Estradiol and selective estrogen receptor modulators differentially regulate target genes with estrogen receptors alpha and beta

Abstract

Estrogens and selective estrogen receptor modulators (SERMs) interact with estrogen receptor (ER) alpha and beta to activate or repress gene transcription. To understand how estrogens and SERMs exert tissue-specific effects, we performed microarray analysis to determine whether ERalpha or ERbeta regulate different target genes in response to estrogens and SERMs. We prepared human U2OS osteosarcoma cells that are stably transfected with a tetracycline-inducible vector to express ERalpha or ERbeta. Western blotting, immunohistochemistry, and immunoprecipitation studies confirmed that U2OS-ERalpha cells synthesized only ERalpha and that U2OS-ERbeta cells expressed exclusively ERbeta. U2OS-ERalpha and U2OS-ERbeta cells were treated either with 17beta-estradiol (E2), raloxifene, and tamoxifen for 18 h. Labeled cRNAs were hybridized with U95Av2 GeneChips (Affymetrix). A total of 228, 190, and 236 genes were significantly activated or repressed at least 1.74-fold in U2OS-ERalpha and U2OS-ERbeta cells by E2, raloxifene, and tamoxifen, respectively. Most genes regulated in ERalpha cells in response to E2, raloxifene, and tamoxifen were distinct from those regulated in ERbeta cells. Only 38 of the 228 (17%) genes were regulated by E2 in both U2OS-ERalpha and U2OS-ERbeta cells. Raloxifene and tamoxifen regulated only 27% of the same genes in both the ERalpha and ERbeta cells. A subset of genes involved in bone-related activities regulated by E2, raloxifene, and tamoxifen were also distinct. Our results demonstrate that most genes regulated by ERalpha are distinct from those regulated by ERbeta in response to E2 and SERMs. These results indicate that estrogens and SERMs exert tissue-specific effects by regulating unique sets of targets genes through ERalpha and ERbeta

Figure 1.

Characterization of the stable U2OSERα and U2OS-ERβ cell lines. (A) Doxycycline produces a time-dependent increase in ERα and ERβ. The U2OS-ERα (left) and U2OS-ERβ (right) cell lines were treated with 1 μg/ml doxycycline for increasing times before performing Western blots. Lanes 7 and 14 show that there is no ERβ detectable in the U2OS-ERα cells and no ERα detectable in the U2OS-ERβ cells, respectively. (B) Immunohistochemistry of U2OS-ERα and U2OS-ERβ cells. Cells were treated with 1 μg/ml doxycycline for 18 h on slides, fixed with formalin, and stained for ERα and ERβ as described in MATERIALS AND METHODS. Cells labeled 1 and 5 were not induced with doxycycline. Cells labeled 2 and 6 were induced with doxycycline but did not receive primary antibody. Cells labeled 3 and 7 were induced with doxycycline and stained with anti-ERβ and anti-ERα, respectively. Cells labeled 4 and 8 were induced with doxycycline and stained with anti-ERα and anti-ERβ, respectively. (C) Immunoprecipitation of ERα and ERβ in the stable cell lines. Cells were treated with 1 μg/ml doxycycline for 18 h and the immunoprecipitated with anti-ERα (lanes 3, 4, and 7) and anti-ERβ (lanes 2, 8, and 9). Lanes 1 and 6 show a positive control from cell lysate of U2OS-ERα and U2OS-ERβ cells, respectively. All three techniques demonstrate that ERα is detected only in the ERα cells, whereas ERβ is detected exclusively in the ERβ cells.

Figure 2.

Regulation of selected genes by E₂ and SERMs in the U2OS-ERα and ERβ cell lines. Doxycycline-induced U2OS-ERα and ERβ cells were treated for 18 h with 10^-11-10^-8 M E₂ (A), 10^-8-10^-6 M raloxifene (B), or 10^-8-10^-6 M tamoxifen (C). The extracted total RNA was analyzed by RT-PCR as described in MATERIALS AND METHODS. The genes examined were the WISP-2, α-AT, NKG2C, cDNA clone image 996282, NKG2E, and G0S2. Glyceraldehyde-3-phosphate dehydrogenase was used as an internal control. The data presented were representative of at least three experiments.

Figure 3.

Effect of ERα protein level on WISP-2 expression in the U2OS-ERα cell line. The U2OS-ERα cells were treated for 18 h with 10 nM E₂ and increasing concentrations of doxycycline (0.1-2.5 μg/ml). (A) The level of ERα expression was determined by Western blot analysis with anti-ERα antibodies. (B) The expression of WISP-2 was evaluated by semiquantitative RT-PCR in control (lanes 1-5) and E₂-treated (lanes 6-10) U2OS-ERα cell line.

Figure 4.

Regulation of SERM-specific genes in the U2OS-ERα cell line. (A) U2OSERα cells were treated with 1 μg/ml doxycycline and 1 μM raloxifene or 1 μM tamoxifen for 3 h. The expression patterns of TGFβ3 (induced by raloxifene), G0S2 (induced by tamoxifen), and thrombin receptor (inhibited by tamoxifen) were evaluated by semiquantitative RT-PCR. (B) U2OS-ERα and ERB cells were treated for 18 h with doxycycline and SERMs, and the expression patterns of TGFβ3, G0S2, and thrombin receptor were measured by real-time qualitative RT-PCR.

Figure 5.

Regulation of keratin 19 in the U2OS-ERα and U2OSERβ cell lines. (A) E₂ increases K19 mRNA levels in U2OS-ERα and U2OS-ERβ cells. Northern blot was performed with 20 μg of total RNAs from doxycycline-induced U2OS-ERα or U2OS-ERβ cells incubated in the absence (-) or presence of 10 nM E₂ (+) overnight. Before transfer to a nylon blot and hybridization with the K19 cDNA probe (left), the gel containing ethidium bromide-stained RNAs (right) was photographed for a loading control. (B) ERα and ERβ bind the ERE enhancer in the endogenous K19 gene. ChIP assays were performed using U2OS-ERα (left) and U2OS-ERβ (right) cells. Anti-ERα- and anti-ERβ-precipitated DNAs were amplified with PCR primers spanning the near consensus ERE and half ERE in the K19 enhancer region (Choi et al., 2000). PCR products from input chromatin before and after immunoprecipitation (IP) are shown.

Figure 6.

Regulation of selected genes by E₂ and SERMs in the breast cancer MCF-7 cell line. ERα expressing MCF-7 cells were treated for 18 h with 10 nM E₂ (A), 1 μM raloxifene (B), or 1 μM tamoxifen (C). The expression patterns of K19, WISP-2, α-AT, NKG2C, unknown cDNA 996282, NKG2E, and G0S2 were determined by semiquantitative RT-PCR.