Ceramide- and Oxidant-Induced Insulin Resistance Involve Loss of Insulin-Dependent Rac-Activation and Actin Remodeling in Muscle Cells
Article Figures & Tables
Figures
- FIG. 1.
Glucose oxidase (GO) and ceramide (C2) reduce GLUT4 translocation and glucose uptake in L6 myotubes. A: Glucose oxidase caused a dose-dependent increase in H2O2 in the medium. B and C: Glucose oxidase dose-dependent inhibition of insulin-induced gain in surface GLUT4myc (B) or in 2-deoxyglucose uptake (C). D: NAC prevents the glucose oxidase–induced decrease in insulin-induced GLUT4 translocation. E and F: C2-ceramide dose-dependent inhibition of insulin-induced gain in surface GLUT4myc (E) or in 2-deoxyglucose uptake (F). Results in B–F are expressed as the percent of maximal insulin response observed in the absence of glucose oxidase or C2-ceramide. #P < 0.001, *P < 0.05, *P < 0.01 (n = 4–13), relative to the insulin response in absence of glucose oxidase or C2-ceramide. Basal surface GLUT4myc levels were not reduced by glucose oxidase or C2-ceramide (see supplemental Tables 1 and 2).
- FIG. 2.
Glucose oxidase treatment prevents insulin-induced actin remodeling. Myotubes pretreated without or with 6.25–100 mU/ml glucose oxidase (GO) were stimulated for 10 min with 100 nmol/l insulin where indicated. Cells were then fixed and actin filaments stained with rhodamine-phalloidin, and then images were collected by confocal fluorescence microscopy. A: Organized actin stress fibers in unstimulated (Basal) cells. B: Insulin-induced actin remodeling. Glucose oxidase doses progressively inhibited insulin-induced actin remodeling (F, H, J, and L) without affecting basal-state stress fibers (E, G, I, and K). Less than half of the fields examined showed any of the small actin bundles illustrated in F. See also supplemental Figs. 2 and 3.
- FIG. 3.
C2-ceramide (C2) treatment prevents insulin-induced actin remodeling. Myotubes pretreated without or with 6.25–100 μmol/l C2-ceramide were stimulated for 10 min with 100 nmol/l insulin where indicated, and otherwise they were processed as in Fig. 2. A: Organized actin stress fibers in unstimulated (Basal) cells. B: Insulin-induced actin remodeling. C2-ceramide doses progressively inhibited insulin-induced actin remodeling (F, H, J, and L) without affecting basal-state stress fibers (E, G, I, and K). See also supplemental Fig. 3.
- FIG. 4.
Dose-dependent inhibition of insulin-induced Rac activation by glucose oxidase (GO) (A) or C2-ceramide (C2) (B). Myotubes were pretreated with glucose oxidase or C2-ceramide and exposed to insulin for 10 min, and then GTP-Rac bound to glutathione-S-transferase–CRIB (Cdc 42/Rac interactive binding) as well as total Rac in lysates were determined by immunoblotting. Representative gels with doses up to 50 mU/ml glucose oxidase or 50 μmol/l C2-ceramide are illustrated. Additional gels tested the effect of higher concentrations. These experiments were repeated 3–12 times, and the averaged results are expressed as a percent of the maximal insulin response in the absence of pretreatments. #P < 0.001, **P < 0.01 relative to the stimulation in the absence of glucose oxidase or C2-ceramide. Basal levels were not altered by glucose oxidase or C2-ceramide (see supplemental Tables 1 and 2).
- FIG. 5.
Glucose oxidase (GO) does not alter whole-cell levels of insulin-induced IRS-1 tyrosine phosphorylation or its association with p85 or PI 3-kinase activity. A: Myotubes pretreated with glucose oxidase or C2-ceramide as in Fig. 1 were incubated without or with insulin for 10 min, followed by lysis, immunoprecipitation with polyclonal anti–IRS-1 or IgG, and immunoblotting for phosphotyrosine or p85. Supernatants were immunoblotted with anti–IRS-1 to ascertain immunoprecipitation efficiency. B: PI 3-kinase activity associated with anti–IRS-1 immunoprecipitates was determined by in vitro incorporation of radioactive ATP into PI. Shown are representative thin-layer chromatography plates and averaged results of seven experiments quantifying the fold insulin response normalized to basal activity in the absence of treatments. ns, not statistically significant.
- FIG. 6.
Glucose oxidase (GO) and C2-ceramide (C2) inhibit insulin-induced Akt phosphorylation on Thr308 and Ser473. Myotubes retreated with glucose oxidase or C2-ceramide were incubated without or with insulin for 10 min, followed by lysis and immunoblotting for phosphorylated Akt (pAkt) on Thr308 and Ser473. A and D: Representative gels illustrate the response to 0–100 mU/ml glucose oxidase or 0–100 μmol/l C2-ceramide. Parallel experiments were carried out with 100 μmol/l C2-ceramide and C2-dihydroceramide. The results from 4–12 experiments analyzing Thr308 phosphorylated Akt (B and E) or Ser473 phosphorylated Akt (C and F) in cells pretreated with glucose oxidase (B and C) or C2-ceramide (E and F) were quantified and expressed as a percent of the maximal insulin response observed in the absence of treatments. **P < 0.01,*P < 0.05, #P < 0.001 (n = 4–12) relative to the stimulation in absence of glucose oxidase or C2-ceramide.
- FIG. 7.
Rac1 expression silencing via siRNA inhibits insulin-induced actin remodeling and GLUT4 translocation, but not Akt phosphorylation. Myotubes were exposed to 400 nmol/l of unrelated siRNA (siUR) or Rac1 siRNA (siRac1) for up to 72 h, incubated without or with insulin for 10 min, and used in the following assays. A: Rac expression assessed by immunoblotting of cell lysates. B: Filamentous actin visualization in fixed and permeabilized myotubes. C: Surface GLUT4myc in 15 experiments, expressed as a percent of the maximal insulin response in unrelated siRNA cells, and corresponding fractional expression of Rac. ▪, surface GLUT4myc; □, fractional expression of Rac. #P < 0.001 relative to unrelated siRNA. D: Phosphorylation of Rac-target PAK on Thr423, phosphorylation of Akt on Ser423 and Thr308, and immunoblotting of actin levels to assess protein loading. NT, no treatment; p-Akt, phosphorylated Akt.
In this Issue
