Age-Dependent Protection of Insulin Secretion in Diet Induced Obese Mice

Abstract

Type 2 diabetes is an age-and-obesity associated disease driven by impairments in glucose homeostasis that ultimately result in defective insulin secretion from pancreatic β-cells. To deconvolve the effects of age and obesity in an experimental model of prediabetes, we fed young and aged mice either chow or a short-term high-fat/high-sucrose Western diet (WD) and examined how weight, glucose tolerance, and β-cell function were affected. Although WD induced a similar degree of weight gain in young and aged mice, a high degree of heterogeneity was found exclusively in aged mice. Weight gain in WD-fed aged mice was well-correlated with glucose intolerance, fasting insulin, and in vivo glucose-stimulated insulin secretion, relationships that were not observed in young animals. Although β-cell mass expansion in the WD-fed aged mice was only three-quarters of that observed in young mice, the islets from aged mice were resistant to the sharp WD-induced decline in ex vivo insulin secretion observed in young mice. Our findings demonstrate that age is associated with the protection of islet function in diet-induced obese mice, and furthermore, that WD challenge exposes variability in the resilience of the insulin secretory pathway in aged mice.

Figure 1

The age-dependent heterogeneity in weight gain after 4 weeks of Western diet in group-housed mice is highly correlated with impaired glucose tolerance. (a) Young mice (4–6 mo, black) and aged mice (22 mo, red) were co-housed in groups of 3–4 animals per cage. Mice were fed Western diet (WD) ad libitum for four weeks. Before and after diet, glucose tolerance and weight were measured. (b) Average weight in young (n = 20) and aged (n = 32) mice before and after WD. (c) Glucose tolerance test (GTT) in young (n = 20) and aged (n = 32) mice quantified by area under the curve (AUC). (d,e) Distribution of weight gain (Δg) (d) and correlation (e) between weight gain and area under the curve (AUC) in a glucose tolerance test (GTT) in young (n = 20) and aged (n = 32) mice. Data are ± SEM and were compared by paired two-way ANOVA with Sidak posttest (b and c), Tukey plot (d), and linear regression (e). NC, not correlated (P > 0.05). ***P < 0.001 and ****P < 0.0001.

Figure 2

Housing-independent correlation of weight gain and impaired glucose tolerance in aged mice fed a Western Diet. (a) Young mice (4 mo, black) and aged mice (22 mo, red) were singly housed for a one-week adjustment period before administration of chow or Western diet (WD) ad libitum for four weeks. Before and after diet, weight, glucose tolerance, and in vivo glucose-stimulated insulin secretion (GSIS) weight were assessed. (b–i) Average weight (b,f), glucose tolerance (c,g), distribution of weight gain (d,h), and the correlation between weight gain and GTT AUC (e,i) were assessed before and after WD (n = 17–24) or chow diet (n = 18–22). Data are ± SEM and were compared by paired two way ANOVA with Sidak posttest (b and c), Tukey plot (d and h), and linear regression (e). *P < 0.05, **P < 0.01, and ****P < 0.0001.

Figure 3

Weight gain correlates with energy expenditure in aged mice fed Western diet, but is not dependent on food consumption or spontaneous activity. (a–d) Average weight (a), food consumption (b), spontaneous activity (c), and average energy expenditure (d) in singly-housed aged mice following administration of chow or Western diet (chow, solid, n = 9; WD, cross-hatch, n = 9). (e–g) Weight gain was compared to food consumption (e), spontaneous activity (f), and energy expenditure (g) to acquire correlations. Data are ± SEM and were compared by paired two way ANOVA with Sidak posttest (a–d) and linear regression (e–g). *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

Figure 4

Compared with young mice, in vivo glucose-stimulated insulin secretion (GSIS) is enhanced in aged mice following Western diet administration. (a–d) Plasma glucose (a) and insulin (b) levels after an overnight fast in young (black) and aged (red) mice on chow (n = 16–17) and WD (n = 13–19) were plotted as a function of weight gain (c). Plasma insulin levels were again assessed 15 min after intraperitoneal injection of glucose (d). (e,f) In vivo GSIS plotted as a function of weight gain (e) and GTT AUC (f). The dashed circle indicates two mice excluded from regression analysis. Data are ± SEM and were compared by paired two-way ANOVA with Sidak posttest (a,b,d) and linear regression (c,e,f). **P < 0.01, ***P < 0.001, and ****P < 0.0001.

Figure 5

β-cell mass is increased by Western diet in both young and aged mice. Representative insulin staining and quantification of β-cell mass (a) and α:β-cell ratio (b) in chow and WD-fed young and aged mice as indicated. β-cell mass, n = 9 mice per group, scale bar = 400 μm; α:β-cell ratio, n = 3–4 mice per group, scale bar = 50 μm. Data are ± SEM and were compared by t-test. *P < 0.05.

Figure 6

Aged mice are protected from Western diet-induced insulin secretory decline in isolated pancreatic islets. (a,b) Ex vivo GSIS (a) and insulin content (b) measured in pancreatic islets from young (black) and aged (red) mice on chow (n = 7–9) and WD (n = 9–11). (c,d) Correlation of ex vivo GSIS as a function of weight gain (c) and in vivo GSIS (d). (E) Summary of correlations (R²) of mice fed a Western Diet in the singly-housed study (Figs 2–6). Abbreviations: Energy Expend, energy expenditure; ND, not determined. Data are ± SEM and were compared by paired two-way ANOVA with Sidak post-test (a and b) or linear regression (c–e). **P < 0.01 and ****P < 0.0001.