Long-term rates of mitochondrial protein synthesis are increased in mouse skeletal muscle with high-fat feeding regardless of insulin-sensitizing treatment

Abstract

Skeletal muscle mitochondrial protein synthesis is regulated in part by insulin. The development of insulin resistance with diet-induced obesity may therefore contribute to impairments to protein synthesis and decreased mitochondrial respiration. Yet the impact of diet-induced obesity and insulin resistance on mitochondrial energetics is controversial, with reports varying from decreases to increases in mitochondrial respiration. We investigated the impact of changes in insulin sensitivity on long-term rates of mitochondrial protein synthesis as a mechanism for changes to mitochondrial respiration in skeletal muscle. Insulin resistance was induced in C57BL/6J mice using 4 wk of a high-fat compared with a low-fat diet. For 8 additional weeks, diets were enriched with pioglitazone to restore insulin sensitivity compared with nonenriched control low-fat or high-fat diets. Skeletal muscle mitochondrial protein synthesis was measured using deuterium oxide labeling during weeks 10-12 High-resolution respirometry was performed using palmitoyl-l-carnitine, glutamate+malate, and glutamate+malate+succinate as substrates for mitochondria isolated from quadriceps. Mitochondrial protein synthesis and palmitoyl- l-carnitine oxidation were increased in mice consuming a high-fat diet, regardless of differences in insulin sensitivity with pioglitazone treatment. There was no effect of diet or pioglitazone treatment on ADP-stimulated respiration or H₂O₂ emission using glutamate+malate or glutamate+malate+succinate. The results demonstrate no impairments to mitochondrial protein synthesis or respiration following induction of insulin resistance. Instead, mitochondrial protein synthesis was increased with a high-fat diet and may contribute to remodeling of the mitochondria to increase lipid oxidation capacity. Mitochondrial adaptations with a high-fat diet appear driven by nutrient availability, not intrinsic defects that contribute to insulin resistance.

Keywords: deuterium oxide; mitochondria; protein turnover; respiration; tracer.

Fig. 1.

Assessment of glucose and insulin sensitivity. Fasting hyperglycemia was detected at weeks 4 and 12 in mice consuming a high-fat diet (HFD) compared with low-fat diet (LFD), with no differences with pioglitazone (PIO) treatment (A). Fasting hyperinsulinemia was detected at week 4 in mice consuming HFD, which was attenuated by week 12 with PIO treatment (B). Blood glucose concentrations during glucose tolerance testing (GTT) at week 12 (C and D). PIO treatment in HFD decreased the area under the curve (AUC) above baseline for glucose concentrations during GTT (E). Blood glucose concentrations during insulin tolerance testing (ITT) at week 12 (F and G). Two mice had a fall in blood glucose during GTT (1 each from HFD+PIO and LFD+CON), and one mouse in the HFD had a drastic rise in blood glucose during ITT; these mice were removed from the respective analyses. The fall in blood glucose from t = 0 to t = 30 min was greater with PIO treatment in HFD mice (H). Blood glucose concentrations during GTT and ITT were analyzed by two-way ANOVA (drug × time) within each diet group. AUC during GTT and ITT were compared within each diet using unpaired t-test. Bars indicate means  ± SD; AU, arbitrary units. n = 8–14 per group. ***P ≤ 0.001 for post-hoc comparison within diet group.

Fig. 2.

Body composition and whole body energy metabolism. Mice underwent body composition analysis by EchoMRI and continuous laboratory animal monitoring in individual cages at week 10. Body weight was greater in HFD+PIO mice because of greater fat mass (A). ANCOVA revealed that PIO treatment did not change the positive relationship between total daily energy expenditure (EE) and lean body mass during 24 h of feeding (B). PIO treatment resulted in greater energy expenditure during 24 h of fasting without an interaction with lean body mass (C). Whole body respiratory exchange ratio (RER) was lower with HFD with no effect of PIO (D). Body composition and RER data were compared with two-way ANOVA (diet × drug) with post hoc analysis performed using Tukey’s test. Bars indicate means  ± SD; n = 8–14 per group. ***P < 0.01 for post hoc comparison within diet group; #P < 0.0001 between intercepts of regression lines.

Fig. 3.

Long-term protein synthesis of skeletal muscle fractions. The fractional synthesis rate (FSR) of skeletal muscle protein pools was measured over 14 days using deuterium oxide labeling. Mitochondrial protein FSR was increased in HFD mice with no difference of PIO treatment (A). Sarcoplasmic protein FSR was increased with HFD, and there was a main effect of PIO treatment, but no diet × drug interaction (B). Myofibrillar protein FSR was not different between HFD or PIO treatment groups (C). Data were compared with two-way ANOVA (diet × drug). Data are means  ± SD; n = 8–14 per group.

Fig. 4.

Skeletal muscle mitochondrial respiration and H₂O₂ emission. High-resolution respirometry was performed on mitochondria isolated from quadriceps muscle using glutamate-malate-succinate (GMS) or palmitoyl-l-carnitine for fatty acid oxidation (FAO) followed by inhibitors and uncoupler. Oxygen consumption (JO₂) during glutamate-malate-succinate-based respiration was not different between diet or PIO groups when expressed relative to mitochondrial protein (mito) abundance (A). Oxygen consumption (JO₂) during fatty acid oxidation (OxPHOS_FAO) was increased with HFD with no effect of PIO when expressed relative to mitochondrial protein abundance (B). The addition of cytochrome c (cyto c) did not alter respiration and indicates that mitochondrial membranes were intact. There were no differences between groups in H₂O₂ emissions measured during glutamate-malate-succinate respiration (JH₂O₂; C). Oxidative damage was not different between diet or PIO treatment groups as measured by carbonyl modification to quadriceps muscle lysates (D). Representative images from blots for antibodies against oxidative damage and tubulin. Samples were analyzed for oxidative damage using nonderivatized (−) or derivatized samples (+) and were not boiled such that tubulin has an apparent molecular weight that is higher than predicted. Data were compared with two-way ANOVA (diet × drug). Data are means  ± SD; n = 8–14 per group.

Fig. 5.

Mitochondrial complex protein abundance. The protein abundance for subunits for complexes I–V was not different between diet or PIO treatment groups as measured using immunoblotting (A–E). Hydroxyacl-CoA dehydrogenase (HADH) protein abundance was increased with HFD (F). Representative images are provided from the same blot for subunit identification and Ponceau staining or tubulin as loading control (G and H). Data are means  ± SD; CTL, interassay control; n = 8–14 per group.

Fig. 6.

Autophagy and Akt activation. Autophagy and Akt activation were measured by immunoblotting of lysates from quadriceps muscle. The conversion of microtubule-associated light chain 3 (LC3)-I to LC3-II was greater with HFD mice and not different with PIO treatment (A–C). Protein content for p62 was decreased with HFD (D). LC3-II content of mitochondria isolated from gastrocnemius muscle was not different between diet or PIO groups (E). Akt phosphorylation was greater with HFD at Ser473 normalized to total Akt (F). PIO treatment did not alter the response to HFD. Representative images are provided with corresponding graphs. Voltage-dependent anion channel (VDAC) was used to normalize LC3-II content in mitochondrial isolations. Data are means  ± SD; (−), negative control for autophagy or phosphorylation; (+), positive control for autophagy or phosphorylation; CTL, interassay control; n = 8–14 per group.