Running behavior and its energy cost in mice selectively bred for high voluntary locomotor activity

Abstract

Locomotion is central to behavior and intrinsic to many fitnesscritical activities (e.g., migration, foraging), and it competes with other life-history components for energy. However, detailed analyses of how changes in locomotor activity and running behavior affect energy budgets are scarce. We quantified these effects in four replicate lines of house mice that have been selectively bred for high voluntary wheel running (S lines) and in their four nonselected control lines (C lines). We monitored wheel speeds and oxygen consumption for 24-48 h to determine daily energy expenditure (DEE), resting metabolic rate (RMR), locomotor costs, and running behavior (bout characteristics). Daily running distances increased roughly 50%-90% in S lines in response to selection. After we controlled for body mass effects, selection resulted in a 23% increase in DEE in males and a 6% increase in females. Total activity costs (DEE - RMR) accounted for 50%-60% of DEE in both S and C lines and were 29% higher in S males and 5% higher in S females compared with their C counterparts. Energetic costs of increased daily running distances differed between sexes because S females evolved higher running distances by running faster with little change in time spent running, while S males also spent 40% more time running than C males. This increase in time spent running impinged on high energy costs because the majority of running costs stemmed from postural costs (the difference between RMR and the zero-speed intercept of the speed vs. metabolic rate relationship). No statistical differences in these traits were detected between S and C females, suggesting that large changes in locomotor behavior do not necessarily effect overall energy budgets. Running behavior also differed between sexes: within S lines, males ran with more but shorter bouts than females. Our results indicate that selection effects on energy budgets can differ dramatically between sexes and that energetic constraints in S males might partly explain the apparent selection limit for wheel running observed for over 15 generations. © 2009 by The University of Chicago. All rights reserved.