Activation of Aromatase Expression by Retinoic Acid Receptor-related Orphan Receptor (ROR) α in Breast Cancer Cells

Estrogen is a key regulator of the proliferation and differentiation of breast cancer cells. In addition to the estrogen supply from the ovary, estrogen is produced locally from androgen by aromatase. However, the regulation of aromatase gene expression in breast cancer has not yet been fully clarified. Retinoic acid receptor-related orphan receptor (ROR) α plays an important role in the differentiation of many organs by regulating the transcription of target genes. Because aromatase and RORα are expressed in breast cancer, the effect of RORα on aromatase gene expression was studied. RORα significantly augmented the expression of aromatase mRNA, particularly those containing exon I.4, in MCF7 cells, and aromatase activities in T47D and MCF7 cells. RORα also stimulated the proliferation of these cells. Transient transfection-based reporter gene assays using the promoter at exon I.4 showed that RORα augmented the transcription. A series of truncated mutation studies revealed that RORα activated the transcription through −147 to +14 bp of the promoter I.4. Furthermore, RORα bound to the fragment containing −119 to −107 bp of the promoter in vitro, indicating that this region may contain a novel ROR response element. Chromatin immunoprecipitation assay showed that RORα bound to the region containing this site of the promoter I.4 in MCF7 cells. Moreover, we examined clinical samples and found a correlation between RORα and aromatase expression. These results suggest that RORα directly activates the aromatase expression to accelerate the local production of estrogen, which results in the proliferation of breast cancer cells.

tant enzyme for the progression of estrogen-dependent breast cancer, because it catalyzes the final limiting step in the biosynthesis of estrogens from androgens (1)(2)(3)(4). The levels of aromatase activity and aromatase mRNA are significantly greater in breast cancer and surrounding adipose tissues than in nonneoplastic breast tissues (5)(6)(7)(8)(9)(10). Recently, Suzuki et al. (11) and Miki et al. (12) have reported the expression of aromatase at the mRNA and protein levels in intratumoral stromal and parenchymal cells in breast cancer tissues. This may explain the higher estradiol level in breast cancer tissues than in the areas considered as morphologically normal (13,14). The levels of intratumoral estradiol are not significantly different between premenopausal and postmenopausal breast cancer patients, but the intratumoral estradiol/estrone ratio is significantly higher in postmenopausal than in premenopausal breast cancer patients (15). Although postmenopausal women have low levels of circulating plasma estrogens, the intratumoral production of estrogens in breast cancer tissue itself can lead to high estrogen levels in the tumor (16). Intratumoral aromatase has been considered as a viable clinical target for the treatment of postmenopausal women with ER-positive breast cancer (17).
The human aromatase gene has a multiplex promoter, followed by untranslated exon I and nine common exons (II-X) (Fig. 1). Each exon I is spliced to exon II ( Fig. 1, solid box). The 5Ј region of exon II (Fig. 1, dotted box) is included in aromatase mRNA only when transcription starts with promoter II (PII). The ATG translation start site is located in coding exon II (18 -21). Expression of the aromatase gene is regulated by the tissuespecific activation of the multiplex promoter. In breast cancer tissues and cell lines, exon I arising from promoters (Ps) I.1, I.3, I.4, and I.7, and a 5Ј region of exon II arising from PII have been confirmed (9,(22)(23)(24)(25). Thus, it appears that the prototype estrogen-dependent breast cancer takes advantage of these promoters for aromatase expression. The sum of aromatase mRNA species arising from these promoters markedly increases the aromatase activity in breast cancer cells compared with the normal breast (9,(22)(23)(24)(25)(26).
Aromatase expression is regulated by many factors, including estradiol, glucocorticoid, cyclic AMP, prostaglandin E2, and estrogen-related receptor ␣ (11,25,(27)(28)(29). However, the transcriptional control mechanism and/or the correlation between nuclear receptor expression and aromatase activity in breast cancer cells have remained largely unknown.
Retinoic acid receptor-related orphan receptor (ROR) ␣ is a member of the steroid/thyroid hormone nuclear receptor superfamily. ROR␣ constitutively activates gene transcription by binding as a monomer to specific DNA sequences termed ROR response elements (ROREs), which are composed of a half-core PuGGTCA motif preceded by a 6-bp A/T-rich sequence (30 -32). By alternative splicing, the ROR␣ gene gives rise to four isoforms, ROR␣1, 2, 3, and 4 (30,33). Distinct ROR␣ isoforms share the common DNA-binding and putative ligandbinding domains but differ in the N-terminal domain, which confers the different DNA binding specificities and transcriptional activities among ROR␣ isoforms. Until recently, ROR␣ has been considered as a true orphan receptor that does not acquire ligands (34 -36). However, a recent study has shown the possibility that cholesterol and its metabolite may be its ligand (37). Because cholesterol and its metabolites are abundantly contained in cells, ROR␣ binds to such compounds constitutively and activates transcription of the target genes. Dai et al. (38) has reported that the ROR␣ mRNA is expressed in MCF7 cells. However, the function of ROR␣ in breast cancer cells has been entirely unknown.
In the present study, we examined the expression of ROR␣ isoforms in a series of human breast cancer cell lines. We then investigated the effect of ROR␣ on the aromatase transcription, aromatase activity, and proliferation of breast cancer cells to study the role of ROR␣ in the aromatase-mediated progression of breast cancer.
Cell Culture-The breast cancer cell lines (T47D, MCF7, BT20, and MDA-MB-231) were obtained from the American Type Culture Collection. The cell lines were maintained in the medium recommended by American Type Culture Collection supplemented with 10% fetal bovine serum. The serum was stripped of hormones by constant mixing with 5% (w/v) AG1-X8 resin (Bio-Rad) and powdered charcoal before ultrafiltration. The cells were cultured without phenol red.
Total RNA Extraction from Cell Lines and cDNA Synthesis-The cells were seeded into six-well plates. After 24 h of transfection, RNA was carefully extracted using the RNeasy plus mini kit (Qiagen). First strand cDNA was prepared from total RNA using the Superscript III cDNA synthesis kit (Invitrogen) in accordance with the manufacturer's instructions.
The annealing and extension temperature was 60°C for 1 min. Each measured value was calculated using each standard. A standard dilution series of control DNA was included in every experiment. All of the experiments were repeated at least three times in triplicate, with independent RNA preparations. The results for each sample were normalized by ␤-actin mRNA level as an internal control. Transient Transfection-Breast cancer cells were seeded into plates and maintained in the stripped fetal bovine serum medium for 24 h. The cells were transiently transfected with each vector using Lipofectamine LTX or 2000 (Invitrogen) in accordance with the manufacturer's instructions.
siRNA Study-The siRNA for ROR␣ (Hs_RORA_8; 5Ј-caggtcttgatatcaatggaa-3Ј) was purchased from Qiagen. The cells were transfected with siRNA using HiPerFect transfection reagent (Qiagen). The expression level of ROR␣ was confirmed by quantitative real time RT-PCR.
Aromatase Activity Assay-The cells were seeded into sixwell plates. After 48 h from transfection, the endogenous aromatase activity was measured by the detection of 3 HOH released during the aromatization of [1␤-3 H] androst-4-ene-3,17-dione, as previously described (41). After dissolving the cells in lysis buffer, the protein concentration was determined by the Bradford method (Bio-Rad). The 3 HOH count was normalized using the amount of protein (dpm/h/mg of protein). All of the experiments were repeated at least three times in duplicate.
Cell Proliferation Assay-The cells were seeded into 24-well plates (1 ϫ 10 5 /well). Sixteen hours after transfection, the culture medium was changed with that with or without 10 Ϫ7 M androstenedione (0 h). The cells were then cultured for 120 h. The cultured cells were counted using the CellTiter 96 AQueous One solution cell proliferation assay kit (Promega) according to the manufacturer's instructions. All of the experiments were repeated at least three times in triplicate.
Transient Transfection-based Reporter Gene Assay-The cells were seeded into 24-well plates. A pGL3-Luc reporter (0.2 g) containing aromatase promoter fragments was cotransfected with the expression vector encoding ROR␣1 (0.05 g) or the empty vector into T47D cells using Lipofectamine LTX (Invitrogen). The empty pGL3-Luc reporter was used as a negative control, and then luciferase activity was calculated as fold basal luciferase activity with the pGL3-Luc reporter alone (n ϭ 3; means Ϯ S.E.). Cytomegalovirus-␤-galactosidase plasmid was cotransfected as an internal control. Forty-eight hours after transfection, the cell extracts were analyzed for both luciferase and ␤-galactosidase activities to correct for transfection efficiency as described previously (42). All of the experiments were repeated at least three times in triplicate.
Chromatin Immunoprecipitation (ChIP) Assays-ChIP assays were performed using the EZ ChIP kit (Upstate Biotechnology) according to the manufacturer's instructions with modification. Briefly, MCF7 cells were cultured in 10-cm dishes, to 40 -50% confluency, and FLAG-ROR␣1 was transfected. After 48 h of incubation, the cells were treated with 1% formaldehyde, and the cell lysates were precipitated using the anti-FLAG antibody (Sigma-Aldrich). An aliquot of the cell lysates was used to isolate total input DNA. Amplifications of the immunoprecipitated DNA were performed by PCR. All of the reactions were performed in triplicate. Primers for human CYP19/aromatase PI.4 sequences were: sense, 5Ј-atgaccaaccaagactaagag-3Ј and antisense, 5Ј-cagttggtcacgttctacttgg-3Ј Amplified bands were loaded on 1.5% agarose gel and stained with ethidium bromide.
Preparing Samples from Breast Cancer Specimens-Twenty fresh surgical specimens were obtained from female patients (mean age was 56.4 years) with invasive breast cancer. Total RNAs were prepared using the RNeasy plus mini kit (Qiagen) in accordance with the manufacturer's instructions. Following quantitative real time RT-PCR procedure was same as above. All of the procedures were approved by the ethics committee on human research of Gunma University Graduate School of Medicine and written informed consent was obtained from each patient.
Statistical Analysis-Treatment effects were analyzed by analysis of variance and by post hoc comparisons using Bonferroni's multiple range test. When p Ͻ 0.05, the effect was considered to be significant.

Expression of ROR␣ Isoforms in Breast Cancer
Cells-To examine the expression of ROR␣ in a series of clonal breast cancer cell lines, we performed quantitative real time RT-PCR analysis (Fig. 2, A-D). MCF7 and T47D cells are ER-positive, whereas BT20 and MDA-MB-231 cells are ER-negative. We detected total ROR␣ ( Fig. 2A) and ROR␣ isoform mRNA (Fig.  2, B-D). The expression levels of total ROR␣ in MCF7 were 3-fold higher than those in T47D. The expression patterns of ROR␣ isoforms are different among cell lines. Of note, the expression level of ROR␣1 in T47D was very low (Fig. 2B), and ROR␣4 in MDA-MB-231 was not detected (Fig. 2D).
Expression and Activity Levels of Aromatase in Breast Cancer Cells-To examine the expression of aromatase in breast cancer cells, we performed Western blotting (data not shown) and quantitative real time RT-PCR (Fig. 2E) analysis. Aromatase was expressed at the mRNA and protein levels in all of the cell lines examined. We then examined the aromatase activity (Fig.  2F). The aromatase activity levels were correlated with the aromatase mRNA levels. This is consistent with previous studies showing that the cytochrome P-450 enzyme activities are induced largely by transcriptional activation (43). Furthermore, when the expression patterns of aromatase in these four cell lines is compared with that of ROR␣ ( Fig. 2A), they are quite similar: MCF7 and BT20 cells express higher levels of aromatase and ROR␣ than the other two cell lines. Thus, we predicted the possible interaction between aromatase expression and ROR␣.
ROR␣1 Up-regulates the Expressions of Aromatase Protein and mRNA in MCF7 Cells-The function of ROR␣ in breast cancer cells has been entirely unknown. However, the results shown in Fig. 2 indicate a possible interaction between aromatase expression and ROR␣. Thus, we examined the effect of ROR␣ on the regulation of aromatase expression. After transient transfection of the expression vector encoding ROR␣1 into MCF7 cells, we carried out Western blot analysis using a polyclonal antibody against aromatase. We also confirmed the protein levels of ROR␣ before and after transfection (Fig.  3A). The expression levels of aromatase protein in MCF7 were augmented by overexpressed ROR␣1 (Fig. 3A). In addition, the expression levels of aromatase mRNA in MCF7 were also augmented by ROR␣1 (Fig. 3B). These results suggest that the transcription of the aromatase gene may be regulated by ROR␣1.
ROR␣1 Activated the Aromatase Activity-We examined the effect of ROR␣1 on the endogenous aromatase activity. Endogenous aromatase activity was measured by the detection of 3 HOH released during aromatization of [1␤-3 H] androst-4-ene-3,17-dione, as previously described (41). The aromatase activities in MCF7 and T47D, both of which are ER-positive, were augmented by increasing the amount of ROR␣1 in a dose-dependent manner (Fig. 3, C and  D). The aromatase activities in ER-negative BT20 and MDA-MB-231 cells were also augmented by ROR␣1 in a dose-dependent manner (data not shown).
ROR␣1 Activated the Proliferation of ER-positive Breast Cancer Cells-We then examined the effect of ROR␣1 on the proliferation of breast cancer cells. In the presence of androstenedione, which is converted to estrone by aromatase, the cell proliferation was stimulated by the overexpressed ROR␣1 in ER-positive MCF7 and T47D cells (Fig. 3, E and F). Without androstenedione, the proliferation was not stimulated by ROR␣1. In ER-negative BT20 and MDA-MB-231 cells, the cell

. Expression of ROR␣ isoforms and aromatase in breast cancer cells.
A-D, quantitative real time RT-PCR analysis of ROR␣ mRNA was performed using four different breast cancer cell lines. First strand cDNA was prepared from total RNA (1 g). Quantitative real time RT-PCR was carried out using the pair of primers common to all the ROR␣ isoforms (ROR␣1-4) (A), the pair of primers for ROR␣1 (B), the pair of primers common to the ROR␣2 and 3 (C), and the pair of primers for ROR␣4 (D). The ROR␣ mRNA expression level was normalized to the ␤-actin mRNA expression level (n ϭ 3; means Ϯ S.E.). E and F, the pattern of aromatase activity levels was similar to that of aromatase mRNA expression levels. E, quantitative real time RT-PCR analysis of aromatase mRNA. Quantitative real time RT-PCR was carried out using the pair of primers for the coding region of aromatase. The aromatase mRNA expression level was normalized to the ␤-actin mRNA expression level (n ϭ 3; means Ϯ S.E.). F, aromatase activity levels of breast cancer cell lines. The 3 HOH count was normalized to the amount of protein (n ϭ 3; means Ϯ S.E.). proliferation was not stimulated by ROR␣1 in the presence of androstenedione (data not shown). These results indicate that the expression of aromatase is augmented by ROR␣1, which may accelerate the conversion of androstenedione to estrone, leading the stimulation of the proliferation of ER-positive breast cancer cells.

Evaluation of Induction Mechanism of Aromatase Expression by ROR␣1-
The expression of the aromatase gene is regulated in a tissuespecific manner by the multiplex promoter (Fig. 1). In a series of breast cancer cell lines, exon I arising from PI.1, PI.3, PI.4, PI.7, and PII has been detected (9,22,23). To examine the mechanism of the regulation of aromatase transcription by ROR␣, we examined which promoter was activated by ROR␣. Using quantitative real time RT-PCR studies, we measured the levels of each nontranslated exon I of aromatase mRNA with or without transfection of ROR␣1 (Fig. 4,  A-D). Only exon I.4 mRNA was upregulated by ROR␣1. This up-regulation was observed 12-48 h after transfection. The mRNAs of exons I.1, I.3, and I.7 were detected in MCF7 cells, but a significant change was not observed. The 5Ј portion of exon II arising from PII was not detected in MCF7. Previously, it was reported that the aromatase promoter I.4 contains the glucocorticoid response element and is regulated by glucocorticoid (27). Thus, we examined whether ROR␣ up-regulates the glucocorticoid receptor (GR) transcription, which then activates aromatase promoter I.4. However, no significant up-regulation of GR mRNA by ROR␣1 was observed (Fig. 4E).
To confirm further the augmentation of aromatase expression by ROR␣, we performed siRNA study. The siRNA against ROR␣ was transfected into MCF7 cells. After 120 h of incubation, we carried out quantitative real time RT-PCR using the pair of primers for ROR␣ and exon I.4. The decrease of ϳ60% in the expression level of ROR␣ mRNA was confirmed by quantitative real time RT-PCR studies (data not shown). Repression of ROR␣ resulted in the decrease in the expression level of exon I.4 mRNA (Fig. 4F). These results indicate that aromatase expression is regulated by ROR␣ through promoter I.4. Thus, a novel RORE was presumed.
Identification of RORE in Aromatase Promoter I.4-To examine whether ROR␣ regulates the aromatase transcription through PI.4, transient transfection-based reporter gene assays  05 g). B, ROR␣1 activated the aromatase mRNA expression in MCF7 cells. The expression vector encoding ROR␣1 (0.5 g) was transfected into MCF7 cells. The cells were then incubated for 24 h. Quantitative real time RT-PCR was carried out using the pair of primers for the coding region of aromatase. The mRNA expression level was normalized to the ␤-actin mRNA expression level (n ϭ 3; means Ϯ S.E.; *, statistically significant, p Ͻ 0.05). C and D, ROR␣1 stimulated the endogenous aromatase activity in T47D (C) and MCF7 (D) cells. The expression vector encoding ROR␣1 was transfected in increasing amounts (0, 0.005, 0.05, and 0.5 g/well of a six-well plate) into the cells. The 3 HOH count was normalized using the amount of protein (n ϭ 2; means Ϯ S.E.; *, statistically significant, p Ͻ 0.005). E and F, ROR␣1 stimulated the proliferation of T47D (E) and MCF7 (F) cells. The cells were cultured in a 24-well plate and transfected with the expression vector encoding ROR␣1 (0.1 g). After 16 h from transfection, the culture medium was changed (0 h) with that with or without 10 Ϫ7 M androstenedione, and the cells were then cultured for 120 h (n ϭ 3; means Ϯ S.E.; *, statistically significant, p Ͻ 0.05). DMSO, dimethyl sulfoxide.
were carried out in T47D cells (Fig. 5). As predicted, ROR␣ activated the transcription of aromatase PI.4 (Ϫ1004/ϩ14) ϳ10-fold compared with the basal level (second column). To evaluate the ROR␣ target region, 5Ј-deletional and mutational constructs of the aromatase PI.4 fused to a luciferase expression reporter gene were transfected. Each construct was cotransfected with or without the ROR␣1 expression vector. The ROR␣ activated the transcription through the regions containing Ϫ1004/ϩ14, Ϫ458/ϩ14, and Ϫ147/ϩ14 (second, third, and fifth columns) but not Ϫ458/Ϫ126 and Ϫ106/ϩ14 (fourth and sixth columns) in T47D cells. These results indicate that the sequence between Ϫ147 and Ϫ106 bp of PI.4 is required for the response of ROR␣1 in T47D cells.
Data base analysis of the aromatase gene sequence revealed the existence of consensus RORE, a 6-bp A/T-rich region fol-lowed by the AGGTCA sequence, at the Ϫ119/Ϫ107-bp region of PI.4. To investigate the role of this sequence in directing the basal promoter activity by ROR␣, the A/Trich sequence (aagattg) at Ϫ119/ Ϫ113-bp, which is located at the 5Ј region of the half-site core motif, was replaced with the G/C-rich sequence (gagctcg). This mutation abolished the promoter activity of the aromatase PI.4 by ROR␣1 in T47D cells (Fig. 6A).
To confirm the interaction of ROR␣1 with the PI.4 Ϫ119/ Ϫ107-bp region, DNA mobility shift and antibody supershift experiments were carried out using in vitro-synthesized ROR␣1 protein and double-strand oligonucleotide containing the Ϫ119/Ϫ107-bp region (Fig. 6B). The ROR␣1 bound to the Ϫ121/Ϫ104-bp probe (Fig.  6B, lane 2). The band was attenuated by the addition of cold probe (lane 3). The antibody supershift using the anti-ROR␣ antibody was observed (lane 4). Moreover, the ROR␣1 did not bind to the mutated DNA probe (lane 6). Together with functional studies, these results indicate that the PI.4 Ϫ119/ Ϫ107-bp of the aromatase gene functions as a novel ROR response element.
ROR␣1 Binds to Aromatase PI.4 in MCF7 Cells-To confirm the binding of ROR␣1 to the aromatase PI.4 in MCF7 cells, ChIP assays were carried out using FLAG-ROR␣1 and anti-FLAG antibody. Anti-RNA polymerase II antibody and normal mouse IgG were used as controls (Fig. 7). ROR␣1 accumulation at the promoter was observed accompanied by an equivalent increase of RNA polymerase II (fourth and sixth lanes), whereas in the absence of transfected ROR␣1, no amplification of DNA fragment was observed (fifth lane). On the other hand, no enrichment of DNA fragments was observed when the immunoprecipitations were carried out using mouse IgG (third lane). These results indicate that ROR␣1 binds to the aromatase PI.4 and a substantial increase in RNA polymerase II as well. These findings indicate that aromatase exon I.4 induction mediated by the ROR␣1 by binding to the aromatase PI.4 likely precedes changes in mRNA output.
The Relationship between ROR␣ and Aromatase Expression in Human Breast Cancer-To investigate the relationship between the expression of ROR␣ and aromatase, 20 breast cancer samples from female breast cancer patients were examined ( Fig. 8). We carried out the quantitative real time RT-PCR studies. There is a correlation between ROR␣ and aromatase mRNA expression levels (correlation coefficient: r ϭ 0.512, p ϭ 0.019).

DISCUSSION
In the present study, we have shown the augmentation of aromatase activity in breast cancer cells by ROR␣, which results in cell proliferation. Then we confirmed further that ROR␣ directly activates the transcription. We have identified a novel RORE in the promoter region of aromatase exon I.4, using the combination of site-directed mutagenesis and in vitro protein-DNA binding assays. We also confirmed the binding using ChIP assay. Finally, we showed that there is a correlation between ROR␣ and aromatase expression in human breast cancer tissues. Initially, we found a possible correlation of expression pattern between aromatase and ROR␣ mRNA in four breast cancer-derived cell lines (Fig. 2). MCF7 and BT20 cells expressed higher levels of both aromatase and ROR␣ mRNA than the other two cells. These results have made us hypothesize a possible interaction between aromatase activity and ROR␣. Because ROR␣ belongs to a steroid/thyroid hormone receptor superfamily, a ligand-dependent transcription factor, it was likely that ROR␣ regulates the transcription of the aromatase gene. The cytochrome P-450 enzyme activities including that of aromatase are induced largely by transcriptional activation (43), which is also confirmed in the present study (Fig. 2, E and  F). To test the hypothesis that ROR␣ is involved in the progression of breast cancer by activating aromatase activity, we further conducted a series of experiments. We then confirmed that the aromatase mRNA and protein levels were augmented by ROR␣1 (Fig. 3, A and B). Aromatase activities were also augmented with the increase in the dose of ROR␣1 in two ER-positive cell lines (Fig. 3, C and D). The proliferation of breast cancer cells that were transfected with ROR␣ was greater than those without transfection (Fig. 3, E and F). Such cell proliferation was observed only in the presence of androstenedione, which is a major substrate of aromatase and is converted to estrone. These results are consistent with our initial hypothesis that ROR␣ activates aromatase activity to increase the local concentration of estrogen to activate proliferation.
Next, we investigated the mechanism of the augmentation of aromatase transcription by ROR␣. We sought the promoter responsible for the augmentation and identified that the exon I.4 mRNA level is increased by ROR␣ (Fig. 4C). Because the expression of GR was not affected by ROR, the possibility that ROR␣ induces GR, which then activates aromatase, is excluded (Fig. 4E). We also confirmed that the exon I.4 mRNA expression level was decreased markedly by suppressing ROR␣ expression using siRNA (Fig. 4F). Thus, we hypothesized that ROR␣ may act at PI.4 to regulate transcription. The deletion of the upstream region of PI.4 revealed that the region responsible for the ROR␣-mediated transcriptional regulation is located in Ϫ147/Ϫ106 bp from the transcriptional start site (Fig. 5). Data search of this region revealed a presumed RORE at Ϫ119/Ϫ107 bp. Mutation of this region abolished the effect of ROR␣ in transient transfection-based reporter gene assay (Fig. 6A). Furthermore, electrophoretic mobility shift assay showed that ROR␣1 binds to the fragment containing Ϫ119/Ϫ107 bp, but not to the mutated RORE (Fig. 6B). These results indicate that Ϫ119/Ϫ107 bp is a novel RORE. Finally, we confirmed the binding of ROR␣1 to the intrinsic chromatin that contains the aromatase PI.4 using ChIP assay in MCF7 cells (Fig. 7). Thus, we concluded that ROR␣ directly activated the aromatase expression through a novel RORE in promoter I. 4.
To confirm the relationship of ROR␣ to aromatase expression in vivo, we carried out the quantitative real time RT-PCR studies using 20 surgically obtained samples from female breast cancer patients (Fig. 8). As stated above, there was a correlation between ROR␣ and aromatase mRNA levels in clinical samples, which is also consistent with the hypothesis that ROR␣ may regulate aromatase in breast cancer cells.
The biology of breast cancer is very complex, but there is no doubt that estrogen plays a central role (44). About 70% of breast cancers are ER␣-positive upon initial diagnosis, and in the majority of those cancers, the ER␣ status serves as a valuable predictive marker for probable response to anti-estrogen ther-FIGURE 5. The region responsible for ROR␣ action is located in ؊147/؊106 of the aromatase I.4 promoter. The truncated mutated promoter-containing constructs are shown on the left. The individual PI.4-pGL3-Luc (0.2 g) reporter was cotransfected with the expression vector encoding ROR␣1 (0.05 g) or the empty vector into T47D cells. The empty pGL3-Luc reporter was used as a negative control. The total amounts of DNA for each well were balanced by adding the vector pcDNA3. Luciferase activity was normalized to ␤-galactosidase activity and then calculated as fold-basal luciferase activity with the pGL3-Luc reporter alone (n ϭ 3; mean Ϯ S.E.).
apy (44,45). On the other hand, to date, the expression of ROR␣ in breast cancer cells has not been considered as a critical factor for prognosis. However, as shown in the present study, ROR␣ can stimulate the proliferation of breast cancer cells. Thus, during the pathological diagnosis of breast cancer, in addition to the examination of ER and progesterone receptor, histochemical or quantitative examination of ROR␣ could be useful to clarify the nature of breast cancers particularly in post-menopausal women. Trials to examine ROR␣ expression in surgically removed breast tumor tissues are currently underway.
In conclusion, we have shown that ROR␣ activates the aromatase I.4 promoter through a novel RORE. Thus, ROR␣ may be useful for the diagnostic classification or prognosis prediction of breast cancers.