The estrogen receptor antagonist ICI 182,780 can act both as an agonist and an inverse agonist when estrogen receptor α AF-2 is modified

Abstract

The bone-sparing effect of estrogen is primarily mediated via estrogen receptor (ER) α, which stimulates target gene transcription through two activation functions (AFs), AF-1 in the N-terminal and AF-2 in the ligand-binding domain. It was recently demonstrated that the ER antagonist ICI 182,780 (ICI) acts as an ER agonist in uterus of mice with mutations in the ERα AF-2. To evaluate the estrogen-like effects of ICI in different tissues, ovariectomized wild-type mice and mice with mutations in the ERα AF-2 (ERαAF-2(0)) were treated with ICI, estradiol, or vehicle for 3 wk. Estradiol increased the trabecular and cortical bone mass as well as the uterine weight, whereas it reduced fat mass, thymus weight, and the growth plate height in wild-type but not in ERαAF-2(0) mice. Although ICI had no effect in wild-type mice, it exerted tissue-specific effects in ERαAF-2(0) mice. It acted as an ERα agonist on trabecular bone mass and uterine weight, whereas no effect was seen on cortical bone mass, fat mass, or thymus weight. Surprisingly, a pronounced inverse agonistic activity was seen on the growth plate height, resulting in enhanced longitudinal bone growth. In conclusion, ICI uses ERα AF-1 in a tissue-dependent manner in mice lacking ERαAF-2, resulting in no effect, agonistic activity, or inverse agonistic activity. We propose that ERα lacking AF-2 is constitutively active in the absence of ligand in the growth plate, enabling ICI to act as an inverse agonist.

Fig. 1.

No effect of ICI on cortical thickness, fat mass, or thymus weight in ERαAF-2⁰ mice. (A) Cortical bone thickness and (B) representative images of cortical bone scans in the middiaphysis of tibia, (C) percentage whole body fat, (D) thymus weight of 12-wk-old ovx female WT, and ERαAF-2⁰ mice treated with Veh, E2, or ICI for 3 wk. **P < 0.01, *P < 0.05 vs. ovx + Veh. ^##P < 0.01, ^#P < 0.05, vs. ovx + E2. Student t test with Bonferroni correction. Values are means ± SEM (n = 8–10).

Fig. 2.

ICI acts as an ERα agonist in trabecular bone and uterus in ERαAF-2⁰ mice. (A) Trabecular volumetric BMD (vBMD), (B) trabecular number, (C) representative µCT images of trabecular bone in the proximal metaphyseal region of tibia, and (D) uterine weight of 12-wk-old ovx female WT and ERαAF-2⁰ mice treated with Veh, E2, or ICI for 3 wk. **P < 0.01, *P < 0.05 vs. ovx + Veh. ^##P < 0.01, ^#P < 0.05, vs. ovx + E2. Student t test with Bonferroni correction. Values are means ± SEM (n = 8–10).

Fig. 3.

ICI acts as an inverse ERα agonist on growth plate height in ERαAF-2⁰ mice. (A) Growth plate height, (B) proliferative zone, (C) hypertrophic zone, and (D) representative images of the proximal tibia of 12-wk-old ovx female WT and ERαAF-2⁰ mice treated with Veh, E2, or ICI for 3 wk. **P < 0.01, *P < 0.05 vs. ovx + Veh. ^##P < 0.01, ^#P < 0.05, vs. ovx + E2. Student t test with Bonferroni correction. Values are means ± SEM (n = 8–10). (Scale bar, 100 μm.)

Fig. 4.

Within the skeleton, ICI acts as an inverse agonist in the growth plate, as a partial agonist in trabecular bone, whereas it has no effect in cortical bone of ovx ERαAF-2⁰ mice. Twelve-week-old ovx female WT and ERαAF-2⁰ mice were treated with Veh, E2, or ICI for 3 wk. **P < 0.01 vs. ovx + Veh. Student t test. The estrogen-like effects of ICI in the different skeletal tissues in ERαAF-2⁰ mice are given as percentage of the E2 response in WT mice. The E2 responses in WT mice are set to 100% (dotted line). n = 8–10.

Fig. 5.

The estrogen-like effects of Ral and Las require a functional ERα AF-2. Tissue-dependent effects of (A) Ral and (B) Las, which require a functional ERα AF-2. Twelve-week-old ovx WT and ERαAF-2⁰ mice were treated with Veh, E2, Ral, or Las for 3 wk. **P < 0.01, *P < 0.05, ^(*)P = 0.058 vs. ovx + Veh. Student t test. The estrogen-like effects of Ral and Las are given as percentage of the E2 response in WT mice. The E2 responses in WT mice are set to 100% (dotted line). n = 8–10.