods(Reference Young and Nestle6). This increase in portion sizes has been accompanied by an increase in food consumption and energy intake(Reference Nielsen, Seiga-Riz and Popkin7), leading to obesity. Furthermore, some authors have suggested that the obesity epidemic observed mainly in the USA may have been the result of high soft drink intake(Reference Malik, Popkin and Bray8). The cafeteria diet is a widely used animal model to study obesity, consisting of a variety of palatable foods associated with the obesity epidemic(Reference Sclafani and Springer9–Reference Sampey, Vanhoose and Winfield11) that produce hyperphagia and high energy intake(Reference Shafat, Murray and Rumsey12), body-weight gain, increased adipose tissue mass, glucose intolerance, insulin resistance, hyperinsulinaemia, and pro-inflammatory response(Reference Sampey, Vanhoose and Winfield11).
Overweight and obese individuals who lose as little as 10 % of their body weight can reduce their risk factors for type 2 diabetes mellitus and CVD(13, Reference Orzano and Scott14). Restriction of energy intake is the most commonly used strategy for weight loss in humans. Guidelines for obesity management emphasise the reduction of energy intake to promote a negative energy balance that results in weight loss(Reference Logue, Thompson and Romanes15). However, weight regain after long-term energy restriction is the main challenge of interventions that restrict overall energy intake. During energy restriction, mice(Reference Hambly and Speakman16) and human subjects(Reference Weyer, Walford and Harper17) have shown sustained reductions in energy expenditure, which may predispose subjects to a positive energy balance when food is made available post-restriction. Moreover, sustained restriction is associated with neuroendocrine changes promoting hunger, leading to hyperphagia(Reference Hambly, Mercer and Speakman18), thus making mass regain even more likely.
Physical exercise is recommended to increase energy expenditure in order to generate a negative energy balance, assisting in weight loss and maintenance during the treatment of obesity(Reference Jakicic and Otto19). In addition, physical activity can alter patterns of substrate utilisation and enhance whole-body insulin sensitivity(Reference Yamanouchi, Shinozaki and Chikada20), improving the overall metabolic profile. In experimental models, exercise has attenuated weight gain in animals fed a high-fat diet(Reference Wang, Chen and Wang21, Reference Medeiros, Frederico and da Luz22) by reducing food intake compared with obese animals that do not perform exercise(Reference Wang, Chen and Wang21). However, there remains considerable debate about the effects of exercise in relation to appetite, and the evidence so far has pointed to the fact that some individuals are capable of tolerating exercise and controlling appetite, while others are resistant to exercise and compensate for the energy expended during exercise by increasing energy intake(Reference Hopkins, King and Blundell23).
Thus, given the general lack of success in treating obesity and the fact that dietary changes and exercise are indicated for the treatment of obesity and metabolic disorders, the present study aimed to evaluate the effect of dietary change and physical exercise on alterations induced by a cafeteria diet.
Experimental methods
Animal procedures
The experimental protocol was approved by the Research Ethics Committee and Animal Care and Use Committee of Universidade Federal do Rio Grande do Sul (protocol no. 200938). All animals were housed in plastic cages (four rats/cage) in a 12 h light–dark cycle at 24–26°C. The animals were housed on shavings and had ad libitum access to chow and water at all times. The experimental design is shown in Fig. 1. A total of fifty-six female Wistar rats (aged 3 weeks) were randomly allocated into five experimental groups to be fed the following: (1) a control chow diet and water alone for 34 weeks (control chow+sedentary; CON-CON+SED; n 11); (2) a control chow diet and water alongside a random selection of highly energetic and palatable human foods for 34 weeks (cafeteria diet+sedentary; CAF-CAF+SED; n 10); (3) a control chow diet and water alongside a random selection of highly energetic and palatable human foods for 26 weeks, followed by a control chow diet and water alone for 8 weeks (food pattern change+sedentary; CAF-CON+SED; n 12); (4) a control chow diet and water alongside a random selection of highly energetic and palatable human foods for 34 weeks plus physical exercise starting at week 26 for 8 weeks (cafeteria diet+physical exercise; CAF-CAF+EX; n 12); (5) a control chow diet and water alongside a random selection of highly energetic and palatable human foods for 26 weeks, followed by a control chow diet and water alone plus physical exercise starting at week 26 for 8 weeks (food pattern change+physical exercise; CAF-CON+EX; n 11). The foods included in the cafeteria diet are described in Table 1. Foods were provided in excess amounts and changed daily, to maintain variety, by replacing four of the foods with new items. Soft drink was provided daily. Hence, the animals did not receive the same foods for more than two consecutive days at a time. The chow-diet and cafeteria-diet foods were individually weighed in and out of the cage between 09.00 and 10.00 hours daily. The daily data obtained for each food item were summed, generating the total consumption per d. Total daily intakes were summed to obtain a total weekly intake, which was then divided by the number of days of the week and the number of animals to reach the amount of food intake per animal per d of the week. Daily intakes of energy, macronutrients and Na were calculated from the manufacturers' data (Table 1). The animals were weighed on alternate days between 09.00 and 10.00 hours.
Schematic diagram of the experimental groups. CON-CON+SED, control chow+sedentary; CAF-CAF+SED, cafeteria diet+sedentary; CAF-CON+SED, food pattern change+sedentary; CAF-CAF+EX, cafeteria diet+physical exercise; CAF-CON+EX, food pattern change+physical exercise.
Food available to the animals in the cafeteria diet
Physical exercise protocol
The exercise protocol was adapted from a previous study(Reference Melton, Hegsted and Keenan24) and consisted of swimming in plastic barrels (diameter, 51 cm). The water depth was approximately 50 cm, and the temperature ranged between 31 and 33°C. The swimming protocol consisted of a 4-week adaptation period and a 4-week training period. The rats swam 5 d/week in the morning at the beginning of the light period. Animals swam for 5 min on the first day, and swimming time was increased by 5 min/d up to 40 min for a 2-week period. Training time was increased to 60 min for 1 week and then to 75 min for the remaining 4 weeks to ensure an aerobic training effect. Rats swam in groups of four in plastic barrels and were constantly supervised and encouraged to keep moving in the water by gently prodding them if needed. Rats swam freely with no additional weights (unloaded). The exercise oxygen consumption of non-weighted swimming rats has previously been determined to be 2·7 times the resting value(Reference McArdle25), which is considered to be moderate-intensity exercise for rats(Reference Baker and Horvath26). At the end of each daily swimming session, each rat was towel-dried and returned to its cage. Rats were watched carefully after being placed in their cages to ensure that they did not shiver, indicating hypothermia stress.
Blood sampling and tissue collection
At the end of the experiment, after 12–13 h of fasting, the animals were decapitated by a guillotine. Blood was collected in previously heparinised tubes and centrifuged at 1609 g for 20 min. After centrifugation, plasma was separated into aliquots and stored at − 80°C until biochemical analysis was performed. Liver and abdominal visceral adipose tissue were removed and weighed immediately after decapitation.
Plasma insulin and glucose
Plasma insulin was measured in duplicate by ELISA using reagents specific for the rat (Millipore), with a detection sensitivity of 0·2 ng/ml. The homeostasis model assessment of insulin resistance (HOMA-IR) was used to calculate approximate insulin resistance (fasting glucose (mg/dl) × fasting insulin (μIU/ml)/2430)(Reference Cacho, Sevillano and de Castro27). Plasma glucose was measured with a colorimetric method (Glucose PAP Liquiform; Labtest).
Statistical analysis
All data were analysed using GraphPad Prism, version 5.04 (GraphPad Software, Inc.). The effects of different treatments on food intake, body weight, adipose tissue, liver and fasting insulin and glucose at the end of the experiment (week 34) were analysed using one-way ANOVA followed by the Student–Newman–Keuls post hoc test. Data on total energy intake before dietary change and/or the introduction of physical exercise (week 26) were analysed using one-way ANOVA followed by the Student–Newman–Keuls post hoc test. The effect of time and treatment (time v. treatment) on total energy intake over the 8 weeks of treatment (weeks 27 to 34) was analysed by repeated-measures ANOVA followed by the Bonferroni post hoc test. Values are expressed as means with their standard errors. P < 0·05 was considered to be statistically significant.
Results
Nutrient intake
Daily chow intake per animal was significantly higher in the CAF-CON+EX group compared with the other groups at the end of the experiment (week 34; P < 0·05; Table 2). Moreover, daily standard chow intake per animal was significantly lower in the CAF-CAF+SED, CAF-CON+SED and CAF-CAF+EX groups compared with the CON-CON+SED group in the same period (P < 0·05; Table 2). Daily water intake per animal at week 34 was significantly higher in the CAF-CON+EX group compared with the other groups (P < 0·05; Table 2). In the same period, daily water intake per animal was significantly higher in the CON-CON+SED and CAF-CON+SED groups compared with the CAF-CAF+SED and CAF-CAF+EX groups (P < 0·05; Table 2). However, water intake was significantly higher in the CAF-CAF+EX group than in the CAF-CAF+SED group at week 34 (P < 0·05; Table 2). Total daily fluid intake per animal was significantly higher in the CAF-CAF+SED and CAF-CAF+EX groups compared with the other groups at week 34 (P < 0·05; Table 2). In addition to higher water intake, the CAF-CAF+EX group also showed significantly higher total fluid intake than the CAF-CAF+SED group (P < 0·05; Table 2). The CAF-CON+EX group showed significantly higher total fluid intake at week 34 compared with the CON-CON+SED and CAF-CON+SED groups (P < 0·05; Table 2). Daily Na intake per animal at week 34 was significantly higher in the CAF-CAF+SED and CAF-CAF+EX groups compared with the other groups (P < 0·05; Table 2), while the CAF-CON+EX group showed a significantly higher Na intake than the CAF-CON+SED group at week 34 (P < 0·05; Table 2).
Daily intake of standard chow, fluids, nutrients and energy at the end of the experiment (week 34) (Mean values with their standard errors)
CON-CON+SED, control chow+sedentary; CAF-CAF+SED, cafeteria diet+sedentary; CAF-CON+SED, food pattern change+sedentary; CAF-CAF+EX, cafeteria diet+physical exercise; CAF-CON+EX, food pattern change+physical exercise.
a,b,c,d Mean values within a row with unlike superscript letters were significantly different (P < 0·05; one-way ANOVA with Student–Newman–Keuls post hoc test).
Daily carbohydrate intake per animal at week 34 of the experiment was significantly higher in the CAF-CAF+EX group compared with the CON-CON+SED, CAF-CAF+SED and CAF-CON+SED groups (P < 0·05; Table 2). In addition, both CAF-CON+EX and CAF-CAF+SED groups had a significantly higher carbohydrate intake compared with the CON-CON+SED and CAF-CON+SED groups at week 34 (P < 0·05; Table 2), the latter showing the lowest carbohydrate intake among all groups. Daily protein intake per animal at week 34 was significantly lower in the CAF-CAF+SED and CAF-CAF+EX groups compared with the other groups (P < 0·05; Table 2). Moreover, protein intake was significantly higher in the CAF-CON+EX group compared with the other groups (P < 0·05; Table 2), while protein intake was the same for the CON-CON+SED and CAF-CON+SED groups at week 34 (P < 0·05; Table 2). Daily fat intake per animal was significantly higher in the CAF-CAF+SED and CAF-CAF+EX groups compared with the other groups at week 34 (P < 0·05; Table 2). Daily fat intake per animal was similar for the CON-CON+SED, CAF-CON+SED and CAF-CON+EX groups at the end of the experiment (P < 0·05; Table 2). Daily energy intake per animal at week 34 was significantly higher in the CAF-CAF+EX group compared with the other groups (P < 0·05; Table 2). The CAF-CAF+SED and CAF-CON+EX groups showed higher daily energy intake per animal at week 34 compared with the CON-CON+SED and CAF-CON+SED groups (P < 0·05; Table 2).
Before the introduction of the treatment (week 26), daily energy intake per animal was significantly higher for all groups compared with the CON-CON+SED group (P < 0·05; Fig. 2(A)). After dietary change and/or the introduction of physical exercise, daily energy intake per animal over the last 8 weeks of the experiment (weeks 27–34) was significantly affected by time and treatment (F(28,210) = 6·19; P < 0·05; Fig. 2(B)). In the CAF-CAF+SED group, daily energy intake per animal was significantly higher compared with the CON-CON+SED (P < 0·05; weeks 27–34; Fig. 2(B)), CAF-CON+SED (P < 0·05; weeks 27–34; Fig. 2(B)) and CAF-CON+EX groups (P < 0·05; except week 34; Fig. 2(B)). In the CAF-CAF+EX group, daily energy intake per animal was significantly higher compared with the CON-CON+SED (P < 0·05; weeks 27–34; Fig. 2(B)), CAF-CAF+SED (P < 0·05; only at week 34; Fig. 2(B)), CAF-CON+SED (P < 0·05; weeks 27–34; Fig. 2(B)) and CAF-CON+EX groups (P < 0·05; weeks 27–34; Fig. 2(B)). In the CAF-CON+SED group, daily energy intake per animal was significantly lower compared with the CON-CON+SED (P < 0·05; only at week 27; Fig. 2(B)) and CAF-CON+EX groups (weeks 32–34; P < 0·05). In the CAF-CON+EX group, daily energy intake per animal was significantly lower than that in the CON-CON+SED group (only at weeks 27 and 29), but energy intake was significantly higher at week 34 (P < 0·05; Fig. 2(B)).
Daily energy intake per animal (kJ/d) (A) before dietary change and/or the introduction of physical exercise (week 26) and (B) over the 8 weeks of treatment (weeks 27–34). Values are means, with standard errors represented by vertical bars. CON-CON+SED, control chow+sedentary (n 11; 
); CAF-CAF+SED, cafeteria diet+sedentary (n 10; 
); CAF-CON+SED, food pattern change+sedentary (n 12; 
); CAF-CAF+EX, cafeteria diet+physical exercise (n 12; 
); CAF-CON+EX, food pattern change+physical exercise (n 11; 
). In (A), the CAF-CAF+SED, CAF-CON+SED, CAF-CAF+EX and CAF-CON+EX groups had access to the cafeteria diet. One-way ANOVA (week 26) and repeated-measures ANOVA with Bonferroni post hoc test (weeks 27–34). * Mean values were significantly different from those of the CON-CON+SED group (up to week 26; P < 0·05). † Mean values were significantly different from those of the CON-CON+SED (weeks 27–34), CAF-CON+SED (weeks 27–34), CAF-CAF+EX (only at week 34) and CAF-CON+EX groups (except week 34) (P < 0·05). ‡ Mean values were significantly different from those of the CON-CON+SED (weeks 27–34), CAF-CON+SED (weeks 27–34) and CAF-CON+EX groups (weeks 27–34) (P < 0·05). § Mean values were significantly different from those of the CON-CON+SED (only at week 27) and CAF-CON+EX groups (weeks 32–34) (P < 0·05). ∥ Mean values were significantly different from those of the CON-CON+SED group (weeks 27, 29 and 34; P < 0·05).
Daily energy intake of solid foods per animal at week 34 was significantly higher in the CAF-CON+EX group compared with the other groups (P < 0·05; Table 2). Conversely, the CAF-CAF+SED group showed significantly lower solid food energy intake than the CON-CON+SED and CAF-CAF+EX groups at week 34 (P < 0·05; Table 2). Solid food energy intake was similar for the CON-CON+SED, CAF-CON+SED and CAF-CAF+EX groups (P>0·05; Table 2).
Body weight and tissue weight
Body weight at week 34 of the experiment was significantly higher in the CAF-CAF+SED and CAF-CAF+EX groups compared with the other groups (P < 0·05; Fig. 3(A)). Body weight was similar for the CON-CON+SED, CAF-CON+SED and CAF-CON+EX groups at the end of the experiment (P>0·05; Fig. 3(A)). At week 34, visceral adipose tissue was significantly higher in the CAF-CAF+SED and CAF-CAF+EX groups compared with the other groups (P < 0·05; Fig. 3(B)), but similar for the CON-CON+SED, CAF-CON+SED and CAF-CON+EX groups (P>0·05; Fig. 3(B)). Liver weight at week 34 was significantly higher in the CAF-CAF+SED and CAF-CAF+EX groups compared with the other groups (Fig. 3(C); P < 0·05), but similar for the CON-CON+SED, CAF-CON+SED and CAF-CON+EX groups (Fig. 3(C); P>0·05).
(A) Body weight, (B) visceral adipose tissue (n 10–12) and (C) liver weight (n 9–12) at the end of the experiment (week 34). Values are means, with standard errors represented by vertical bars. CON-CON+SED, control chow+sedentary (n 11); CAF-CAF+SED, cafeteria diet+sedentary (n 10); CAF-CON+SED, food pattern change+sedentary (n 12); CAF-CAF+EX, cafeteria diet+physical exercise (n 12); CAF-CON+EX, food pattern change+physical exercise (n 11). a,bMean values with unlike letters were significantly different (P < 0·05; one-way ANOVA with Student–Newman–Keuls post hoc test).
Plasma glucose, insulin and homeostasis model assessment of insulin resistance
Fasting glucose concentration was similar for all groups at week 34 of the experiment (P>0·05; Fig. 4(A)). Fasting insulin concentration at week 34 was significantly higher in the CAF-CAF+SED group compared with the other groups (P < 0·05; Fig. 4(B)), but similar for the CON-CON+SED, CAF-CON+SED, CAF-CAF+EX and CAF-CON+EX groups (P>0·05; Fig. 4(B)). HOMA-IR was significantly higher in the CAF-CAF+SED group compared with the other groups at week 34 (P < 0·05; Fig. 4(C)). There was no significant difference among the CON-CON+SED, CAF-CON+SED, CAF-CAF+EX and CAF-CON+EX groups regarding HOMA-IR at week 34 of the experiment (P>0·05; Fig. 4(C)).
(A) Fasting glucose, (B) fasting insulin and (C) insulin resistance index (homeostasis model assessment of insulin resistance; HOMA-IR) at the end of the experiment (week 34). Values are means, with standard errors represented by vertical bars. CON-CON+SED, control chow+sedentary (n 11); CAF-CAF+SED, cafeteria diet+sedentary (n 10); CAF-CON+SED, food pattern change+sedentary (n 11); CAF-CAF+EX, cafeteria diet+physical exercise (n 11); CAF-CON+EX, food pattern change+physical exercise (n 11). a,b Mean values with unlike letters were significantly different (P < 0·05; one-way ANOVA with Student–Newman–Keuls post hoc test).
Discussion
Dietary change by replacing the cafeteria diet and allowing animals exclusive access to the chow diet caused an increase in the consumption of chow in relation to food intake observed in the cafeteria diet-fed animals, although chow intake was slightly lower compared with the control animals. However, dietary change did not prevent the animals from maintaining a normal intake of carbohydrates, proteins, lipids and Na, indicating that withdrawal of the cafeteria diet was sufficient to improve the nutritional quality of the animals' diet. The lower chow intake, compared with the control animals, after the withdrawal of the cafeteria diet may have been the result of reduced palatability of the diet compared with soft drinks and solid foods offered in the cafeteria diet, leading to decreased motivation to consume chow even after 8 weeks without cafeteria-diet foods. Similar experiments involving withdrawal of cafeteria-diet foods after chronic access have also reported lower energy intake when animals are exposed only to standard chow, suggesting that these animals show increased brain reward thresholds compared with those observed when they had access to the cafeteria diet(Reference Johnson and Kenny28). Thus, maintenance of increased reward thresholds may influence the intake of foods that do not cause the same sensations of pleasure caused by palatable foods.
After 8 weeks of treatment (weeks 27–34), animals subjected to dietary change showed an energy intake similar to that of the controls, except for the first week of the withdrawal of cafeteria-diet foods, when there was a dramatic reduction in energy intake, probably as a result of the lower palatability of the standard chow(Reference Johnson and Kenny28). Likewise, body weight, visceral adipose tissue and liver weight were also similar to that of the controls at the end of the experiment. Dietary change consisting of the withdrawal of palatable foods and access only to standard chow ad libitum showed that the energy intake of animals was not aimed at maintaining body weight achieved with the consumption of cafeteria-diet foods, suggesting that energy depots accumulated in the form of visceral adipose tissue might have been used, thus failing to produce a compensatory increase in energy intake. This contrasts with approaches that voluntarily restrict food intake, which may predispose subjects to a positive energy balance when food is made available post-restriction due to sustained reductions in energy expenditure(Reference Hambly and Speakman16, Reference Weyer, Walford and Harper17) and neuroendocrine changes that stimulate hunger and hyperphagia(Reference Hambly, Mercer and Speakman18), favouring mass regain. In human subjects, a dietary regimen known as the Paleolithic diet, consisting mainly of ad libitum intake of foods such as nuts, vegetables, fruits, lean meats, fish and eggs, was able to increase the feeling of satiety and promote lower energy intake(Reference Jönsson, Granfeldt and Erlanson-Albertsson29).
Regarding fasting insulin, animals subjected to dietary change showed levels similar to those of the controls probably as a result of reduced insulin resistance as evaluated by HOMA-IR. Similar insulin sensitivity may have been a reflection of both the nutritional quality of food after withdrawal of the cafeteria diet and the reduction of visceral adipose tissue, which may have triggered a reduction in TNF-α plasma concentrations, a cytokine secreted by adipose tissue that appears to cause insulin resistance, at least in part, by inhibiting intracellular signalling from the insulin receptor(Reference Feinstein, Kanety and Papa30, Reference Hotamisligil, Murray and Choy31). Moreover, removal of high-glycaemic foods, such as soft drinks, and consumption of chow components containing complex carbohydrates probably resulted in a decreased release of insulin during the postprandial period in response to small increases in plasma glucose caused by low-glycaemic index carbohydrate.
The introduction of exercise associated with dietary change for the last 8 weeks of the experiment (weeks 27–34) caused an increase in chow intake at the end of the experiment. However, exercise associated with the cafeteria diet exerted no effect on chow intake. Additionally, physical exercise associated with dietary change or the cafeteria diet increased water intake. When associated with the cafeteria diet, exercise also slightly increased soft drink intake. The effect of exercise on fluid intake may have been an attempt by the body to maintain plasma osmolality within physiological values, since exercise may cause water loss and thus increase extracellular osmotic pressure, activating the physiological mechanisms of thirst and fluid intake. Regarding Na intake, animals treated with dietary change plus physical exercise showed an increase in Na intake, which was basically a reflection of greater chow intake. Carbohydrate intake was influenced by physical exercise associated with both the cafeteria diet and dietary change, with an increased intake of this macronutrient, whereas protein intake was only increased by exercise associated with dietary change; however, in any of the regimens, fat intake was not affected by exercise. Increased carbohydrate and protein intake caused by exercise associated with dietary change as well as increased carbohydrate intake caused by exercise associated with the cafeteria diet led to increased energy intake only at the end of the experiment (week 34), but did not affect body weight, liver weight and visceral adipose tissue in either case at the end of the study. However, during the other weeks of the treatment period (dietary change and/or physical exercise), energy intake was similar to that of animals receiving palatable food. When associated with dietary change, for the same 8-week treatment period, exercise increased energy intake compared with sedentary animals subjected to dietary change after 6 weeks of treatment. The exercise-induced increase in energy intake observed in the present study disagrees with some studies reporting that exercise does not increase hunger and energy intake(Reference King, Lluch and Stubbs32), and also with reports that food intake of obese rats eating palatable food is decreased by exercise compared with obese rats not undergoing physical exercise(Reference Wang, Chen and Wang21).
After 8 weeks of physical exercise associated with dietary change or the cafeteria diet, insulin sensitivity assessed by HOMA-IR and fasting insulin plasma concentrations showed significant improvements. Exercise-mediated fasting insulin levels similar to those of the controls even in animals subjected to the cafeteria diet were probably a consequence of an improvement in insulin sensitivity in peripheral tissues. Exercise has been implicated in the improvement of insulin resistance through an increased expression of GLUT4 in skeletal muscle(Reference Christ-Roberts, Pratipanawatr and Pratipanawatr33, Reference Wang, Simar and Fiatarone Singh34), leading to increased glucose transport into the cytosol and thus reducing the incentive for β-pancreatic cells to secrete insulin. Moreover, exercise has been reported to increase insulin action in cells by affecting the transcriptional regulation of insulin receptor substrate 1(Reference Sato, Nagasaki and Nakai35, Reference Sigal, Kenny and Wasserman36) and appears to improve insulin-stimulated intracellular signalling pathways. Thus, exercise exerts a protective effect on animals fed the cafeteria diet and appears to be a useful therapeutic tool for treating diseases such as type 2 diabetes mellitus, which is characterised by insulin resistance.
Conclusion
The present study concluded that exposure to industrially processed foods, represented here by cafeteria-diet foods, resulted in obesity and metabolic changes after prolonged exposure. Contrary to what is widely recommended, dietary change alone, without energy restriction or exercise, was able to prevent, in the short term, outcomes produced by the cafeteria diet. Moreover, exercise was able to improve insulin sensitivity even in obese rats that were resistant to insulin and were eating processed foods of low nutritional value. These findings may provide a basis for the development of studies focusing on a change in eating habits for the treatment of obesity and on the mechanisms by which exercise improves insulin sensitivity independently of a reduction in visceral adipose tissue.
Acknowledgements
The present study was funded by the Conselho Nacional de Pesquisa (CNPQ, Brazil). J. F. G and G. L. S designed the experiment, performed the experimental and laboratory analyses, collected the data, and drafted the manuscript. J. F. G, M. B. C. F and G. L. S performed the statistical analyses and revised the manuscript. All authors read and approved the manuscript as submitted and take full responsibility for the reported findings. There are no conflicts of interest.
