Acutely decreased thermoregulatory energy expenditure or decreased activity energy expenditure both acutely reduce food intake in mice

Abstract

Despite the suggestion that reduced energy expenditure may be a key contributor to the obesity pandemic, few studies have tested whether acutely reduced energy expenditure is associated with a compensatory reduction in food intake. The homeostatic mechanisms that control food intake and energy expenditure remain controversial and are thought to act over days to weeks. We evaluated food intake in mice using two models of acutely decreased energy expenditure: 1) increasing ambient temperature to thermoneutrality in mice acclimated to standard laboratory temperature or 2) exercise cessation in mice accustomed to wheel running. Increasing ambient temperature (from 21 °C to 28 °C) rapidly decreased energy expenditure, demonstrating that thermoregulatory energy expenditure contributes to both light cycle (40 ± 1%) and dark cycle energy expenditure (15 ± 3%) at normal ambient temperature (21 °C). Reducing thermoregulatory energy expenditure acutely decreased food intake primarily during the light cycle (65 ± 7%), thus conflicting with the delayed compensation model, but did not alter spontaneous activity. Acute exercise cessation decreased energy expenditure only during the dark cycle (14 ± 2% at 21 °C; 21 ± 4% at 28 °C), while food intake was reduced during the dark cycle (0.9 ± 0.1 g) in mice housed at 28 °C, but during the light cycle (0.3 ± 0.1 g) in mice housed at 21 °C. Cumulatively, there was a strong correlation between the change in daily energy expenditure and the change in daily food intake (R(2) = 0.51, p<0.01). We conclude that acutely decreased energy expenditure decreases food intake suggesting that energy intake is regulated by metabolic signals that respond rapidly and accurately to reduced energy expenditure.

Figure 1. Study schematic.

Initial temperature for both groups of mice (n = 8 mice/group) was 21°C. Starting on day -10 all mice were acclimatized to wheel running with the wheel locked for 3 days followed by 7 days of spontaneous running. Subsequently, continuous measurement of outcome variables was performed in metabolic cages for 10days (day 0–10). Experiment 1: Baseline measurement period (day 0–2) was equal in both groups of mice (n = 8) in that both groups were studied at 21°C. Ambient temperature was raised to 28°C for mice in the G28C group starting on day 3 and continued to the end of the study. Day 3–5 was evaluated as the temperature change (T. Change) period for Experiment 1. For Experiment 2 both groups had free running wheel access for spontaneous exercise. For both G21C and G28C day 3–5 was the exercise (EX) period evaluated. Running wheels were locked on day 6 and day 6–10 was evaluated as the sedentary (Sed) period for both groups. Throughout Experiment 2 ambient temperature was 21°C for G21C mice and 28°C for G28C mice. Arrows indicate the two days that body composition measurements were performed.

Figure 2. Effect of increased ambient temperature (T_a) on total energy expenditure (EE) in G28C and G21C groups.

A. G28C group (n = 8). Mean EE (kcal/hr) rate every 10 min during the entire photoperiod at baseline (21°C) and after increasing T_a to 28°C. Insert: Cumulative mean dark cycle and light cycle EE in G28C mice at baseline (Base 21°C) and after increasing T_a to 28°C (28°C). B. G21C group (n = 8). Mean EE (kcal/hr) rate every 10 min during the entire photoperiod at baseline (21°C) and after maintaining T_a at 21°C. Insert: Cumulative mean dark cycle and light cycle EE in G21C mice at baseline (Base 21°C) and after maintaining T_a at 21°C (21°C). All data shown as means ± SEM for 3 consecutive days, * signifies p<0.05 by pair-wise t-test.

Figure 3. Effect of increased ambient temperature (T_a) on food intake (FI) in G28C and G21C groups.

A. G28C group (n = 8). Mean hourly FI (g) during the entire photoperiod at baseline (21°C) and after increasing T_a to 28°C. Insert: Cumulative mean dark cycle and light cycle FI in G28C mice at baseline (Base 21°C) and after increasing T_a to 28°C (28°C). B. G21C group (n = 8). Mean hourly FI (g) during the entire photoperiod at baseline (21°C) and after maintaining T_a at 21°C. Insert: Cumulative mean dark cycle and light cycle FI in G21C mice at baseline (Base 21°C) and after maintaining T_a at 21°C (21°C). All data shown as means ± SEM for 3 consecutive days, * signifies p<0.05 by pair-wise t-test.

Figure 4. Effect of exercise cessation on total energy expenditure (EE) in G28C and G21C groups.

A. G28C group (n = 8) housed at 28°C. Mean EE (kcal/hr) rate every 10 min during the entire photoperiod with Exercise and after running wheels were locked (Sedentary). Insert: Cumulative mean dark cycle and light cycle EE in G28C mice during the Exercise period (EX) and after running wheels were locked (SED). B. G21C group (n = 8) housed at 21°C. Mean EE (kcal/hr) rate every 10 min during the entire photoperiod with Exercise and after running wheels were locked (Sedentary). Insert: Cumulative mean dark cycle and light cycle EE in G21C mice during the Exercise period (EX) and after running wheels were locked (SED). All data shown as means ± SEM for 3(EX) or 4(SED) consecutive days, * signifies p<0.05 by pair-wise t-test.

Figure 5. Effect of exercise cessation on food intake (FI) and ambulatory activity (AA) in G28C and G21C groups.

A. G28C group (n = 8) housed at 28°C. Mean hourly FI (g) during the entire photoperiod with Exercise and after running wheels were locked (Sedentary). Insert: Cumulative mean dark cycle and light cycle FI in G28C mice during the Exercise period (EX) and after running wheels were locked (SED). B. G21C group (n = 8) housed at 21°C. Mean hourly FI (g) during the entire photoperiod with Exercise and after running wheels were locked (Sedentary). Insert: Cumulative mean dark cycle and light cycle FI in G21C mice during the Exercise period (EX) and after running wheels were locked (SED). C. Mean hourly ambulatory activity (Y-axis beam breaks) in both G28C and G21C groups (n = 8/group) during exercise period (Ex) and following exercise cessation (Sed). All data shown as means ± SEM for 3(EX) or 4(SED) consecutive days; for A and B * signifies p<0.05 by pair-wise t-test within groups, for C * signifies p<0.05 by one-way ANOVA for all hours indicated by the two lines (solid gray line G21C, dashed black line G28C).

Figure 6. Correlation between change in energy expenditure (EE) and change in food intake (FI) or wheel running (WR) in individual mice.

A Decrease in EE and FI for individual mice shown as % of value during the study period prior to the intervention (Baseline for Temp. change period (n = 15) and Exercise for the Sedentary period (n = 14). EE and FI data used were 3–4day averages for each distinct time period. Linear regression line (black) is parallel to line of identity (gray). B Decrease in dark cycle EE with exercise cessation vs. (negative) mean dark cycle WR distance during exercise period for mice housed at 21°C (open squares, dashed line) or 28°C (closed squares, solid line; note WR did not record for one mouse in the G28C group).

Figure 7. Change in body composition.

Change in fat mass and lean body mass at the end of the study relative to baseline body composition. Mice housed at 21°C (white bars, n = 8) throughout the study are compared to mice housed at 28°C (black bars, n = 8) throughout all phases of the study. Data are shown as means ± SEM with * showing p<0.05 by both paired (within-group relative to baseline) and unpaired (between group) Student's t-test and # showing p<0.05 by paired Student's t-test.