Estrogen Receptor-Alpha (ER-Α)

Home , Estrogen receptor alpha

CORE Metadata, citation and similar papers at core.ac.uk

Provided by Elsevier - Publisher Connector

FEBS Letters 583 (2009) 1368–1374

journal homepage: www.FEBSLetters.org

Estrogen receptor-alpha (ER-a) suppresses expression of its variant ER-a36

Yi Zou 1, Ling Ding, Megan Coleman, Zhaoyi Wang *

Department of Medical Microbiology and Immunology, Creighton University School of Medicine, Omaha, NE 68178, USA

article info abstract

Article history: Recently, we identified and cloned a membrane-based estrogen receptor (ER), ER-a36, which is tran- Received 16 January 2009 scribed from a previously unidentified promoter in the first intron of the original ER-a gene. We Revised 10 March 2009 cloned the 50 flanking region of ER-a36 and found the cloned DNA sequence possessed strong pro- Accepted 19 March 2009 moter activity. A single transcription initiation site was mapped and a number of putative regula- Available online 26 March 2009 tory elements for various transcription factors such as the Wilms’ tumor suppressor, WT1 and Edited by Robert Barouki estrogen receptor (ER) were identified in this region. Transient co-transfection experiments further demonstrated that ER-a suppresses ER-a36 promoter activity in an estrogen-independent manner, which can be released by ER- 36 itself. Keywords: a Promoter Ó 2009 Published by Elsevier B.V. on behalf of the Federation of European Biochemical Societies. Estrogen receptor ER-a36 Breast cancer

1. Introduction unidentified promoter located in the first intron of the original ER-a (ER-a66) gene [12,13]. This novel isoform of ER-a, ER-a36 Diverse functions of estrogens are mediated by estrogen recep- lacks both transcriptional activation domains of ER-a (AF1 and tors: ER-a and ER-b. Both ER-a and ER-b function as ligand-depen- AF2), but retains the DNA-binding domain and partial dimerization dent transcription factors and consist of an N-terminal activation and ligand- binding domains. ER-a36 is predominantly expressed function 1 (AF1) domain, a DNA-binding domain, and a C-terminal on the cell surface and mediates the membrane-initiated estrogen ligand-binding domain that harbors an activation function 2 (AF2) signaling [13], indicating that ER-a36 is a membrane-based estro- domain [1–4]. The liganded ERs readily form homodimers or het- gen receptor and a novel player in mitogenic estrogen signaling. erodimers that interact with the palindromic estrogen response- Our recent study also revealed that ERa36 is not only expressed element (ERE) in the promoter regions of estrogen responsive in ER-positive breast cancer but also expressed in ER-negative genes and stimulate gene transcription [1–4]. Alternatively, ER-a breast cancer that lacks expression of ER-a66 [14], further indicat- may act indirectly by tethering to other transcription factors, such ing that ER-a36 is subjected to a totally different transcriptional as Sp1 and AP1 to modulate activities of these transcription factors, regulation from the original ER-a. As an important mediator of which in turn regulates downstream gene expression [1–5]. estrogen signaling, ER-a36 must be regulated dynamically and For the past 30 years, accumulating evidence demonstrates a ra- strictly to maintain a normal estrogen signaling. Dysregulation of pid (within seconds or minutes) estrogen action that cannot be ex- its expression may be involved in development of a number of hu- plained by the genomic signaling pathway that usually requires man diseases including breast cancer. hours to reach maximal gene activation [6–11]. This pathway is In order to elucidate the molecular mechanisms underlying reg- also called the ‘‘non-classical”, ‘‘extra-nuclear”, ‘‘non-genomic” or ulation of this biologically important gene, we isolated and cloned ‘‘membrane-associated” estrogen signaling pathway [8,9]. How- the 50-flanking region of ER-a36, and characterized the general fea- ever, the identity of the membrane-based estrogen receptor that tures of this region. mediates these rapid estrogen effects has not been well established. Recently, we identified and cloned a variant of ER-a with a 2. Materials and methods molecular weight of 36 kDa that is transcribed from a previously 2.1. Cell culture and luciferase reporter assay

* Corresponding author. Fax: +1 402 280 3543. MCF10A cells were obtained from Karmanos Cancer Institute, E-mail address: [email protected] (Z. Wang). Detroit, MI. Human embryonic kidney (HEK) 293 cells, human 1 Present address: Department of Pathology and Microbiology, University of Nebraska Medical School, 987810 Nebraska Medical Center, Omaha, NE 68198, USA. breast cancer cells, MCF7, T47D, MDA-MB-231, MDA-MB-436,

0014-5793/$36.00 Ó 2009 Published by Elsevier B.V. on behalf of the Federation of European Biochemical Societies. doi:10.1016/j.febslet.2009.03.047 Y. Zou et al. / FEBS Letters 583 (2009) 1368–1374 1369

HB3396 and human lung cancer cells H226 were obtained from Table 2 0 American Type Culture Collection (ATCC). All cells were main- Oligonucleotides used in the 5 RACE PCR. tained at 37 °C in a 5% CO2 atmosphere in appropriate tissue cul- AP1 (S) 50 CCATCCTAATACGACTCACTATAGGGC 30 ture medium. AP2 (S) 50 ACTCACTATAGGGCTCGAGCGGC 30 0 0 Expression vectors, pCR3.1-ER-a66 and pCR3.1-ER-a46 were GSP1 (AS) 5 CCTTGCAGCCCTCACAGGACCAGACTCC 3 GSP2 (AS) 50 TCTCCTTGGCAGATTCCATAGCC 30 obtained from Dr. Zafar Nawaz at the University of Miami Sylvester Comprehensive Cancer. The expression vectors pCR3.1-ER-a36 was (S): sense; (AS): anti-sense. API (S), GSP1 (AS) and AP2 (S), GSP2 (AS) were two sets constructed as described before [12]. Amounts of the expression of primers used for the 50 RACE. vectors used in each transfection assay were normalized by the empty expression vector pCR3.1. The luciferase assays were performed using the Dual Luciferase Assay kit from Promega accord- cific primer 2 (GSP2) and nested primer AP2 from the kit. The PCR ing to the manufacturer’s recommendation. products from the 50 RACE were cloned into the TA cloning pCR2.1 vector and examined by DNA sequencing. 2.2. Molecular cloning of the 50-flanking region of ER-a36 2.5. RNA extraction, RT-PCR and semi-quantitative analysis According to computer analysis, ER-a36 mRNA is initiated from a promoter located in the first intron of ER-a genomic DNA. Using a Total RNA was isolated with Trizol (Invitrogen) from cells human placenta genomic DNA library (Clontech Laboratories Inc.) according to the manufacture’s instructions. RNA integrity was as a template, the 50-flanking region of ER-a36 was amplified by examined by gel-electrophoresis with 1% formaldehyde gel. a genomic PCR kit (Clontech Laboratories Inc.), using a primer set 500 ng total RNA was reverse transcribed with Protoscript II RT- containing XhoI or HindIII restriction sites at the ends, respec- PCR kit (New England BioLabs). PCR was performed in a final vol- tively: forward primer 50-CCCTCGAGGGTACCCGCGCCCGC-30 and ume of 50 ll containing 2 ll RT transcript, 0.2 lM of each primer, reverse primer 50-CCAAGCTTAAAAATTCGTGAACACGAAGG-30. The 25 ll Taq 2X Master Mix. The following primers were used: ER-a66 amplified 751 bp DNA product was cloned into a TA cloning forward primer: 50-CACTCAACAGCGTGTCTCCGA-30; ER-a66 re- pCR2.1 vector (Invitrogen) and verified by DNA sequencing on both verse primer: 50-CCAATCTTTCTCTGCCACCCTG-30; ER-a36 forward strands. The putative transcription factor binding sites in the 50 primer 50-CACTCAACAGCGTGTCTCCGA-30; ER-a36 reverse primer flanking region of ER-a36 gene were analyzed using the AliBaba 50-CCAATCTTTCTCTGCCACCCTG-30; b-actin forward primer 50- 2.1 promoter analysis software. TGACGGGGTCACCCACACTGTGCCCATCTA-30; b-actin reverse primer 50-CTAGAAGCATTTGCGGTGGACGATGGAGGG-30. Samples 2.3. Construction of the 50 deletion mutants of ER-a36 promoter were amplified: ER-a36 35 cycles, ER-a66 35 cycles and b-actin, 25 cycles. PCR products (20 ll) were electrophoresed on 1.2% aga- The DNA fragment generated by genomic PCR was cloned up- rose gel and visualized under ultraviolet light after ethidium bro- stream of a luciferase reporter gene in a promoter-less pGL2-basic mide staining. Fragment sizes were 286 bp for ER-a36, 200 bp for luciferase reporter vector (Promega Co.). The resulted plasmid was ER-a66 and 661 bp for b-actin. The band densities of ER-a36, named as pGL2-ER36. The 50-deletion mutants of ER-a36 promoter ER-a66 and b-actin were determined with a video-densitometry were produced by PCR method from the pGL2-ER36 plasmid. Prim- system (Molecular Imager ChemiDoc XRS System, Bio-Rad). Subse- ers were designed with HindIII and XhoI sites at the ends of the quently, the ER-a36:b-actin ratio and ER-a66:b-actin ratio (b-actin products (Table 1). The PCR products were purified and cloned into mRNAs were detected in the same RT sample) were calculated the HindIII and XhoI sites upstream of a luciferase reporter gene in with the Quantity One software. the pGL2-basic vector (Promega Co.). All plasmids were then verified by restriction enzyme digestion and DNA sequencing. The se- 2.6. Statistical analysis ven deletion mutants were named pGL2-ER36 [1–7]. Data were summarized as the mean ± standard error (S.E.) using 2.4. Rapid amplification of 50 cDNA end (50 RACE) GraphPad InStat software program. The student t-test was also used, and the significance was accepted for P-values of less than The transcription initiation site in the 50 flanking sequence of 0.05. ER-a36 was identified with the 50 RACE method. Human placenta ‘‘marathon-ready” cDNA from Clontech Inc. was used to perform 3. Results 50 RACE according to the manufacturer’s instruction. An ER-a36 gene-specific primer 1 (GSP1) was designed (Table 2) for the 3.1. Genomic cloning and sequence analysis of the 50 flanking region of touch-down PCR, and paired with an adaptor primer 1 (AP1) pro- ER-a36 gene vided by the kit. Secondary PCR was performed with the gene spe- ER-a36 is a membrane-based estrogen receptor that mediates non-genomic estrogen signaling and stimulates cell growth [13]. Table 1 ER-a36 is expressed in specimens from ER-negative breast cancer Primers used to generate 50 deletion constructs of human ER-a36 promoter. patients that lack expression of ER-a66 [14], suggesting that its 50P36-1 S 50-CCCTCGAGCCGCGGGACGCGCGACCCGA-30 expression is regulated independently and differently from the 50P36-2 S 50-CCCTCGAGCCCGGGGGTTCTGCGTGCAGC-30 ER-a66 gene. However, the molecular mechanisms by which 50P36-3 S 50-CCCTCGAGTACTCTCCACCCACTTACCGTC-30 ER-a36 gene transcription is regulated are unknown. We decided 0 0 0 5 P36-4 S 5 -CCCTCGAGGCAAATAAACACGGGGCGCTTTG-3 to clone the promoter region of ER-a36 and to study its transcrip- 50P36-5 S 50-CCCTCGAGGAAGGTCTCGCTCTTGGCATTT-30 tional regulation. To this aim, we designed primers corresponding 50P36-6 S 50-CCCTCGAGTTTTATTTATCCTTTTAATGTTTTTGTTTAAT-30 0 50P36-7 S 50-CCCTCGAGCCTTTTAATGTTTTTGTTTAATGTGCTCCCC-30 to the 5 flanking sequence upstream of the ER-a36 cDNA accord- 30P36-1 AS 50-CCAAGCTTGTGCTTCCATATGGCTTGTTTTAAAC-30 ing to the published genomic sequence of the ER-a66 gene (ESR1, Gene bank AY425004). A genomic PCR assay was performed using All the 50 deletion mutants were generated with the common anti-sense primer (AS) but different sense primers (S). Restriction enzyme sites incorporated for easy human genomic DNA from placenta. An expected 751 bp PCR prod- cloning were underlined. uct was cloned into a TA cloning vector (pCR2.1), and completely 1370 Y. Zou et al. / FEBS Letters 583 (2009) 1368–1374 sequenced in both directions. Sequence analysis revealed that the the front of the firefly luciferase gene of the pGL2-basic vector to 50 flanking region contains a high G + C content, a non-canonical generate the plasmid pGL2-ER36. The plasmid pGL2-ER36 was TATA box, but lacks the CCAAT box. A number of SP1 sites and transiently transfected into HEK293 cells and the promoter activity AP1 sites were identified in the 50 flanking region of ER-a36 was assessed by measuring luciferase activity in the transfected (Fig. 1). The region also contains putative binding sites for the fol- HEK293 cells. As shown in Fig. 2, pGL2-ER36 exhibited about 14- lowing transcription factors: AhR (aryl hydrocarbon receptor), Erg- fold higher promoter activity compared to the empty vector that 1 (ETS-related gene 1), NF-jB (nuclear factor kappa B), WT1 (the produced barely detectable luciferase activity. The basal activity Wilms’ tumor suppressor), PU.1, GATA-1, Elk-1 and GR (glucocorti- of the ER-a36 promoter is about half of that of the SV-40 pro- coid receptor). This region lacks consensus palindromic ERE site moter/enhancer activity in HEK293 cells (Fig. 2). (50-GGTCAnnnTGACC-30); however, it contains an imperfect ERE half site (50-AGTCA-30, located at 369 to 365, relative to the 3.3. Deletion analysis of the ER-a36 promoter transcription initiation site) that is overlapped with an AP-1 binding site (Fig. 1). To further characterize the 50 flanking region of the ER-a36 gene, we constructed a set of seven reporter vectors, containing 3.2. Transcription initiation site of ER-a36 gene and promoter activity successive 50 deletion of the ER-a36 promoter (Fig. 3). These deletion mutants were then transfected into three cell lines: HEK293, To map the transcription start site of ER-a36, we used human ER-negative breast cancer MDA-MB-231 cells and ER-positive placenta ‘‘marathon-ready” cDNA as templates to perform 50 RACE breast cancer MCF7 cells, and the luciferase activities of these con- assay and cloned PCR products into the TA cloning pCR2.1 vector. structs were then measured. In HEK293 cells, a significant increase The DNA clones were sequenced from both directions. We found of promoter activity was observed with the removal of 50 sequence all DNA clones contained sequences matched exactly to the se- up to nucleotide 584 bp (relative to the transcription initiation quences of the ER-a66 genomic DNA. Two clones contained identi- site), indicating the presence of negative regulatory element(s) in cal sequences corresponding to the 50 end of the ER-a36 cDNA. the region from 736 bp to 584 bp in HEK 293 cells. Further trun- After comparing these two DNA sequences with the 50 flanking se- cation to 513 bp returned the promoter activity to the full-length quence of ER-a36 gene, we mapped the first nucleotide of the tran- promoter activity, suggesting the presence of positive regulatory scription initiation at the nucleotide A (+1) 238 bp upstream of the element(s) in the region 584 bp to 513 bp. In ER-positive breast translation initiation site (Fig. 1). cancer MCF7 cells, however, a decrease of promoter activity was To determine that the 50 flanking region of ER-a36 gene con- observed in the deletion of the DNA sequence 736 bp to tains promoter activity, we placed the 751 bp DNA fragment in 675 bp, indicating the presence of enhancer element(s) in this

Fig. 1. Nucleotide sequence of the 50 ﬂanking region of ER-a36 gene and putative regulatory elements. Potential binding sites of AP-1, Sp1, AhR, NF-jB, and WT1, and an ERE half site, are underlined. The putative TATA box is shown as an open box. Curved arrow indicates the transcription start site mapped by the 50 RACE. The numbers shown at the left are relative to the transcription start site of ER-a36. Y. Zou et al. / FEBS Letters 583 (2009) 1368–1374 1371

promoter activity (Fig. 4). These results suggested that ER-a66 suppresses ER-a36 promoter activity in an estrogen-independent manner presumably through the AF1 domain, which can be released by co-expression of either ER-a46 or ER-a36 itself both of which lack the AF1 domain. We also examined expression patterns of ER-a66 and 36 in established breast cancer cell lines with RT-PCR analysis. ER-a36 was not expressed in normal mammary epithelial cells MCF10A, weakly expressed in ER-positive breast cancer cells such as MCF7 and T47D both of which express high levels of ER-a66. ER-a36 was highly expressed in ER-negative breast cancer cells such as MDA-MB-231 cells that lack ER-a66 expression (Fig. 5). However, the ER-negative breast cancer MDA-MB-436 cells express low levels of ER-a36. ER-a36 was also highly expressed in the ER-positive breast cancer cells H3396 that have high levels of ER-a46 expression (Fig. 5, data not shown). ER-a36 was also detected in lung cancer H226 cells that lack ER-a66 expression (Fig. 5). The different expression patterns of ER-a66 and 36 in these cancer cells further indicated that ER-a36 is transcriptionally regulated differently from ER-a66.

4. Discussion

In this study, we have isolated and cloned a DNA fragment containing the promoter region of ER-a36 gene that is located in the first intron of the human ER-a66 gene. Functional analysis demonstrated that this DNA sequence exhibited strong promoter activity, indicating that this region contains ER-a36 promoter. We also further determined a single transcription start site at the nucleotide A 238 bp upstream of the translation initiation site. Sequence analysis revealed that the ER-a36 promoter region contains a high G + C content, a non-canonical TATA box and multiple Sp1 binding sites. Fig. 2. The promoter activity in the 50 flanking region of ER-a36. The pGL2-ER36, Deletion of these Sp1 sites and the non-canonical TATA box re- the promoter-less pGL2 plasmid and the positive pSV-Luc were transfected into sulted in significantly decreased promoter activity, suggesting HEK293 cells, respectively. The luciferase reporter activity was measured. Relative these DNA sequences are critical for the optimum ER-a36 pro- luciferase activities are shown with the luciferase activity in pGL2 empty vector transfected cells was set as 1. Columns: mean of more than three independent moter activity. experiments; bars, S.E. *P < 0.05 vs. the promoter-less pGL2 plasmid transfected To further analyze ER-a36 promoter, we constructed a series of cells. 50 truncations of the ER-a36 promoter region and tested promoter activities of these truncated promoters with transient transfection in three cell lines: HEK293 cells, ER-positive breast cancer MCF7 region. The promoter activity was without significant changes in cells and ER-negative breast cancer MDA-MB-231 cells. HEK293 the remaining deletion constructs. In ER-negative breast cancer cells express undetectable levels of both ER-a66 and 36; MCF7 MDA-MB-231 cells, there are no significant changes of the pro- cells weakly express ER-a36 but highly express ER-a66 while ER- moter activities of all deletion mutants. In three cell lines, the pro- negative breast cancer MDA-MB-231 cells only express ER-a36. moter activity was dramatically reduced after the putative TATA These cell lines presumably provide different genetic background box was deleted, indicating its essential role in the optimal pro- to test ER-a36 promoter activity. In HEK293 cells, the highest tran- moter activity of ER-a36 gene. Taken together, these results dem- scriptional activity was observed for the promoter deletion 2 (re- onstrated that both negative and positive regulatory regions moval of DNA sequence from 675 bp to 584 bp), suggesting controlled the basal promoter activity of ER-a36 gene depending existence of negative element(s) in this region. We observed a Wil- on cell context. ms’ tumor suppressor WT1 binding site within this region. In pre- liminary study, we found that WT1 expression inhibited ER-a36 3.4. Suppression of the ER-a36 promoter activity by ER-a66 promoter activity through this binding site (unpublished data), suggesting that deletion of the WT1 binding site may release Since there is one imperfect ERE half site (369 bp to 365 bp) WT1 suppression activity in HEK293 cells that highly express the in the ER-a36 promoter, we decided to test whether ER-a66 regu- transcription suppressor WT1. It is worth noting that loss of WT1 lates ER-a36 promoter activity using HEK293 cells that express no expression was implicated in development of breast cancer [15]. detectable level of endogenous ER-a. The results from transient co- Aberrant methylation of WT1 promoter was also found in breast transfection indicated that co-transfection of ER-a66 expression cancer [16,17]. Our laboratory previously demonstrated that vector with pGL-ER36 reporter plasmid resulted in about 2-fold over-expression of recombinant WT1 in MDA-MB-231 cells inhib- repression of ER-a36 promoter activity (Fig. 4) and this suppres- its malignant growth of breast cancer cells [18]. Thus, the finding sion was not changed in the presence of 17b-estrodial (data not here that deletion of the WT1 binding site increased ER-a36 pro- shown). However, the combination of ER-a66 and ER-a36 or ER- moter activity suggested that WT1 is involved in negative regula- a46, another ER-a variant that lacks the AF-1 domain [23] released tion of ER-a36 expression and aberrant WT1 expression may the suppression activity of ER-a66 while co-transfection with ER- lead to the increased ER-a36 expression observed in breast cancer. a36 or ER-a46 expression vector alone has no effect on ER-a36 We also noticed that several deletion mutants of the ER-a36 flank- 1372 Y. Zou et al. / FEBS Letters 583 (2009) 1368–1374

Fig. 3. Promoter activities of serial deletion mutants of ER-a36 promoter in HEK293, MDA-MB-231, MCF7 cells. The schematic diagram of the full-length and 50 serial truncation constructs of ER-a36 promoter (left panel) and their corresponding luciferase activities in different cell types are shown. Positions of the deletions are indicated and the curved arrow indicates the transcription start site. P0: the full-length promoter in the pGL2-ER36 plasmid; P1–7: seven deletion mutants of ER-a36 promoter, pGL2- ER36 (1–7). Relative luciferase activities are shown with the luciferase activities in pGL2-ER36 vector transfected cells were set as 1. Columns: mean of more than three independent experiments; bars, S.E. *P < 0.01 vs. the full-length promoter, the pGL2-ER36 plasmid transfected MCF7 cells; NP < 0.01 vs. the pGL2-ER36 plasmid transfected MDA-MB-231 cells; jP < 0.01 vs. the pGL2-ER36 plasmid transfected HEK293 cells.

Fig. 4. Regulation of ER-a36 promoter activity by ER-a isoforms. HEK293 cells were transiently co-transfected with the pGL2-ER36 luciferase reporter plasmid alone or with the expression vectors of ER-a66, ER-a46 or ER-a36, respectively, or in different combinations. Relative luciferase activities are shown with the luciferase activity in pGL2- ER36 vector transfected cells was set as 1. Columns: mean of more than three independent experiments; bars, S.E. *P < 0.05 vs. the empty vector transfected cells. **P < 0.05 vs. the pCR3.1-ER-a66 transfected cells.

ing region exhibited higher promoter activity in HEK293 cells com- itively regulatory elements. In MDA-MB-231 cells, most deletion pared to MCF7 and MDA-MB-231 cells although HEK293 cells ex- mutants display similar promoter activity. press undetectable levels of ER-a36 expression [13], suggesting It is well established that ER-a binds to the palindromic estro- that other mechanisms such as DNA methylation may occur in gen response-element (ERE) in the target genes and regulates gene the G + C rich promoter region of ER-a36 and contribute to the reg- transcription [4]. It was also reported that ER-a regulates promot- ulation of ER-a36 expression. ers containing ERE half sites [19]. Other studies suggested that ER- In MCF7 cells, the full-length promoter exhibited highest pro- a binds to a single ERE half-site closely located with Sp1 binding moter activity; deletion of the DNA sequence from 736 bp to sites in the presence of Sp1 binding in the promoters of certain 675 bp that harbors several Sp1 binding sites dramatically re- estrogen regulated genes such as hsp27 [20], TGF-alpha [21] and duced the promoter activity, suggesting these Sp1 sites act as pos- PR [22]. ER-a36 promoter contains an imperfect ERE half site in Y. Zou et al. / FEBS Letters 583 (2009) 1368–1374 1373

Fig. 5. Expression of ER-a66 and 36 in different cancer cells. Total RNA was prepared from cultured cancer cells. RT-PCR was used to examine mRNA levels of ER-a66 and ER- a36 in cells. Housekeeping gene b-actin was used as an internal control. RT-PCR products were electrophoresed on a 1.2% agarose gel and visualizer by staining with ethidium bromide (A). The levels of mRNA expression of ER-a36, ER-a66 and b-actin were determined with a video-densitometry system and the ER-a36:b–actin ratio and ER-a66:b- actin ratio was calculated with the Quantity One software (Bio-Rad Laboratories). The band densities relative to the density of b–actin band were shown (B).

adjacent to an Sp1 binding site proximal to the TATA box (Fig. 1). In AF1 function of ER-a66, may explain this phenomenon. It was re- this study, we found that expression of the original 66-kDa ER-a ported that BRCA1 mediates the ligand-independent transcrip- (ER-a66) suppressed ER-a36 promoter activity in an estrogen- tional repression activity of the ER-a66 through its AF-1 domain independent manner, presumably through binding to the ERE half [24]. Thus, it may be possible that BRCA1 is involved in transcrip- site. Flouriot et al. [23] cloned a 46-kDa isoform of ER-a and dem- tional suppression activity mediated by ER-a66 and BRCA1 muta- onstrated that the 46-kDa isoform lacking the ﬁrst 173 amino acids tions may lead to upregulation of ER-a36 expression and its (A/B or AF-1 domain) is derived from alternative splicing of the ER- activity in BRCA1-related breast cancer. a gene by skipping exon 1. This isoform of ER-a was named as ER- Recently, we found that singling pathways mediated by EGFR, a46 and the original one as ER-a66 [23]. ER-a46 forms homodi- HER2 and GPR30 regulates ER-a36 expression at both transcrip- mers and heterodimers with ER-a66 [23]. Furthermore, the ER- tional and post-transcriptional levels (Kang and Wang, unpub- a46/66 heterodimers form preferentially over the ER-a66 homodi- lished observations), suggesting as an important player in mers and ER-a46 acts competitively to inhibit transcription activ- estrogen signaling, ER-a36 is regulated strictly at both mRNA ity mediated by the AF-1 domain of ER-a66 but without effect on and protein levels. This may also provide an explanation for the ob- an AF-2-dependent transactivation [23]. Here, we found that co- served inconsistence of ER-a36 expression in MDA-MB-436 cells; expression ER-a46 or ER-a36 itself released suppression activity low mRNA levels but high protein levels observed before [13]. Fur- mediated by ER-a66. The molecular mechanism underlying this re- ther study of the molecular mechanisms by which ER-a36 is regu- lease of suppression is currently unknown. The competitive bind- lated at the transcription and post-transcription levels will provide ing to the ERE half site by these ER-a isoforms, or heterodimer novel and important insights for the biological function of ER-a36 formation between ER-a66 and ER-a36 or ER-a46 to block the in many physiological and pathological processes. 1374 Y. Zou et al. / FEBS Letters 583 (2009) 1368–1374

Acknowledgement [13] Wang, Z.Y., Zhang, X.T., Shen, P., Loggie, B.W., Chang, Y.C. and Deuel, T.F. (2006) A variant of estrogen receptor-{alpha}, hER-{alpha}36: transduction of estrogen- and antiestrogen-dependent membrane-initiated mitogenic Grant support: NIH grant DK070016 (Z.Y. Wang) and the Ne- signaling. Proc. Natl. Acad. Sci. USA 103, 9063–9068. braska Tobacco Settlement Biomedical Research Program Award [14] Lee, L.M.-J., Cao, J., Chen’, P., Gatalica, Z. and Wang, Z.-Y. (2008) ER-a36, a novel (LB-595) to (Z.Y. Wang). variant of ER-a, is expressed in ER-a positive and negative human breast carcinomas. Anticancer Res. 28, 479–484. [15] Silberstein, G.B., Van Horn, K., Strickland, P., Roberts Jr., C.T. and Daniel, C.W. References (1997) Altered expression of the WT1 wilms tumor suppressor gene in human breast cancer. Proc. Natl. Acad. Sci. USA 94, 8132–8137. [1] Zhang, W.H., Andersson, S., Cheng, G., Simpson, E.R., Warner, M. and [16] Huang, T.H., Laux, D.E., Hamlin, B.C., Tran, P., Tran, H. and Lubahn, D.B. (1997) Gustafsson, J.A. (2003) Update on estrogen signaling. FEBS Lett. 546, 17– Identification of DNA methylation markers for human breast carcinomas using 24. the methylation-sensitive restriction fingerprinting technique. Cancer Res. 57, [2] Tsai, M.J. and O’Malley, B.W. (1994) Molecular mechanisms of action of 1030–1034. steroid/thyroid receptor superfamily members. Annu. Rev. Biochem. 63, 451– [17] Laux, D.E., Curran, E.M., Welshons, W.V., Lubahn, D.B. and Huang, T.H. (1999) 486. Hypermethylation of the Wilms’ tumor suppressor gene CpG island in human [3] Osborne, C.K. and Schiff, R. (2005) Estrogen-receptor biology: continuing breast carcinomas. Breast Cancer Res. Treat. 56, 35–43. progress and therapeutic implications. J. Clin. Oncol. 23, 1616–1622. [18] Zhang, T.F., Yu, S.Q., Guan, L.S. and Wang, Z.Y. (2003) Inhibition of breast [4] Klinge, C.M. (2001) Estrogen receptor interaction with estrogen response cancer cell growth by the Wilms’ tumor suppressor WT1 is associated with a elements. Nucl. Acids Res. 29, 2905–2919. destabilization of beta-catenin. Anticancer Res. 23, 3575–3584. [5] Kushner, P.J., Agard, D.A., Greene, G.L., Scanlan, T.S., Shiau, A.K., Uht, R.M. and [19] Anderson, I. and Gorski, J. (2000) Estrogen receptor a interaction with Webb, P. (2000) Estrogen receptor pathways to AP-1. J. Steroid Biochem. Mol. estrogen response element half sites from the rat prolactin gene. Biochemistry Biol. 74, 311–317. 39, 3842–3847. [6] Pietras, R.J. and Szego, C.M. (1977) Specific binding sites for oestrogen at the [20] Porter, W., Saville, B., Hoivik, D. and Safe, S. (1997) Functional synergy outer surfaces of isolated endometrial cells. Nature 265, 69–72. between the transcription factor Spl and the estrogen receptor. Mol. [7] Levin, E.R. (2002) Cellular functions of plasma membrane estrogen receptors. Endocrinol. 11, 1569–1580. Steroids 67, 471–475. [21] Vyhlidal, C., Samudio, I., Kladde, M.P. and Safe, S. (2000) Transcriptional [8] Segars, J.H. and Driggers, P.H. (2002) Estrogen action and cytoplasmic activation of transforming growth factor alpha by estradiol: requirement for signaling cascades. Part I: membrane-associated signaling complexes. Trends both a GC-rich site and an estrogen response element half-site. J. Mol. Endocrinol. Metab. 13, 349–354. Endocrinol. 24, 329–338. [9] Driggers, P.H. and Segars, J.H. (2002) Estrogen action and cytoplasmic [22] Petz, L.N. and Nardulli, A.M. (2000) Sp1 binding sites and an estrogen response signaling pathways. Part II: the role of growth factors and phosphorylation element half-site are involved in regulation of the human progesterone in estrogen signaling. Trends Endocrinol. Metab. 13, 422–427. receptor A promoter. Mol. Endocrinol. 14, 972–985. [10] Wehling, M. (1997) Specific, nongenomic actions of steroid hormones. Annu. [23] Flouriot, G., Brand, H., Denger, S., Metivier, R., Kos, M., Reid, G., Sonntag-Buck, Rev. Physiol. 59, 365–393. V. and Gannon, F. (2000) Identification of a new isoform of the human [11] Levin, E.R. (2005) Integration of the extranuclear and nuclear actions of estrogen receptor-alpha (hER-alpha) that is encoded by distinct transcripts estrogen. Mol. Endocrinol. 19, 1951–1959. and that is able to repress hER-alpha activation function 1. EMBO J. 19, 4688– [12] Wang, Z.Y., Zhang, X.T., Shen, P., Loggie, B.W., Chang, Y.C. and Deuel, T.F. 4700. (2005) Identification, cloning, and expression of human estrogen receptor- [24] Zheng, L., Annab, L.A., Afshari, C.A., Lee, W.H. and Boyer, T.G. (2001) BRCA1 alpha36, a novel variant of human estrogen receptor-alpha66. Biochem. mediates ligand-independent transcriptional repression of the estrogen Biophys. Res. Commun. 336, 1023–1027. receptor. Proc. Natl. Acad. Sci. USA 98, 9587–9592.