Alternate-Day Wheel Access

Alternate-Day Wheel Access

Physiology & Behavior, Vol. 62, No. 4, pp. 905–908, 1997 © 1997 Elsevier Science Inc. All rights reserved. Printed in the U.S.A. 0031-9384/97 $17.00 1...

146KB Sizes 1 Downloads 111 Views

Physiology & Behavior, Vol. 62, No. 4, pp. 905–908, 1997 © 1997 Elsevier Science Inc. All rights reserved. Printed in the U.S.A. 0031-9384/97 $17.00 1 .00

PII S0031-9384(97)00266-7

Alternate-Day Wheel Access: Effects on Feeding, Body Weight, and Running1 DEVIN T. MUELLER, ARTHUR LOFT AND ROELOF EIKELBOOM2 Department of Psychology, Wilfrid Laurier University, Waterloo, Ontario N2L 3C5, Canada Received 3 March 1997; Accepted 13 May 1997 MUELLER, D. T., A. LOFT AND R. EIKELBOOM. Alternate-day wheel access: Effects on feeding, body weight, and running. PHYSIOL BEHAV 62(4) 905–908, 1997.—For rats, access to a running wheel results in a pronounced but temporary suppression of feeding. The reasons for the feeding suppression and its temporary nature are unclear. The effects of alternate-day wheel access were explored by comparing feeding and running in 25 male Sprague–Dawley rats given either no wheel access, continuous wheel access, or alternate-day wheel access. With alternate-day wheel access food intake was suppressed on wheel days and elevated on non-wheel days for the full 32 days of the experiment. Body weight decreased on wheel days and showed a large increase on non-wheel days. Acquisition of running over days was similar in both wheel groups and plateaued at the same level, but running was elevated, compared to continuous-access rats, for the first few hours when alternate-day rats were returned to the wheel. These results suggest that wheel-induced feeding suppression is not due to the novelty of the wheel and that this suppression can be extended by providing periods with no wheel access. The temporary nature of the feeding suppression in chronic access conditions may be due to secondary longer term motivational changes. © 1997 Elsevier Science Inc. Body weight

Feeding

Feeding suppression

Rat

Wheel running

MAKING a running wheel available to rats has profound effects on energy expenditure and food intake. Wheel access results in running that starts low and rises over a period of a few weeks to several thousand wheel turns, mostly occurring at night (9). This pattern of wheel running has been well known since it was first reported (13). Additionally, when the wheel is introduced, there is a marked decrease in feeding that lasts for ca. 10 days before food intake gradually increases to prewheel, or above, levels (1,3,6,8,9,12,16). The reasons for these changes in feeding and wheel running are unclear. One explanation for the decrease and gradual recovery of food intake is response competition, when running rats cannot eat. In this case one would expect similar suppression in drinking as seen with feeding, but fluid consumption, if anything, increases when wheels are made available (8). It is possible that rats find access to a wheel novel and experience ‘‘novelty stress’’ (6,14). Initially this reduces intake but the novelty decreases over time, allowing food intake to recover. A third explanation is that access to an alternate reinforcer, that is wheel running (2,5,7), reduces the feeding motivation directly (12). Using progressive ratio schedules, it has been reported that deprivation or satiation of either food or the wheel changes the reinforcing value of the other (10). Over days, however, the increasing energy expenditure and the decreased weight (relative to home cage controls) increases food motivation. This then results in an elevation of feeding and the initial feeding suppression disappears. With continuous wheel access, animals have no opportunity to restore their energy balance and so the 1 2

wheel-access-induced suppression may gradually be masked by an increasing level of food motivation. One way to test these explanations is to give the rats alternateday wheel access, providing them with an opportunity to restore their energy balance during a period when the wheel is not available. If novelty is responsible for the decrease in food consumption, alternate-day access should have no effect except perhaps to slow the loss of novelty. In this case the food consumption on wheel-access days should show the initial decrease but then return to normal as novelty is lost. Eventually feeding on wheel days should be similar to that on days when the wheel is not available. However, if the cumulative effects on energy balance are responsible for the increased feeding, then periods of no wheel access could permit the restoration of the animals’ energy balance and result in a maintenance of the wheel-induced decrease in feeding. In this case animals should continue to eat less on wheel-access days and might show increased consumption on the intervening non-wheel-access days. It is also possible to ask how alternate-day wheel access affects running. It has been suggested (3) that activity, as indexed by wheel running, may be a regulated behavior. For many behaviors, restricting access changes the amount and temporal patterning of the behavior. Eating and drinking are the classic example of behaviors that are more likely to occur after deprivation. Alternateday access to alcohol and saccharin solutions results in increased consumption (11,17). A previous study (9) showed that with experienced runners 10 days of wheel deprivation resulted in a decrease in running, but other research has shown an increase in

This research was supported by funds from the WLU Office of Research to R. E. To whom requests for reprints should be addressed. E-mail: [email protected]

905

906

MUELLER, LOFT AND EIKELBOOM

running with shorter wheel deprivation (4,15). Thus it is unclear how alternate-day access will affect wheel running, in terms of both acquisition and the final running plateau. In the present experiment food intake, body weight, and wheel running are compared in rats with either alternate-day wheel access, no wheel access, or continuous wheel access. METHOD

Subjects Twenty-five male Sprague–Dawley rats (Charles River Canada) weighing 270 –315 g at the start of the experiment were housed in individual standard wire cages (25 3 17 3 20 cm, unless otherwise specified). The colony room was maintained at 21–22°C with a 12:12 L/D lighting cycle (lights off at 2200 hours). Rat chow pellets and tap water were available ad lib. Apparatus Twelve wire cages (25 3 17 3 20 cm) were set up with constant access to attached running wheels (11 cm wide, 33 cm in diameter) mounted on a three-tiered metal rack. The rats were able to see the other animals when in the wheels. The number of wheel revolutions were recorded using a magnetic contact closure system, in 10-s bins by Dataquest III, a Mini-Mitter Co. data collection system. Procedure Animals were habituated to the colony conditions for ca. 2 weeks before a 4-day baseline period. Animals were weighed daily, usually between 1700 and 1800 hours, a period of low running and away from the light– dark transitions. Food intake was measured by weighing lab chow and weighing the remaining pellets and any large crumbs 24 h later (spillage was assessed during the baseline period and was found to be similar across animals). Water intake was measured by weighing the water bottle upon filling and weighing the bottle with the remaining water 24 h later. A third of the animals (WHEEL group, n 5 8) were placed in cages with constant access to running wheels. Another third of the animals (HOMECAGE group, n 5 9) were left in their home cages. The remaining rats (ALTDAY group, n 5 8) received alternate-day wheel exposure. The rats were assigned to groups so that mean group weights were similar the last day of baseline. Because there were only 12 wheels available, 4 of the wheels were shared by ALTDAY animals. The first day of the experiment was a wheel day and so the lightest 4 animals in the ALTDAY group started the experiment 1 day later than the other 21 animals. Wheel turns, food intake, water intake, and body weight were recorded for a further 32 days. RESULTS

There were no differences among the three groups of rats over the 4 baseline days for the mean body weight and food consumption. Two Group 3 Day analyses of variance (ANOVA) revealed that no Group effect or interaction was significant (largest F 5 1.19, p . 0.3). As the ALTDAY group was given access to wheels on odd days, and not on even days, the odd and even days were analyzed separately. Food consumption was averaged over 2 odd (or even) day blocks for animals in each of the three groups and results are shown in Fig. 1A,B. Two Group 3 Blocks ANOVAs revealed that for both odd and even days the interaction (Odd F(14, 154) 5 11.16, p , 0.001; Even F(14, 154) 5 9.58, p , 0.001) and the Blocks effect (Odd F(7, 154) 5 28.23, p , 0.001; Even F(7, 154) 5 13.31, p , 0.001) were significant. Only on odd days was the

FIG. 1. Food consumption (6SEM) averaged over 2-day odd or even blocks for the 32 days of the experiment. (A) Mean food consumption on the odd-day blocks (when ALTDAY rats were in the wheel cages) for animals in the three groups. (B) Mean food consumption on the even-day blocks (when ALTDAY rats were in the non-wheel cages) for animals in all three groups. (C) Mean food consumption of animals in the ALTDAY group on days when they had access to the wheels (odd blocks) and when they did not (even blocks). These two lines are the same as the ALTDAY data in (A) and (B).

group effect significant (Odd F(2, 22) 5 6.92, p , 0.01; Even F(2, 22) 5 1.15, p . 0.25). For both odd and even blocks HOMECAGE group rats showed stable food consumption over the experiment (about 32 g/day). Rats in the WHEEL group initially showed a suppressed food intake but were eating more than the HOMECAGE group rats by the end of the experiment. Given the significant interactions, further ANOVAs (interaction comparisons) were carried out. Comparisons between the two groups with wheel access on the odd days revealed a significant group difference (F(1, 14) 5 6.41, p , 0.05), a significant Blocks effect (F(7, 98) 5 37.22, p , 0.001), and a significant interaction (F(7, 98) 5 2.69, p , 0.05). On the odd days, ALTDAY group rats (when in wheel cages) showed an initial decrease similar to the WHEEL group rats but maintained this suppressed intake over the experiment. Comparisons between rats in the HOMECAGE and ALTDAY groups on the even days revealed only a significant interaction (F(7, 105) 5 5.79, p , 0.001). On even days, the ALTDAY group rats (when in cages with no wheel access) were similar to the HOMECAGE group rats initially but exceeded HOMECAGE food intake by the end of the experiment, eating an amount similar to WHEEL group rats. Figure 1C shows the odd and even blocks for animals in the

ALTERNATE-DAY WHEEL ACCESS

907

FIG. 3. Mean number of wheel turns per hour (6SEM) averaged over 2 odd days for animals in the WHEEL and ALTDAY groups (ALTDAY rats were only in the wheel on odd days. (A) Average of 21-h running (excluding the 3 h during which data were collected). (B) Average running in the 3 h after data were collected (1900 to 2200 hours).

FIG. 2. Daily weight change (6SEM) averaged over 2-day odd or even blocks for the 32 days of the experiment. (A) Mean daily weight change on the odd-day blocks (when ALTDAY rats were in the wheel cages) for animals in the three groups. (B) Mean daily weight change on the even-day blocks (when ALTDAY rats were in the non-wheel cages) for animals in all three groups. (C) Mean daily weight change of animals in the ALTDAY group on days when they had access to the wheels (odd blocks) and when they did not (even blocks). These two lines are the same as the ALTDAY data in (A) and (B).

ALTDAY group and it is evident that they ate more on the home cage days (even days) than on the wheel cage days (odd days). In an Odd–Even 3 Blocks within-subjects ANOVA, only the main effects were found to be significant (Odd–Even F(1, 7) 5 18.98, p , 0.01; Blocks F(7, 49) 5 23.07, p , 0.001), showing that whereas consumption increased over the experiment, the feeding difference in the wheel and non-wheel cages was maintained. The daily weight changes of animals in the three groups were analyzed in a manner similar to the food data. Daily weight change for odd and even days grouped over 2-day blocks are shown in Fig. 2. For both odd and even days only the main effects were significant (Groups, Odd days F(2, 22) 5 24.77, p , 0.001; Even days F(2, 22) 5 13.40, p , 0.001; Blocks, Odd days F(7, 154) 5 4.04, p , 0.001; Even days F(7, 154) 5 9.51, p , 0.001). After the initial period, rats in the HOMECAGE and WHEEL groups gained weight at a similar rate. The ALTDAY rats, however, lost weight on wheel days and gained more than rats in the other two groups on their home cage days. On the last day of the experiment (Day 32), the weights of rats in the three groups differed significantly (F(2, 22) 5 6.08, p , 0.01). Animals in the HOMECAGE group (461 6 10.1 SEM g) were heavier than animals in the WHEEL group (420 6 13.8 g), which did not differ from ALTDAY group rats (414 6 7.3 g).

As the ALTDAY group animals had access to wheels only on odd days, their running on these days was compared to odd-day running in WHEEL group animals using 2-day blocks. Figure 3A shows the mean hourly running for the two groups for the 21 h from 1900 to 1600 hours (excluding the 3-h block when animals were weighed, moved, etc.). A Group 3 Blocks ANOVA revealed that for the 21-h running only the main effect of blocks was significant (F(7, 98) 5 22.30, p , 0.001). This reflected the increase in running seen over the experiment. The mean hourly running for the 3 h after the ALTDAY animals were rehoused (1900 to 2200 hours) for the two groups is shown in Fig. 3B. In this case the Group 3 Block ANOVA revealed the Group effect (F(1, 14) 5 10.93, p , 0.01), the Block effect (F(7, 98) 5 6.98, p , 0.001), and interaction were all significant (F(7, 98) 5 7.71, p , 0.001). This was due to the increase in running in these 3 h that occurred over the course of the experiment for animals in the ALTDAY group. DISCUSSION

Food consumption of rats with alternate-day access to running wheels showed the same decrease, on wheel days, that rats with continuous access displayed initially. On the intervening days (with no wheel access) food consumption in these rats was similar to, and later higher than, that of animals maintained in cages with no wheels. The wheel-induced decrease was maintained for rats with alternate-day access throughout the experiment whereas it was only temporary for rats with continuous wheel access. If the feeding decrease had been due to novelty, it should have disappeared in both groups, although perhaps at different rates. However, if the decrease was due to novelty, the rapidly occurring elevation of feeding seen on non-wheel-access days would not be expected so soon. Finally, the change seen in the running pattern over days (Fig. 3B) argues that animals have clearly adapted to the alternate-day procedure. This makes it unlikely that novelty of the wheel cages (6,14) is responsible for the decrease in feeding seen at wheel introduction. It suggests the wheel-induced suppression is due to the interactions among reinforcers (10,12).

908

MUELLER, LOFT AND EIKELBOOM

The fact that animals with alternate-day wheel access ate less on wheel days throughout the experiment suggests that presence of the wheel reduces feeding motivation chronically. The temporary nature of the decrease in feeding in animals with chronic wheel access may reflect a second additional change in the feeding motivation. The combined effect of the weight change and the increasing energy expenditure from running may combine to increase hunger over days. If the rat has the opportunity to restore its energy balance, the effects of a running wheel on food intake can be maintained. The weight of the animals in the alternate-day wheel access group raises two issues. First, the weight went up when the animals did not have access to a wheel and decreased on days the rats had access to the wheel. The fact that the weight went down each time the wheel was available (Fig. 2A) suggests that the animals are chronically putting themselves into a negative energy balance. On the days the wheel was not available, their weight gain was larger than that observed for the other two groups (Fig. 2B), suggesting these days restore some aspect of energy balance. Given that the difference in feeding on these two types of days averaged only ca. 6 – 8 g (Fig. 1C), the large difference in weight gain (from 10 to 15 g) on these two types of days (Fig. 2C) suggests the energy costs of running are not trivial. This conclusion is supported by the data of the WHEEL group rats, which showed increased food consumption but similar weight gain to HOMECAGE rats, thus never reducing the weight difference between the groups. The second interesting fact is that at the end of the experiment the ALTDAY and WHEEL group animals were still similar in weight and both were lower than animals in the HOMECAGE group (the averaged weights of rats in the two wheel groups were similar throughout the experiment). Thus the recovery from the feeding suppression in the WHEEL group rats is unlikely to be solely due to the increased motivation induced by the weight decrease. If increased motivation was responsible, the food intake of the ALTDAY group should also have increased on wheel days as their weight was similar to that of the WHEEL group rats. The large weight increases on the non-wheel days of the ALTDAY rats would suggest that they are responding to a real imbalance but their suppressed weight relative to HOMECAGE rats suggests that HOMECAGE rats may be overweight. Perhaps the ‘‘normal’’ control rat housing condition results in animals that are both chronically underactive and overweight. Running in the two groups did not differ significantly when the

21-h average was compared. Both WHEEL and ALTDAY groups showed an increase over the course of the experiment. In this experiment the running seems to have reached its maximum by the 17th day (fifth block of odd days). Whereas in the ALTDAY group rats running seemed to increase less and lag behind that of the animals in the WHEEL group (Fig. 3A), these differences were not significant. Thus acquisition of running and its maximum seem not to be affected by the alternate-day exposure to the wheel. Averaged over the whole day, it seems that there is no elevation or depression of running if the animals have only alternate-day access to the wheel. These results suggest that if running is a regulated behaviour (3), the time period of regulation is about a day and repeated 24-h deprivation does not produce an elevated response the day the wheel is reintroduced. When the first 3 h after the ALTDAY animals are introduced to the wheel is looked at, a different picture emerges (Fig. 3B). Over the course of the experiment the introduction of animals to the wheel cages results in an increase in running relative to animals with continuous access. This suggests that repeated 24-h wheel access results in a deprivation-induced running. Clearly, if the total running is similar and the first 3 h is elevated, the running at some other time is suppressed relative to the continuous-access condition. Thus the initial elevation in some way produces a subsequent decrease. In conclusion, our results are inconsistent with the novelty explanation for wheel-induced feeding suppression (6,14). They suggest that the temporary nature of the feeding suppression seen with wheel access may be due to a second longer term change in energy balance. If animals are given periods with no wheel access, their feeding pattern adjusts so that it remains low on wheel-access days and is elevated on non-wheel-access days. This pattern seemed stable over the 32 days of the experiment. Note that running only plateaus after ca. 14 –18 days so the 16 days of wheel exposure for the ALTDAY group rats may have been too limited to show any long-term changes after running plateaus. It is, however, much longer than the normal feeding suppression seen in the WHEEL group rats (a total of ca. 10 days). Repeated 1-day wheel deprivation does not result in global elevations of running when the wheel is again made available. On a smaller time scale, restored access does result in a few hours of elevated running, suggesting activity is regulated within a short time window. It might be worthwhile to further explore the regulation of wheel running in terms of short and longer term wheel deprivation.

REFERENCES 1. Bauman, R. A. The effects of wheel running, a light/dark cycle, and the instrumental cost of food on the intake of food in a closed economy. Physiol. Behav. 52:1077–1083; 1992. 2. Belke, T. W.; Heyman, G. M. A matching law analysis of the reinforcing efficacy of wheel running in rats. Anim. Learn. Behav. 22: 267–274; 1994. 3. Collier, G. H. Work: A weak reinforcer. Trans. NY Acad. Sci. 32: 557–576; 1970. 4. Hill, W. F. Activity as an autonomous drive. J. Comp. Physiol. Psychol. 49:15–19; 1956. 5. Iversen, I. H. Techniques for establishing schedules with wheel running as reinforcement in rats. J. Exp. Anal. Behav. 60:219 –238; 1993. 6. Jennings, W. A.; McCutcheon, L. E. Proximity of food, sex, and access to running wheels: Effects on food intake in rats. J. Comp. Physiol. Psychol. 87:106 –109; 1974. 7. Kagan, J.; Berkun, M. The reward value of running activity. J. Comp. Physiol. Psychol. 47:108; 1954. 8. Levitsky, D. A. Feeding patterns of rats in response to fasts and changes in environmental conditions. Physiol. Behav. 5:291–300; 1970.

9. Looy, H.; Eikelboom, R. Wheel running, food intake, and body weight in male rats. Physiol. Behav. 45:403– 405; 1989. 10. Pierce, W. D.; Epling, W. F.; Boer, D. P. Deprivation and satiation: The interrelations between food and wheel running. J. Exp. Anal. Behav. 46:199 –210; 1986. 11. Pinel, J. P. J.; Huang, E. Effects of periodic withdrawal on ethanol and saccharin selection in rats. Physiol. Behav. 16:693– 698; 1976. 12. Premack, D.; Premack, A. J. Increased eating in rats deprived of running. J. Exp. Anal. Behav. 6:209 –212; 1963. 13. Richter, C. P. Animal behavior and internal drives. Q. Rev. Biol. 2:307–343; 1927. 14. Routtenberg, A. ‘‘Self-starvation’’ of rats living in activity wheels: Adaptation effects. J. Comp. Physiol. Psychol. 66:234 –238; 1968. 15. Shirley, M. Spontaneous activity. Psychol. Bull. 26:341–365; 1929. 16. Tokuyama, K.; Saito, M.; Okuda, H. Effects of wheel running on food intake and weight gain of male and female rats. Physiol. Behav. 28:899 –903; 1982. 17. Wayner, M. J.; Greenberg, I.; Tartaglione, R.; Nolley, D.; Fraley, S.; Cott, A. A new factor affecting the consumption of ethyl alcohol and other sapid fluids. Physiol. Behav. 8:345–362; 1972.