Hepatic glucose intolerance precedes hepatic steatosis in the male aromatase knockout (ArKO) mouse

Abstract

Estrogens are known to play a role in modulating metabolic processes within the body. The Aromatase knockout (ArKO) mice have been shown to harbor factors of Metabolic syndrome with central adiposity, hyperinsulinemia and male-specific hepatic steatosis. To determine the effects of estrogen ablation and subsequent replacement in males on whole body glucose metabolism, three- and six-month-old male ArKO mice were subjected to whole body glucose, insulin and pyruvate tolerance tests and analyzed for ensuing metabolic changes in liver, adipose tissue, and skeletal muscle. Estrogen-deficient male ArKO mice showed increased gonadal adiposity which was significantly reduced upon 17β-estradiol (E2) treatment. Concurrently, elevated ArKO serum leptin levels were significantly reduced upon E2 treatment and lowered serum adiponectin levels were restored to wild type levels. Three-month-old male ArKO mice were hyperglycemic, and both glucose and pyruvate intolerant. These phenotypes continued through to 6 months of age, highlighting a loss of glycemic control. ArKO livers displayed changes in gluconeogenic enzyme expression, and in insulin signaling pathways upon E2 treatment. Liver triglycerides were increased in the ArKO males only after 6 months of age, which could be reversed by E2 treatment. No differences were observed in insulin-stimulated ex vivo muscle glucose uptake nor changes in ArKO adipose tissue and muscle insulin signaling pathways. Therefore, we conclude that male ArKO mice develop hepatic glucose intolerance by the age of 3 months which precedes the sex-specific development of hepatic steatosis. This can be reversed upon the administration of exogenous E2.

Figure 1. Serum glucose, insulin and HOMA-IR levels.

(a) Fasted and 20 min post glucose infused serum glucose (b) Fasted serum insulin and (c) HOMA-IR (fasted serum glucose (mmol/l) × fasted serum insulin (µU/l)/22.5) from 3 and 6 month-old fasted male wildtype (WT) and aromatase knockout (KO) and KO treated with 2.5 µg/day estrogen (KOE). Expression data from samples (n =  shown on corresponding bar) per genotype are shown, and are presented from replicate analysis as the mean ± SD. *p<0.05 versus WT mice. ^# p<0.05 versus KO.

Figure 2. Three month-old male whole body glucose, insulin and pyruvate tolerance.

Whole body tolerance tests were completed on fasted three month-old male wildtype (WT) and aromatase knockout (KO) and 2.5 µg/day 17β-estradiol-treated KO (KOE) (a) glucose tolerance test and corresponding area under curve; (b) insulin tolerance test and corresponding area under curve; (c) pyruvate tolerance test and corresponding area under curve Data are presented from replicate analysis (n =  shown on corresponding bar) as the mean ± SD. *p<0.05, **p<0.01, ***p<0.001, versus expression in age-matched WT samples and ^#p<0.05 and ^##p<0.01 versus KO samples.

Figure 3. Six month-old male whole body glucose, insulin and pyruvate tolerance.

Whole body tolerance tests were completed on fasted 6 month-old male wildtype (WT), aromatase knockout (KO) and 2.5 µg/kg 17β-estradiol-treated KO (KOE). (a) Glucose tolerance test and corresponding area under curve; (b) insulin tolerance test and corresponding area under curve; (c) pyruvate tolerance test and corresponding area under curve. Data are presented from replicate analysis (n =  shown on corresponding bar) as the mean ± SD. *p<0.05, **p<0.01, versus expression in age-matched WT samples and ^#p<0.05 and ^##p<0.01 versus KO samples

Figure 4. Liver Triglyceride levels.

Liver triglyceride assay were performed on 3 and 6 month-old fasted male wildtype (WT), aromatase knockout (KO) and 17β-estradiol-treated KO (KOE) mice. Expression data from 7–8 samples per genotype are shown following and presented from replicate analysis as the mean ± SD. **p<0.01, versus expression in age-matched WT samples and ^###p<0.001 versus age-matched KO samples and ∧∧p<0.01, ∧∧∧p<0.001, versus 3mth old KO samples.

Figure 5. Western blot analyses of insulin stimulated and basal tissue.

Protein phosphorylation analyses of Akt and AMPK levels were performed on protein extracted from insulin stimulated (a) gonadal adipose tissue, (b) gastrocnemius muscle, (c) liver in insulin stimulated liver of 6 month-old male wildtype (WT) and aromatase knockout (KO) and 2.5 µg/day estrogen-replaced KO (KOE) mice. Expression data from 6–8 samples per genotype are shown, and presented from replicate analysis as the mean ± SD. *p<0.05, **p<0.01, ***p<0.001 versus expression in age-matched WT samples and ^#p<0.05, ^###p<0.001 versus KO samples.

Figure 6. Real-time PCR analysis of gluconeogenesis mRNA expression in the liver.

Real-time-PCR analyses of G6Pase, PEPCK, IRS1, IRS2 and GSK3β mRNA expression were performed on cDNA derived from total RNA prepared from fasted liver tissue of (a) 3 and (b) 6 month-old male wildtype (WT, n = 8), aromatase knockout (KO, n = 8) and 2.5 µg/day 17β-estradiol-treated KO (KOE, n = 7) mice. Expression data from 7–8 samples per genotype are shown following normalization for cyclophilin mRNA expression, and presented from replicate analysis as the mean ± SD. *p<0.05,**p<0.01 and ***p<0.001 versus expression in age-matched WT samples and ^##p<0.01, ^###p<0.05 versus KO samples.

Figure 7. Real-time PCR analysis of mRNA lipid profile in the liver.

Real-time-PCR analyses of Fasn, ACCα and Scd-1 mRNA expression were performed on cDNA derived from total RNA prepared from fasted liver tissue of (a) 3 and (b) 6 month-old male wildtype (WT, n = 8), aromatase knockout (KO, n = 8) and 2.5 µg/day 17β-estradiol-treated KO (KOE, n = 7) mice. Expression data from 7–8 samples per genotype are shown following normalization for cyclophilin mRNA expression, and presented from replicate analysis as the mean ± SD. *p<0.05, **p<0.01 and **p<0.001, versus expression in age-matched WT samples and ^##p<0.01, ^###p<0.001 versus KO samples.