Enhanced mitochondrial superoxide scavenging does not improve muscle insulin action in the high fat-fed mouse

Abstract

Improving mitochondrial oxidant scavenging may be a viable strategy for the treatment of insulin resistance and diabetes. Mice overexpressing the mitochondrial matrix isoform of superoxide dismutase (sod2(tg) mice) and/or transgenically expressing catalase within the mitochondrial matrix (mcat(tg) mice) have increased scavenging of O2(˙-) and H2O2, respectively. Furthermore, muscle insulin action is partially preserved in high fat (HF)-fed mcat(tg) mice. The goal of the current study was to test the hypothesis that increased O2(˙-) scavenging alone or in combination with increased H2O2 scavenging (mtAO mice) enhances in vivo muscle insulin action in the HF-fed mouse. Insulin action was examined in conscious, unrestrained and unstressed wild type (WT), sod2(tg), mcat(tg) and mtAO mice using hyperinsulinemic-euglycemic clamps (insulin clamps) combined with radioactive glucose tracers following sixteen weeks of normal chow or HF (60% calories from fat) feeding. Glucose infusion rates, whole body glucose disappearance, and muscle glucose uptake during the insulin clamp were similar in chow- and HF-fed WT and sod2(tg) mice. Consistent with our previous work, HF-fed mcat(tg) mice had improved muscle insulin action, however, an additive effect was not seen in mtAO mice. Insulin-stimulated Akt phosphorylation in muscle from clamped mice was consistent with glucose flux measurements. These results demonstrate that increased O2(˙-) scavenging does not improve muscle insulin action in the HF-fed mouse alone or when coupled to increased H2O2 scavenging.

Fig 1. Summary of effects of diet and transgenic expression of SOD2 and/or catalase on oxidant production and cellular redox state.

* Data adapted from Kang et al. 2012 [4].

Fig 2. SOD2 overexpression does not augment glucose infusion rates in HF-fed mice during an insulin clamp.

(A-D) Blood glucose was maintained at ~150 mg/dL in all groups (diet and genotype) during the insulin clamp by variable venous infusion of 50% glucose. (E-H) Glucose infusion rate (GIR) over the 120-minute duration of the insulin clamp.

Fig 3. Rates of endogenous glucose production (EndoRa) and glucose disappearance (Rd) during the insulin clamp.

Rd was calculated using the tracer [3-³H]glucose dilution method in SOD2 overexpressing WT (A) and mcat ^tg (B) mice. EndoRa was determined by subtracting the glucose infusion rate from total Ra in sod2 overexpressing WT (C) and mcat ^tg (D) mice. Insulin-stimulated increase in Rg (E and F) and suppression of EndoRa (G and H) are presented as percentage (%) of basal rates. *p<0.05 compared with sod2 ^tg overexpressing mice (WT or mcat ^tg) within a diet; ^# p<0.05 compared with chow within a genotype.

Fig 4. Divergent effects of SOD2 overexpression on muscle glucose uptake (Rg) and insulin signaling in HF-fed WT and mcat ^tg mice.

Non-metabolizable glucose analog [¹⁴C]2-deoxyglucose was administered as an intravenous bolus to determine muscle glucose uptake (Rg) using liquid scintillation counting in SOD2 overexpressing WT (A) and mcat ^tg (B) mice. Insulin signaling was measured in tissue homogenates extracted from gastrocnemius, applied to 4–12% SDS-PAGE gel and Western blotted with anti-phospho-Akt (Ser⁴⁷³) or anti-total Akt antibodies. Values are expressed as mean ± SEM of integrated intensity and representative bands are presented (C). Ratio of phosphorylated-Akt to total Akt was calculated in SOD2 overexpressing WT (D) and mcat^tg (E) mice. n = 4–6. *p<0.05 compared with sod2 ^tg or WT within a diet; ^# p<0.05 compared with chow within a genotype. SVL, superficial vastus lateralis.

Fig 5. The relationship between glutathione redox state and muscle glucose uptake in chow and HF-fed mice.

Rg determined by 2[¹⁴C]deoxyglucose during the insulin clamp plotted as a function of glutathione redox state (GSH/GSSG) in gastrocnemius of chow- (A) and HF-fed (B) mice. GSH/GSSG values were adapted from Kang et al. [4].