Caloric Restriction per se Rather Than Dietary Macronutrient Distribution Plays a Primary Role in Metabolic Health and Body Composition Improvements in Obese Mice

Abstract

Caloric restriction (CR) is of key importance in combating obesity and its associated diseases. We aimed to examine effects of dietary macronutrient distribution on weight loss and metabolic health in obese mice exposed to CR. Male C57BL/6J mice underwent diet-induced obesity for 18 weeks. Thereafter mice were exposed to a 6-week CR for up to 40% on either low-fat diet (LFD; 20, 60, 20% kcal from protein, carbohydrate, fat), low-carb diet (LCD; 20, 20, 60% kcal, respectively) or high-pro diet (HPD; 35, 35, 30% kcal, respectively) (n = 16 each). Ten mice on the obesogenic diet served as age-matched controls. Body composition was evaluated by tissue dissections. Glucose tolerance, bloods lipids and energy metabolism were measured. CR-induced weight loss was similar for LFD and LCD while HPD was associated with a greater weight loss than LCD. The diet groups did not differ from obese controls in hindlimb muscle mass, but showed a substantial decrease in body fat without differences between them. Glucose tolerance and blood total cholesterol were weight-loss dependent and mostly improved in LFD and HPD groups during CR. Blood triacylglycerol was lowered only in LCD group compared to obese controls. Thus, CR rather than macronutrient distribution in the diet plays the major role for improvements in body composition and glucose control in obese mice. Low-carbohydrate-high-fat diet more successfully reduces triacylglycerol but not cholesterol levels compared to isocaloric high-carbohydrate-low-fat weight loss diets.

Keywords: diets; high-protein; low-carbohydrate; low-fat; weight loss.

Figure 1

Changes in body mass (a,b), hindlimb muscle mass (c,d) and combined body fat (e,f) during a 6-week caloric restriction on low-fat (LFD), low-carbohydrate (LCD), high-protein (HPD) diets compared to the age-matched obese controls. Data are means with each dot represented by a single mouse. Statistical analysis: (a) Two-way repeated measures ANOVA with Bonferroni’s post hoc analysis was performed for effects of diet group and time, respectively; (b–f) One-way ANOVA with Bonferroni’s post hoc analysis was performed for diet group effect. N.S. indicates no statistical significance between compared means. ^††† p < 0.001 for effects of diet group (D), time (T), subject (S) and interaction (I). * p < 0.05 between diet groups connected by line. ^§§§ p < 0.001 pre vs. post values. ^# p < 0.05, ^### p < 0.001 vs. obese controls.

Figure 2

Blood glucose kinetics (a), blood glucose area under the curve (AUC) (b), blood lipid levels (c) and correlation between total blood cholesterol and body mass (d) following a 6-week caloric restriction on low-fat (LFD), low-carbohydrate (LCD), high-protein (HPD) compared to the age-matched obese controls. Abbreviations for (c): TAG, triacylglycerol; CHOL, cholesterol; HDL, high density lipoprotein. Data are means ± SD (a,c) or means with plotted individual values (b). Statistical analysis: (a) Two-way repeated measures ANOVA with Bonferroni’s post hoc analysis was performed for effects of diet group and time, respectively; (b,c^TAG) One-way ANOVA with Bonferroni’s post hoc analysis was performed for diet group effect; (c^CHOL,HDL) Non-parametric Kruskal-Wallis with Dunn’s post hoc analysis was performed for diet group effect. ^† p < 0.05, ^††† p < 0.001 for effects of diet group (D), time (T), subject (S) and interaction (I). ^# p < 0.05, ^## p < 0.01, ^### p < 0.001 vs. obese controls. * p < 0.05, ** p < 0.01 vs. LCD group.

Figure 3

Energy intake (a,b), energy expenditure (c,d), physical activity level (e) and respiratory quotient (f) following a 6-week caloric restriction on low-fat (LFD), low-carbohydrate (LCD), high-protein (HPD) diets compared to the age-matched obese controls. Data are means with each dot represents a single mouse (white dots in Figure b indicate mice which had some food leftovers after the final day of CR). Statistical analysis: (a,c,f) One-way ANOVA with Bonferroni’s post hoc analysis was performed for diet group effect; (b,e) Non-parametric Kruskal-Wallis with Dunn’s post hoc analysis was performed for diet group effect. N.S. indicates no statistical significance between compared means. ^†† p < 0.01, ^††† p < 0.001 for effect of diet group (D). ^## p < 0.01, ^### p < 0.001 vs. obese controls. * p < 0.05 vs. LCD group.