N-terminal transactivation function, AF-1, of estrogen receptor alpha controls obesity through enhancement of energy expenditure

Abstract

Objective: Studies using the estrogen receptor alpha (ERα) knock-out (αERKO) mice have demonstrated that ERα plays a crucial role in various estrogen-mediated metabolic regulations. ERα is a ligand dependent transcription regulator and its activity is regulated by estrogenic compounds. ERα consists of two transcriptional activation domains, AF-1 and AF-2. The activities of these domains are regulated through different mechanisms; however, the specific physiological role in metabolic regulation by these domains is still unclear.

Methods: We utilized an ERα AF-2 mutant knock-in mouse (AF2ERKI) to evaluate the physiological functionality of ERα transactivation domains. Due to the estrogen insensitive AF-2 mutation, the phenotypes of AF2ERKI mice are seemingly identical to the global αERKO including obesity in the females. Distinct from the αERKO, the AF-1 function of AF2ERKI mice can be activated by tamoxifen (Tam). Ovariectomized (OVX) AF2ERKI and WT females were treated with Tam and fed a high-fat diet (HFD) for 10 weeks. Additionally, indirect calorimetric analysis was performed using metabolic chambers with food intake and locomotor activity recorded for Tam-treated AF2ERKI and αERKO females.

Results: Obesity in HFD-fed AF2ERKI females was prevented by Tam treatment; particularly, inguinal fat accumulation was strongly blocked by Tam treatment. Alterations in fat metabolism genes, however, were not found in either inguinal fat nor visceral fat to be Tam-regulated, even though fat accumulation was strongly reduced by Tam treatment. Indirect calorimetric analysis revealed that without alteration of food intake and locomotor activity Tam treatment increased energy expenditure in AF2ERKI but not αERKO females.

Conclusions: These results suggest that the activation of ERα AF-1 prevents fat accumulation. The prevention of obesity through AF-1 is mediated by induction of energy expenditure rather than ERα AF-1 functionality of lipid metabolism gene regulation in fat tissues.

Keywords: Domain functionality; Energy expenditure; Estrogen receptor alpha; Obesity; Subcutaneous fat; Tamoxifen; Visceral fat.

Figure 1

Inactivation of ERα AF-2 leads to obesity and pre-diabetic condition in females. (A) Body weight in WT and AF2ERKI (KI) male (M; square symbols; WT n = 7, KI n = 7) and female (F; circle symbols; WT n = 7, KI n = 7) mice fed regular diet (RD). The body weight is represented as mean ± S.E.M.. Two-way ANOVA was performed to indicate significant difference between genotype. (B) Body fat percentage of 6-month-old RD-fed WT and AF2ERKI (KI) males (WT n = 5, KI n = 5) and females (WT n = 5, KI n = 5). The value of mean ± S.E.M. is also indicated. Two-way ANOVA was performed to indicate significant difference between genotype and sex. (C) Glucose tolerance tests in 6-month-old RD-fed WT and AF2ERKI (KI) male (left; WT n = 5, KI n = 5) and female (right; WT n = 5, KI n = 5). The blood glucose level is represented as mean ± S.E.M.. Dotted lines show initial glucose levels of fasting animals. Two-way ANOVA was performed to indicate significant difference against initial glucose level. *p < 0.05, **p < 0.01, ****p < 0.0001; ns, non-significant difference.

Figure 2

Differential fat deposition in the fat tissues of high fat diet fed AF2ERKI females. (A) Body weight in WT and AF2ERKI (KI) female (WT n = 5, KI n = 5) mice fed high fat diet (HFD) for 10 weeks. The line indicates the period of HFD feeding. The body weight is represented as mean ± S.E.M.. Two-way ANOVA was performed to indicate significant difference against initial body weight. (B) Body fat percentage of HFD-fed WT and AF2ERKI (KI) females. The value of mean ± S.E.M. is also indicated. (C) Total tissue weight (liver, inguinal fat, visceral fat, and interscapular fat) of HFD-fed WT and AF2ERKI (KI) females. The value of mean ± S.E.M. is also indicated. (D) Proportion of tissue weight in HFD-fed WT and AF2ERKI (KI) females. The value of mean ± S.E.M. is also indicated. Unpaired t test was performed to indicate significant difference between genotype (B, C and D). (E) Representative tissue sections of liver, inguinal fat, visceral fat, and interscapular fat from HFD-fed WT and AF2ERKI females. (F) Glucose tolerance tests (GTT) in HFD fed WT and AF2ERKI (KI) females. (G) Insulin tolerance tests (ITT) in HFD fed WT and AF2ERKI (KI) females. The blood glucose level is represented as mean ± S.E.M. Two-way ANOVA was performed to indicate significant difference against initial glucose level. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001; ns, non-significant difference.

Figure 3

Activation of ERα AF-1 prevents AF2ERKI obesity. (A) Body weight in HFD-fed ovariectomized (OVX) WT and AF2ERKI (KI) females which implanted with placebo (P; WT n = 8, KI n = 8), tamoxifen (T; WT n = 8, KI n = 8) or estradiol (E; WT n = 8, KI n = 8) pellet for 10 weeks. The line indicates the period of HFD feeding. The body weight is represented as mean ± S.E.M.. Dotted lines indicate the body weight in HFD-fed intact WT and AF2ERKI females which are cited from Figure 2. (B) Body fat percentage of placebo (P), tamoxifen (T), or estradiol (E) treated OVX WT and AF2ERKI (KI) females fed HFD for 10 weeks. The value of mean ± S.E.M. is also indicated. (C) Total tissue weight (liver, visceral fat, inguinal fat and interscapular fat) of placebo (P), tamoxifen (T) or estradiol (E) treated OVX WT and AF2ERKI (KI) females fed HFD for 10 weeks. The value of mean ± S.E.M. is also indicated. (D) Proportion of tissue weight in placebo (P), tamoxifen (T), or estradiol (E) treated OVX WT and AF2ERKI (KI) females fed HFD for 10 weeks. The value of mean ± S.E.M. is also indicated. Two-way ANOVA was performed to indicate significant difference between treatment in each genotype (B, C and D). Average values of HFD-fed intact WT and AF2ERKI females cited from Figure 2 are shown in B, C and D. (E) Representative tissue sections of liver, visceral fat, inguinal fat, and interscapular fat from placebo (P) or tamoxifen (T) treated OVX WT and AF2ERKI (KI) females fed HFD for 10 weeks. (F) GTT in HFD fed OVX WT and AF2ERKI (KI) females which implanted with placebo (P), tamoxifen (T), or estradiol (E) pellet for 8 weeks. Dotted lines indicate the GTT response in HFD-fed intact WT and AF2ERKI females which are cited from Figure 2. (G) ITT in HFD fed OVX WT and AF2ERKI (KI) females which implanted with placebo (P), tamoxifen (T), or estradiol (E) pellet for 9 weeks. Data represented as mean ± S.E.M.. Dotted lines indicate the ITT response in HFD-fed intact WT and AF2ERKI females which are cited from Figure 2. ITT was not performed for E2-treated OVX WT females, because the mice were hypoglycemic. (H) Blood glucose levels in 16 h fasting animals which used for GTT and ITT. (I) Serum insulin levels in 4 h fasting animals. Two-way ANOVA was performed to indicate significant difference between treatments (E and F). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001; ns, non-significant difference.

Figure 4

Differential expression profile of fat metabolism related genes in inguinal and visceral fats. (A) mRNA levels of lipolysis rate limiting enzyme coding genes in visceral fat and inguinal fat. (B) mRNA levels of triglyceride synthesis key enzyme coding genes in visceral fat and inguinal fat. (C) mRNA levels of fatty acid synthesis and cholesterol synthesis related genes in visceral fat and inguinal fat of HFD-fed intact WT and AF2ERKI (KI) females and in HFD-fed OVX WT and AF2ERKI (KI) females implanted with placebo (Plac), tamoxifen (Tam), or estradiol (E2) pellet for 10 weeks. Data represented as relative mRNA levels compared to the level of placebo treated OVX WT female set as 1. The data shown as mean ± S.E.M.. Two-way ANOVA was performed to indicate significant difference between treatments. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 5

Expression profile of fat metabolism related genes in liver. (A) mRNA levels of lipolysis rate limiting enzyme coding genes in liver. (B) mRNA levels of triglyceride synthesis key enzyme coding genes in liver. (C) mRNA levels of fatty acid synthesis and cholesterol synthesis related genes in liver of HFD-fed intact WT and AF2ERKI (KI) females, and HFD-fed OVX WT and AF2ERKI (KI) females implanted with placebo (Plac), tamoxifen (Tam), or estradiol (E2) pellet for 10 weeks. Data represented as relative mRNA levels compared to the level of placebo treated OVX WT female set as 1. The data shown as mean ± S.E.M.. Two-way ANOVA was performed to indicate significant difference between treatments. *p < 0.05, ****p < 0.0001.

Figure 6

ERα AF-1 activation improved energy expenditure. (A) Body weight of WT and AF2ERKI (KI) females implanted with placebo (P; WT n = 6, KI n = 6) or tamoxifen (T; WT n = 6, KI n = 6) pellet for 14 days. (B) Lean mass of WT and KI females. (C) Food consumption of WT and KI females in the day (7:00–18:59) and night (19:00–6:59) during the experiment (48 h). (D) Ambulatory activity of WT and KI females in the day and night during the experiment. (E) Sum of energy expenditure in the day and night periods. (F) Energy expenditure at the time. Black bars indicate night period. (G) Sum of respiratory exchange ratio (RER) in the day and night during the experiment. (H) Body weight of WT and αERKO (KO) females implanted with placebo (P; WT n = 6, KO n = 6) or tamoxifen (T; WT n = 6, KO n = 6) pellet for 14 days. (I) Lean mass of WT and KO females. (J) Food consumption of WT and KO females in the day (7:00–18:59) and night (19:00–6:59) during the experiment (48 h). (K) Ambulatory activity of WT and KO females in the day and night during the experiment. (L) Sum of energy expenditure in the day and night periods. (M) Energy expenditure at the time. Black bars indicate night period. (N) Sum of RER in the day and night during the experiment. Data represented as mean ± S.E.M.. Two-way ANOVA was performed to indicate significant difference. *p < 0.05, **p < 0.01, ****p < 0.0001; ns, non-significant difference.