High-fat diet-induced changes in liver thioredoxin and thioredoxin reductase as a novel feature of insulin resistance

Abstract

High-fat diet (HFD) can induce oxidative stress. Thioredoxin (Trx) and thioredoxin reductase (TrxR) are critical antioxidant proteins but how they are affected by HFD remains unclear. Using HFD-induced insulin-resistant mouse model, we show here that liver Trx and TrxR are significantly decreased, but, remarkably, the degree of their S-acylation is increased after consuming HFD. These HFD-induced changes in Trx/TrxR may reflect abnormalities of lipid metabolism and insulin signaling transduction. HFD-driven accumulation of 4-hydroxynonenal is another potential mechanism behind inactivation and decreased expression of Trx/TrxR. Thus, we propose HFD-induced impairment of liver Trx/TrxR as major contributor to oxidative stress and as a novel feature of insulin resistance.

Keywords: 4-HNE, 4-hydroxynonenal; ASK-1, apoptosis signal-regulating kinase-1; Gpx, glutathione peroxidase; HFD, high-fat diet; High-fat diet; IRS-1, insulin receptor substrate-1; ITT, insulin tolerance test; Insulin resistance; OGTT, oral glucose tolerance test; PTP-1B, protein-tyrosine phophatase-1B; S-acylation; Thioredoxin; Thioredoxin reductase; Trx, thioredoxin; TrxR, thioredoxin reductase.

Fig. 1

Insulin resistance occurred in the mice fed on HFD for 13 weeks. (A and B) OGTT in HFD and control groups. The mice in HFD group operated at lower ability to metabolize glucose (Fig. 1A, upper curve) compared with control group (Fig. 1A, lower curve) (n = 8). (C and D) ITT in HFD and control groups. (C) The mice in HFD group were less sensitive to insulin (Fig. 1C, upper curve) than controls (Fig. 1C, lower curve) (n = 8). (E) Whole body glucose utilization (clamp glucose infusion rate, GIR) and (F) hepatic insulin-dependent glucose uptake, analyzed by hyperinsulinemic euglycemic-clamp (n = 6). Values are mean ± SD. ^∗P < 0.05; ^∗∗P < 0.01; ^∗∗∗P < 0.001.

Fig. 2

HFD-induced decrease in liver Trx/TrxR. (A) Specific activity of total TrxR and Trx, determined by super-insulin assay (n = 8). (B) Upper panel: protein levels of TrxR1 and Trx1, detected by Western blotting with anti-TrxR1 or anti-Trx1 monoclonal antibody. GADPH as an internal reference (n = 3). Lower panel: density quantification of TrxR1 and Trx1 immunoblots. (C) mRNA levels of TrxR1 and Trx1, determined by quantitative RT-PCR (n = 6). All values are means ± SD. ^∗P < 0.05.

Fig. 3

TG content and Trx/TrxR S-acylation. (A) TG levels in mice livers (n = 6). (B) Upper panel: S-acylated TrxR1/hTrx1 were separated from liver extracts of either control or HFD group using Acyl-Biotin Exchange assay protocol, followed by Western blotting with anti-TrxR1/hTrx1 monoclonal antibodies. Methods are detailed under Materials and Methods. Lanes 1 and 3, no acylated TrxR1/hTrx1 bound to streptavidin-agarose. Lanes 2 and 4, S-acylated TrxR1/hTrx1 detected in streptavidin-enriched biotin-labeled proteins. Lower panel: density quantification of TrxR1/hTrx1 immunoblots. Data represent mean ± SD from three mice per group. ^∗P < 0.05, and ^∗∗∗P < 0.001.

Fig. 4

Altered proteins detected by ELISA. (A) PPARγ and (B) PPARα levels in the nuclear protein extracts from frozen mice livers. (C) Serum leptin levels. (D) Leptin receptor level and (E) PTP-1B level in the liver protein extracts. Data represent mean ± SD from eight mice per group. All experiments were done in triplicate. ^∗P < 0.05.

Fig. 5

Representative examples of the liver proteins separated on 2-DE gels. The liver extracts either from control group or from HFD group (n = 3) were separated by 2-DE, respectively. The indicated protein spots have been identified by mass spectrometry. The images were analyzed using PDQuest7.3.0 software.

Fig. 6

4-HNE-mediated effect of HFD on Trx/TrxR. (A) Upper panel: a representative of 4-HNE levels in liver extracts from control and HFD groups examined by Western blot. Lower panel: average levels of 4-HNE detected by Western blot assays, representing mean ± SD from three mice per group. (B) Upper panel: a representative of TrxR1 and Trx1 expression in the presence or absence of 50 μM 4-HNE. Lower panel: average levels of TrxR1/Trx1 detected by Western blot assays. Values are means ± SD (n = 3). ^∗P < 0.05; ^∗∗P < 0.01.

Fig. 7

Molecular mechanism of HFD-induced decrease in liver Trx/TrxR. HFD results in decrease of liver Trx/TrxR, owing to increased degree of Trx/TrxR S-acylation and increased accumulation of 4-HNE. The use of proteomic analysis and ELISA has allowed the validation of involved proteins in lipid metabolism and insulin signaling, which mediate the effect of HFD on Trx/TrxR.