The TSC complex is required for the benefits of dietary protein restriction on stress resistance in vivo

Abstract

Protein restriction (PR) is important for the benefits of dietary restriction on longevity and stress resistance, but relevant nutrient sensors and downstream effectors in mammals remain poorly defined. We used PR-mediated protection from hepatic ischemia reperfusion injury to probe genetic requirements for the evolutionarily conserved nutrient sensors GCN2 and mTORC1 in stress resistance. One week of PR reduced free amino acids and circulating growth factors, activating GCN2 and mTORC1 repressor tuberous sclerosis complex (TSC). However, although GCN2 was dispensable for PR-induced protection, hepatic TSC1 was required. PR improved hepatic insulin sensitivity in a TSC1-dependent manner prior to ischemia, facilitating increased prosurvival signaling and reduced apoptosis after reperfusion. These benefits were partially abrogated by pharmacological PI3K inhibition or genetic deletion of the insulin receptor in hepatocytes. In conclusion, improved insulin sensitivity upon short-term PR required TSC1, facilitated increased prosurvival signaling after injury, and contributed partially to PR-mediated resistance to clinically relevant ischemia reperfusion injury.

Copyright © 2014 The Authors. Published by Elsevier Inc. All rights reserved.

Figures

Figure 1. Dietary protein restriction mediates stress resistance independent of GCN2

(A) Serum LDH, ALT and AST activity before ischemia (0 hr) and 3 hrs after reperfusion in male mice preconditioned on the indicated diets for 1 wk prior to hepatic IRI; n = 5 mice/group. Complete: ad libitum access to complete diet; 35% DR: 65% of ad libitum food allowance on complete diet; 55% DR: 45% of ad libitum food allowance on complete diet, PR: 65% of ad libitum food allowance on protein-free diet. Asterisks indicate the significance of the difference vs. the Complete group by 1-way ANOVA with Dunnett’s test for multiple comparisons, multiplicity adjusted p-values; *p < 0.05, **p < 0.005, ***p < 0.0005, ****p < 0.0001. (B) Serum ALT activity before ischemia (0 hr) and 3 hrs after reperfusion in female GCN2KO mice preconditioned on the indicated diets for 1 wk prior to hepatic IRI; n = 4–5 mice/group. Complete: ad libitum access to complete diet; PR: 65% of ad libitum food allowance on protein-free diet; Trp−: 65% of ad libitum food allowance on tryptophan-free diet. Asterisk indicates the significance of the difference vs. the Complete group by student’s t-test; *p < 0.05. (C) Immunoblots of markers of mTORC1 signaling in extracts from primary dermal fibroblasts or hepatocytes isolated from WT or GCN2KO mice and treated with the indicated media in the presence of 10% dialyzed FBS (2 hrs for MDFs, 3 hrs for hepatocytes). Complete: complete DMEM; AA-free: amino acid-free DMEM; MEM: minimum essential medium amino acid mix; AA: all amino acids added back to 1× DMEM concentration; Rapamycin: 20 nM. Data in all panels are shown as means ± SD. See also Figure S1.

Figure 2. The TSC complex is required for inhibition of mTORC1 upon PR in vivo

(A) Body and liver weights of the indicated genotype after 1 wk on the indicated diet; n = 4–5 mice/group. Body weight on the last day of treatment is represented as percentage of initial weight (top). Liver weights on the last day of treatment are represented as absolute values (middle) and percentages of body weight (bottom). Asterisks indicate the significance of the difference between diets within genotype by student’s t-test; **p

Figure 3. Differential reduction of essential amino acids and growth factors in vivo upon PR

(A) Immunoblots of mTORC1 targets from extracts of primary hepatocytes isolated from WT or LTsc1KO mice and starved of amino acids in the presence of 10% dialyzed serum (dFBS) for 2 hrs prior to lysis, coincident with 1× MEM addition +/− 20nM rapamycin (Rap) as indicated. (B, C) Free EAA levels in serum (B) and liver (C) of WT or LTsc1KO mice on the indicated diet expressed as fold change relative to the WT complete diet group; n = 4–5 mice/group. Asterisks/number signs indicate the significance of the difference between diets within genotype by student’s t-test; */#p < 0.05. (D–F) Blood glucose, serum insuin (E) and IGF-1 (F) levels of WT or LTsc1KO mice on the indicated diet; n = 4–5 mice/group for each experiment, with data pooled from 3 separate experiments (D). Asterisks indicate the significance of the difference between diets within genotype by student’s t-test; *p

Figure 4. The TSC complex is required for benefits of PR against acute hepatic stress

(A) Serum LDH, ALT and AST activity before ischemia (0 hr) and 3 or 24 hrs after reperfusion in female WT or LTsc1KO mice preconditioned on the indicated diets for 1 wk prior to hepatic IRI; n = 4–6 mice/group for each experiment, with data pooled from 4 separate experiments for 0 and 3 hrs and a single experiment for 24 hrs. Asterisks indicate the significance of the difference between diets within genotype by student’s t-test; **p −8. (B) Representative images of H&E stained sections from non-ischemic (left) or ischemic livers 24 hrs after reperfusion (right) from WT or LTsc1KO mice preconditioned on the indicated diets for 1 wk prior to hepatic IRI. Lightly stained areas with loss of hepatocyte morphology indicate necrotic tissue. Right: representative images of whole livers 24 hrs after IRI and fixed with formalin, with areas of hemorrhagic necrosis appearing as dark red. Scale bar: 400 µm. (C) Quantification of hepatic cell death expressed as percentages of necrotic region/field in 3–5 mice/group, 3 sections/sample; Asterisks indicate the significance of the difference between diets within genotype by student’s t-test; **p

Figure 5. The TSC complex is required for improved hepatic insulin sensitivity upon PR

(A) Insulin sensitivity as determined by immunoblotting for markers of Akt pathway activation in liver extracts from mice fasted for 6 hrs and then stimulated with 0.5 U/kg insulin by portal vein injection 3 min before harvest. (B) Akt activation status as determined by immunoblotting of liver extracts 3 hrs after reperfusion from mice preconditioned for 1 wk on the indicated diet prior to induction of hepatic IRI. (C) Serum insulin levels from tail blood of WT mice preconditioned for 1 wk on the indicated diet taken prior to ischemia (0 hr), at the end of the ischemic period (30 min I, n=3–4), 30 min after reperfusion period (30 min R, n=3) or 3 hrs after reperfusion (3 hrs R, n=11–12). Asterisk indicates the significance of the indicated comparison according to a Kruskal-Wallis test followed by Dunn’s multiple comparisons test; *p

Figure 6. Increased prosurvival signaling and reduced apoptosis contribute to PR-mediated protection

(A, B) Immunoblots of pro-apoptotic Bcl-2 family members in liver extracts from WT or LTsc1KO mice preconditioned on the indicated diets for 1 wk and harvested without ischemia (A) or 3 hrs after reperfusion (B). NI: non-ischemic. (C, D) TUNEL staining of non-ischemic or ischemic livers from the indicated diet-genotype combinations 3 hrs after reperfusion. (C) Quantification was performed by blind scoring of the number of TUNEL+ nuclei/200× field in 10 random fields from each sample, with n = 5 mice/group. Asterisks indicate the significance of the difference between diets within genotype by student’s t test; *p < 0.05. (D) Representative images of TUNEL-stained sections. Scale bar: 100 µm. (E) Immunoblot of Akt phosphorylation in liver extracts from WT mice injected with vehicle or 1 mg/kg wortmannin 1 hr prior to harvest. (F) Serum ALT and AST activity 3 hrs after reperfusion in WT mice preconditioned on the indicated diets for 1 wk and injected with vehicle or 1 mg/kg wortmannin (Wmn) 1 hr prior to hepatic IRI; n = 8–10 mice/group. Statistical significance was assessed by student’s t-test between Complete-V and P-RV groups (***p #p < 0.05). (G) Immunoblot of liver extracts from WT mice on complete diet, or WT and LIrKO mice preconditioned with PR for 1 wk and harvested 3 hrs after reperfusion. (H) Serum LDH and AST activity before ischemia (0 hr) and 3 hrs after reperfusion in male WT and LIrKO mice preconditioned on the indicated diets for 1 wk prior to hepatic IRI; n = 4–5/group. Statistical significance was assessed by student’s t-test between WT-Complete and WT-PR groups (***p ##p < 0.001). Data in all panels are shown as means ± SD. See also Figure S6.

Figure 7. Model of TSC/mTORC1 function and insulin sensitivity in PR-mediated hepatic stress resistance

A model for the effects of dietary protein or tryptophan restriction on improved hepatic insulin sensitivity prior to an acute ischemia event. After reperfusion, increased insulin levels and increased hepatic insulin sensitivity facilitate pro-survival signaling and contribute to protection from injury.