Glucose substitution prolongs maintenance of energy homeostasis and lifespan of telomere dysfunctional mice

Abstract

DNA damage and telomere dysfunction shorten organismal lifespan. Here we show that oral glucose administration at advanced age increases health and lifespan of telomere dysfunctional mice. The study reveals that energy consumption increases in telomere dysfunctional cells resulting in enhanced glucose metabolism both in glycolysis and in the tricarboxylic acid cycle at organismal level. In ageing telomere dysfunctional mice, normal diet provides insufficient amounts of glucose thus leading to impaired energy homeostasis, catabolism, suppression of IGF-1/mTOR signalling, suppression of mitochondrial biogenesis and tissue atrophy. A glucose-enriched diet reverts these defects by activating glycolysis, mitochondrial biogenesis and oxidative glucose metabolism. The beneficial effects of glucose substitution on mitochondrial function and glucose metabolism are blocked by mTOR inhibition but mimicked by IGF-1 application. Together, these results provide the first experimental evidence that telomere dysfunction enhances the requirement of glucose substitution for the maintenance of energy homeostasis and IGF-1/mTOR-dependent mitochondrial biogenesis in ageing tissues.

Figure 1. Glucose feeding extends lifespan of telomere dysfunctional mice.

(a) G3 mTerc^−/− mice (12- to 15-month old) were fed with a glucose-rich diet when mice lost 10–15% of body weight on normal ad libitum diet (n=13 mice). The dot plot shows body weights of G3 mTerc^−/− mice at the indicated time points before (black line) and after switching the mice from a normal to a glucose-enriched diet (red line). (b) Survival curves of mTerc^+/+ and G3 mTerc^−/− mice under the different diets. G3 mTerc^−/− mice (12- to 15-month old) that exhibited weight loss on normal diet were shifted to a glucose-enriched diet (red line) or continuously fed with normal diet (black line). (c) Kaplan–Meier survival curve: mTerc^+/+ and G3 mTerc^−/− mice under the different diets. G3 mTerc^−/− mice (13.5-month old), which did not yet exhibit weight loss on normal diet, were shifted to a glucose-enriched diet (red dotted line) or continuously fed with normal diet (red straight line) (n=50 mice). Log-rank test, **=P<0.01. (d–f) Analysis of daily food intake normalized to body weight (n=9–11 mice per group). (d) Food intake in relation to body weight in grams per day. Energy intake (e) as well as energy excretion (f) in relation to the body weight of the respective groups per day. All statistical data were assessed using Student’s t-test and are presented as mean±s.d. WT, wild type. *P<0.05, **P<0.01, ***P<0.001.

Figure 2. Glucose supplementation rescues thymus atrophy and IGF-1 and glucose levels of telomere dysfunctional mice.

12–15 month old G3 mTerc^−/− mice with weight loss on normal ad libitum diet and age-matched mTerc^+/+ mice were analyzed under continuous exposure to normal ad libitum diet or 2 weeks after switching to a glucose enriched diet (+Glc): (a) Percentage of double-positive CD4⁺/CD8⁺ naive T cells in thymus (n=3 per group). (b) Blood plasma levels of IGF-1 and growth hormone (GH) in ng ml⁻¹ (n=4–7 mice per group). (c) Diurnal plasma glucose levels were measured every 4 h (n=4–5 mice per group). G3 mTerc^−/− mice (12- to 15-month old) on normal diet exhibit significantly lower glucose levels compared with all other groups (P<0.001). Glucose levels were measured in mg dl⁻¹. (d) Intracellular glucose levels in liver (n=4–5 mice per group). All statistical data were assessed using Student’s t-test and are presented as mean±s.e.m. NS, non–significant. *P<0.05, **P<0.01, ***P<0.001.

Figure 3. Telomere dysfunctional mice show elevated glucose depletion rates and glucose supplementation rescues catabolic metabolism in this context.

12–15 month old G3 mTerc^−/− mice that exhibited (a,c–f) weight loss on normal ad libitum diet or a regane in body weight (2 weeks after initiation of a glucose enriched diet) or (b) a reappearance of body weight loss 2 month after initiation of the glucose enriched diet were compared to age-matched mTerc^+/+ mice. (a) Ratio of U¹³C-labelled glucose at 180 min versus 10 min after i.p. injection of labelled glucose (n=3 mice per group). (b) Glucose tolerance test. The histogram shows blood glucose levels in mg dl⁻¹ after a bolus injection of glucose after overnight fasting (n=4–9 mice per group). (c) Insulin tolerance test. The histogram shows blood glucose levels in mg dl⁻¹ after a bolus injection of insulin (n=4–9 mice per group). (d) Representative pictures of Oil-red-O fat staining in liver sections. Red droplets indicate hepatic fat storage. (e) Serum levels of the essential amino acids (tryptophan (Trp), tyrosine (Tyr) and methionine (Met)) in aged mice with and without glucose supplementation (n=3 pools per group, three mice per pool). (f) The western blot analysis shows the expressions of phosphorylated mTOR in liver tissues of mice of the indicated genotype, fed with ad libitum normal diet or high-glucose diet (n=4–5 mice per group). All statistical data were assessed using Student’s t-test and are presented as mean±s.e.m. *P<0.05, **P<0.01, ***P<0.001.

Figure 4. Glucose supplementation improves energy homeostasis of telomere dysfunctional mice by increasing glycolysis and oxidative glucose metabolism.

12–15 month old G3 mTerc^−/− mice with weight loss on normal ad libitum diet and age-matched mTerc^+/+ mice were analyzed under continuous exposure to normal ad libitum diet or 2 weeks after switching to a glucose enriched diet (+Glc): (a,b) The bar graphs show serum levels of pyruvate (in pmol μl⁻¹) (a) and lactate (in nmol μl⁻¹) (b) in the indicated groups (n=4–6 mice per group). (c) Lactate production. For mice on normal diet, the amount of plasma M3 lactate isotopologues 20 min after injection of U¹³C₆ glucose normalized by the amount of plasma M6 glucose after 10 min (n=4–5 mice per group). For the group on the glucose-enriched diet, two times the amount of plasma M2 lactate isotopologues 20 min after injection of 1,2¹³C₂ glucose normalized by the amount of plasma M2 glucose after 10 min (n=4–5 mice per group). (d) The bar graph shows the ratio of 1-M0 citrate and M3-labelled lactate in liver tissue. Higher levels indicate increased relative flux into the TCA cycle (n= 4–7 mice per group). (e) The graph shows differences in the fractional gluconeogenesis (GNG) of the respective groups. Higher fractional GNG means a relative higher contribution of GNG to glucose flux (n=6 mice per group). (f) Aged human fibroblasts show a higher dependency on glucose. The graph shows the proliferation rate of BJ fibroblasts in early (PD 38) and late passage (PD 50) under different glucose concentrations (in g l⁻¹) of the medium. All statistical data were assessed using Student’s t-test and are presented as mean±s.e.m.*P<0.05, **P<0.01, ***P<0.001.

Figure 5. Glucose supplementation rescues mitochondrial mass and increases oxygen consumption and ATP levels in G3 mTerc^−/− mice.

12–15 month old G3 mTerc^−/− mice with weight loss on normal ad libitum diet and age-matched mTerc^+/+ mice were analyzed under continuous exposure to normal ad libitum diet or 2 weeks after switching to a glucose enriched diet (+Glc): (a–d,g,f) Histograms showing (a) the mitochondrial DNA (mtDNA) copy number in the indicated tissues, (b) the expression of PGC-1α and PGC-1β in the indicated tissues, (c) hepatic citrate synthase (CS) expression, (d) hepatic cytochrome c oxidase levels (COX), (f) oxygen consumption of freshly isolated haemtopoietic cells (Lin-negative), and (g) ATP-levels of freshly isolated haemtopoietic cells (Lin-negative). Data in all histograms are shown as relative expression levels/numbers with data for mTerc^+/+ mice on normal ad libitum diet being set to 1. (a-d,g) n=4–6 mice per group, (f) n=9–17 mice per group. (e) Representative western blot analysis of OXPHOS enzymes ((NDUFA (CI), 30 kDa su (CII), ATPα (CV), Core 2 (CIII)) in liver homogenates. All statistical data were assessed using Student’s t-test and are presented as mean±s.e.m. *P<0.05, **P<0.01, ***P<0.001.

Figure 6. DNA damage signalling impairs mitochondrial biogenesis and function in ageing G3 mTerc^−/− mice.

12–15 month old G3 mTerc^−/− mice with weight loss on normal ad libitum diet and age-matched mTerc^+/+ mice were analyzed under continuous exposure to normal ad libitum diet or 2 weeks after switching to a glucose enriched diet (+Glc): (a) Relative expression of the DNA damage signalling members (p21 and PUMA) in liver and skeletal muscle (n=4–5 mice per group). Activation of p53 targets occurs in skeletal muscle but not in liver of G3 mTerc^−/− mice. Glucose supplementation partially rescued p53 activation amplified in skeletal muscle of G3 mTerc^−/− mice, suggesting that starvation responses in G3 mTerc^−/− mice with weight loss on normal diet could amplify the activation of p53 in tissues. Data for mTerc^+/+ mice were set to 1. (b,c) There is a direct relation between the activity of DNA damage signalling and mitochondrial biogenesis induced by glucose supplementation. The scatter plots show the relation between the mitochondrial DNA (mtDNA) copy number rescue rate and the relative expression of p21 (b) and PUMA (c) in both muscle and liver tissue of ageing G3 mTerc^−/− mice. The mtDNA copy number rescue rate was calculated by dividing the delta in mtDNA copy numbers in G3 mTerc^−/− mice on glucose-enriched diet compared with G3 mTerc^−/− mice on normal diet through the delta in mtDNA copy numbers in G3 mTerc^−/− mice on normal diet compared with wild-type mice on normal diet (n=7 mice per group). (d) KO of p53 downstream targets rescue impaired mitochondrial biogenesis in G3 mTerc^−/− mice. The bar graph shows relative mtDNA copy number in liver of 12- to 15-month-old G3 mTerc^−/− mice and controls, as well as age-matched G3 mTerc^−/− mice, p21^−/− double KO mice and controls (n=4–5 mice per group). Data for mTerc^+/+ mice were set to 1. (e,f) Western blot analysis of protein levels of phosphorylated ACC (p-ACC; as a phosphorylated AMPK (p-AMPK; target) and phosphorylated AMPK in frozen liver (e) and skeletal muscle (f) extracts of the indicated groups (n=4–6 mice per group). All statistical data were assessed using Student’s t-test and are presented as mean±s.e.m. *P<0.05, **P<0.01.

Figure 7. Suppression of IGF-1 and mTOR contributes to impairments in mitochondrial biogenesis and function in ageing G3 mTerc^−/− mice.

(a–d) G3 mTerc^−/− mice (12- to 15-month old) on glucose-enriched diet for 2 weeks were shifted to normal diet and separated into two groups: one saline-treated group (n=4 mice) and one IGF-1-treated group (n=5 mice). (a) The graph shows the relative body weight loss of the respective groups at the indicated time points. (b) Western blot analysis of the respective groups for members of the respiratory chain (NDUFA (CI), 30 kDa su (CII), CIV, ATPα (CV)) as well as PGC1α and TFAM. (c,d) The bar graphs show the relative mitochondrial citrate synthase (CS) (c) and cytochrome c oxidase (COX) expression (d) in liver homogenates of mice with or without IGF-1 treatment. (e) The bar graphs show the relative mitochondrial DNA (mtDNA) copy number in liver of 12- to 15-month-old G3 mTerc^−/− mice with weight loss as well as mTerc^+/+controls with and without glucose supplementation and rapamycin treatment (n=5–6 mice per group). Data for mTerc^+/+ mice were set to 1. (f) The bar graph shows the relative oxygen consumption rate (OCR) of haematopoietic cells (Lin-negative, Lin−) of the depicted mouse cohorts on the indicated diets with and without rapamycin treatment (n=4 mice per group). Data for mTerc^+/+ mice were set to 1. (g) The bar graph shows relative ATP levels of freshly isolated haematopoietic cells (Lin−) on the different diets with and without rapamycin treatment. (n=4 mice per group). Data for mTerc^+/+ mice were set to 1. (h) The reduction of mtDNA copy number was significantly increased in late-passage fibroblasts when compared with early passage at 0.25 g l⁻¹ glucose concentration in culture medium. Data are shown as relative values with mtDNA copy number at 1gl⁻¹ glucose concentration set to 1. Data for mTerc^+/+ mice were set to 1. All statistical data were assessed using Student’s t-test and are presented as mean±s.e.m. *P<0.05, **P<0.01, ***P<0.001.

Figure 8. Model of telomere dysfunction-induced metabolic changes that accelerate tissue ageing.

Telomere dysfunction enhances energy consumption in ageing tissues leading to compensatory increases in ATP production, glucose metabolism (via glycolysis and TCA cycle) and oxygen consumption. Under these conditions, the dietary energy content becomes limiting for the maintenance of energy homeostasis. Impairments in energy homeostasis result in catabolic changes and suppression of IGF-1/mTOR-dependent mitochondrial biogenesis, which is aggravated by the activation of p53/p21-dependent DNA damage responses in telomere dysfunctional tissues. Glucose supplementation stops this vicious circle by increasing bioavailability of substrates for glycolytic and oxidative glucose metabolism resulting in improvements of energy homeostasis and activation of IGF-1/mTOR-dependent mitochondrial biogenesis. This leads to further enhancement of oxidative glucose metabolism and ATP levels, thus prolonging the maintenance of functional tissues in the context of telomere dysfunction.