Hyperandrogenism and insulin resistance contribute to hepatic steatosis and inflammation in female rat liver

Figures

• 

  Figure 1. Alterations of metabolic and endocrine parameters in rats treated with insulin and/or hCG Comparison of body weight (A), liver weight (B), ratio of liver weight to body weight (C), fasting blood glucose (D), OGTT (E), area under the curve (AUC) for glucose (F), fasting blood insulin (G), HOMA-IR (H), serum E2 (I), T (J), DHT (K), and A4 (L). AUC was calculated by the formula [0.5 × (BG0 + BG30)/2 + 0.5 × (BG30 + BG60)/2 + 0.5 × (BG60 + BG120)/2 + 0.5 × (BG120 + BG180)/2] where the BG terms are the blood glucose levels at 0 min, 30 min, etc. Data in A–L represent the means of n = 12/group ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001. In the OGTT, #p < 0.05 (the insulin group versus the control group) and *p < 0.05 (the insulin+hCG group versus the control group) were determined by analysis of variance (ANOVA) comparing controls to the different treatments. In the HOMA-IR, **p < 0.01 versus the insulin+hCG group.

• 

  Figure 2. Effects of insulin and/or hCG on major markers of lipid profile, and liver morphology Comparison of circulating TC (A), TG (B), HDL-C (C), non-HDL (D), FFA (E), and liver TG (H). Data represent the means of n = 12/group ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001. Representative photomicrographs are of liver tissue sections stained with hematoxylin and eosin (H&E) staining (F, upper panels) and Oil Red O staining (F, lower panels) with quantification (G). Control rats show a normal arrangement of hepatocytes (H) and sinusoids (s) as well as normal central canal and phagocytic Kupffer cells (k). The investigators were blinded to the allocation for morphological analyses (n = 6–9/group). Scale bars are indicated in the photomicrographs. CV, central vein.

• 

  Figure 3. Effects of insulin and/or hCG on the expression of genes involved in hepatic lipid metabolism Relative levels of key regulator mRNAs of hepatic lipid metabolism were determined by qRT-PCR. Data represent means of n = 5–6/group ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001.

• 

  Figure 4. Effects of insulin and/or hCG on the insulin-mediated IRS–PI3K–Akt signaling pathway in the rat liver Representative images with immunoblotting quantification (n = 9/group) are shown. Total proteins served as the loading control. Values are expressed as means ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001.

• 

  Figure 5. Effects of insulin and/or hCG on hepatic inflammation in the rat liver Representative photomicrographs are of liver tissue sections stained with H&E staining (A) with quantification (B). Solid bold arrows and rings (yellow) indicate hepatocytes with inflammatory cells. Enhanced magnifications are shown at the right corner of A. The investigators were blinded to the allocation for morphological analyses (n = 6–9/group). CV, central vein; PT, portal tract; HA, hepatic artery; HPV, hepatic portal vein; BD, bile duct. Scale bars are indicated in the photomicrographs. Relative levels of mRNAs (C) of hepatic inflammatory factors were determined by qRT-PCR. Data in C represent means of n = 6/group ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001. Immunohistochemical detection of TNFα (D) and IL-1β (E) in rats treated with insulin and/or hCG. Representative images are shown. The investigators were blinded to the allocation for immunohistochemical analysis (n = 5 or 6/group). Scale bars are indicated in the photomicrographs.

• 

  Figure 6. Effects of insulin and/or hCG on adipose tissue distribution, morphological changes, and gene and protein expression in the rat adipose tissue Comparison of total fat depots (A), and the ratio of total fat depots to body weight (B). Total fat depots include inguinal, parametrical, retroperitoneal, and mesenteric fat portions. Data in A–B are means of n = 12/group ± SEM. Relative protein (C) and mRNA (E) levels of the phosphorylation of Akt and activated inflammatory factors were determined by Western blot analysis and qRT-PCR. For immunoblotting quantification, total proteins served as the loading control. Data in C represent means of n = 9/group ± SEM, and data in E are means of n = 6/group ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001. H&E staining of mesenteric fat tissue sections (D) indicates that “crown like” structures are present in rats treated with insulin and hCG (solid bold arrows). Scale bars are indicated in the photomicrographs.

• 

  Figure 7. Effects of insulin and/or hCG on apoptosis and autophagy in the rat liver Relative levels of mRNAs (A) and proteins (B) of pro-apoptotic and anti-apoptotic genes and autophagy-related proteins were determined by qRT-PCR and Western blot analysis. For immunoblotting quantification, total proteins served as the loading control. Data in A represent means of n = 6/group ± SEM, and the data in B are means of n = 9/group ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001.

• 

  Figure 8. Effects of insulin and/or hCG on fibrosis-related gene and protein expression, and liver functional markers in the rat liver Relative levels of mRNAs (A) and protein (B) of major markers for hepatic stellate cell activation were determined by qRT-PCR and Western blot analysis. For immunoblotting quantification, total proteins served as the loading control. Data in A represent means of n = 6/group ± SEM, and data in B are means of n = 9/group ± SEM. Comparison of circulating PIIINP (C), HA (D), ALT (E), AST (F), RBP4 (G), and liver RBP4 (H) are shown. Data in C–H represent means of n = 12/group. *p < 0.05; **p < 0.01; ***p < 0.001.