help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology, doi:10.1210/en.2003-1682
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
145/7/3473    most recent
Author Manuscript (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Stossi, F.
Right arrow Articles by Katzenellenbogen, B. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Stossi, F.
Right arrow Articles by Katzenellenbogen, B. S.
Endocrinology Vol. 145, No. 7 3473-3486
Copyright © 2004 by The Endocrine Society

Transcriptional Profiling of Estrogen-Regulated Gene Expression via Estrogen Receptor (ER) {alpha} or ERß in Human Osteosarcoma Cells: Distinct and Common Target Genes for These Receptors

Fabio Stossi, Daniel H. Barnett, Jonna Frasor, Barry Komm, C. Richard Lyttle and Benita S. Katzenellenbogen

Departments of Molecular and Integrative Physiology (F.S., J.F., B.S.K.) and Cell and Structural Biology (D.H.B., B.S.K.), University of Illinois at Urbana-Champaign, Urbana, Illinois 61801; and Women’s Health Research Institute (B.K., C.R.L.), Wyeth Research, Collegeville, Pennsylvania 19426

Address all correspondence and requests for reprints to: Dr. Benita S. Katzenellenbogen, University of Illinois, Department of Molecular and Integrative Physiology, 524 Burrill Hall, 407 South Goodwin Avenue, Urbana, Illinois 61801-3704. E-mail: katzenel{at}life.uiuc.edu.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Estrogens exert many important effects in bone, a tissue that contains both estrogen receptors {alpha} and ß (ER{alpha} and ERß). To compare the actions of these receptors, we generated U2OS human osteosarcoma cells stably expressing ER{alpha} or ERß, at levels comparable with those in osteoblasts, and we characterized their response to 17ß-estradiol (E2) over time using Affymetrix GeneChip microarrays to determine the expression of approximately 12,000 genes, followed by quantitative PCR verification of the regulation of selected genes. Of the approximately 100 regulated genes we identified, some were stimulated by E2 equally through ER{alpha} and ERß, whereas others were selectively stimulated via ER{alpha} or ERß. The E2-regulated genes showed three distinct temporal patterns of expression over the 48-h time course studied. Of the functional categories of the E2-regulated genes, most numerous were those encoding cytokines and factors associated with immune response, signal transduction, and cell migration and cytoskeleton regulation, indicating that E2 can exert effects on multiple pathways in these osteoblast-like cell lines. Of note, E2 up-regulated several genes associated with cell motility selectively via ERß, in keeping with the selective E2 enhancement of the motility of ERß-containing cells. On genes regulated equally by E2 via ER{alpha} or ERß, the phytoestrogen genistein preferentially stimulated gene expression via ERß. These studies indicate both common as well as distinct target genes for these two ERs, and identify many novel genes not previously known to be under estrogen regulation.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
ESTROGEN IS A PLEIOTROPIC hormone with multiple actions in reproductive tissues (such as breast, uterus, and ovary) and in many nonreproductive tissues including bone, the central nervous system, and the cardiovascular system. In the skeleton, estrogen effects range from regulation of bone growth during puberty to bone remodeling in the adult (reviewed in Refs.1, 2, 3, 4). The pivotal role of estrogens in the maintenance of bone tissue has long been known from clinical studies where 17ß-estradiol (E2) deficiency in postmenopausal or ovariectomized women caused a rapid loss of trabecular bone; moreover, in both men and women, estrogen deficiency is associated with an age-related sustained bone loss that can lead to osteoporosis. Both situations can be reversed by hormone replacement therapy (1, 2, 3, 4).

Estrogen exerts its effects on bone and other target tissues by interacting with two different members of the nuclear receptor superfamily of hormone-regulated transcription factors, named estrogen receptor (ER) {alpha} and ERß (5, 6, 7). These two receptors are encoded by different genes, on human chromosome 6 and 14, respectively. ER{alpha} and ERß have similar modular domain structures and very high amino acid identity in their DNA-binding domains (97%), whereas they are more divergent in their N-terminal A/B domains (only 18% amino acid identity) and in their ligand-binding domains (59% amino acid identity). After the binding of hormone to these receptors, the hormone-receptor complexes bind to specific sequences on the DNA [estrogen response elements (EREs)] or interact with other transcription factors without direct ER{alpha} or ERß binding to DNA (i.e. at activator protein-1, Sp1, and other sites) (8, 9, 10, 11, 12, 13). In both cases, liganded ERs recruit coregulator proteins and components of the transcriptional machinery to regulate the transcription of target genes (14, 15, 16). Recently, the importance of membrane-initiated signaling in the actions of estrogens has been highlighted by studies in vitro and in vivo, where membrane/cytoplasmic ERs as well as nuclear ERs seem to be involved in regulating E2 action and gene transcription in bone and other target cells (1, 17, 18, 19).

In bone, ERs are present in osteoblasts and chondrocytes, and at somewhat lower levels in osteoclasts, bone marrow stromal cells, osteocytes, and bone cell precursors (20, 21). The levels of ERs in bone are generally about 10-fold lower than in reproductive tissues, such as uterus, and their levels can be affected by many parameters including cell differentiation state (22, 23, 24). Estrogen seems to act directly on osteoblasts and probably, as well, on osteoclasts and precursors of both cell types, and indirectly by regulating cytokine production in osteoblasts and bone marrow stromal cells, which in turn affects the actions and formation of osteoclasts (1, 25).

The aims of this study were to determine the effects of E2 on gene expression in a human osteoblastic cell line and, via transcriptional profiling, to better understand the comparative roles of ER{alpha} and ERß in mediating the effects of this hormone. In particular, our interest was to elucidate whether the two receptors elicit the same and/or different transcriptional responses on a range of endogenous cellular genes. For this purpose, we have generated U2OS human osteosarcoma cells stably expressing either ER{alpha} or ERß and examined several clonal cell lines containing similar, relatively low, physiological levels of functional receptor. Using Affymetrix GeneChip microarrays, which allow the examination of >12,000 genes, we observed that the hormone-regulated genes divided into three distinct temporal patterns and into several functional categories, indicating that E2 exerts effects on multiple regulatory pathways in these cells. In this study, we report the identification of genes that were commonly regulated by E2 through ER{alpha} and ERß, as well as some that were preferentially or exclusively regulated by one or the other ER subtype, and many novel genes not previously known to be under regulation by this hormone, and we discuss the relationship of some of these genes to the biological effects of E2 we observed in these cells.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Cell culture
U2OS human osteosarcoma cells were stably transfected with human ER{alpha} (encoding amino acids 1–595) or ERß (encoding amino acids 1–530). Each ER cDNA (26) was subcloned into the pcDNA3.1+ expression vector (Invitrogen, Carlsbad, CA), which contains a neomycin resistance gene. Clones were selected with the antibiotic G418 (800 µg/ml) and 20 clones per ER subtype were screened for ER expression and for transactivation ability with an ERE-containing reporter gene [2ERE-tk-CAT (chloramphenicol acetyltransferase)]. Four clones of each (denoted as ER{alpha} clones 1–4 and ERß clones 1–4) were chosen for further characterization and for the gene expression studies. The U2OS-ER cells were routinely grown in MEM with phenol red (Sigma, St. Louis, MO) supplemented with 15% fetal bovine serum (Hyclone, Logan, UT), 100 U penicillin/ml, 100 µg streptomycin/ml, and 400 µg G418/ml. Before use in experiments, cells were grown in MEM without phenol red and supplemented with 5% charcoal-dextran-treated fetal bovine serum for at least 4 d before the start of E2 treatment. MCF-7 human breast cancer cells were grown as previously described (27).

Hormone binding and Western immunoblot assays
Whole cell extracts were prepared in cell lysis buffer [20 mM Tris (pH 7.4), 0.5 M NaCl, 1 mM dithiothreitol, 10% glycerol, 50 µg/ml leupeptin, 50 µg/ml aprotinin, 2.5 µg/ml pepstatin A, and 0.2 mM phenylmethylsulfonyl fluoride] using freeze/thaw procedure. Total protein concentration was determined using the BCA kit (Pierce Biotechnology, Rockford, IL). Whole cell extracts were incubated in duplicate with a range of 3H-E2 concentrations alone or with 100-fold excess unlabeled E2 for 1 h on ice. Hydroxylapatite slurry was added and incubated for an additional 15 min on ice. The slurry was washed twice and its radioactivity then determined by scintillation counting. The amount of E2 binding in picomoles was calculated per milligram of protein.

Western blots of whole cell extracts used the ER{alpha}-specific antibody H222 and the ERß-specific antibody UCG40 (kindly provided by Geoffrey L. Greene, University of Chicago, Chicago, IL) and were done as previously described (28).

Hormone treatments, DNA microarrays, and analysis of microarray data
U2OS-ER{alpha} or ERß containing cells were maintained in culture for 48 h before cell harvest and RNA collection. During this period, cells were treated with 10 nM E2 (Sigma) for 4, 8, 24, or 48 h, and three separate samples were collected for each time point. Control cell samples were also treated with 0.1% ethanol control vehicle for 48 h. In this way, cell density was similar in all samples, consistent with our observations (as discussed in Results), that E2 does not affect proliferation of these cells. Additional independent time course experiments were conducted to generate RNA samples for quantitative RT-PCR verification of gene regulation, as detailed in the next section below. Total RNA was prepared using Trizol Reagent (Invitrogen) according to the manufacturer’s instructions. RNA was further purified using RNeasy columns (QIAGEN, Valencia, CA) and treatment with ribonuclease-free deoxyribonuclease I (QIAGEN).

Total RNA from each sample was used to generate cRNA, which was labeled with biotin as recommended by Affymetrix (Santa Clara, CA). Each cRNA was then hybridized on an Affymetrix human Hu-U95A GeneChip, which contains oligonucleotide probe sets representing approximately 12,500 human genes and expressed sequence tags. After washing, the chips were scanned and analyzed using MicroArray Suite 5.0 software (Affymetrix, Santa Clara, CA). Average intensities for each GeneChip were globally scaled to a target intensity of 150. Data were then analyzed using GeneSpring version 5.0.1 software (Silicon Genetics, San Carlos, CA). Data were first normalized on a per chip basis by dividing each measurement by the 50th percentile of all measurements on that chip, and then E2-treated samples were normalized to the mean of the vehicle-treated control samples.

We applied a confidence score (CS) to evaluate which genes were estrogen regulated (as described in Ref.29 ; adapted from Ref.30). The CS is based on four parameters: fold change (FC), P value (PV), percentage of present calls (PC) and expression level (EL), (CS = FC + PV + PC + EL). For each parameter, arbitrary scores were assigned. For FC, a score of 5 was assigned if the FC was greater than 1.95, 2 if the FC was between 1.5 and 1.95 and a penalty of –0.5 if it was under 1.5. For P value, a score of 3 was assigned if P value was less than 0.05, 2 if was between 0.05 and 0.1, and a penalty of –0.5 if it was greater than 0.1. If present calls were assigned to more than 50% of the samples, the score was 3, between 25–50% was 1, and if less than 25% a penalty of –0.5 was applied. For the expression level, a score of 3 was applied if it was greater than 30, a score of 1 if it was between 15 and 30 and a penalty of –0.5 if expression level was less than 15. The confidence scores ranged from –2 to 14, and genes with a CS value of 11 or higher were considered to be significantly E2-regulated genes. To determine ER{alpha} or ERß preference in gene regulation, the average FC for one of the receptors had to be 2.0 or greater, whereas FC was less than 1.3-fold for the other receptor at all time points.

Estrogen-regulated genes were assigned to functional categories according to LocusLink, OMIM, PubMed (http://www.ncbi.nlm.nih.gov), GeneCards (http://bioinfo.weizmann.ac.il/cards) and GenMAPP databases. The entire microarray data set will be available through the Gene Expression Omnibus (GEO, http://www.ncbi.nlm.nih.gov/geo).

Quantitative real-time PCR
Real-time PCR was carried out on approximately 25 genes to verify E2 regulation as assessed by microarray data analysis, and to evaluate mRNA levels of ER{alpha} or ERß in U2OS stably transfected cells. Three independent time-course experiments using three separate cell samples each for ER{alpha} or ERß containing cells either E2 or control vehicle-treated at each time point, were conducted. The primers used are listed in Table 1Go. One microgram of total RNA from each sample was reverse transcribed in a total volume of 20 µl using 200 U reverse transcriptase, 50 pmol random hexamers, and 1 mM deoxynucleotide triphosphates (New England Biolabs, Beverly, MA). The resulting cDNA was then diluted to a total volume of 100 µl. Each real-time PCR consisted of 5 µl of diluted reverse transcription product, 1x SYBR Green PCR Master Mix (Applied Biosystems, Foster City, CA), and 50 nM of forward and reverse primers. Reactions were carried out in an ABI Prism 7700 Sequence Detection System (Applied Biosystems) for 40 cycles (95 C for 15 sec, 60 C for 1 min) after an initial 10-min incubation at 95 C. The FC in expression was calculated using the {Delta}{Delta} Ct comparative threshold cycle method (31) with the ribosomal protein 36B4 mRNA as an internal control. As described in Ref.31 , gene expression is normalized to an endogenous reference gene (36B4) and the FC in gene expression is then determined relative to the vehicle-treated control.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Primers used for real-time PCR

 
Cell motility and cell adhesion assays
Cell motility assays used 48-well Boyden microchemotaxis chambers according to the manufacturer’s instructions (Neuroprobe, Cabin John, MD). Briefly, the cells were grown in 5% charcoal-dextran-treated fetal bovine serum MEM without phenol red for at least 4 d until approximately 70–80% confluency. Cells were treated for either 24 h or 48 h with E2 or control (0.1%) ethanol vehicle, harvested, and 50 µl of cell suspension (105 cells/50 µl serum-free MEM minus phenol red and containing 0.5% BSA) were placed in the open-bottom wells of the upper compartment. Each pair of wells was separated by a poly-vinylpyrrolidone-free polycarbonate porous membrane (8-µm pores) precoated with gelatin (0.2 mg/ml in PBS). Twenty-eight microliters of the chemoattractant [MEM with 5% fetal bovine serum (FBS) and MEM with no FBS as control] were placed into the lower compartment wells of the chamber. The Boyden chambers were then incubated for 5 h at 37 C. Cells were then fixed on the membrane with methanol, stained with Diff-Quick staining kit (Baxter Health Corp., McGraw Park, IL), mounted onto glass slides, and counted under a x100 microscope objective.

Cell adhesion assays were performed in 96-well plates coated for 2 h at 37 C with 20–40 µg/ml of different coating substrates (poly-L-lysine, BSA, collagen type I, fibronectin, or laminin). Cells were treated with 10 nM E2 for 24 or 48 h, and after blocking nonspecific sites with 1% BSA for 2 h at 37 C, cells were harvested, resuspended in MEM minus phenol red containing 0.1% BSA at 105 cells/ml, and 100 µl of cell suspension were seeded per well. Cells were allowed to adhere for 30 min and then the medium was gently removed, the wells washed once with MEM minus phenol red and the cells then fixed with methanol. Cells were stained with 0.1% crystal violet for 30 min at room temperature and washed multiple times with double distilled water. Stain was then extracted from the cells with 5% Triton X-100 overnight at room temperature and absorbance monitored at 570 nm.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Generation and characterization of U2OS cell lines stably expressing ER{alpha} or ERß
U2OS human osteosarcoma cells, an ER-negative osteoblast-like cell line, were stably transfected with either ER{alpha} or ERß using a pcDNA3.1+ plasmid containing a neomycin resistance gene as a vector for stable integration. Positive clones were selected with G418 (800 µg/ml), and after two rounds of selection, various clones were tested for the presence of ER mRNA by real-time PCR and for ER protein by Western blot with ER{alpha}-specific or ERß-specific antibodies (Fig. 1Go). The levels of ER{alpha} and ERß mRNA in U2OS-ER cell clones, as determined by real-time quantitative PCR using standard curves with known amounts of ER{alpha} and ERß cDNA (Fig. 1AGo), were compared with the parental U2OS cells and with the ER{alpha}-containing human breast cancer cell line MCF-7. In the eight clones (ER{alpha} 1–4, ERß 1–4) used for most of the studies reported in this paper, the number of copies of ER{alpha} and ERß mRNA were similar: 8410 ± 777 per 10 ng total RNA for ER{alpha} and 9112 ± 1304 for ERß. These ER{alpha} mRNA levels are approximately 20% of that measured in our MCF-7 breast cancer cells. In addition, as seen in Fig. 1AGo, no ERß mRNA was detected in the ER{alpha}-expressing U2OS cells and likewise no ER{alpha} mRNA was present in the U2OS ERß-expressing cells. The parental U2OS cells lacked detectable levels of either receptor mRNA.



View larger version (27K):
[in this window]
[in a new window]
  		FIG. 1. Characterization of U2OS human osteosarcoma cells stably expressing ER{alpha} or ERß. A, ER mRNA copy number/10 ng total RNA as assessed by real-time PCR in ER{alpha} clones 1–4 and in ERß clones 1–4. B, E2 binding sites/milligram protein evaluated by E2 binding assay in parental U2OS and U2OS-ER cell lines (ER{alpha} clone 1 and ERß clone 1) in comparison with the MCF-7 breast cancer cell line. The mean of closely corresponding duplicate determinations is shown. C, Western blot for ER{alpha} and ERß protein in different clones of stably transfected U2OS cells. D, Motility (Boyden chamber assay) of cells (ER{alpha} clone 1 and ERß clone 1) was assayed with 5% FBS as a general chemoattractive factor. E2 treatment was for 48 h. Similar findings were obtained with three other ER{alpha} and ERß clones. E, Adhesion assay was performed in 96-well plates coated with various substrates. Shown is 30 min adhesion on collagen type I measured as absorbance at 570 nm. E2 treatment was for 48 h. No effect of E2 on cell adhesion was also seen in repeat experiments with three other ER{alpha} and ERß clones. *, P < 0.01 for E2-treated cells vs. cells treated with control vehicle.

 
Hormone binding (Scatchard) assays were performed on extracts from ER{alpha} clone 1 and ERß clone 1 cells to ensure the presence of receptor able to bind hormone. These assays demonstrated similar levels of 3H-E2 binding in these U2OS-ER{alpha} and U2OS-ERß cells (~0.1 pmol/mg protein, Fig. 1BGo). These levels are about 20% of that found in MCF-7 cells. Western blots (Fig. 1CGo) with H222 monoclonal antibody for ER{alpha} and UCG40 polyclonal antibody for ERß showed ER{alpha} or ERß of the correct size (~65 and 56 kDa, respectively), and the presence of only ERß in the ERß clones (ß clones 1–4 shown) and only ER{alpha} in the ER{alpha} clones ({alpha} clone 1 shown). Further characterization by estrogen-responsive ERE-tk-CAT reporter assays demonstrated that U2OS-ER{alpha} and ERß cells possessed E2-dependent transcriptionally functional receptors (data not shown). The level of ER{alpha} and ERß in these clones is consistent with previous reports indicating the approximately 10-fold lower ER level in bone vs. uterine tissue (23, 24).

We examined whether E2 affected the motility and adhesion properties of the U2OS-ER{alpha} and ERß-containing cells. Motility was assessed using a Boyden chamber chemotactic assay and 5% FBS as a general chemotactic stimulus. With E2 treatment for 24 or 48 h, the ERß-containing cells showed a 3-fold increase in motility; in contrast, although the ER{alpha} and ERß-containing cells showed similar motility, E2 elicited no change in motility of the ER{alpha}-containing cells (Fig. 1DGo). We also evaluated whether E2 treatment for 24 or 48 h affected the adhesive properties of these cells and found that it did not. This employed an adhesion assay that allows a comparison of the rapid adhesion (in 30 min) of the cells to various substrates (poly-L-lysine, BSA, collagen type I, fibronectin, or laminin). Adhesion to collagen type I is shown as an example (Fig. 1EGo) because it is a major component of bone matrix, and it is seen that E2 had no effect on the adhesion of the cells. This was the case with all of the substrates tested, indicating that this property is largely E2 independent.

Gene expression profiling and time courses of gene regulation using Affymetrix GeneChip microarrays
To characterize changes in gene expression in response to E2 treatment, U2OS-ER{alpha} (clone 1) and U2OS-ERß (clone 1) cells were treated with 10 nM E2 for times from 0–48 h, and RNA was harvested, purified, and derivatized for gene profiling. Affymetrix Hu-U95A GeneChips, which contain probes for approximately 12,000 human genes, were used. As described in Materials and Methods, a CS was applied to discriminate genes robustly and reproducibly regulated by E2. The CS takes into account several parameters for each gene and weights the FC increase or decrease vs. control, the reproducibility of the output (P value), the expression level related to the intensity of the signal on the chip and a fidelity score based on a decision matrix, which compares hybridization of the perfect match probes vs. the mismatch for each probe set.

We considered E2-regulated genes to be those with a confidence score equal to or greater than 11 (of a maximum score of 14), and we included in our further studies only genes up- or down-regulated more than 2-fold. Using these criteria, we identified 105 genes that were E2-regulated through ER{alpha} and/or ERß. This list of genes was subjected to gene tree cluster analysis using a standard correlation algorithm available in the GeneSpring software.

In the gene cluster analysis (Fig. 2Go), up-regulated genes are shown in red, down-regulated genes in blue, and unchanging genes in yellow. We observed that many more genes were stimulated than inhibited (85 vs. 20) in ER{alpha}- or ERß-containing cells, and in further analyses, we focused on the up-regulated genes because these were more numerous, were the most robustly regulated, and gave good reproducibility of validation by real-time PCR.



View larger version (51K):
[in this window]
[in a new window]
  		FIG. 2. Cluster analysis of E2-regulated genes in U2OS-ER{alpha}- or -ERß-containing cells. After applying normalization and confidence score analyses, all the E2-regulated genes in U2OS-ER containing cells (ER{alpha} clone 1 and ERß clone 1) were clustered using a standard correlation algorithm (GeneSpring software). Up-regulated genes are shown in red, down-regulated genes in blue, and nonchanging genes in yellow. The color scale corresponding to FC in gene expression is shown at the left. The different time points of the E2 (10 nM) time-course treatment are indicated at the right. The gene expression values shown are the average FC of independent samples, each run on a separate microarray chip, for each time point.

 
When the E2-regulated genes were categorized according to their time patterns of expression, three major groupings were identified: early (stimulated only at 4–8 h), early and late (stimulated from 4–48 h) and late (stimulated only at 24–48 h) (Fig. 3Go). For ER{alpha}-containing cells, the early and late pattern was the most frequently represented (49% of E2-stimulated genes through ER{alpha}), followed by the late (46%) and early (5%) patterns. In U2OS-ERß cells, the distribution was slightly different with 43%, 41%, and 16%, respectively, in these three time patterns.



View larger version (18K):
[in this window]
[in a new window]
  		FIG. 3. Time-course patterns of E2-stimulated genes in U2OS-ER{alpha} and U2OS-ERß cells as identified from microarray analysis. All E2 up-regulated genes were assigned to one of three categories. For "Early only" regulated genes (dotted broken line with diamonds), the FC was 2-fold or greater at 4 and/or 8 h only. For "Early and Late" regulated genes (dashed line with triangles), the FC was 2-fold or greater at 4 and/or 8 h and at 24 and/or 48 h. And for "Late only" regulated genes (solid line with squares), the FC was 2-fold or greater at 24 and/or 48 h only. Once genes were assigned to one of the three time-course patterns, the mean FC ± SEM was calculated and plotted for all genes in that pattern. On some points, error bars are too small to be visible.

 
Of the 85 up-regulated genes, 52 were commonly regulated by ER{alpha} and ERß (Fig. 4Go, top). These genes, their functions, and their time patterns of regulation are listed in Table 2AGoGo. As shown in Fig. 4Go (top), 24 genes were stimulated selectively by ER{alpha}, whereas nine genes were stimulated selectively by ERß. Thus, the two receptors showed substantial overlap in regulated genes (61% of the genes), but both also have some different target genes. The estrogen-stimulated genes regulated selectively through one of the two ER subtypes are listed in Table 3Go, along with their temporal pattern of regulation and their function. Many new E2 target genes were also identified, as indicated in Tables 2aGo and 3Go.



View larger version (33K):
[in this window]
[in a new window]
  		FIG. 4. Functional classification of E2-stimulated genes. E2 up-regulated genes were classified as ER-subtype selective when the FC from the microarray data for one receptor was 2.0-fold or greater, whereas for the other receptor it was less than 1.3-fold at all time points. All the genes identified were then categorized in functional groups according to their main known function based on LocusLink, OMIM, PubMed, GeneCards, and GenMAPP databases.

 

View this table:
[in this window]
[in a new window]
  		TABLE 2. Genes up-regulated by estradiol through ER{alpha} and ERß in U2OS cells

 

View this table:
[in this window]
[in a new window]
  		TABLE 2A. Continued

 

View this table:
[in this window]
[in a new window]
  		TABLE 3. Estrogen-stimulated genes regulated selectively through one ER subtype

 
Of note, our method of analysis demonstrated E2 regulation of some known E2-target genes [i.e. pS2/TFF1 (trefoil factor 1), PDZK1 (PDZ domain containing 1), GREB1 (gene regulated by estrogen in breast cancer 1), WISP2 (WNT1-inducible signaling pathway protein 2), keratin 19, angiotensinogen, NHERF (Na+/H+ exchanger regulatory factor)/SLC9A3R2, bone morphogenetic protein 6 (BMP6), connexin 43/GJA1], suggesting the validity of our analyses. Furthermore, regulation of several bone-related factors [BMP6, CD34, CD164, PTH-like hormone (PTHLH)] by E2 was observed (Tables 2aGo and 3Go).

Functional classification of E2-regulated genes
To classify E2 up-regulated genes into functional groups (Fig. 4Go), we chose eight categories: cytokines/immune response, signal transduction, cell motility/cytoskeleton regulation, growth factors/hormones, apoptosis/cell proliferation, housekeeping, nucleic acid processing, and other/unknown. We then assigned each gene to one major category representing its primary function (even though many genes encode proteins that have several physiological functions), based on mining of several databases (LocusLink, PubMed, GeneCards, OMIM, GenMapp).

Many of the genes commonly regulated through ER{alpha} and ERß encode proteins associated with cytokine/immune response, signal transduction, and cell motility and cytoskeleton regulation (56% of all genes). Several ER{alpha}-selective genes were associated with nucleic acid processing. Very few genes regulated by the ERs were related to cell cycle, proliferation, or apoptosis, supporting our observations (not shown) that E2 had no effect on the proliferation rate of these cells.

Hormone-regulated genes encoding cytokines and factors associated with immune response
Three members of the Natural Killer cell lectin-type receptors (NKG family), NKG2-C, -E, and -F (KLRC2, KLRC3, and KLRC4) were robustly regulated by E2 through both receptors (Table 2AGo). NKGs are activating receptors of natural killer cells that mediate HLA-E recognition (32, 33, 34). CD34 (common to both ER{alpha} and ERß up-regulation; Table 2AGo) and sialomucin/CD164 (ER{alpha} selective; Table 3Go) are cell surface antigens characteristic of human hematopoietic progenitors. CD164 is an adhesion receptor that inhibits proliferation of CD34+ hematopoietic progenitors (35, 36). IL-8, a cytokine known for its angiogenic and chemotactic properties, is thought to be involved in bone cell function under IL-1ß and TNF{alpha} control in bone marrow stromal cells and osteoblasts (37, 38). It was also regulated by E2 in both ER{alpha}- and ERß-containing U2OS cells (Table 2AGo).

Cyclooxygenase-2 (Cox-2), a key enzyme regulating the production of prostaglandins, and also regulating angiogenesis, tumor cell invasion and inflammatory responses (39), was markedly up-regulated by E2 via both ER{alpha} and ERß (Table 2AGo). Cox-2 has been shown recently to be under estrogenic regulation mediated through the phosphoinositide 3-kinase/Akt pathway (40).

Hormone-regulated genes associated with cell motility and cytoskeletal function
This category is of particular interest because, using cell motility assays, we observed an increase in motility of U2OS-ERß containing cells after E2 treatment (Fig. 1DGo). Among the ERß selectively regulated genes, there are two candidates that might be involved in this response. The first is autotaxin (ENPP2), a known tumor autocrine motility factor that has been shown to stimulate the migration of melanoma cells via a G protein-coupled phosphoinositide 3-kinase {gamma} and Cdc42- and Rac1-dependent pathway. This regulation possibly involves Paxillin and focal adhesion kinase, two key players in cell adhesion and motility control (41, 42). Autotaxin has 5-nucleotide pyrophosphatase and phosphodiesterase activity, and it acts in the extracellular space as a lysophospholipase D with production of lysophosphatidic acid, which seems to be the tumor cell motility inducer (43). The experiments described in the section below demonstrate that autotaxin is very selectively regulated by ERß. Another candidate associated with the motility increase by ERß might be Rap1GEF (GFR), a guanine nucleotide exchange factor that constitutively activates Rap1 (44) and has been shown to be involved in cell motility by modulating integrin function (45, 46).

Quantitative real-time PCR validation and verification of gene expression regulation in multiple cell clones and using ER subtype-selective ligands
Table 1Go shows the primers used, and Tables 2AGo and 3Go show the approximately 25 genes that we verified for their E2 regulation by quantitative PCR (names in italics). It is of note that many of these are genes newly identified as being regulated by E2. In addition, we verified their E2 regulation in several clones (four each) of U2OS ER{alpha} and/or ERß-containing cells (clones 1–4) and used ER subtype-selective ligands to confirm common regulation by both ERs or selective gene regulation by ER{alpha} or ERß. The tables indicate generally good correspondence in gene regulation between the microarray and the quantitative RT-PCR analyses using several ER{alpha}- or ERß-containing cell clones. RT-PCR, however, is often more sensitive, with a greater dynamic range, such that FCs found by RT-PCR were often greater than those obtained from microarray analysis.

Figure 5Go shows real-time PCR validation of some genes that were commonly regulated by E2 through both ER subtypes. The time course of stimulation for four of these genes (GREB1, oligophrenin-1, IL-24, and carbonic anhydrase XII) is presented. All of these genes showed up-regulation by E2 by 4 h. There were robust, and very similar FCs in mRNA for GREB1 mediated via ER{alpha} or ERß over time (Fig. 5Go). Oligophrenin-1 also showed a pattern of mRNA change that was similar for ER{alpha} and ERß in which a plateau in stimulation was achieved by 4 h for ER{alpha} and 8 h for ERß. Interestingly, IL-24 was preferentially up-regulated by ER{alpha}, with ERß being a much less effective mediating receptor. In contrast, the stimulation via ERß was substantially greater for the carbonic anhydrase XII gene than was the stimulation via ER{alpha}. The ER mediation of all of these E2-induced changes is clearly suggested by the fact that the stimulation is fully reversed by an excess of the antiestrogen ICI 182,780. In addition, and as expected, no stimulation was observed with this pure antiestrogen alone (Fig. 5Go).



View larger version (30K):
[in this window]
[in a new window]
  		FIG. 5. Real-time PCR validation of genes regulated commonly by both ER{alpha} and ERß. E2 (10 nM) time-course and ICI 182,780 (1 µM) treatments were performed in three independent experiments to confirm DNA microarray data and assess the ER-dependent mechanism of the gene regulation through reversal by a 100-fold excess of the antiestrogen (ICI + E2). Values are mean FC + SEM. *, P < 0.05 for gene stimulation by E2 vs. vehicle control.

 
Figure 6Go shows the time course changes for two genes selectively regulated through ERß: autotaxin and cystatin D. Their expression was highly stimulated via ERß, with no significant change being elicited via ER{alpha}. The ICI antiestrogen reversal of the E2 stimulation of these genes supports the mediation by ER (Fig. 6Go).



View larger version (13K):
[in this window]
[in a new window]
  		FIG. 6. Real-time PCR validation of genes regulated selectively through ERß. E2 (10 nM) time-course and ICI 182,780 (1 µM) treatments were performed in three independent experiments to confirm DNA microarray data and assess the ER-dependent mechanism of the gene regulation through reversal by a 100-fold excess of the antiestrogen (ICI + E2). Values are mean FC + SEM. *, P < 0.05; {dagger}, P = 0.06 for gene stimulation in ERß- vs. ER{alpha}-containing cells.

 
To further examine the selectivity of gene regulation via ER{alpha} or ERß, cells were treated with the ER{alpha}-selective ligand propylpyrazoletriol [PPT (47, 48, 49, 50)], the ERß-selective ligand diarylpropionitrile [DPN (51, 52)], or the soy phytoestrogen genistein which has a higher affinity for ERß than ER{alpha} (53) (Fig. 7Go). The PDZK1 gene, which is equivalently stimulated by E2 via ER{alpha} or ERß, showed stimulation by PPT in the ER{alpha}-containing cells and by DPN in the ERß-containing cells, as expected. Interestingly, genistein, which has preferential binding affinity for ERß (53), was considerably more effective in stimulating PDZK1 gene expression via ERß (Fig. 7Go). At the cystatin D gene, E2 exclusively stimulated via ERß, and the stimulation by genistein and DPN was only observed in the ERß-containing cells, with no stimulation by PPT or genistein via ER{alpha}. These findings confirm the common, ER{alpha} and ERß regulation for PDZK1 observed with E2, and the ERß selectivity for E2 in regulation of cystatin D expression.



View larger version (12K):
[in this window]
[in a new window]
  		FIG. 7. Real-time PCR analysis of gene regulation by E2, the ER{alpha}-selective ligand PPT, the ERß-selective ligand DPN, and Genistein. E2 (10 nM), PPT (100 nM), DPN (100 nM), and Genistein (1 µM) treatments were performed for 48 h in three separate experiments to examine the ability of ER-subtype selective ligands to regulate gene expression in cells containing either ER{alpha} or ERß. Values are mean FC + SEM. *, P < 0.05; {dagger}, P = 0.07 for gene stimulation in ERß- vs. ER{alpha}-containing cells.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Estrogens exert profound effects on bone, a tissue that expresses ER{alpha} and ERß. Yet the individual contributions of ER{alpha} and ERß to gene regulation and physiological function in bone remain only partially understood. In this study, we have characterized a model system for examining the actions of the ER subtypes, ER{alpha} and ERß, in a human osteoblast-like cell type. These studies have enabled us to evaluate changes in the expression of approximately 12,000 genes by E2 through ER{alpha} and/or ERß. We have identified novel estrogen-regulated genes and genes regulated commonly as well as exclusively or preferentially through ER{alpha} or ERß. Furthermore, the identification of regulated genes allows some insight into the pathways and gene networks regulated by E2 through these two ERs.

Estrogen regulation of genes associated with cytokines, motility, cytoskeleton, and bone remodeling
Because estrogen plays important roles in bone remodeling and the control of osteoblast and osteoclast numbers (1), it is of note that E2 showed significant regulation of cytokines and of genes encoding proteins associated with regulation of the immune response. Our findings are consistent with the reports of estrogen regulation of genes encoding cytokines, growth factors, and bone matrix proteins in wild-type mouse trabecular bone and the importance of ER{alpha} and/or ERß in the mediation of these effects, as evidenced by the use of ER knockout mice (54, 55, 56). We observed E2 up-regulation of genes encoding three members of the Natural Killer cell lectin-type receptors (NKG family, KLRC2, KLRC3, and KLRC4). These NKG family genes were all regulated by E2 at early as well as later times. Other factors associated with cytokine/immune function observed to be under estrogen regulation included CD34 antigen, defensin ß1, IL-13 receptor {alpha}2, IL-8, phospholipase A2, selenoprotein P1, and sialomucin/CD164. Many of these genes, as indicated in Tables 2AGo and 3Go, represent new E2-regulated genes, ones not previously reported to be regulated by this hormone.

Another category of genes, in which many members were regulated by E2, were those associated with regulation of cell motility and the cytoskeleton. We observed a 3-fold increase in the motility of the ERß-containing cells in response to E2 treatment. Hence, it is of note that two genes preferentially regulated by E2 via ERß were autotaxin (ENPP2), a known tumor autocrine motility factor and phosphodiesterase, and Rap1 GEF, the guanine nucleotide exchange factor that constitutively activates Rap1 and is known to modulate integrin function (45, 46).

WISP-2, a connective tissue growth factor isolated from osteoblasts (57), was regulated by E2 via both receptor subtypes, and it might play a role in cytoskeletal regulation in these U2OS cells. Its stimulation by E2 has been reported in MCF-7 breast cancer cells (29, 58). In addition, oligophrenin-1, keratin 19, integrin {alpha} 6, cyritestin 1, connexin 43, and cadherin 19, all of which play roles in cell motility and the cytoskeleton, were for the most part not previously known to be regulated by estrogen in bone, although several are known to be under estrogen regulation in reproductive target cells (59, 60).

BMP6, a potent osteogenic factor believed to play an important role in the bone-protective actions of estrogens, was observed to be up-regulated through both ERs, as was PTH-like hormone (PTHLH/PTHrP), a homolog of PTH that functions as an autocrine growth inhibitor for osteoblast-like cells and as an anabolic agent in osteoporosis (61, 62). Neurotensin was another growth factor/hormone found to be up-regulated by E2 via ER{alpha} and ERß. Cystatin D, an E2-responsive gene newly identified in these studies, was stimulated very preferentially via ERß. Because this gene encodes a protein that is a secreted inhibitor of cysteine peptidases (cathepsins S and H), it might function in several roles in bone physiology.

ER regulation of other cellular functions: similarities and differences in estrogen target genes and temporal patterns of gene regulation in osteosarcoma and breast cancer cells
E2 up-regulated the expression of several genes encoding proteins associated with nucleic acid processing, such as the ER coregulators RIP140 (receptor-interacting protein 140) and TRAP240 (thyroid hormone receptor-associated protein, 240-kDa subunit). However, in these U2OS cells, E2 regulated very few genes associated with apoptosis or cell proliferation, consistent with our findings that E2 had no effect on the proliferation rate of either the ER{alpha}- or ERß-containing U2OS cells (data not shown). In contrast, in a recently reported study in ER-positive human breast cancer cells, in which E2 stimulates cell proliferation, we observed that E2 stimulated the expression of positive proliferation regulators including multiple growth factors and genes involved in cell cycle progression, and E2 also down-regulated transcriptional repressors and antiproliferative and proapoptotic genes (29). Hence, the categories of genes stimulated by E2 in the U2OS cells and in breast cancer cells are very different, and they reflect the quite different physiological effects of E2 on these target cells. Furthermore, E2 down-regulated the expression of few genes in these ER-containing osteosarcoma cells, whereas E2 down-regulated the expression of more genes in ER-positive MCF-7 breast cancer cells than it up-regulated (29). Despite these marked differences in genes regulated in these two different types of estrogen target cells, some genes were stimulated by E2 in both U2OS and MCF-7 breast cancer cells. These included pS2, keratin 19, NHERF, GREB1, PDZK1, receptor interacting protein 140, WISP2, and connexin 43, all previously shown to be under estrogen regulation, as referenced in Tables 2AGo and 3Go. Some have important roles in cell cytoarchitecture and cytoskeletal regulation (e.g. NHERF, connexin 43), supporting the known effects of estrogen on the cytoskeleton of both mammary (63) and bone cells.

It is of interest that our time-course analysis in the U2OS cells (Fig. 3Go) revealed that genes regulated by E2 divided into three temporal patterns, denoted E, E + L, or L in Tables 2AGo and 3Go: those regulated at early times only (4 and/or 8 h), those regulated at both early and late times, and those showing increased expression only at late times (24 and/or 48 h). In both MCF-7 and in ER-containing U2OS cells, approximately equal numbers of stimulated genes fell into the early and late, and late-only categories, whereas relatively few genes showed early-only regulation in the U2OS cells compared with the breast cancer cells, where approximately one third of the total stimulated genes showed the early-only expression pattern. We have used the protein synthesis inhibitor, cycloheximide, to begin to distinguish primary vs. secondary effects of E2-ER complexes on regulation of some of the interesting genes we have identified in the U2OS cells. These studies reveal that some of the genes turned on early that show elevated expression over time and hence fall into the "early and late" category (such as GREB1, autotaxin, oligophrenin-1, and KLRC4) likely represent primary estrogen response genes because their increase in mRNA expression is not prevented by cycloheximide. In contrast, cycloheximide blocked the estrogenic stimulation of cystatin D and PDZK1, a late and an early and late gene, respectively, suggesting that these are probably secondary response genes.

Novel estrogen-regulated genes and genes regulated selectively by ER{alpha} or ERß
These studies have enabled us to compare the actions of ER{alpha} and ERß. Although the majority of E2-stimulated genes were regulated through both ER subtypes, we identified a number of genes selectively regulated either by ER{alpha} or by ERß, as well as many novel estrogen-regulated genes (64). Most of the E2-regulated genes involved in cytoskeleton regulation and motility, signal transduction, cytokine, and immune response and growth factors/hormones were commonly regulated through both receptors. These included a marked up-regulation of Cox-2, cyritestin 1, keratin 19, several ILs, and integrin {alpha}6. The latter factor, integrin {alpha}6, forms laminin receptors when heterodimerized with integrin ß1 or ß4 subunits, and it is one of the factors involved in tumor invasion through basal membranes (65, 66, 67).

Important growth factors and hormones regulated by E2 through both receptors included several that can act in either an autocrine or paracrine manner, including BMP6 and PTHLH, to effect the bone-protective actions of E2 (68). PTHLH, a homolog of PTH, acts as an anabolic agent in osteoporosis (62), and its overproduction in breast and prostate cancers is associated with bone metastasis of these tumors (69, 70). Hence, its up-regulation by E2 via either ER{alpha} or ERß is of interest. Of the genes we observed to be selectively regulated by one ER subtype, nearly three fourths were ER{alpha} selective. These ER{alpha}-regulated genes included several associated with nucleic acid processing, two zinc finger proteins, the coregulator TRAP240, and the cytochrome P450 family member 2B6 (CYP2B6). Of the genes exclusively regulated by ERß were autotaxin and cystatin D, factors associated with motility/cytoskeleton and cytokine/immune function.

A recent report of related studies in U2OS cells identified some similar and some different gene targets for ER{alpha} and ERß (71) as we observed in this report. There are several important methodological differences between the two studies that could account for some of the differences in findings. First, the levels of ER{alpha} and ERß in the U2OS cells used by Monroe et al. (71) were very high, namely three times that of MCF-7 cells, whereas we chose to use cells that contained ER at levels more comparable to those in osteoblasts, about 20% that of MCF-7 cells. Also, the studies in Ref.71 examined E2 regulation only at 24 h, whereas we examined gene regulation over a time course of E2 treatment up to 48 h, so that we identified both early, early and late, and late-only responding genes; hence, some different sets of genes would likely be identified in the two studies. In addition, we used a confidence score that includes considerations of FC, P values, expression level and present calls in defining estrogen-regulated genes, whereas Monroe et al. (71) used FC only. Furthermore, the studies reported in Ref.71 used an earlier version Affymetrix microarray chip that contains probes for about half of the genes on the U95A array we used, and also uses fewer and different probe sets per gene. Hence, it is not surprising that there are some differences in the observed pattern of gene regulations between our study and that presented in Ref.71 .

The differences in gene targets of ER{alpha} vs. ERß, or of progesterone receptor-A vs. progesterone receptor-B (72), may reflect differences in the activation functions of these receptors, in particular their N-terminal activation regions, as well as the complexity of gene regulation by hormone occupied-nuclear receptors. The latter involves a combination of aspects (7, 16, 73): the diverse nature of target gene promoters (with different response elements that may be simple or composite) where receptors may bind directly to the DNA or act through tethering to other DNA-bound transcription factors; the nature of the coregulators recruited to the hormone-receptor-gene promoter; and the receptor protein conformations induced by different ligands in the context of associated coregulators and transcription factors.

These studies, which have allowed a comparison of the gene regulatory activities of ER{alpha} and ERß, should assist in the characterization of genes and cellular signaling pathways regulated by E2 through ER{alpha} and/or ERß. They highlight commonality but also significant differences in gene targets for these two ERs and have identified many novel genes not previously known to be under estrogen regulation. Several of these are likely to play significant roles in the bone maintenance and antiosteoporosis effects of estrogens. It will be of interest in subsequent studies to use the preferentially regulated genes we have identified to explore the molecular basis for the ER subtype selectivity in the regulation of these genes.


   Acknowledgments
 
We thank Karen Weis and Paolo Martini, former members of the Katzenellenbogen laboratory, for initial work on the generation and characterization of these cells, and acknowledge Mark Band, Lei Liu, and Dennis Akan of the University of Illinois Biotechnology Center for their advice and assistance.


   Footnotes
 
This work was supported by National Institutes of Health Grants CA18119 (to B.S.K.) and T32ES07326 (to D.H.B.), and The Breast Cancer Research Foundation (to B.S.K.).

F.S. and D.H.B. contributed equally and should be considered equal first authors.

Abbreviations: BMP6, Bone morphogenetic protein 6; CAT, chloramphenicol acetyltransferase; Cox-2, cyclooxygenase-2; CS, confidence score; DPN, diarylpropionitrile; E2, 17ß-estradiol; ERE, estrogen response element; ER, estrogen receptor; FBS, fetal bovine serum; FC, fold change; GREB, gene regulated by estrogen in breast cancer 1; NHERF, Na+/H+ exchanger regulatory factor; NKG, natural killer cell lectin-type receptors; PDZK1, PDZ domain containing 1; PPT, propylpyrazoletriol; PTHLH, PTH-like hormone, also known as PTHrP; WISP2, WNT1-inducible signaling pathway protein 2.

Received December 11, 2003.

Accepted for publication March 9, 2004.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Manolagas SC, Kousteni S, Jilka RL 2002 Sex steroids and bone. Recent Prog Horm Res 57:385–409[Abstract/Free Full Text]
  2. Manolagas SC 2000 Birth and death of bone cells: basic regulatory mechanisms and implications for the pathogenesis and treatment of osteoporosis. Endocr Rev 21:115–137[Abstract/Free Full Text]
  3. Riggs BL, Khosla S, Melton 3rd LJ 2002 Sex steroids and the construction and conservation of the adult skeleton. Endocr Rev 23:279–302.[Abstract/Free Full Text]
  4. Riggs BL, Khosla S, Melton 3rd LJ 1998 A unitary model for involutional osteoporosis: estrogen deficiency causes both type I and type II osteoporosis in postmenopausal women and contributes to bone loss in aging men. J Bone Miner Res 13:763–773[CrossRef][Medline]
  5. Katzenellenbogen BS, Montano MM, Ediger TR, Sun J, Ekena K, Lazennec G, Martini PGV, McInerney EM, Delage-Mourroux R, Weis K, Katzenellenbogen JA 2000 Estrogen receptors: selective ligands, partners, and distinctive pharmacology. Recent Prog Horm Res 55:163–195
  6. Pettersson K, Gustafsson JA 2001 Role of estrogen receptor ß in estrogen action. Annu Rev Physiol 63:165–192[CrossRef][Medline]
  7. McDonnell DP, Norris JD 2002 Connections and regulation of the human estrogen receptor. Science 296:1642–1644[Abstract/Free Full Text]
  8. McDonnell DP 1999 The molecular pharmacology of SERMs. Trends Endocrinol Metab 10:301–311[CrossRef][Medline]
  9. Klinge CM 2001 Estrogen receptor interaction with estrogen response elements. Nucleic Acids Res 29:2905–2919[Abstract/Free Full Text]
  10. Kushner PJ, Agard DA, Greene GL, Scanlan TS, Shiau AK, Uht RM, Webb P 2000 Estrogen receptor pathways to AP-1. J Steroid Biochem Mol Biol 74:311–317[CrossRef][Medline]
  11. Safe S 2001 Transcriptional activation of genes by 17ß-estradiol through estrogen receptor-Sp1 interactions. Vitam Horm 62:231–252[Medline]
  12. Saville B, Wormke M, Wang F, Nguyen T, Enmark E, Kuiper G, Gustafsson JA, Safe S 2000 Ligand-, cell-, and estrogen receptor subtype ({alpha}/ß)-dependent activation at GC-rich (Sp1) promoter elements. J Biol Chem 275:5379–5387[Abstract/Free Full Text]
  13. Paech K, Webb P, Kuiper GGJM, Nilsson S, Gustafsson J-A, Kushner PJ, Scanlan TS 1997 Differential ligand activation of estrogen receptors ERa and ERb at AP1 sites. Science 277:1508–1510[Abstract/Free Full Text]
  14. Hall JM, Couse JF, Korach KS 2001 The multifaceted mechanisms of estradiol and estrogen receptor signaling. J Biol Chem 276:36869–36872[Free Full Text]
  15. Shang Y, Hu X, DiRenzo J, Lazar MA, Brown M 2000 Cofactor dynamics and sufficiency in estrogen receptor-regulated transcription. Cell 103:843–852[CrossRef][Medline]
  16. McKenna NJ, O’Malley BW 2002 Combinatorial control of gene expression by nuclear receptors and coregulators. Cell 108:465–474[CrossRef][Medline]
  17. Kousteni S, Bellido T, Plotkin LI, O’Brien CA, Bodenner DL, Han L, Han K, DiGregorio GB, Katzenellenbogen JA, Katzenellenbogen BS, Roberson PK, Weinstein RS, Jilka RL, Manolagas SC 2001 Nongenotropic, sex-nonspecific signaling through the estrogen or androgen receptors: dissociation from transcriptional activity. Cell 104:719–730[Medline]
  18. Kousteni S, Han L, Chen JR, Almeida M, Plotkin LI, Bellido T, Manolagas SC 2003 Kinase-mediated regulation of common transcription factors accounts for the bone-protective effects of sex steroids. J Clin Invest 111:1651–1664[CrossRef][Medline]
  19. Kelly MJ, Levin ER 2001 Rapid actions of plasma membrane estrogen receptors. Trends Endocrinol Metab 12:152–156[CrossRef][Medline]
  20. Eriksen EF, Colvard DS, Berg NJ, Graham ML, Mann KG, Spelsberg TC, Riggs BL 1988 Evidence of estrogen receptors in normal human osteoblast-like cells. Science 241:84–86[Abstract/Free Full Text]
  21. Hoyland JA, Mee AP, Baird P, Braidman IP, Mawer EB, Freemont AJ 1997 Demonstration of estrogen receptor mRNA in bone using in situ reverse-transcriptase polymerase chain reaction. Bone 20:87–92[Medline]
  22. Waters KM, Rickard DJ, Riggs BL, Khosla S, Katzenellenbogen JA, Katzenellenbogen BS, Moore J, Spelsberg TC 2001 Estrogen regulation of human osteoblast function is determined by the stage of differentiation and the estrogen receptor isoform. J Cell Biochem 83:448–462[CrossRef][Medline]
  23. Bodine PV, Henderson RA, Green J, Aronow M, Owen T, Stein GS, Lian JB, Komm BS 1998 Estrogen receptor-{alpha} is developmentally regulated during osteoblast differentiation and contributes to selective responsiveness of gene expression. Endocrinology 139:2048–2057[Abstract/Free Full Text]
  24. Arts J, Kuiper GG, Janssen JM, Gustafsson JA, Lowik CW, Pols HA, van Leeuwen JP 1997 Differential expression of estrogen receptors {alpha} and ß mRNA during differentiation of human osteoblast SV-HFO cells. Endocrinology 138:5067–5070[Abstract/Free Full Text]
  25. Spelsberg TC, Subramaniam M, Riggs BL, Khosla S 1999 The actions and interactions of sex steroids and growth factors/cytokines on the skeleton. Mol Endocrinol 13:819–828[Free Full Text]
  26. McInerney EM, Weis KE, Sun J, Mosselman S, Katzenellenbogen BS 1998 Transcription activation by the human estrogen receptor subtype ß (ERß) studied with ERß and ER{alpha} receptor chimeras. Endocrinology 139:4513–4522[Abstract/Free Full Text]
  27. Ediger TR, Kraus WL, Weinman EJ, Katzenellenbogen BS 1999 Estrogen receptor regulation of the Na+/H+ exchange regulatory factor. Endocrinology 140:2976–2982[Abstract/Free Full Text]
  28. Lazennec G, Alcorn JL, Katzenellenbogen BS 1999 Adenovirus-mediated delivery of a dominant negative estrogen receptor gene abrogates estrogen-stimulated gene expression and breast cancer cell proliferation. Mol Endocrinol 13:969–980[Abstract/Free Full Text]
  29. Frasor J, Danes JM, Komm B, Chang K, Lyttle CR, Katzenellenbogen BS 2003 Profiling of estrogen up- and down-regulated gene expression in human breast cancer cells: insights into gene networks and pathways underlying estrogenic control of proliferation and cell phenotype. Endocrinology 144:4562–4574[Abstract/Free Full Text]
  30. Jelinsky SA, Harris HA, Brown EL, Flanagan K, Zhang X, Tunkey C, Lai K, Lane MV, Simcoe DK, Evans MJ 2003 Global transcription profiling of estrogen activity: estrogen receptor {alpha} regulates gene expression in the kidney. Endocrinology 144:701–710[Abstract/Free Full Text]
  31. Livak KJ, Schmittgen TD 2001 Analysis of relative gene expression data using real-time quantitative PCR and the 2(-{Delta}{Delta} C(T)) method. Methods 25:402–408[CrossRef][Medline]
  32. Braud VM, Allan DS, O’Callaghan CA, Soderstrom K, D’Andrea A, Ogg GS, Lazetic S, Young NT, Bell JI, Phillips JH, Lanier LL, McMichael AJ 1998 HLA-E binds to natural killer cell receptors CD94/NKG2A, B and C. Nature 391:795–799[CrossRef][Medline]
  33. Lazetic S, Chang C, Houchins JP, Lanier LL, Phillips JH 1996 Human natural killer cell receptors involved in MHC class I recognition are disulfide-linked heterodimers of CD94 and NKG2 subunits. J Immunol 157:4741–4745[Abstract]
  34. Lanier LL, Corliss B, Wu J, Phillips JH 1998 Association of DAP12 with activating CD94/NKG2C NK cell receptors. Immunity 8:693–701[CrossRef][Medline]
  35. Zannettino AC, Buhring HJ, Niutta S, Watt SM, Benton MA, Simmons PJ 1998 The sialomucin CD164 (MGC-24v) is an adhesive glycoprotein expressed by human hematopoietic progenitors and bone marrow stromal cells that serves as a potent negative regulator of hematopoiesis. Blood 92:2613–2628[Abstract/Free Full Text]
  36. Watt SM, Chan JY 2000 CD164—a novel sialomucin on CD34+ cells. Leuk Lymphoma 37:1–25[Medline]
  37. Chaudhary LR, Avioli LV 1994 Dexamethasone regulates IL-1ß and TNF-{alpha}-induced interleukin-8 production in human bone marrow stromal and osteoblast-like cells. Calcif Tissue Int 55:16–20[CrossRef][Medline]
  38. Akoum A, Lawson C, McColl S, Villeneuve M 2001 Ectopic endometrial cells express high concentrations of interleukin (IL)-8 in vivo regardless of the menstrual cycle phase and respond to oestradiol by up-regulating IL-1-induced IL-8 expression in vitro. Mol Hum Reprod 7:859–866[Abstract/Free Full Text]
  39. Nithipatikom K, Isbell MA, Lindholm PF, Kajdacsy-Balla A, Kaul S, Campell WB 2002 Requirement of cyclooxygenase-2 expression and prostaglandins for human prostate cancer cell invasion. Clin Exp Metastasis 19:593–601[CrossRef][Medline]
  40. Pedram A, Razandi M, Aitkenhead M, Hughes CC, Levin ER 2002 Integration of the non-genomic and genomic actions of estrogen. Membrane-initiated signaling by steroid to transcription and cell biology. J Biol Chem 277:50768–50775[Abstract/Free Full Text]
  41. Lee HY, Bae GU, Jung ID, Lee JS, Kim YK, Noh SH, Stracke ML, Park CG, Lee HW, Han JW 2002 Autotaxin promotes motility via G protein-coupled phosphoinositide 3-kinase {gamma} in human melanoma cells. FEBS Lett 515:137–140[CrossRef][Medline]
  42. Jung ID, Lee J, Yun SY, Park CG, Choi WS, Lee HW, Choi OH, Han JW, Lee HY 2002 Cdc42 and Rac1 are necessary for autotaxin-induced tumor cell motility in A2058 melanoma cells. FEBS Lett 532:351–356[CrossRef][Medline]
  43. Umezu-Goto M, Kishi Y, Taira A, Hama K, Dohmae N, Takio K, Yamori T, Mills GB, Inoue K, Aoki J, Arai H 2002 Autotaxin has lysophospholipase D activity leading to tumor cell growth and motility by lysophosphatidic acid production. J Cell Biol 158:227–233[Abstract/Free Full Text]
  44. de Rooij J, Rehmann H, van Triest M, Cool RH, Wittinghofer A, Bos JL 2000 Mechanism of regulation of the Epac family of cAMP-dependent RapGEFs. J Biol Chem 275:20829–20836[Abstract/Free Full Text]
  45. Reedquist KA, Ross E, Koop EA, Wolthuis RM, Zwartkruis FJ, van Kooyk Y, Salmon M, Buckley CD, Bos JL 2000 The small GTPase, Rap1, mediates CD31-induced integrin adhesion. J Cell Biol 148:1151–1158[Abstract/Free Full Text]
  46. Bos JL, De Bruyn K, Enserink J, Kuiperij B, Rangarajan S, Rehmann H, Riedl J, De Rooij J, Van Mansfeld F, Zwartkruis F 2003 The role of Rap1 in integrin-mediated cell adhesion. Biochem Soc Trans 31:83–86[Medline]
  47. Stauffer SR, Coletta CJ, Tedesco R, Nishiguchi G, Carlson K, Sun J, Katzenellenbogen BS, Katzenellenbogen JA 2000 Pyrazole ligands: structure-affinity/activity relationships and estrogen receptor-{alpha}-selective agonists. J Med Chem 43:4934–4947[CrossRef][Medline]
  48. Kraichely DM, Sun J, Katzenellenbogen JA, Katzenellenbogen BS 2000 Conformational changes and coactivator recruitment by novel ligands for estrogen receptor-{alpha} and estrogen receptor-ß: correlations with biological character and distinct differences among SRC coactivator family members. Endocrinology 141:3534–3545[Abstract/Free Full Text]