SERMs have substance-specific effects on bone, and these effects are mediated via ERαAF-1 in female mice

Abstract

The bone-sparing effect of estrogens is mediated primarily via estrogen receptor (ER)α, which stimulates gene transcription through activation function (AF)-1 and AF-2. The role of ERαAF-1 for the estradiol (E2) effects is tissue specific. The selective ER modulators (SERMs) raloxifene (Ral), lasofoxifene (Las), and bazedoxifene (Bza) can be used to treat postmenopausal osteoporosis. They all reduce the risk for vertebral fractures, whereas Las and partly Bza, but not Ral, reduce the risk for nonvertebral fractures. Here, we have compared the tissue specificity of Ral, Las, and Bza and evaluated the role of ERαAF-1 for the effects of these SERMs, with an emphasis on bone parameters. We treated ovariectomized (OVX) wild-type (WT) mice and OVX mice lacking ERαAF-1 (ERαAF-1(0)) with E2, Ral, Las, or Bza. All three SERMs increased trabecular bone mass in the axial skeleton. In the appendicular skeleton, only Las increased the trabecular bone volume/tissue volume and trabecular number, whereas both Ral and Las increased the cortical bone thickness and strength. However, Ral also increased cortical porosity. The three SERMs had only a minor effect on uterine weight. Notably, all evaluated effects of these SERMs were absent in ovx ERαAF-1(0) mice. In conclusion, all SERMs had similar effects on axial bone mass. However, the SERMs had slightly different effects on the appendicular skeleton since only Las increased the trabecular bone mass and only Ral increased the cortical porosity. Importantly, all SERM effects require a functional ERαAF-1 in female mice. These results could lead to development of more specific treatments for osteoporosis.

Keywords: activation function-1 of estrogen receptor-α; estrogen; estrogen receptor; mouse; osteoporosis; selective estrogen receptor modulators.

Fig. 1.

Effects of selective estrogen receptor modulator (SERM) treatment on the axial skeleton in ovariectomized (OVX) wild-type (WT) mice and OVX mice lacking activation function (AF)-1 of estrogen receptor-α (ERαAF-1⁰). OVX WT and ERαAF-1⁰ mice were treated with vehicle (Veh), estradiol (E₂), raloxifene (Ral), lasofoxifene (Las), or bazedoxifene (Bza) for 3 wk. A: lumbar spine (LS) areal bone mineral density (aBMD) was analyzed by dual-energy X-ray absorptiometry. B and C: trabecular thickness (Tb.Th.) in the lumbar vertebra 5 (B) and trabecular bone volume/tissue volume (BV/TV; C) were analyzed by microcomputed tomography (μCT). *P < 0.05 vs. vehicle-treated OVX mice, Student's t-test Bonferroni corrected. Values are given as means ± SE (n = 7–11).

Fig. 2.

Effects of SERM treatment on the appendicular skeleton in OVX WT and ERαAF-1⁰ mice. OVX WT and ERαAF-1⁰ mice were treated with vehicle (Veh), estradiol (E₂), raloxifene (Ral), lasofoxifene (Las), or bazedoxifene (Bza) for 3 wk. Trabecular bone volume/tissue volume (BV/TV; A), trabecular number (Tb. N.; B), and cortical thickness (Ct. Th; C) in the femur were analyzed by high-resolution μCT. Maximal load at failure (Max.Load) of the femur was analyzed by 3-point bending (D), and femoral cortical porosity (Ct. Po) was analyzed by high-resolution μCT (E). *P < 0.05 vs. Veh-treated OVX mice, Student's t-test Bonferroni corrected. Values are given as means ± SE (n = 7–10).