Contribution of impaired myocardial insulin signaling to mitochondrial dysfunction and oxidative stress in the heart

Abstract

Background: Diabetes-associated cardiac dysfunction is associated with mitochondrial dysfunction and oxidative stress, which may contribute to left ventricular dysfunction. The contribution of altered myocardial insulin action, independent of associated changes in systemic metabolism, is incompletely understood. The present study tested the hypothesis that perinatal loss of insulin signaling in the heart impairs mitochondrial function.

Methods and results: In 8-week-old mice with cardiomyocyte deletion of insulin receptors (CIRKO), inotropic reserves were reduced, and mitochondria manifested respiratory defects for pyruvate that was associated with proportionate reductions in catalytic subunits of pyruvate dehydrogenase. Progressive age-dependent defects in oxygen consumption and ATP synthesis with the substrate glutamate and the fatty acid derivative palmitoyl-carnitine were observed. Mitochondria also were uncoupled when exposed to palmitoyl-carnitine, in part as a result of increased reactive oxygen species production and oxidative stress. Although proteomic and genomic approaches revealed a reduction in subsets of genes and proteins related to oxidative phosphorylation, no reductions in maximal activities of mitochondrial electron transport chain complexes were found. However, a disproportionate reduction in tricarboxylic acid cycle and fatty acid oxidation proteins in mitochondria suggests that defects in fatty acid and pyruvate metabolism and tricarboxylic acid flux may explain the mitochondrial dysfunction observed.

Conclusions: Impaired myocardial insulin signaling promotes oxidative stress and mitochondrial uncoupling, which, together with reduced tricarboxylic acid and fatty acid oxidative capacity, impairs mitochondrial energetics. This study identifies specific contributions of impaired insulin action to mitochondrial dysfunction in the heart.

Figure 1. Age-dependent changes in mitochondrial respiration and ATP synthesis rates in CIRKO and wild-type mice

(Data are means±SEM). (A), (D) and (G): V_ADP, V_Oligo, ATP synthesis rates and ATP/O ratio in the presence of glutamate-malate in 8- (n=4), 24- (n=4) and 54- (n=5) week-old CIRKO (black bars) and wild-type (WT) (white bars) respectively. Statistics (2-Way ANOVA): Panel A: V_ADP –WT vs. CIRKO, p<0.01, 54-weeks vs. 8-weeks, p<0.01, 24-weeks vs. 54-weeks, P=0.06. Panel D and G: No statistical differences between groups. (B), (E), and (H) Same as (A), (D), (G) but in the presence of pyruvate-malate. Statistics (2-Way ANOVA): Panel B: V_ADP – WT vs. CIRKO, p<0.005; Panel E: V_Oligo – 54-weeks vs. 24 or 8-weeks, p<0.01; Panel H: ATP – WT vs. CIRKO, p<0.05. (C), (F), and (I) Same as (A), (D), (G) but in the presence of palmitoyl-carnitine and malate. Statistics (2-Way ANOVA): Panel C: V_ADP – WT vs. CIRKO (at all ages), p<0.05, Decline in CIRKO with age (p<0.005, after multiple comparison adjustments). Panel F: V_Oligo – WT vs. CIRKO (at 8-weeks), p<0.005, Increase in WT with age (p<0.005, after multiple adjustments). Panel I: ATP – WT vs. CIRKO, p<0.01 and p<0.05 at 24 and 54-weeks respectively. ATP/O – WT vs. CIRKO, p<0.01.

Figure 2. Mitochondrial function and electron transport chain complex activity

(A) and (B) Respiration traces of mitochondria (0.4 mg/ml) isolated from 24-week-old wild-type (n = 5) and CIRKO mice (n = 5) respectively, using palmitoyl-carnitine as substrate. (C) and (D) ATP and ATP/O ratio (means±SEM) measured on isolated mitochondria from wild-type and CIRKO mice at 24-weeks of age using palmitoyl-carnitine as substrate. (E) Representative blue-native polyacrylamide gel electrophoresis (BN-PAGE) Arrows indicate separation of complexes I, II, III, IV, and V. (F) and (G) Representative in-gel complex activity stainings, and densitometric quantification (means±SEM) of complex I, IV, and V activities, of wild-type and CIRKO mice at 8- and 76-weeks of age (n = 4 – 5). * p < 0.05 vs. age-matched wild-type controls.

Figure 3. Increased oxidative stress in CIRKO hearts

(Data are means±SEM). A) Aconitase activity measured in mitochondria isolated from 8- and 24-week-old CIRKO (black bars, n=5) and wild-type (white bars, n=6) hearts. Statistics (2-Way ANOVA): WT vs. CIRKO, p<0.05 and p<0.005 at 8 and 24-weeks respectively. B) Hydrogen peroxide (H₂O₂) production in isolated mitochondria obtained from 8 and 24-week-old CIRKO (n=7) and wild-type (n=7). Mitochondria were incubated with palmitoyl-carnitine and L–carnitine in the presence of oligomycin as described in the methods. Statistics (2-Way ANOVA): WT vs. CIRKO, p<0.005 C) Mitochondrial aconitase activity determined in saline-treated wild-type (open bars, n=6 per group), and CIRKO mice (black bars, n=6), and MnTBAP-treated wild-type (dotted bars, n=6) and CIRKO mice (hatched bars, n=6), age 8-weeks. D) Oligomycin-insensitive respiration (V_oligomycin) in saponin-permeabilized cardiac fibers from wild-type (saline and MnTBAP) and CIRKO mice (saline and MnTBAP), age 8-weeks. * p<0.05, ** p < 0.005 versus wild-type; ^a p < 0.005 versus CIRKO saline; ^b p < 0.005 versus wild-type MnTBAP (1-Way ANOVA).

Figure 4. Mitochondrial number and morphology in CIRKO and wild-type hearts

A) Representative electron micrographs of heart sections (80–100 nm) from 3-, 8-, 24-, and 54-week-old wild-type (1, 2, 3 and 4) and CIRKO (5, 6, 7 and 8) mice (three per group). B) Quantification of mitochondrial number in CIRKO hearts (fold change vs. wild-type normalized to 1±SEM). Statistics (2-Way ANOVA): WT vs. CIRKO, p<0.005. C) Mitochondrial volume density relative to cell area in wild-type and CIRKO mice at 8- and 24-weeks of age. Statistics (2-Way ANOVA): WT vs. CIRKO, p<0.005.