Glutathione peroxidase 3 mediates the antioxidant effect of peroxisome proliferator-activated receptor gamma in human skeletal muscle cells - PubMed (original) (raw)
Glutathione peroxidase 3 mediates the antioxidant effect of peroxisome proliferator-activated receptor gamma in human skeletal muscle cells
Sung Soo Chung et al. Mol Cell Biol. 2009 Jan.
Abstract
Oxidative stress plays an important role in the pathogenesis of insulin resistance and type 2 diabetes mellitus and in diabetic vascular complications. Thiazolidinediones (TZDs), a class of peroxisome proliferator-activated receptor gamma (PPARgamma) agonists, improve insulin sensitivity and are currently used for the treatment of type 2 diabetes mellitus. Here, we show that TZD prevents oxidative stress-induced insulin resistance in human skeletal muscle cells, as indicated by the increase in insulin-stimulated glucose uptake and insulin signaling. Importantly, TZD-mediated activation of PPARgamma induces gene expression of glutathione peroxidase 3 (GPx3), which reduces extracellular H(2)O(2) levels causing insulin resistance in skeletal muscle cells. Inhibition of GPx3 expression prevents the antioxidant effects of TZDs on insulin action in oxidative stress-induced insulin-resistant cells, suggesting that GPx3 is required for the regulation of PPARgamma-mediated antioxidant effects. Furthermore, reduced plasma GPx3 levels were found in patients with type 2 diabetes mellitus and in db/db/DIO mice. Collectively, these results suggest that the antioxidant effect of PPARgamma is exclusively mediated by GPx3 and further imply that GPx3 may be a therapeutic target for insulin resistance and diabetes mellitus.
Figures
FIG. 1.
Effect of troglitazone on the insulin signaling pathway and impairment of glucose uptake by H2O2 in human skeletal muscle cells. (A) Human skeletal muscle cells were prepared and differentiated as described in Materials and Methods. The cells were treated with glucose oxidase (100 mU/ml) for 3 h and then with insulin (100 nM) for 30 min. For troglitazone treatment, the cells were pretreated with troglitazone for 48 h and then with glucose oxidase and insulin. The cells were harvested, and 20 μg of proteins was subjected to SDS-polyacrylamide gel electrophoresis and Western blot analysis. (B) Glucose uptake was measured as described in Materials and Methods. The cells were treated with glucose oxidase (100 mU/ml) for 3 h and then incubated with or without insulin (100 nM) for 30 min. For troglitazone treatment, the cells were pretreated with troglitazone for 48 h and then with glucose oxidase and insulin. The glucose uptake of untreated cells was set as 100, and the other values were expressed relative to it. The bar graph represents the mean ± standard error of four independent experiments; *, P < 0.01 versus the basal value of control cells not treated with troglitazone; **, P < 0.01 versus the corresponding value of control cells not treated with troglitazone, †, P < 0.01 versus the corresponding value of cells treated with glucose oxidase but not with TZDs.
FIG. 2.
Antioxidant effects of TZDs are PPARγ dependent. Western blot analysis. For each figure, at least three independent experiments were performed, and similar results were obtained. (A) Cells were treated with troglitazone in the presence or absence of GW9662 (10 μM), a PPARγ antagonist, for 48 h and then with glucose oxidase for 3 h. (B) The cells were transfected with siRNAs of PPARγ (siPPARγ) or control (siNS) using Lipofectamine 2000. The cells were treated with rosiglitazone 24 h after siRNA transfection and were incubated for an additional 48 h. (C) The cells were treated with troglitazone for 3 h or 48 h, and then glucose oxidase was added. (D) The cells were treated with troglitazone for 48 h, and a set of cells (+*) was washed with PBS and incubated with fresh medium prior to the addition of glucose oxidase.
FIG. 3.
Effect of TZDs on extracellular H2O2 concentration. (A) The amount of H2O2 in the medium was measured after the human skeletal muscle cells were treated with troglitazone for 48 h and different doses of glucose oxidase for 1 h. The graph represents the mean ± standard error of three independent experiments. *, P < 0.01; **, P < 0.05 (versus the control). (B) Cells were treated with troglitazone for the indicated periods of time and with glucose oxidase for 1 h. The bar graph represents the mean ± standard error of three independent experiments. *, P < 0.01 versus the H2O2 concentration at 0 h. (C) Cells were treated with various PPARγ agonists, namely, troglitazone, rosiglitazone, or pioglitazone, for 48 h and with glucose oxidase for 1 h (n = 5); *, P < 0.01 versus the value of control cells. (D) Cycloheximide (10 μg/ml) was added to the cells along with TZDs (n = 5). *, P < 0.01 versus the value of control cells. (E) After the cells were treated with glucose oxidase for 1 h in the absence or presence of TZDs (48 h), the intracellular ROS level was detected by DCF-DA as described in Materials and Methods. The value of cells not treated with glucose oxidase was set to 100, and the other values were presented in relation to that value. The bar graph shows the mean ± standard errors of four independent experiments. *, P < 0.05 versus the value of control cells not treated with glucose oxidase; **, P < 0.05 versus the value of cells treated with glucose oxidase but not with TZDs. (F) Cells treated with troglitazone or rosiglitazone for 48 h and then treated with glucose oxidase for 1 h. The cells were harvested, and the proteins were subjected to Western blot analysis using antibodies against phospho-JNK, p38, Erk1/2, IKKα/β, and Ser307 IRS-1 or total JNK, p38, Erk1/2, IKKα/β, and IRS-1.
FIG. 4.
TZDs induce GPx3 expression in human skeletal muscle cells. For Northern blot analysis in panels A to D, at least three independent experiments were performed, and similar results were obtained. (A) Human skeletal muscle cells were infected with Ad-βGal or Ad-PPARγ and then treated with troglitazone for 48 h. Total RNA was prepared and subjected to Northern blot analysis. The band intensity of GPx3 was normalized to that of GAPDH and presented as a bar graph. The GPx3 mRNA level of Ad-βGal- and troglitazone-treated cells was set as 100, and the other values were represented in relation to it. The graph represents the mean ± standard error of three independent experiments. *, P < 0.05 versus the value obtained from cells treated with Ad-βGal and troglitazone. (B) Human skeletal muscle cells were treated with 10 μM of troglitazone, rosiglitazone, pioglitazone, or wy14643 (a PPARα agonist) for 48 h. (C) Human skeletal muscle cells were treated with rosiglitazone for the indicated periods of time. (D) Cells were treated with troglitazone (10 μM) and 9-_cis_-retinoic acid (1 μM) for 48 h. (E) After treatment with troglitazone for 48 h, extracellular GPx3 was determined by ELISA. The bar graph shows the mean ± standard error of four independent experiments. *, P < 0.01.
FIG. 5.
Identification of a PPRE in the hGPx3 gene. (A) COS-7 cells were transfected with the indicated vectors containing the hGPx3 promoter (hGPx3 (−2294) Luc or hGPx3 (−809) Luc), the expression vectors of PPARγ, RXRα, and β-galactosidase. Rosiglitazone (Rosi) (10 μM) and 9-_cis_-retinoic acid (RA) (1 μM) were added for 48 h. β-Galactosidase activity was used as an internal control to monitor the transfection efficiency. The luciferase activity of hGPx3 (−2294) Luc in the absence of PPARγ/RXRα expression and agonists was set as 1, and other activities were expressed relative to it. The bar graph represents the mean ± standard error of six independent experiments. *, P < 0.01 versus the activity of hGPx3 (−2294) Luc without PPARγ/RXRα and their agonists. (B) ChIP was performed on human skeletal muscle cells treated or not treated with rosiglitazone. Immunoprecipitation was performed using control immunoglobulin G or anti-PPARγ antibody (α-PPARγ). PCR was performed using the primers described in Materials and Methods, and 10% of the cell lysates used for immunoprecipitation was used as the input. Similar results were obtained from three independent experiments. (C) EMSA was performed using an oligomer representing hGPx3 PPRE at −2186 as a probe and nuclear extracts of human skeletal muscle cells treated with rosiglitazone. For competition assays, unlabeled oligomers representing the hGPx3 (−2186) PPRE (self) (lanes 3 and 4), PPRE mutant (mt) (lanes 5 and 6), or a consensus PPRE (PPRE) (lanes 7 and 8) were used at a 50- or 100-fold molar excess. Lane 1 shows only the probe, and lane 2 shows the control without the competitor. (D) COS-7 cells were transiently transfected with pGL2P, pGL2P (−2186) PPRE, or pGL2P (−2186) PPREmt, expression vectors of PPARγ and RXRα, and pCMV-β-galactosidase. Luciferase activity was normalized by β-galactosidase activity, the value of pGL2P Luc without any treatment was set as 1, and other activities were expressed in relation to it. The bar graph represents the mean ± standard error of four independent experiments. *, P < 0.05 versus the activity of pGL2P (−2186) Luc without PPARγ/RXRα and their agonists.
FIG. 6.
Effect of GPx3 overexpression on extracellular ROS and insulin resistance induced by glucose oxidase. For each panel, at least three independent experiments were performed. (A) Human skeletal muscle cells were infected with Ad-GPx3 for 48 h. The increase in GPx3 mRNA was determined by reverse transcription-PCR. (B) The cells were infected with Ad-GFP or Ad-GPx3 for 48 h, and the extracellular H2O2 concentrations were measured 1 h after incubation with glucose oxidase. The amount of H2O2 in the medium of glucose oxidase-treated cells was set as 100, and other values were expressed in relation to it. The bar graph represents the mean ± standard error of three independent experiments. *, P < 0.01 versus the value of cells transfected with the same multiplicity of infection (moi) of Ad-GFP. (C) The cells were treated with the adenovirus, i.e., with Ad-GFP (control) or Ad-GPx3, at a multiplicity of infection of 50 for 48 h and then with glucose oxidase for 3 h. The cellular proteins were prepared and subjected to immunoblot analysis. (D) The cells were treated with the adenovirus and glucose oxidase as described for panel C and then with insulin for 30 min. The cellular proteins were prepared and subjected to immunoblot analysis.
FIG. 7.
Effect of GPx3 siRNA on the extracellular H2O2 concentration and the insulin signaling pathway. For each panel, at least three independent experiments were performed, and similar results were obtained. (A) Human skeletal muscle cells were transfected with siRNAs of NS (control), PPARγ, or GPx3 for 24 h and then treated with troglitazone or rosiglitazone for an additional 48 h. RNAs were prepared, and the reduction of mRNAs of PPARγ or GPx3 was confirmed by reverse transcription-PCR. (B) The cells were transfected with siRNAs and TZDs as described for panel A. The H2O2 level was measured 1 h after treatment with glucose oxidase (G/Oxidase). The bar graph represents the mean ± standard error of five independent experiments. *, P < 0.01 versus the value of control (NS) siRNA-treated cells treated in the same way with TZDs. (C) The cells were transfected with siRNAs of NS (siNS) or GPx3 (siGPx3) for 24 h and treated with rosiglitazone for an additional 48 h. The cells were harvested after treatment with glucose oxidase (3 h), and the cellular proteins were subjected to SDS-polyacrylamide gel electrophoresis and Western blot analysis. (D) The cells were treated with siRNAs, rosiglitazone, and glucose oxidase sequentially as described above and then with insulin for 30 min. Western blot analysis was then performed. (E) Glucose uptake was measured using siRNA-treated cells. The value from the cells that had been transfected with NS siRNA and not treated with either glucose oxidase or troglitazone was set as 100, and the other values were shown in relation to it. The bar graph shows the mean ± standard error of five independent experiments. *, P < 0.05 versus troglitazone- and glucose oxidase-treated cells of siNS.
FIG. 8.
The expression of GPx3 in diabetic patients and animal models. (A) Plasma GPx3 levels among subjects with NGT (n = 57), IGT (n = 48), and type 2 DM (n = 49) were measured by ELISA. *, P < 0.01 versus NGT. (B) Total RNA was prepared from normal or db/db mouse muscles, and Northern blot analysis was performed using GPx3 and GAPDH probes. The intensity of the GPx3 band was normalized with that of the GAPDH band, and the mean value from the control mice was set as 100. The bar graph represents the mean ± standard error of six normal and six db/db mice. *, P < 0.05. (C) Total RNA was prepared from the skeletal muscle of standard chow-fed (control, n = 5), high-fat diet-fed (HF; n = 5), and high-fat diet-fed/rosiglitazone-treated (HF + Rosi; n = 5) mice, and Northern blot analysis was performed using GPx3 and GAPDH probes. The intensity of the GPx3 band was normalized with that of the GAPDH band, and the mean value from the control mice was set as 100. The bar graph represents the mean ± standard error. *, P < 0.05 versus the control mice; **, P < 0.05 versus high-fat diet-fed mice.
References
- American Diabetes Association. 2006. Diagnosis and classification of diabetes mellitus. Diabetes Care 29S43-S48. - PubMed
- Antunes, F., and E. Cadenas. 2000. Estimation of H2O2 gradients across biomembranes. FEBS Lett. 475121-126. - PubMed
- Berry, M. J., L. Banu, Y. Y. Chen, S. J. Mandel, J. D. Kieffer, J. W. Harney, and P. R. Larsen. 1991. Recognition of UGA as a selenocysteine codon in type I deiodinase requires sequences in the 3′ untranslated region. Nature 353273-276. - PubMed
- Blair, A. S., E. Hajduch, G. J. Litherland, and H. S. Hundal. 1999. Regulation of glucose transport and glycogen synthesis in L6 muscle cells during oxidative stress. Evidence for cross-talk between the insulin and SAPK2/p38 mitogen-activated protein kinase signaling pathways. J. Biol. Chem. 27436293-36299. - PubMed
- Brigelius-Flohe, R. 1999. Tissue-specific functions of individual glutathione peroxidases. Free Radic. Biol. Med. 27951-965. - PubMed
Publication types
MeSH terms
Substances
LinkOut - more resources
Full Text Sources
Medical
Molecular Biology Databases
Miscellaneous