Locomotor activity, core body temperature, and circadian rhythms in mice selected for high or low heat loss.

Daily locomotor activity, core body temperature, and their circadian rhythms were measured in lines of mice selected for high (MH) or low (ML) heat loss and unselected controls (MC). Lines were created by selecting for 16 generations in each of three replicates. Collection of locomotor activity and core temperature data spanned Generations 20 and 21 for a total of 352 mice. Physical activity and core body temperature data were accumulated using implanted transmitters and continuous automated collection. Measurement for each animal was for 3 d. Activity was recorded for each half hour and then averaged for the day; temperature was averaged daily; circadian rhythm was expressed in 12-h (light vs dark) or 6-h periods as well as by fitting cyclic models. Activity means were transformed to log base 2 to lessen heterogeneity of variance within lines. Heat loss for a 15-h period beginning at 1630 and feed intake for 7 d were measured on 74 additional mice in order to estimate the relationship between locomotor activity and heat loss or feed intake. Selection lines were different (P < 0.01) for both locomotor activity and core body temperature. Differences were due to selection (MH-ML, P < 0.01), and there was no evidence of asymmetry of response (P > 0.38). Retransformed from log base 2 to the scale of measurement, mean activity counts were 308, 210, and 150 for MH, MC, and ML, respectively. Mean core temperatures were 37.2, 36.9, and 36.7 degrees C for MH, MC, and ML (P < 0.01), respectively. Females had greater physical activity (P < 0.01) and body temperature (P < 0.01) than males. There was no evidence of a sex x selection criterion interaction for either activity or temperature (P > 0.20). Overall phenotypic correlation between body temperature and log base 2 activity was 0.43 (P < 0.01). Periods during the day were different for both 12- and 6-h analyses (P < 0.01), but there were no period x selection criterion interactions (P > 0.1) for physical activity or body temperature. More sensitive cyclic models revealed significant (P < 0.01) 24-, 12-, 8-, and 6-h cycles that differed (P < 0.01) among lines. Estimated differences between MH and ML mice in feed intake and heat loss due to locomotor activity were 36 and 11.5%, respectively. Variation in activity thus contributed to variation in feed intake.