The balance between food consumption and energy expenditure is essential in maintaining body mass (7). Past studies in male and female rodents have not consistently shown that increased physical activity has a positive effect on body mass, particularly when the training periods are of short duration (i.e., <12 wk) (8,10,20). However, in many cases, a decrease or an attenuation of the increase in body mass or body fat as a result of increased physical activity has been shown in both male and female rodents (3,4,23,25). Thus, the effect of activity on body mass or body fat loss seems to be multifactorial and does not seem to be limited to a single species or gender. Factors affecting the degree of weight loss include, but are not limited to, the duration of the study period, gender, age, rodent species, genetic influences on activity level, and/or energy intake (1,5,6,9,17,24).
In several studies, regardless of the effect on body mass, increased physical activity through use of running wheels or treadmills resulted in increased food consumption in rodents when food was provided ad libitum (4,6,7,12,24,25). However, many of these studies have used mouse strains that were highly active (ARC Swiss albino) or bred to be highly active (24). Other studies using mouse strains that tend to be low-active, such as C57BL/6 mice, have found no difference in food consumption, or time spent feeding, in mice exposed to running wheels (8,10). Thus, the innate running wheel activity level of the rodents may influence food consumption. In addition, mice that were forced to work for food tend to increase their activity (18,27). When the forced work was of low intensity or volume, mice consumed a similar amount of food regardless of the activity required to obtain the food (18). However, if the activity required to obtain food was too high, it seems mice lost weight because the food consumption could not accommodate the high level of activity (18,27).
The interaction of running wheel activity and food consumption is an understudied area. Therefore, the purpose of this study was to determine the effect of innate activity level and running wheel access on food consumption in two inbred mouse strains: a high-active strain (SWR/J) and a low-active strain (DBA/2J).
All procedures were approved by the University Institutional Animal Care and Use Committee, and all procedures adhered to the animal care standards of the American College of Sports Medicine. Ten female SWR/J mice and 10 female DBA/2J mice were used in this study (Jackson Laboratories, Bar Harbor, ME). Female mice were used because of their ease in handling, compliance with physical activity tasks, and higher daily activity levels compared with male mice in these two strains (15,26). SWR/J mice were selected as the high-active strain, whereas DBA/2J mice were selected as the low-active strain (15,26).
All mice were housed in the university's vivarium with 12-h light-dark cycles and room temperatures and relative humidity standardized to 18°C-22°C and 20%-40%, respectively. All mice were provided with water and standard chow (Harlan Teklad 8604 Rodent Diet, 24.5% protein, 4.4% fat, 3.7% fiber, and 48.6% nitrogen-free extract; Madison, WI) ad libitum.
At 9 wk, all mice were housed in individual cages. After 1 wk of acclimation, half of each of the SWR/J (n = 5) and the DBA/2J mice (n = 5) received running wheels in their cages to measure daily running wheel activity. The remaining SWR/J (n = 5) and DBA/2J mice (n = 5) were under identical conditions but did not have running wheels in their cages. Each solid surface running wheel (145 mm; Ware Manufacturing, Phoenix, AZ) was interfaced with a magnetic sensor and bicycle computer (BC600; Sigma Sport, Olney, IL) that counted the total number of wheel revolutions and total duration of exercise for each mouse (15,26). Wheel revolutions and time spent exercising were used to measure distance (km) and duration (min), respectively, which were recorded daily for each mouse for 13 wk beginning at 14 wk of age. Technical difficulties associated with the running wheels necessitated data collection begin at 14 wk of age. Average daily running speed (m·min−1) was calculated by dividing the daily distance by the daily duration (13,15,26). Weight of food was measured twice weekly to determine average daily food consumption. Food was manually removed from the solid surface cage floor, and every attempt was made to obtain all remaining food in the cage. Body weight was measured once weekly throughout the study period.
Food consumption and body weight were each analyzed using a group × time repeated-measures ANOVA to determine differences between the four groups in average daily food consumption and average weekly body weight, respectively. Multivariate analyses were used to determine differences between groups at each time point. All post hoc analyses were performed using the Games-Howell test. Running distance, duration, and speed were each compared between the two running groups using a group × time repeated-measures ANOVA to determine differences between groups across time. Multivariate analyses were used to determine differences between groups each week. The relationship between running wheel activity and food consumption for the SWR/J runners and DBA/2J runners was analyzed using bivariate correlations for distance, duration, and speed, respectively. The α value was set a priori at 0.05 for all analyses.
One mouse died from each group during the study period, with the exception of the SWR/J runner group. In addition, the wheel activity-monitoring device associated with one mouse in the SWR/J runner group malfunctioned for extended periods, so these data were not included in the analysis. Therefore, the following data represent four mice per group. SWR/J runners consumed a significantly higher amount of food (6.0 ± 0.4 g·d−1) compared with SWR/J nonrunners (4.7 ± 0.2 g·d−1, P = 0.03), DBA/2J runners (4.6 ± 0.2 g·d−1, P = 0.02), and DBA/2J nonrunners (4.2 ± 0.2 g·d−1, P = 0.006) on average and over time (Figs. 1A, B). SWR/J nonrunners consumed a significantly higher amount of food compared with DBA/2J nonrunners (P = 0.03). There was no difference in food consumption between SWR/J nonrunners and DBA/2J runners (P > 0.05) or between DBA/2J runners and DBA/2J nonrunners (P = 0.13).
Daily distance (P < 0.001; Fig. 2A) and duration (P < 0.001; Fig. 2B) on the running wheel were significantly greater for the SWR/J runners (6.4 ± 0.7 and 333.6 ± 40.5 min·d−1, respectively) for each week of the study period compared with the DBA/2J runners (1.6 ± 0.4 and 91.3 ± 23.0 min·d−1, respectively). Speed throughout the study period between SWR/J runners (19.0 ± 0.6 m·min−1) and DBA/2J runners (16.9 ± 1.1 m·min−1) was not significantly different (P = 0.07; Fig. 2C).
There was a significant correlation between average daily distance and average daily food consumption (r = 0.74, P < 0.001; Fig. 3A), between average daily duration and average daily food consumption (r = 0.68, P < 0.001; Fig. 3B), and between average daily speed and average daily food consumption (r = 0.58, P < 0.001; Fig. 3C) in all mice. However, when considering the individuals strains, the relationship between running wheel activity and food consumption was significant only for the SWR/J mice for distance and speed (Figs. 3A-C).
Table 1 shows correlations between running wheel activity and body mass and between average daily food consumption and body mass. Finally, although the SWR/J mice ate more, there was no difference in body mass between groups across time or at any time point (P > 0.05; Fig. 4).
The purpose of this study was to determine the effect of innate activity level and running wheel access on food consumption in two inbred mouse strains: a high-active strain (SWR/J) and a low-active strain (DBA/2J). The data revealed significantly higher food consumption for the SWR/J runners compared with that for the DBA/2J runners, for the SWR/J nonrunners compared with that for the DBA/2J nonrunners, and for the SWR/J runners compared with that for the SWR/J nonrunners (Fig. 1). In addition, there was a strong positive correlation between daily food intake and daily wheel activity, as measured by average daily distance, duration, and speed (Fig. 3).
Previous data have shown that increased wheel activity is associated with increased food consumption, suggesting that mice or rats that run more require higher energy intake (4,7,12,24,25). In our study, the SWR/J runners consumed 27% more food than the SWR/J nonrunners did. This is similar to the 20%-25% increase in food consumption found in high-active strains exposed to a functional running wheel (24). The increased energy expenditure from the wheel likely resulted in increased food consumption, as suggested by Swallow et al. (24). The SWR/J runners in our study consumed 30% more food than the DBA/2J runners did (Fig. 1), which is likely due in part to wheel activity being nearly fourfold higher in the SWR/J mice (Fig. 2). In addition, the SWR/J nonrunners consumed 12% more food than the DBA/2J nonrunners did, which has been reported previously (2). This suggests either that SWR/J mice inherently eat more compared with DBA/2J mice or that the SWR/J strain was more active even without a running wheel compared with the less active DBA/2J strain, resulting in greater food consumption. This is contrary to the findings of Swallow et al. (24), who found no difference in food consumption between high-active and control strains of mice that did not have access to a functional running wheel.
A previous study using a control strain not bred for high activity showed a 19.5% increase in food consumption compared with mice that had wheel access (24). In contrast, our study showed no significant difference in food consumption (<10%) between DBA/2J runners DBA/2J nonrunners, although it is recognized that the strain used by Swallow et al. (24) may have been more active than the DBA/2J strain, possibly resulting in the higher food consumption in their study. We have previously shown that caloric intake of C57BL/6 mice fed either a high-fat or a high-CHO diet (ad libitum) did not differ between mice that had access to a running wheel compared with those that did not have access to a running wheel (10). Assuming food intake is related to energy expenditure, as supported in the literature, it is possible that, because the DBA/2J mice have such a low level of running wheel activity, the DBA/2J nonrunners were able to achieve a similar activity level (energy expenditure) within their cage as those with a wheel, and thus, the food consumption levels were similar. Our results support previous findings, which suggest that energy expenditure of wheel activity accounts for only a small percentage of the total daily energy expenditure (12) and that access to a wheel may decrease other activities that result in energy expenditure, such as climbing on the cage lid (8,21). This idea is further supported by studies showing that mice increase home cage activity when housed without wheels (16). Thus, the DBA/2J nonrunners, in the absence of wheels, possibly increased home cage activity to similar levels of DBA/2J runners, as evidenced by similar levels of food consumption and weight gain. On the other hand, when using a prediction equation to estimate the oxygen cost of wheel running (19), the DBA/2J runners would have expended approximately 1.78 kcal·d−1 to perform their running wheel activity. Although daily food consumption was not significantly different between DBA/2J runners and nonrunners, the DBA/2J runners consumed an average of 0.4 g·d−1 more food than the DBA/2J nonrunners did (Fig. 1). The standard chow used in this study provided 3.1 kcal·g−1 of metabolizable energy; thus, when accounting for this additional 0.4 g of food, the DBA/2J runners consumed an average of 1.24 kcal·d−1 more than the DBA/2J nonrunners did. This additional caloric intake (1.24 kcal) is similar to the caloric expenditure of the wheel activity (1.78 kcal). On the basis of these values, it seems that the DBA/2J runners consumed a greater amount of food to support the running, although the difference was not significant.
There was no difference in body mass between the groups in our study (Fig. 4), suggesting that SWR/J and DBA/2J mice may have instinctively balanced their food consumption with their energy expenditure. However, previous results are mixed as to the effect of wheel access on body mass in both male and female rodents, with some finding decreased or attenuated body mass with wheel use (3,4,17,23,25) and others finding that exercise had no effect on body mass (8,10,20). It is interesting to note that, in several studies that suggested running wheel activity decreased or attenuated weight gain, the mouse strains used tended to be highly active (3,4,23), whereas those that found wheel access had no effect on body mass used low-active strains (8,10). However, Swallow et al. (22) found access to a running wheel for 7-8 wk did not attenuate weight gain in the high-active strain to a greater degree compared with weight gain in the controls. Similarly, our study showed that running wheel activity affected body mass for the high-active or the low-active strains in a similar manner, suggesting that the increased running wheel activity of the high-active strain was compensated by increased food consumption or vice versa. However, high-active hamsters that had access to a running wheel for 1 yr increased food consumption by the second week compared with a nonwheel group, but no difference was found in body mass between groups until the 16th week (6). In this case, the wheel group was significantly heavier than the nonwheel group (6). Nehrenberg et al. (17) found that higher wheel activity was not always associated with weight loss. They concluded that changes in body fat may not always be due to the amount of wheel activity, further suggesting the influence of genetic background (17). However, it seems that the mice in the study of Nehrenberg et al. (17) that had greater wheel activity also ate a greater volume of food compared with those that were less active.
Although there were no differences in body mass between groups, it is unclear whether there were differences in lean mass or fat mass because body composition was not measured. However, data from our laboratory using C57BL/6 mice showed that access to a running wheel had no effect on lean body mass. Any differences in body mass between mice with access to a running wheel compared with mice without access to a running wheel were due to differences in fat mass (10). It should also be noted that SWR/J mice have been shown to be obesity-resistant (28); thus, it is possible that the lack of difference in body mass between the SWR/J runners and SWR/J nonrunners could be due to obesity resistance rather than increased activity due to the running wheel. However, it is equally plausible that access to the running wheel resulted in greater energy expenditure, thereby balancing the increased food intake.
It is interesting to note the strong, positive relationship between body mass and average daily food intake for the DBA/2J mice (Table 1) compared with the nonsignificant relationship between food intake and body mass for the SWR/J mice. In addition, in the case of the DBA/2J mice, there was no relationship between activity level and food consumption. However, there was a significant relationship between activity level and food consumption for the SWR/J mice. These data seem to support the idea that the increased food consumption in the SWR/J mice was driven by the increased running wheel activity.
Harri et al. (8) found that low-active mice (∼2 km·d−1 on a running wheel) exposed to a running wheel expend less time, and thus energy, in other activities (e.g., climbing on the cage lid). On the basis of our finding that DBA/2J runners and nonrunners had similar food consumption and body mass values, it is logical to conclude that access to a running wheel did not substantially increase the daily physical activity of DBA/2J mice. This assumption seems reasonable because the DBA/2J mice averaged only ∼90 min·d−1 (<2 km·d−1) on the running wheel during the study. On the other hand, the SWR/J mice averaged slightly <6 h·d−1 (∼6.5 km·d−1) on the running wheel throughout the duration of the study. It seems reasonable to conclude that the SWR/J mice without a running wheel were unable to achieve the activity level and energy expenditure of the SWR/J runners with cage activity alone. Thus, the SWR/J nonrunners had the same body mass as the SWR/J runners, but they consumed significantly fewer calories.
One possible limitation to our study is the effect of differing estrous cycles in the female mice and its effect on running wheel activity because it is known that sex hormones have an influence on running wheel activity (14). However, because wheel activity was recorded daily and averaged each week for 13 wk, we think that any effect of differing estrous cycles between the strains would be dampened and would have minimal effect on the results. In addition, because cage activity was not monitored outside the wheel, it is difficult to estimate total energy expenditure. Finally, Koteja et al. (11) have suggested that food wastage should be considered when calculating food consumption. Although every attempt was made to measure all of the food remaining in the cage, we recognize that there was a possibility of error by missing small pellets. Nonetheless, the average variance in measured food consumed between animals was very small (SE = ±0.2 g for each strain).
In conclusion, our findings suggest that, at least during a 13-wk trial early in the life span, SWR/J and DBA/2J mice tended to balance energy intake with energy expenditure. However, increased running wheel activity in mice was associated with increased food consumption only in the SWR/J mice. There was no difference in body mass between any of the groups despite access to a running wheel and genetic differences in activity level. Food consumption of the high-active mouse strain (SWR/J) with a running wheel was increased above that of a high-active control, likely due to the increased activity level on the wheel, resulting in similar weight gain compared with control mice. However, in the low-active strain (DBA/2J), addition of a running wheel did not result in significantly increased food consumption, suggesting that energy expenditure of cage activity in the control DBA/2J mice was similar to the energy expenditure of the wheel activity because body mass was similar between the two groups. Finally, the high activity of the SWR/J mice was associated with greater food intake compared with the DBA/2J mice in both the running wheel and the control setting.
This research study was partially supported by the following grants: DK61635 from the National Institutes of Health/National Institute of Diabetes and Digestive and Kidney Diseases (J.T. Lightfoot), AR050085 from the National Institute of Arthritis and Musculoskeletal and Skin Diseases (J.T. Lightfoot), and AG022417 from the National Institutes of Health/National Institute on Aging (M.J. Turner).
Results of the present study do not constitute endorsement by the American College of Sports Medicine.
1. Applegate EA, Upton DE, Stern JS. Food intake, body composition and blood lipids following treadmill exercise in male and female rats. Physiol Behav
2. Bachmanov AA, Reed DR, Beauchamp GK, Tordoff MG. Food intake, water intake, and drinking spout side preference of 28 mouse strains. Behav Genet
3. Bell RR, Spencer MJ, Sheriff JL. Diet-induced obesity in mice can be treated without energy restriction using exercise and/or a low fat diet. J Nutr
4. Bell RR, Spencer MJ, Sheriff JL. Voluntary exercise and monounsaturated canola oil reduce fat gain in mice fed diets high in fat. J Nutr
5. Donnelly J, Smith B. Is exercise effective for weight loss with ad libitum
diet? Energy balance, compensation, and gender differences. Exerc Sport Sci Rev
6. Gattermann R, Weinandy R, Fritzsche P. Running-wheel activity and body composition in golden hamsters (Mesocricetus auratus
). Physiol Behav
7. Goodrick CL. Effect of voluntary wheel exercise on food intake, water intake, and body weight for C57BL/6J mice and mutations which differ in maximal body weight. Physiol Behav
8. Harri M, Lindblom J, Malinen H, et al. Effect of access to a running wheel on behavior of C57BL/6J mice. Lab Anim Sci
9. Hoffman-Goetz L, MacDonald MA. Effect of treadmill exercise on food intake and body weight in lean and obese rats. Physiol Behav
10. Jung AP, Luthin DR. Wheel access does not attenuate weight gain in mice fed high-fat or high-CHO diets. Med Sci Sports Exerc
11. Koteja P, Carter PA, Swallow JG, Garland T Jr. Food wasting by house mice: variation among individuals, families, and genetic lines. Physiol Behav
12. Koteja P, Swallow JG, Carter PA, Garland T Jr. Energy cost of wheel running in house mice: implications for coadaptation of locomotion and energy budgets. Physiol Biochem Zool
13. Lerman I, Harrison BC, Freeman K, et al. Genetic variability in forced and voluntary endurance exercise performance in seven inbred mouse strains. J Appl Physiol
14. Lightfoot JT. Sex hormones' regulation of rodent physical activity: a review. Int J Biol Sci
15. Lightfoot JT, Turner MJ, Daves M, Vordermark A, Kleeberger SR. Genetic influence on daily running wheel activity level. Physiol Genomics
16. Malisch JL, Breuner CW, Kolb EM, et al. Behavioral despair and home-cage activity in mice with chronically elevated baseline corticosterone concentrations. Behav Genet
17. Nehrenberg DL, Hua K, Estrada-Smith D, Garland T Jr, Pomp D. Voluntary exercise and its effects on body composition depend on genetic selection history. Obesity (Silver Spring)
18. Perrigo G, Bronson FH. Foraging effort, food intake, fat deposition and puberty in female mice. Biol Reprod
19. Rezende EL, Kelly SA, Gomes FR, Chappell MA, Garland T. Effects of size, sex, and voluntary running speeds on costs of locomotion in lines of laboratory mice selectively bred for high wheel-running activity. Physiol Biochem Zool
20. Schemmel RA, Hannum SH, Rosekrans JA, Heusner WW. Moderate exercise in young female S5B/P1 rats does not reduce body fat. Physiol Behav
21. Stevenson JA. Exercise, food intake and health in experimental animals. Can Med Assoc J
22. Swallow JG, Garland T, Carter PA, Zhan W-Z, Sieck GC. Effects of voluntary activity and genetic selection on aerobic capacity in house mice (Mus domesticus
). J Appl Physiol
23. Swallow JG, Koteja P, Carter PA, Garland T. Artificial selection for increased wheel-running activity in house mice results in decreased body mass at maturity. J Exp Biol
24. Swallow JG, Koteja P, Carter PA, Garland T. Food consumption and body composition in mice selected for high wheel-running activity. J Comp Physiol B
25. Tokuyama K, Saito M, Okuda H. Effects of wheel running on food intake and weight gain of male and female rats. Physiol Behav
26. Turner MJ, Kleeberger SR, Lightfoot JT. Influence of genetic background on daily running-wheel activity differs with aging. Physiol Genomics
27. Vaanholt LM, De Jong B, Garland T Jr, Daan S, Visser GH. Behavioural and physiological responses to increased foraging effort in male mice. J Exp Biol
. 2007;210(Pt 11):2013-24.
28. West DB, Waguespack J, McCollister S. Dietary obesity in the mouse: interaction of strain with diet composition. Am J Physiol
. 1995;268(3 Pt 2):R658-65.
Keywords:©2010The American College of Sports Medicine
VOLUNTARY EXERCISE; RUNNING WHEEL; SWR MICE; DBA MICE