Defects in mitochondrial efficiency and H2O2 emissions in obese women are restored to a lean phenotype with aerobic exercise training

Abstract

The notion that mitochondria contribute to obesity-induced insulin resistance is highly debated. Therefore, we determined whether obese (BMI 33 kg/m(2)), insulin-resistant women with polycystic ovary syndrome had aberrant skeletal muscle mitochondrial physiology compared with lean, insulin-sensitive women (BMI 23 kg/m(2)). Maximal whole-body and mitochondrial oxygen consumption were not different between obese and lean women. However, obese women exhibited lower mitochondrial coupling and phosphorylation efficiency and elevated mitochondrial H2O2 (mtH2O2) emissions compared with lean women. We further evaluated the impact of 12 weeks of aerobic exercise on obesity-related impairments in insulin sensitivity and mitochondrial energetics in the fasted state and after a high-fat mixed meal. Exercise training reversed obesity-related mitochondrial derangements as evidenced by enhanced mitochondrial bioenergetics efficiency and decreased mtH2O2 production. A concomitant increase in catalase antioxidant activity and decreased DNA oxidative damage indicate improved cellular redox status and a potential mechanism contributing to improved insulin sensitivity. mtH2O2 emissions were refractory to a high-fat meal at baseline, but after exercise, mtH2O2 emissions increased after the meal, which resembles previous findings in lean individuals. We demonstrate that obese women exhibit impaired mitochondrial bioenergetics in the form of decreased efficiency and impaired mtH2O2 emissions, while exercise effectively restores mitochondrial physiology toward that of lean, insulin-sensitive individuals.

Figure 1

Experimental design. Blood and skeletal muscle biopsy samples were obtained in the fasted state from lean women (n = 14) after 3 days of a standardized weight-maintaining diet. Obese women were randomized to AET (n = 12) or SED (n = 13) and participated in study days A and B before and after the interventions. Obese women also consumed a standardized, weight-maintaining diet 3 days before and during their study days. Study day A consisted of a 3-h hyperinsulinemic-euglycemic pancreatic clamp. Blood samples were obtained every 10 min to adjust the GIR to maintain euglycemia (∼90 mg/dL). After the clamp was completed, standardized meals were provided to keep participants weight stable. Study day B consisted of a fasted skeletal muscle biopsy, followed by consumption of a high-fat mixed meal. Blood samples were collected every 30 min for 4 h after the meal. A second muscle biopsy was obtained 4 h after the meal as previously described (18). *Blood sample.

Figure 2

Obese women exhibited lower mitochondrial bioenergetics efficiency and increased mtH₂O₂ emissions. Whole-body oxygen consumption (A), maximal citrate synthase (CS) activity (B), and mitochondrial oxidative capacity (C) were not different between sedentary obese and lean women. Maximal mitochondrial oxidative capacity was determined during state 3 respiration with substrates targeting CI, CI+II, and CII. However, greater basal and state 4 respiration (C) and lower RCR (state 3/state 4) (D) indicate decreased mitochondrial coupling efficiency in obese women. Obese women also demonstrated reduced phosphorylation efficiency (ADP:O) (E). mtH₂O₂ emissions were increased in obese compared with lean women during state 2 conditions with glutamate (10 mmol/L), malate (2 mmol/L), and succinate (10 mmol/L). Mitochondrial energetics in obese and lean women were evaluated in isolated mitochondria and expressed relative to mitochondrial protein content (micrograms of protein). Data presented as mean ± SEM. #Statistical differences from lean (P < 0.05).

Figure 3

Mitochondrial efficiency and mtH₂O₂ emission are restored with exercise training. Whole-body oxygen consumption (A), maximal citrate synthase (CS) activity (B), and mitochondrial oxidative capacity (D and E) in obese women were increased after 12 weeks of AET and mtDNA was unaltered (C). Moreover, deficiencies in mitochondrial coupling (RCR) and phosphorylation efficiency (ADP:O) in obese women were improved after exercise and no longer different than in lean women (F and G). mtH₂O₂ emission in isolated mitochondria was reduced after exercise training, reaching values that were not different than lean women (H). The finding that exercise lowers mtH₂O₂ emission was recapitulated in permeabilized myofibers during a stepwise succinate titration protocol (0.25–12 mmol/L) during state 4 conditions (2 μg/μL oligomycin) (I and J). Baseline comparisons between groups (AET versus SED) via an unpaired Student t test revealed no statistical differences (P = 0.17). SED did not alter any mitochondrial parameters. Data from isolated mitochondria were expressed relative to mitochondrial protein content (micrograms of protein). mtH₂O₂ emission from permeabilized myofibers were expressed relative to fiber bundle wet weight (milligrams). Data are mean ± SEM. *Statistical differences from before intervention (PRE) (P < 0.05). AU, arbitrary units.

Figure 4

Catalase activity is increased and DNA oxidative damage is reduced after exercise training. Maximal catalase buffering capacity assessed in skeletal muscle homogenates was increased after AET (A). Skeletal muscle DNA oxidative damage was reduced when determined from 8-oxo-dG, an abundant adduct formed by DNA oxidation (B). Protein oxidative damage was preserved with exercise, while the SED group accumulated damaged proteins (C). Assessment of protein carbonyl provides a crude assessment of cellular protein oxidative damage as seen in the representative blot. Vinculin was used to confirm equal protein loading (D). Maximal activities of cytoplasmic (SOD1) and mitochondrial (SOD2) SOD were unaltered with exercise training or sedentary behavior (E and F). mRNA expression of antioxidants catalase, Sod2, Gpx4, and Txrnd2 and mitochondrial marker CoxIV was not different between groups (G and H). Sod1 mRNA expression decreased after AET. Antioxidant activity is expressed relative to muscle homogenate protein content (milligrams of protein). Data are mean ± SEM. *Statistical differences from before intervention (PRE) (P < 0.05).

Figure 5

mtH₂O₂ emissions were increased after acute, high-fat feeding in the trained state. During SED conditions (before intervention [PRE] AET, PRE and POST SED), acute high-fat feeding did not alter mtH₂O₂ emissions. However, after exercise training (POST AET), feeding increased mtH₂O₂ production (A). mtH₂O₂ emissions were evaluated in isolated mitochondria during state 2 conditions. Acute, high-fat feeding did not alter mitochondrial respiration before or after either intervention (B). Mitochondrial oxygen consumption presented in Figure 5 is during state 3 respiration through CI+II supported by glutamate, malate, succinate, and ADP. Data are presented as mean ± SEM. $Effect of feeding (P < 0.05); *difference from PRE (P < 0.05). hrs, hours.

Figure 6

Exercise improves insulin sensitivity. The GIR required to maintain euglycemia during the hyperinsulinemic clamp was increased after exercise training and was not altered during SED (A). A similar trend (P = 0.06) was observed for postprandial insulin sensitivity using the oral minimal model (B). AET lowered the glucose AUC after a high-fat mixed meal (C). Insulin AUC after the meal was also lower (P = 0.05) (D), but C-peptide was not altered (P = 0.17) with exercise (E). ΔAbsolute change in the AUC. Data are means ± SEM. *Difference from before intervention (PRE) (P < 0.05); ^difference from PRE (P ≤ 0.06).