Advertisement
JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M704777200 on July 17, 2007

J. Biol. Chem., Vol. 282, Issue 37, 27478-27492, September 14, 2007
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
282/37/27478    most recent
M704777200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
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 arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kundu, P.
Right arrow Articles by Toro, L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kundu, P.
Right arrow Articles by Toro, L.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Regulation of Mouse Slo Gene Expression

MULTIPLE PROMOTERS, TRANSCRIPTION START SITES, AND GENOMIC ACTION OF ESTROGEN*Formula

Pallob Kundu{ddagger}1, Abderrahmane Alioua{ddagger}, Enrico Stefani{ddagger}§, and Ligia Toro{ddagger}||

From the {ddagger}Department of Anesthesiology, Division of Molecular Medicine, the ||Department of Molecular and Medical Pharmacology and §Department of Physiology, Cardiovascular Research Laboratories and Brain Research Institute, UCLA, Los Angeles, California 90095

Received for publication, June 11, 2007 , and in revised form, July 13, 2007.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The large conductance, voltage- and Ca2+-activated K+ channel plays key roles in diverse body functions influenced by estrogen, including smooth muscle and neural activities. In mouse (m), estrogen up-regulates the transcript levels of its pore-forming {alpha}-subunit (Slo, KCNMA1), yet the underlying genomic mechanism(s) is (are) unknown. We first mapped the promoters and regulatory motifs within the mSlo 5'-flanking sequence to subsequently identify genomic regions and mechanisms required for estrogen regulation. mSlo gene has at least two TATA-less promoters with distinct potencies that may direct mSlo transcription from multiple transcription start sites. These qualities mark mSlo as a prototype gene with promoter plasticity capable of generating multiple mRNAs and the potential to adapt to organismal needs. mSlo promoters contain multiple estrogen-responsive sequences, e.g. two quasi-perfect estrogen-responsive elements, ERE1 and ERE2, and Sp1 sites. Accordingly, mSlo promoter activity was highly enhanced by estrogen and blocked by estrogen antagonist ICI 182,780. When promoters are embedded in a 4.91-kb backbone, estrogen responsiveness involves a classical genomic mechanism, via ERE1 and ERE2, that may be complemented by Sp factors, particularly Sp1. Simultaneous but not individual ERE1 and ERE2 mutations caused significant loss of estrogen action. ERE2, which is closer to the proximal promoter, up-regulates this promoter via a classical genomic mechanism. ERE2 strategic position together with ERE1 and ERE2 independence and Sp contribution should ensure mSlo estrogen responsiveness. Thus, the mSlo gene seems to have uniquely evolved to warrant estrogen regulation. Estrogen-mediated mSlo genomic regulation has important implications on long term estrogenic effects affecting smooth muscle and neural functions.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The large conductance, voltage- and Ca2+-activated K+ (MaxiK)2 channels link cell excitability with cell metabolism, as they are exquisite sensors and controllers of both voltage and intracellular Ca2+. As a consequence, MaxiK channels are key regulators of vital body functions such as, blood flow, neural function, and uresis (1-3). MaxiK channels are formed by four pore-forming {alpha}-subunits (Slo) that can associate with regulatory beta-subunits (4). The {alpha}-subunit is expressed from a single copy gene in all studied species known as KCNMA1 or Slo. In mouse, it is present in chromosome 14 and has 27 constitutive exons, which are evolutionarily conserved.

Knock-out mouse models have highlighted the importance of Slo expression in blood pressure regulation (5), hearing (6), urinary bladder contraction (7), erectile function (8), and neurological disorders (9). Furthermore, a mutation in the human gene has been linked to coexistent generalized epilepsy and paroxysmal dyskinesia (10).

Defining the mechanisms that rule Slo gene expression is key to understanding the basis of Slo channel function. Several publications in recent years show that Slo transcript levels can be under the control of sex hormones and pregnancy. For example, mSlo transcript levels were up-regulated with the advancement of pregnancy in mouse myometrium (11, 12) likely due to the change in sex hormone levels. Consistent with this idea, mSlo transcript levels in myometrium were increased by 17-beta-estradiol (estrogen) treatment of normal and ovariectomized mice (13).3 Also, in the aorta of guinea pig, chronic treatment with estradiol up-regulated Slo mRNA levels (14). However, the underlying mechanism of this transcript up-regulation is unknown. Thus, we questioned whether these long term effects of estrogen on Slo transcript expression could be due to a genomic mechanism. Answering this question had as a prerequisite the establishment of the promoter(s) architecture. In this regard, studies of the Drosophila homolog gene, slowpoke (dSlo), have shown that this gene contains multiple promoters that regulate dSlo expression in a tissue-specifie (15-18) and developmental stage-specific manner (19). However, no studies have addressed the architecture and potential presence of multiple promoters and start sites in mammalian Slo. Therefore, our experimental strategy was designed to define the following: (i) the transcription start site(s) and promoter(s) of mSlo, (ii) the estrogen-regulatory regions in mSlo promoter(s), and (iii) the molecular mechanism(s) of estrogen-mediated regulation of mSlo.

Here we present evidence for the existence of multiple promoters and transcription start sites in the mSlo gene, and for their unique characteristics ensuring estrogen-mediated activation via classical and nonclassical genomic mechanisms.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Animals—Young adult (2-5 months) mice (C57BL/6) were used. Animal protocols were approved by UCLA, Chancellor's Animal Research Committee. Tissues were collected and cleaned in phosphate-buffered saline (10 mM sodium phosphate buffer, pH 7.4, 138 mM NaCl, 2.7 mM KCl) prior processing.

Reagents—TRIzol reagent (Invitrogen) was used for total RNA preparation. Gene Racer RACE kit and Superscript III reverse transcriptase (RT) were from Invitrogen. AMV RT was from Promega. Restriction and modifying enzymes were from New England Biolabs, and HotStar TaqDNA polymerase with Q-solution was from Qiagen. Plasmids were prepared using Qiagen kits, and clones were sequenced with BigDye Terminator version 3.1 (Applied Biosystems). 17-beta-Estradiol was from Sigma, and ICI 182,780 was from Tocris Bioscience. [{gamma}-32P]ATP and poly(dI-dC)-poly(dI-dC) were purchased from GE Healthcare.

Cell Culture—HeLa (human epithelial), NIH-3T3 (mouse fibroblast), and A7r5 (rat aortic smooth muscle) cells were purchased from American Type Culture Collection (ATCC). Cells were maintained in cell culture media consisting of Dulbecco's modified Eagle's medium (high glucose with L-glutamine) (Invitrogen) supplemented with 50 units/ml penicillin, 50 µg/ml streptomycin, and 10% fetal bovine serum (fetal calf serum for NIH-3T3 cells) under 5% CO2/air mixture at 37 °C. Cells were subcultured as recommended by the ATCC.

Plasmid Constructs—All plasmids were generated using standard recombinant DNA techniques. A P1-clone was used to obtain an mSlo gene fragment spanning -6950 nt from A(+1)TG of third possible translation initiation codon (encoding methionine 3, M3) (20) plus 183 nt of exon 1 and 771 nt of intron 1. The fragment was subcloned in SalI and BamHI sites of pGEM3 vector (Promega). Subsequent clones were generated by PCR amplification from this clone.

Constructs to detect promoter activity were subcloned in pGL3-Basic vector, which contains luciferase coding sequence, a polyadenylation signal, and no promoter for luciferase expression (Promega). For 5'-promoter-deletion constructs, mSlo sequences were maintained up to position +7 and included 17 nucleotides of the vector multiple cloning site in-frame with luciferase sequences. Thus, translation produces luciferase protein with an additional eight amino acids at its amino terminus (MDVDSCGS, where MD are from mSlo). Gene fragments were amplified with respective forward primers and mSloR reverse primer (Table 1). PCR products were digested with AscI and BamHI and subcloned into MluI, BglII sites of pGL3-Basic vector. To generate the -236/+7 construct, the -1262/+7 construct was digested with SacI (two sites, one in mSlo -231 and another in the vector). The band of interest was gel-purified and religated. The PCR product of -467/+7 construct was digested with MluI (at -82 nt) and BamHI and ligated in the corresponding sites in pGL3-Basic vector to obtain the -83/+7 construct. To make the -765{Delta}(-636 to -84) construct, the -765/+7 was digested with MluI (two sites at -635 and -82), and the larger product was religated.


View this table:
[in this window]
[in a new window]

  		TABLE 1
Primers used in generating deletion constructs and for primer extension

M3 indicates methionine encoded by the third possible translation codon in mSlo (20). Nucleotides in uppercase are homologous to mSlo gene; nucleotides in lowercase are sequences for cloning purposes, and restriction sites are underlined. PE1 to PE4 are primers used in primer extension. Mutated nucleotides are italicized and underlined.

 
For 3'-promoter-deletion constructs, gene fragments were amplified with -4910 forward primer and respective 3'-reverse primers (Table 1). PCR products were digested with AscI and BamHI and subcloned into MluI and BglII sites of pGL3-Basic vector. These constructs yield luciferase protein without additional amino acids.

Site-directed mutagenesis of EREs was done using two-step PCR. The PCR products containing mutated sequences were subcloned as cassettes for -4910/+7 ERE1mut (see Fig. 6A) using the EcoRV (-3204 nt)-NdeI (-2000 nt) sites and for -4910/+7 ERE2mut using NdeI (-2000 nt)-XhoI (-228 nt) sites. Equivalent strategy was used for the -2924/+7 ERE2mut construct. For the -1056/+7 ERE2 mutant, the whole construct was amplified in the second PCR using -1056/+7 forward primer and mSloR reverse primer.

To generate the double mutant -4910/+7 ERE1 and ERE2mut, the EcoRV-NdeI fragment from -4910/+7 ERE1mut was ligated in the same sites of -4910/+7 ERE2mut. Deletion mutant -4910/+7 {Delta}ERE1 and ERE2mut was obtained by deletion of EcoRI (-3109 and -1973) fragment from -4910/+7 ERE2mut construct followed by re-ligation.

Human ER{alpha} DNA-binding domain mutant (DBDM; E203A/G204A) was generated using one-step PCR with primer GAGTCTGGTCCTGTGCAGCTTGCAAGGCCTTCTTC (mutations, underlined). Restriction mapping and subsequent sequencing confirmed the accuracy of all clones.

Rapid Amplification of cDNA 5'-Ends (RACE) Analysis—The Gene Racer RACE kit was used, which uses the principle of RNA ligase-mediated 5'-RACE. After isolation of mRNA, truncated and non-mRNA were eliminated from the RACE reaction by dephosphorylation of their 5' phosphates with calf intestinal phosphatase. Capped mRNA was then decapped with tobacco acid phosphatase, exposing 5'-phosphates required for oligo ligation. The ligated mRNA was reverse-transcribed with a gene-specific primer designed from the sequences +27 to +47 from A(+1)TG of M3 and Superscript III reverse transcriptase (55 °C, 1 h). cDNAs were amplified by PCR (5'RACE-PCR) using GeneRacer 5' forward primer (from RACE kit) and a gene-specific reverse primer (-255 to -233), followed by seminested PCR using the same forward primer and a nested gene-specific reverse primer (-274 to -254). In this way, only mRNA that had the GeneRacer RNA oligo ligated to the 5'-end and that is completely reverse-transcribed undergoes amplification. Several PCR amplification mixtures were tested, and amplification with HotStar TaqDNA polymerase and Q-solution gave the best results, including amplification of GC-rich regions. Products were identified by sequencing.

Primer Extension—Primers (Table 1) were end-labeled with [{gamma}-32P]ATP using T4 polynucleotide kinase (New England Biolabs). Total RNA (30 µg) was used in primer extension reaction. About 4 x 105 cpm primer (PE1, Fig. 1A and Table 1) was annealed with the RNA in 30 µl of denaturation buffer (40 mM PIPES, pH 6.4, 1 mM EDTA, 0.4 mM NaCl, 80% deionized formamide) overnight at 45 °C. RNA-oligo complexes were purified by ethanol precipitation. Primer extension reaction was with AMV RT or with Superscript III RT for GC-rich regions. Reaction of AMV RT was done with 10 units of enzyme and 1x AMV RT buffer supplemented with 1 mM dNTP mix and 2.8 mM sodium pyrophosphate in a 20-µl reaction volume incubating at 42 °C for 1 h. For primer extension with Superscript III RT, RNA and primers (PE2, PE3, and PE4; Fig. 1A and Table 1) were mixed and heated at 70 °C and cooled to 50 °C for annealing. To the RNA/primer mixture, RT-buffer, dNTP, and 200 units of Superscript III RT were added and incubated at 50 °C for 1 h for primer extension. Reactions were stopped by heating samples at 70 °C for 15 min. Samples were ethanol-precipitated and analyzed by 8% urea-PAGE. As a reference, a sequencing reaction was run in parallel using the femtomole sequencing kit (Promega).

Cell Transfection and Luciferase Assay—Cells were grown in cell culture media to about 70% confluence and transfected with either the empty vector as control or with the promoter-luciferase constructs using Lipofectamine 2000 reagent following the manufacturer's instructions (Invitrogen). In all experiments, pRL-TK vector, which directs Renilla luciferase (RLuc) expression, was used as the transfection control to normalize firefly luciferase (Luc) activity. Transfection was carried out for 16 h in serum-free media, followed by regular culture for 30 h prior cell lysis. Cells were lysed with passive lysis buffer and assayed using Promega dual luciferase assay kit and the Veritas luminometer (Turner BioSystems). Results are expressed as Luc/RLuc ratio normalized to the maximum value in each experiment. In experiments with constructs of different sizes, an equal molar amount of DNA was transfected, and the total amount of DNA transfected was kept constant using the pGem3Zf-vector. For estradiol induction assays, cells were cotransfected with the human estrogen receptor {alpha} (hER{alpha}) expressing vector. Unless otherwise stated, hER{alpha} DNA was 0.125 times the concentration of the promoter constructs. Phenol red-free media were always used, and transfection was carried out overnight (16 h) in serum-free media. The next day, cells were induced with estradiol or carrier (ethanol <0.01%) in cell culture media but with charcoal-dextran-stripped serum (HyClone). Luciferase assays were performed 24 h after induction. Results are presented as average of at least three transfections.

RNA Interference—Pre-designed siRNAs against human Sp1, Sp3, and Sp4 transcription factors were purchased from Dharmacon, Inc. Cotransfection of siRNA (40 nM) with reporter DNA and estrogen responsiveness assay were carried out as described above. Efficiency of transfection was more than 80% as determined from transfection with a fluorescein-conjugated control siRNA. To measure the efficiency of gene down-regulation, real time PCR analysis was performed with cDNA prepared from total RNA of transfected cells. Glyceraldehyde phosphodehydrogenase mRNA levels were used as reference. Relative expression was calculated using 2-{Delta}{Delta}CT method (21).

Electrophoretic Mobility Shift Assay (EMSA)—Nuclear extract from HeLa cells transfected with hER{alpha} and induced with 10 nM estrogen was prepared as described previously (22). Estrogen-treated MCF7 cell nuclear extract was from Active Motif. Sense- and antisense-oriented oligonucleotides spanning mSlo ERE2 or mutated ERE2 were synthesized and annealed to make a duplex probe. ERE consensus and Sp1 duplexes were from Santa Cruz Biotechnology and Promega Inc., respectively. mSlo ERE2 duplex (3.5 pmol) was end-labeled with polynucleotide kinase and [{gamma}-32P]ATP. Labeled probe was purified, and ~20,000 cpm (~3 fmol) of the probe was used in a 10-µl binding reaction. The binding reaction was performed in a buffer containing 4% glycerol, 1 mM MgCl2, 0.5 mM EDTA, 0.5 mM dithiothreitol, 50 mM NaCl, 10 mM Tris-HCl, pH 7.5, and 500 ng of poly(dI-dC)-poly(dI-dC). The nuclear extract (5 µg) was preincubated in binding buffer for 15 min prior addition of probe and incubated for another 20 min at room temperature. For competition assay, competitor molecules were added during the preincubation stage. DNA-protein complexes were resolved in a 0.5x TBE-5% PAGE (10 °C, 220 V) and processed for autoradiography.

Western Blot Analysis—Fifty micrograms of nuclear extract from HeLa cells transfected with hER{alpha} was resolved in a 4-20% gradient SDS gel and immunoblotted using a polyclonal antibody raised against hER{alpha} (ER-{alpha}66, CHI Scientific) as described previously (12). Nontransfected HeLa cell nuclear extract was used as control.

Statistics—All experiments were performed with three different RNA preparations (three animals per each preparation of aorta RNA, two animals per preparation of bladder RNA, and one animal per preparation of uterus and intestine RNA) or three different batches of transfected cells. Results are expressed as means ± S.E. Two-tail Student's t test was used, and p < 0.05 was considered significant.

Software—To find TSSs we have used Neural Network Promoter Prediction (NNPP2.2), Promoter version 2.0 (23), Eponine (24), and FirstEF (25) software. Homology search and alignments were done using NCBI BLAST and VISTA tools (26). Transcription factor-binding sites were identified using Transcription Element Search System (TESS) (27) and Mat-Inspector from Genomatrix Software (GmbH, München, Federal Republic of Germany).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Identification of mSlo Transcription Start Site(s)
Initial analysis of mSlo sequences available in GenBankTM showed cDNAs with various sizes of 5'-untranslated regions. One possible explanation to this variety of 5'-UTR sizes is that mSlo transcription initiates at different start sites. Three independent methods were used to examine this possibility as follows: (i) computational analysis, (ii) RACE, and (iii) primer extension. For practical purposes position +1 was assigned to A in ATG of the third possible translation initiation codon (encoding M3) (20).

Computational Analysis—Analysis of 1.5-kb gene segment (GenBankTM, NC_000080 [GenBank] ) upstream of ATG encoding M3 predicted that indeed transcription could start from a number of different transcription start sites (TSSs) (Fig. 1A, colored arrows). We used for this purpose NNPP2.2, Eponine, Promoter 2.0, and FirstEF software. These programs utilize various operating principles including the presence of CpG islands (~200-500 bp with C+G content >50-55%), TATA, CCAAT, and Inr sequences (28). All but one start site predicted by NNPP2.2 (confidence ≥0.8) and Eponine (confidence ≥0.999) were within the mSlo 5'GC-rich region (~1 kb upstream M3) (red and yellow arrows in Fig. 1A). Promoter 2.0 (Fig. 1A, blue arrows) pointed a single TSS at -271 nt with very high confidence (1.0) within this GC-rich sequence, whereas FirstEF predicted a TSS at -661 nt (confidence 1.0) (Fig. 1A, green arrow) and a promoter between -1161 nt and -592 nt (P-FirstEF, dark gray box). Besides these TSSs in the GC-rich region, there were two more possible TSSs at -1282 (Fig. 1A, red arrow) and -1170 nt (blue arrow) predicted by NNPP2.2 and Promoter 2.0, respectively. All these predictions supported the presence of multiple start sites; however, the predicted positions did not coincide with each other precluding the precise definition of mSlo TSSs by computational means (28).


Figure 1
View larger version (32K):
[in this window]
[in a new window]

  		FIGURE 1.
In silico and 5'-RACE analyses for detection of TSSs in mSlo gene. A, scheme showing computational predicted TSSs (colored arrows), position of selected RACE products (black arrows), and predicted promoter by FirstEF program (P-FirstEF). In silico analysis for TSSs was with NNPP2.2, Eponine, Promoter 2.0, and FirstEF. Arbitrary base pair numbering assigns +1 to the adenosine of ATG encoding M3 in mSlo. Filled box, 5'-flanking sequence; CDS, coding sequence. PE1-PE4, primers used for primer extension (arrowheads); dotted arrow, exon 1 of antisense transcript. B, RACE-PCR product of mouse aorta and uterus. Marked bands (a-e) were purified and re-amplified. C, internal PCR of bands (a-e) in A. Prominent and right size products were gel-purified and cloned for sequencing.

 
RACE Analysis—To prove or disprove the presence of multiple TSSs in the mSlo gene, we initially performed 5'-RACE analysis. We used an RNA ligase-mediated 5'-RACE method that selects capped RNA ensuring the amplification of full-length mRNA molecules. A gene-specific reverse primer spanning from +27 to 47 nt of the coding sequence was used for reverse transcription to ensure amplification of all possible coding transcripts. After reverse transcription, amplified cDNAs were subjected to PCR amplification with another gene-specific reverse primer (-255 to -233 nt) and the forward primer designed to anneal to the RNA-oligo ligated to the mRNA (see "Materials and Methods"). When the RACE products were separated in agarose gel, we found four bands in samples of aorta (Fig. 1B, bands a-d). In the uterus, all four bands were present, but only the larger one was sufficiently strong to capture (Fig. 1B, band e). Bands a-e were gel-purified and re-amplified using semi-internal PCR confirming product specificity (Fig. 1C). Similar results were also obtained from the 5'-RACE analysis of brain tissue (data not shown). Cloning and subsequent sequencing of these four bands showed different transcription start sites with a variety of cDNA lengths in each band; for example, band a yielded products spanning positions -1217 to -1114 (from 14 sequenced clones, only 1 clone predicted a TSS at position -1217 and 6 clones predicted TSSs around -1189). Based on the longest and more frequent products of the sequenced clones from each band, we speculated that at least four TSSs may exist at positions near -1189, -1064, -956, and -726 nt (Fig. 1A, black arrows).

Primer Extension—To further validate the expression of mSlo transcripts from the RACE-inferred TSSs, we used as an alternative approach primer extension analysis with gene-specific reverse primers starting from ATG of M3 (Fig. 2). Experiments were performed using three different RNA preparations of various smooth muscles and with reverse primers designed to detect the various putative TSSs identified by RACE. Primer PE1 (Table 1) complementary to sequences (-1127 to -1107 nt) was used to confirm the -1189-nt TSS. Primer extension with AMV reverse transcriptase (sequences in that region were not GC-rich) using primer PE1 (see Fig. 1A for PE1-PE4 positions) produced a strong signal in all four tissues tested (Fig. 2A) positioning a TSS at -1187 nt in close approximation to the result obtained with RACE. Note that a faint product at -1209 (Fig. 2A, arrowhead) was also extended by PE1 primer in close agreement with RACE product at -1217. Confirmation of TSSs at the other three putative sites was more complex likely because of the high GC content in that region. In fact, two primers designed to detect the -1064 RACE-inferred TSS did not yield any product; thus, this potential TSS could not be corroborated by primer extension. Primer PE3 (-570 to -590) (Table 1) designed to detect the -726 RACE-inferred TSS gave a strong band mapped at -826 nt in bladder and faint bands at the same position in uterus and intestine (Fig. 2B), but no product was observed at -726. However and consistent with RACE-inferred TSS at -956 nt, PE3 primer also extended to another longer weaker product at approximately -950 nt, which although not positioned with accuracy could correspond to the RACE-inferred TSS (Fig. 2B, arrowhead). We also used a primer spanning -708 to -681 (PE2) that produced a smeared signal of ±20 nt long with a strong product mapped at -804 nt (Fig. 2C) or distinct bands encompassing -826 (not shown). Interestingly, PE3 could only detect -826 but not -804 nt TSS as was expected. One possible explanation is that this region has high secondary structure and thus primers selectively anneal to defined mRNAs.

Because NNPP2.2 and Promoter 2.0 software predicted the presence of TSSs close to M3 (Fig. 1A), we designed five different primers to scan up to -500 nt upstream of M3. Two primers extended beyond the resolution of the sequencing gel and two primers failed extension. However, PE4 annealing to -108 to -133 nt mapped strong TSSs at -184 nt and -186 nt; the latter being absent in bladder (Fig. 2D). The existence of true TSSs proximal to M3 (downstream of the first possible translation start codon encoding M1) was supported with promoter analysis as shown later in Fig. 3. Fig. 2E summarizes the main TSSs identified by primer extension (arrows).


Figure 2
View larger version (60K):
[in this window]
[in a new window]

  		FIGURE 2.
Mapping mSlo transcription start sites by primer extension. A-D, autoradiograms of primer extension products using mouse total RNA from uterus (U), intestine (I), bladder (B), and aorta (A). {Phi}X, end-labeled {varphi}X174-HinfI DNA ladder. a, c, g, and t represent the nucleotides in the sequencing reaction. The transcription start nucleotides (complementary strand of the sequence read, mRNA sequence) are in boldface and marked with an arrow. Start sites at -1187 nt (A), -826 nt (B), -804 nt (C), and -186/-184 nt (D) are shown. Box in A marks two adjacent Inr sequences. Dashed box in B marks an Inr-like sequence. Consensus Inr adenosine start sites are in italics (YYA(+1)NA/TYY) in A and B. Arrowheads in A and B mark additional TSSs (see text). E, gene map of TSSs defined by primer extension analysis and relative positions of transcription factor-binding sites related to smooth muscle specific regulation and estrogen responsiveness; 6944 nt upstream of the third possible translation initiation codon encoding M3 are mapped. Filled ovals, almost perfect EREs; unfilled ovals, half-estrogen receptor elements (1/2ERE); diamonds, Sp1-binding sites; trapezoids, AP-1-binding sites; open box, coding sequence (CDS). Gray bars mark smooth muscle specific transcription factor-binding sites as follows: *, serum-response factor (SRF); Figure 2, myocyte enhancer factor 2 (MEF2); and #, MyoD-binding sequences.

 
In summary, RACE and primer extension are consistent with the view that mSlo transcription can start at multiple sites. Although not with precision, computational analysis also supported this view. In particular, the TSS near -1187 nt was confirmed with RACE and primer extension. In addition, primer extension revealed dominant TSSs at positions -826, -804, and -186 to -184, and at least two other weaker TSSs (-1217 and approximately -950) that could be related to RACE products.

Core Promoter Elements and Regulatory Motifs in mSlo 5'-Flanking Sequence
After experimentally establishing the most probable TSSs, we analyzed sequences for the presence of classical core promoter elements that could support transcription like TATA box, Inr, BRE, DPE, MTE, and GC box (29-32). In agreement with a dominant TSS near -1187, two adjacent and perfect Inrs (YYA(+1)NA/TYY) were found surrounding this TSS, which predicted start sites at -1189 (as found by RACE) and -1182 (Fig. 2A, box; see supplemental Fig. 1). These results are consistent with previous findings showing that transcription does not necessarily need to start at the +1 position of the Inr (31). Transcription beginning at -826 is also supported by a partially conserved Inr sequence (with C instead of A/T at position +3) and by the presence of a pyrimidine-purine dinucleotide (-1,+1, CA) initiation site (Fig. 2B, dashed box), which has been recently recognized as a key sequence for active TSS (33). This property also applied for -186 TSS (-1,+1, CG) (Fig. 2D). In contrast, sequences -804 or -184 nt TSSs did not match with an Inr-related sequence nor could we locate any strong TATA box, BRE, DPE, MTE, a pyrimidine-purine dinucleotide (-1, +1), or a GC box (for -804) that could explain transcription initiation at these sites. Nevertheless, both TSSs could be supported by the existence of an imperfect 5'-ATG deficient region (ATG-desert) from -1137 to -287 nt, which is thought to enable transcription (34), where-as-184 may also rely on the presence of a GC box (at -245 to -232 nt) for transcription. Furthermore, promoter activity studies (Fig. 3) support their usage. ATG-desert could also support transcription from -826 and -186 TSS. Therefore, mSlo gene promoter falls into the category of TATA-less promoters with multiple TSSs (predominantly -1187, -826, -804, -184, and -186) and multiple promoter elements.


Figure 3
View larger version (32K):
[in this window]
[in a new window]

  		FIGURE 3.
mSlo basal promoter activity in different cell types. A, scheme of deletion constructs used to measure basal (unstimulated) activity in relation to TSSs and estrogen-responsive sequences map (top) (in this and following figures symbols are as in Fig. 2. B, normalized promoter activity in NIH-3T3, A7r5, and HeLa cell lines (n = 3). Luc/RLuc activities were normalized to maximum value in each group. *, significantly different from the -1262/+7-nt construct.

 
Because Slo is abundantly expressed in smooth muscle, we also looked for muscle-specific transcription factor-binding sites within 6.9 kb upstream from ATG of M3. Most of the smooth muscle expressing genes are controlled by serum-response factor, myocyte enhancer factor-2 (MEF-2), and by muscle-specific Myo-D (35). Indeed, we found one consensus (CArG box, CC(A/T)6GG; -1751CCTTTAAAGG-1742) and five additional semi-perfect serum-response factor-binding sites (Fig. 2E, *). We also located two probable sites for MEF-2 (CTAA/TA(A/T)2TA) at -1960CTATAACTA-1952 and -4153CTACTTTTA-4145 (Fig. 2E, Formula ). E-boxes (CANNTG) that can bind basic helix-loop-helix protein, particularly muscle-specific Myo-D, are also common in mSlo promoter (Fig. 2E, #). Its worth mentioning that most of these sites are clustered near the -1187-nt TSS. Precise locations can be found in supplemental Fig. 1.

In line with our hypothesis of estrogen-mediated regulation of mSlo, computational analyses (using TESS and MatInspector) detected multiple common cis-acting motifs that are known to mediate estrogen response through estrogen receptors via "direct" and "indirect mechanisms." The direct or classical pathway of ER{alpha} action involves the direct binding of the receptor to the target DNA. Few mammalian genes are known to be activated by ER{alpha} binding to a perfect estrogen-responsive element (ERE, GGTCANNNTGACC) being more common the activation via imperfect EREs or 1/2EREs. Indirect or nonclassical mechanisms do not involve direct binding of ER{alpha} itself to the target gene but require another DNA-binding transcription factor, like Sp1, to aid ER{alpha} association with DNA (36). In -6.9 kb of mSlo 5'-flanking sequence, motifs that support classical and nonclassical mechanisms are as follows: (a) two quasi-perfect EREs, -905GGTCACAGaGACC-893 and -2207GGTgATTTTGACC-2195, which are not perfect palindromic sequences as they have one mismatch each (lowercase); (b) 12 1/2 EREs along with >10 Sp1-binding sites, 5 signal transducers and activators of transcription, and 3 CCAAT/enhancer-binding protein sites; (c) 4 AP-1 sites present upstream of -2.3-kb region; and (d) a single NF{kappa}B site that starts at -4297 nt. Fig. 2E shows Sp1 (diamonds), AP-1 (trapezoids), ERE (filled ovals), and 1/2ERE (open ovals) sites (sequences up to -3411 are high-lighted in supplemental Fig. 1).

Identification of a Gene Region That Exhibits Strong Promoter Activity in Different Cell Context
To identify mSlo 5' regions necessary for basal promoter activity, we initially subcloned four fragments of 5'-flanking sequence fused with the firefly luciferase gene and analyzed their activity. All four constructs (-6944/+7, -4910/+7, -2924/+7, and -1262/+7) (Fig. 3A) were designed to direct luciferase expression from the third probable translation start site A(+1)TG of mSlo. We speculated that -1262/+7 construct would direct expression of mSlo if the TSSs at -184, -186, -804, and -826, found by primer extension, represents true TSSs.

Plasmids containing mSlo gene fragments were transfected into a human epithelial cell line (HeLa), a mouse fibroblast cell line (NIH-3T3), and a smooth muscle cell line of rat aortic origin (A7r5). Normalized luciferase (Luc/RLuc) activity showed that the relative transcriptional activity of these four mSlo gene fragments was similar in all cell types tested (Fig. 3B), although the -2924 fragment appeared to have somewhat stronger activity in A7r5 cells. The latter may be explained by the island of smooth muscle-specific transcription factor-binding sites near the -1187 start site (Fig. 2E). In agreement with the presence of proximal TSSs, the -1262 construct retained promoter activity in the three cell lines. In general, the longer the gene fragments were the lower the luciferase activity was, indicating that there may be some cis-acting inhibitory elements present upstream of -2924 nt.

Delineation of Promoter Boundary
To circumscribe promoter regions, we made 5' and 3' deletion constructs (Fig. 4, A and C). Short 5'-serial deletions of about 200 nt apart were made starting from -1967 nt. All these constructs (Fig. 4A) had the same 3'-end as constructs in Fig. 2. Fig. 4, A and B, shows that the minimal region for maximal promoter activity is located within the -467-nt construct as the -236-nt construct had significantly reduced activity that was practically lost using the -83-nt construct. The loss of activity in the -83-nt construct was observed in both HeLa and A7r5 cells. Also, a strong reduction of promoter activity by an internal deletion of -636 to -84 nt within the -765-nt-long construct ({Delta}-636-84) support the presence of a strong proximal promoter in this region, which would drive transcription from the -186/-184-nt TSSs.


Figure 4
View larger version (34K):
[in this window]
[in a new window]

  		FIGURE 4.
Promoter boundaries defined by 5' and 3' deletion constructs. A, scheme of constructs with sequentially deleted 5'-flanking sequences and an internal deletion in perspective to the map of TSSs and estrogen-related transcription factor-binding sites (top; see Fig. 2 for details). Numbers at left of each construct indicate the length of the 5'-flanking sequence analyzed for activity. Arrows mark the relative positions of TSSs identified by primer extension. B, corresponding normalized promoter activity of the various 5'-deleted constructs in HeLa and A7r5 cells (n = 3). Activity of vector alone corresponds to 0 nt upstream of ATG in the bar plot. *, significantly different from constructs ≥-467. C, scheme of constructs with 3'-sequences deleted (downstream of -1187 nt TSS) maintaining -1172 nt to -4910 nt of the 5'-flanking sequence constant. Numbers above the dotted lines describe the length of the deletion from the ATG of M3. Arrows indicate the relative position of the TSSs identified by primer extension. D, normalized promoter activity as a function of nt deleted measured in HeLa, NIH-3T3, and A7r5 cell lines (n = 3). Two breaks in the slope of the graph suggest the presence of at least two promoters in mSlo flanking sequence.

 
Next, we designed a series of plasmids with common 5'-extremities beginning at -4910 nt and variable 3'-ends (Fig. 4C). All these constructs were transiently transfected in HeLa, NIH-3T3, and A7r5 cells and have an ATG codon with a Kozak consensus sequence to drive luciferase translation. As shown in Fig. 4D, deletion constructs {Delta}-1171, {Delta}-1070 and {Delta}-966 had undetectable promoter activity, although they contain the -1187 TSS along with 5'-flanking sequences. However, the {Delta}-780 construct with an additional 186 nt with respect to {Delta}-966 showed significant promoter activity highlighting the functional relevance of this gene stretch. This activity could result from either the -1187- or -826/-804-nt TSSs or both. As expected, the {Delta}-473 construct had no additional activity confirming the absence of active TSS within the -781- to -474-nt region. An additional 278 nt in the {Delta}-195 construct (with respect to the {Delta}-473 construct) had no additional effect on promoter activity in HeLa and NIH-3T3 cells, but it increased the activity in A7r5 cells. One possible explanation to this result is that in A7r5 cells, cell-specific factors might help in the assembly of the transcription machinery on that region uncovering additional TSS between -196 and -473 nt. Nevertheless, comparison of {Delta}-195 construct activity with that of the full-length showed that addition of sequences containing TSSs at -186/-184 dramatically increased the overall promoter activity (Fig. 4D) by about 3-5 times. Taken together the results from 5' and 3' deletion experiments demonstrate that the mSlo gene is under the control of at least two promoters, one distal upstream -780 nt and one much stronger proximal promoter within -467 nt from M3.

mSlo Promoter Activation by Estrogen
Constructs -6944/+7, -4910/+7, and -2924/+7 in Fig. 3A were used to determine the action of 17-beta-estradiol (estrogen) on mSlo promoter activity. HeLa cells were transfected along with hER{alpha}. hER{alpha} was expressed from the cytomegalovirus promoter, which is not responsive to estrogen. Three parameters were measured to assess the specificity of estrogen-mediated promoter activation as follows: (i) estrogen dependence, (ii) estrogen-receptor {alpha} (ER{alpha}) dependence, and (iii) pharmacological response to the classical estrogen antagonist ICI 182,780.

Estrogen dependence was examined by stimulating cells with different doses of estradiol and measuring luciferase activity 48 h post-induction. In general, activation was observed with doses as low as 0.1 nM estradiol, and activity was increased with addition of increasing doses of the hormone (Fig. 5A) reaching maximal activity at about 30 nM estradiol in all constructs. Saturable dose-response curves were fitted to a Hill function as follows: (Luc/RLuc) - B = max/(1 + (EC50/[estrogen])N), where Luc/RLuc is the promoter activity reported by luciferase (Luc) normalized by the activity of a transfection control vector reported by RLuc; B, indicates basal promoter activity; EC50 indicates concentration of 50% of maximum effect; max = maximum effect; [estrogen] indicates estrogen concentration; and N = Hill coefficient. Independently fitted curves showed that the Hill coefficient was about 1, and the apparent estradiol affinity was within the same range (p > 0.1) for the three constructs as follows: for -6944/+7, EC50 was 0.22 ± 0.09 nM (n = 3 different transfections); for -2924/+7, EC50 was 0.4 ± 0.13 nM (n = 3 different transfections); and for -4910/+7, EC50 was 0.34 ± 0.14 nM (n = 3 different transfections). On the other hand, the efficacy or maximal activity of the -6944/+7 construct in response to estrogen was consistently lower (8.3 ± 0.1) compared with -4910/+7 (16 ± 3, p = 0.07) and -2924/+7 (12 ± 0.2, p = 0.0001) constructs suggesting the presence of inhibitory sequences in this construct. Although not significantly different, the slight tendency of -4910/+7 construct to yield higher maximal activity than the -2924/+7 construct could be explained by the presence of four additional estrogen response sequences (1/2EREs) in the longer construct (see Fig. 3A).


Figure 5
View larger version (30K):
[in this window]
[in a new window]

  		FIGURE 5.
Influence of estradiol on mSlo promoter activity, role of ER{alpha}, and demarcation of a region mediating estrogen responsiveness. A, estrogen regulation of mSlo promoter constructs in HeLa cells (n = 3). Constructs used are as described in Fig. 3A. -6944/+7 ({diamondsuit}), -4910/+7 ({blacktriangleup}), and -2924/+7 ({square}) constructs responded to estrogen in a dose-dependent manner. Values were fitted to a Hill function. EC50 values are given in the text. bullet, no promoter, transfected with vector alone. B, estrogen-induced up-regulation of mSlo promoter is dependent on the amount of transfected ER{alpha} clone. The -2924/+7 construct was used, and different DNA molar ratios of ER{alpha} plasmid were cotransfected. Note that in the absence of added estradiol ER{alpha} can constitutively stimulate promoter activity. This constitutive activity was subtracted from all experiments. C, ICI 182,780, an estrogen antagonist inhibits estrogen-induced up-regulation of mSlo transcription in a dose-dependent manner. The -2924/+7 construct was transfected in HeLa cells and stimulated with 1 nM estrogen or 1 nM estrogen plus the indicated doses of ICI 182,780. Activity value with estrogen alone was taken as maximal activity for fitting purposes. IC50 mean value is given in the text. D, detection of minimal estrogen-responsive region in mSlo promoter. Constructs used included the -2924/+7 construct (Fig. 3) and constructs in Fig. 4A with the exception of the internal deletion construct {Delta}-636-84. The -83-nt construct was used to determine background activity. Scheme shows positions of putative EREs (filled ovals) with respect to the constructs used.

 
Estrogen-mediated activation of mSlo promoter was also dependent on the presence of estrogen receptor {alpha} (ER{alpha}). Construct -2924/+7 was transfected with different amounts of human ER{alpha} and treated with 0.1 and 1 nM estradiol. Results in Fig. 5B show a saturable dose-dependent curve conclusively demonstrating that ER{alpha} is required for estrogen response of the mSlo gene. Furthermore, cells treated with different doses of ICI 182,780 along with estradiol (1 nM) showed that the drug successfully antagonized estradiol-mediated up-regulation in a dose-dependent manner (Fig. 5C). The IC50 was 3.2 ± 0.3 nM (n = 3 different transfections; ± S.D. of the fit) with 10 nM ICI 182,780 completely blocking the up-regulatory action of estrogen. In summary, the mSlo gene is highly responsive to estrogen, and this regulation is facilitated by ER{alpha} supporting a genomic mechanism of ER{alpha} on mSlo gene promoter element(s) as shown below.

Promoter Region Required for mSlo Transcriptional Regulation by Estrogen
To determine the elements in the mSlo promoter-regulatory region required for estrogen responsiveness, the activities of 5'-deletion mutants were evaluated in HeLa cells. From the experiment in Fig. 5A, we knew that the -2924/+7 construct contains most of estrogen-responsive elements. Therefore, we made sequential deletion constructs of this region. Estrogen responsiveness assay showed that hormone-mediated promoter activation fell with the progression of deletions toward ATG of M3 (Fig. 5D). Deletion of 956 nt (from -2924 to -1967) caused a clear drop of estrogen responsiveness. On the other hand, serial deletions of ~300 nt from -1967 to -1056 maintained a relatively constant slope of activity decay, whereas deletion of another ~300 nt in -765/+7 construct caused again an abrupt change in the slope completely obliterating estrogen responsiveness. Therefore, we hypothesized that a minimum estrogen-responsive sequence is present within the -1056- and -766-nt region, and that active estrogen-responsive sequences also exist upstream of this region, especially upstream of -1967 that contributes to the overall regulation of mSlo transcription. These predictions are consistent with computational analysis showing two quasi-perfect EREs at -2207 to -2195 nt and at -905 to -893 nt called here for simplicity ERE1 and ERE2, respectively (Fig. 5D, diagram; symbols are as in Fig. 2D).

Requirement of Putative EREs in Estrogen Responsiveness
To evaluate the role of ERE1 and ERE2 sites in estrogen responsiveness, we mutated the longest construct that shows maximal estrogen response (-4910/+7, Fig. 5A). The effect of ERE2 mutation was also analyzed in the context of the minimal estrogen-responsive construct (-1056/+7, Fig. 5D). Diagrams of wild type (WT) and mutant/deletion constructs are displayed in Fig. 6A. Estrogen-mediated activation results are shown in Fig. 6, B and C.

In the context of the -4910/+7 construct, mutation of ERE1 had an insignificant effect on estrogen response with respect to the WT construct, whereas ERE2 mutation had a seemingly inhibitory effect (although statistically insignificant). At 10 nM estrogen, Luc/RLuc for WT was 31 ± 4, for ERE1 was 28 ± 1 (p = 0.4), and for ERE2 was 23 ± 2 (p = 0.1). In contrast, mutating both sites reduced the maximal activation to about half; Luc/RLuc for ERE1 and ERE2 mut was 15 ± 1 (p = 0.01). Similarly, the ERE1-deleted and the ERE2-mutated construct (-4910/+7 {Delta}ERE1 and ERE2mut) also caused about half-reduction in estrogen response (Luc/RLuc = 13 ± 0.1, p = 0.01). Because the results in Fig. 5A indicated that sequences within the -4910- and -2924-nt region may contribute to estrogen response, we examined whether deletion of this region could further decrease estrogen responsiveness in the ERE2 mutant. Indeed, the ERE2 mutant construct lacking the -4910 to -2925 region (-2924/+7 ERE2mut) showed additional reduction in Luc/RLuc from 23 ± 2 to 17 ± 1 (p = 0.02), supporting a complementary role of upstream sequences in estrogen regulation.


Figure 6
View larger version (42K):
[in this window]
[in a new window]

  		FIGURE 6.
Role of putative EREs in estrogen responsiveness of mSlo promoter. A, scheme of constructs used in mutational analysis. ERE1 and ERE2 are marked with red and green ovals, respectively; 1/2 EREs are shown as open ovals; and Sp sites are shown as blue diamonds. Transparent white box shows the region of deletion. B, estrogen responsiveness of mutated constructs in the context of -4910/+7 and -2924/+7 constructs. C, same as in B but in the context of the minimal estrogen-responsive region (-1056/+7). Experimental maximum values at 10 nM estrogen are given in the text.

 
In the context of the minimal estrogen-responsive construct (-1056/+7), mutation of ERE2 resulted in a practical loss of responsiveness to doses of estrogen (Fig. 6C). Because the -1056/+7 construct contains sequences of the proximal and distal promoters (Fig. 4), we directly assessed the role of ERE2 on the proximal promoter by testing estrogen regulation of the {Delta}-780 construct (Fig. 4C) lacking the proximal promoter. This construct containing the full distal promoter but lacking the proximal promoter failed to produce any estrogen response at concentrations from 0.1 to 10 nM (n = 3) (not shown).


Figure 7
View larger version (24K):
[in this window]
[in a new window]

  		FIGURE 7.
mSlo promoter activity after siRNA-mediated down-regulation of Sp transcription factors in HeLa cells. Promoter activity in HeLa cells cotransfected with siRNA directed against Sp1, Sp3, and Sp4. Promoter activity was assayed in the -4910/+7 construct in WT (A) or with EREs mutated (B). *, significantly different with respect to no estrogen treatment (sham); #, significantly different from control in the absence of estrogen. Inset, percent of Sp1, Sp3, and Sp4 mRNAs remaining in cells 48 h after siRNA treatment compared with control (Ctr, no siRNA). Relative levels were determined using 2-{Delta}{Delta}CT method with glyceraldehyde phosphodehydrogenase as reference gene (21).

 
Overall, these results indicate that ERE2 has a major role in estrogen responsiveness of the proximal promoter, whereas both ERE1 and ERE2 partially contribute to the estrogen response with ERE2 having a somewhat preferential role when embedded in the backbone of the 4.91-kb sequence. Partial inhibition of estrogen response by mutation/deletion of both ERE1 and ERE2 and the presence of plenty of putative Sp factor-binding sites in mSlo promoter prompted us to investigate whether Sp factors could contribute to estrogen responsiveness in the context of -4910/+7 construct.

Role of Sp Factors in Estrogen Responsiveness of mSlo Promoter
We compared estrogen responsiveness of the -4910/+7 wild type (WT) construct (Fig. 7A) with that of the ERE-mutated construct (-4910/+7 ERE1andERE2mut, Fig. 7B) in cells cotransfected with 40 nM siRNA directed against Sp1, Sp3, or Sp4 transcription factors. Successful gene silencing was examined by real time PCR (Fig. 7A, inset), which showed ~75% reduction in expression for all three Sp factors.

All three siRNAs reduced mSlo nonstimulated promoter activity regardless of ERE1 and ERE2 being mutated (Fig. 7, A and B, compare unfilled bars in Control versus siSp1-siSp4, # indicates significantly different; inhibition of the WT construct by Sp3 and Sp4 siRNAs was statistically insignificant).

On the other hand, the significant estrogen-mediated stimulation of WT promoter activity (Fig. 7A,*, unfilled versus black bars of Control) was practically abolished by Sp1 siRNA treatment (Fig. 7A, siSp1 unfilled versus gray bars). This effect was independent of the integrity of ERE1 and ERE2 (Fig. 7B, compare estrogen stimulation in Control, unfilled versus black bar with respect to lack of stimulation in siSp1, unfilled versus gray bar).

Sp3 or Sp4 siRNA treatments were less effective in preventing estrogen-mediated stimulation of the WT construct (Fig. 7A, stimulation was still significant, *, compare siSp3 and siSp4 unfilled versus gray bars with respect to Control). However, in the absence of active ERE1 and ERE2 sites treatment with Sp3 and Sp4 siRNAs significantly inhibited estrogen-mediated up-regulation (Fig. 7B, compare siSp3 and siSp4 unfilled versus gray bars with respect to Control, unfilled versus black bar).


Figure 8
View larger version (32K):
[in this window]
[in a new window]

  		FIGURE 8.
DNA-binding mutant ER{alpha} and a direct genomic mechanism for mSlo estrogen-mediated activation. A-F, left, maps of the constructs used; right, Luc/RLuc activities. *, significantly different from activity with WT ER{alpha} and estrogen (1 nM) treatment (black bars) in the context of the WT -4910/+7 (A), -1520/+7 (D), and -1056/+7 (E) constructs. **, significantly different from activity with WT ER{alpha} and estrogen (1 nM) treatment (black bars) in the context of the -4910/+7 mutant constructs, ERE1andERE2mut (B) and {Delta}ERE1andERE2mut (C). Double arrow-dotted line, classical (c) component of estrogen action (because of direct ER{alpha} binding to DNA); black double arrow, nonclassical (nc) component of estrogen action (portion of estrogen action that does not involve ER{alpha} direct binding to DNA); gray double arrow, estrogen response because of ERE1 and ERE2. Dashed line in E and F, see text for explanation.

 
In summary, Sp1 seems to be the main contributor to estrogen-mediated regulation of mSlo promoters, whereas Sp3 and Sp4 seem not to play a major role under normal conditions but could have salvaging roles when ERE sites are no longer active. The robust inhibition of estrogen responsiveness by the deficiency of Sp1 factor likely results from the following combinations: 1) the low nonstimulated promoter activity, and 2) the disruption of Sp factor-mediated nonclassical pathway of estrogen action. Our next experiments with mutant ER{alpha} clarified that only a portion of estrogen responsiveness is contributed by a nonclassical pathway.

Physical Interaction of ER{alpha} with the ERE
Mutagenesis experiments showed the possibility of direct binding of ER{alpha} with mSlo EREs (Fig. 6). To investigate this possibility, we undertook two different approaches as follows: (i) estrogen responsiveness assay with a well characterized DNA-binding domain mutant (DBDM) of ER{alpha} that lacks direct binding to ERE capability but still can interact with other transcription factors (37), and (ii) electrophoretic mobility shift assay (EMSA).

Results in Fig. 8 show that the -4910/+7 (A), -1520/+7 (D), and -1056/+7 (E) WT constructs all showed a significant decrease of estrogen responsiveness when the mutant ER{alpha}, DBDM, was used (*, compare black with corresponding gray bars). This decrease would correspond to the classical (c) pathway marked for clarity only for the -4910/+7 WT construct (Fig. 8A, dotted double arrow). A decrease in the estrogen responsiveness of ERE mutants in -4910/+7 construct with DBDM-ER{alpha} (Fig. 8, B and C, compare black versus corresponding gray bars, **) suggests that the DNA binding property of ER{alpha} to activate mSlo promoter is not limited to the two EREs (the contribution of the two EREs is marked with gray double arrow in Fig. 8B).

Note that estrogen stimulation of WT or ERE mutants in the -4910/+7 construct was not completely abolished by expression of the DBDM-ER{alpha} (Fig. 8, A-C); the remnant activation is consistent with a partial role of nonclassical mechanisms (nc) in estrogen-mediated stimulation, likely Sp1 as shown in Fig. 7. For illustrative purposes the nonclassical contribution of estrogen-mediated stimulation is marked with a black double arrow only for the WT construct (Fig. 8A).

Because ERE2 is the single estrogen receptor-binding site in the -1056/+7 construct, we expected to observe no differential estrogen stimulation between WT (Fig. 8E) and ERE2mut (Fig. 8F) constructs when DBDM-ER{alpha} was used. However, the ERE2mut construct showed less estrogen activation in the presence of DBDM-ER{alpha} than the WT construct (dashed lines). This reduction in estrogen activation is unlikely to come from ERE-like sequences in this construct as none were detected under close inspection, indicating that its origin is likely from the vector itself that is known to possess weak ERE-like sequences. Overall, these results indicate that a mechanism involving direct binding of ER{alpha} to EREs (classical mechanism) is a significant component of estrogen responsiveness of the mSlo gene.

As mutational experiments suggested an ERE2 preferential role (Fig. 6B), we examined whether ER{alpha} binds to this site in mSlo gene using EMSA. Nuclear extracts from estrogen-treated HeLa (transfected with human ER{alpha}) and MCF7 cell lines were used as source of dimerized ER{alpha}. Expression of ER{alpha} in HeLa cell nuclear extract was confirmed by Western blot, which shows labeling of a band of the expected molecular mass (~67 kDa) and its absence in nuclear extracts of mock-transfected cells (control, Ctr) (Fig. 9A, left panel). Ponceau S staining of the same blot served as control for loading (Fig. 9A, right panel). Duplexed DNA probe (mSlo ERE2) and the competitor sequences (mSlo mutated ERE2, consensus ERE, Sp1) are shown in Fig. 9B. The EMSA in Fig. 9C shows that a prominent DNA-protein complex was formed with HeLa-ER{alpha} nuclear extract (left panel) incubated with the mSlo ERE2 probe, whereas three distinguishable complexes were detected with MCF7 cell nuclear extract (right panel). This indicates that in MCF7 cells additional protein interactions with the DNA-ER{alpha} complex must occur. Most importantly, all these bands were efficiently competed in a dose-dependent manner with the consensus ERE containing duplex (Fig. 9C, lines 3 and 6) but not with a mutant form of mSlo ERE (lines 4 and 7) or with an Sp1 consensus duplex (lines 5 and 8). These results conclusively demonstrate that ER{alpha} can directly bind to mSlo ERE2.


Figure 9
View larger version (66K):
[in this window]
[in a new window]

  		FIGURE 9.
Physical interaction of ER{alpha} with the ERE of mSlo gene minimal estrogen-responsive segment. A, Western blot (WB) developed with an anti-hER{alpha} antibody (left panel) showing the presence of ER{alpha} in the HeLa cell nuclear extract used in EMSA. Arrow marks the band of appropriate molecular size of 67 kDa. Right panel showing the same blot stained with Ponceau S as loading control. Ctr, control; ER{alpha}, cells transfected with hER{alpha}; M, molecular mass markers. B, duplex oligos used in EMSA. Box, delineates the quasi-perfect ERE2 in mSlo promoter and the perfect (consensus) ERE of the commercial oligo (Santa Cruz Biotechnology). Dashed line marks the mSlo-mutated ERE2 (specific mutated bases are underlined and italicized). Lowercase letters indicate mismatch from consensus ERE. C, EMSA showing binding of ER{alpha} with ERE2 of mSlo (DNA-ER{alpha} complex). Left panel shows EMSA done with ER{alpha}-transfected and estrogen-treated HeLa cell nuclear extract, and right panel shows EMSA done with nuclear extract of estrogen-treated MCF7 cells. Different concentrations of unlabeled competitor duplexes are indicated in the figure. Nonradioactive (cold) consensus ERE (30x, line 6) but not mutated mSlo ERE2 (line 7) or Sp1 duplexed oligos (line 8) could compete with radiolabeled mSlo ERE2 for protein binding.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
We have shown the first functional analysis of mSlo promoter regions setting the basis for under-standing the regulatory mechanisms ruling the expression of the gene by estrogen. The existence of multiple transcription start sites and promoters along with the redundancy of estrogen response mechanisms depicts the "fail-safe" nature of mSlo estrogenic transcriptional control.

Multiple TSSs in mSlo—In general, computational analysis predicted two main features of the mSlo gene as follows: (i) that mSlo transcription could initiate from more than one site, and (ii) that the gene may contain a CpG island spanning -1 to -1199 nt (CpG island searcher (38)). CpG islands enclose promoter regions in ~40% of human genes and are present in all ubiquitously expressed genes (39), a characteristic of mSlo.

Multiple TSSs were determined experimentally using 5'-RACE (Fig. 1) and primer extension (Fig. 2). The fact that we could only corroborate two of the RACE-predicted TSS (-1189 and -956) with primer extension may be explained by a GC-rich RNA secondary structure region that negatively affects the reaction (only high temperature primer extension worked with certain primers; see "Experimental Procedures"). In this regard, primer PE3 (Fig. 1A) should have recognized -804 and -826 but only detected the latter. Nevertheless, primer extension revealed the existence of four additional TSSs that were not detected by RACE, -826, -804, and surprisingly close to ATG of M3, -186, and -184 sites that are downstream of ATG encoding M1. Furthermore, the presence of TSSs close to ATG of M3 seems real as it was corroborated by promoter-deletion luciferase assays (Fig. 4B) and predicted by NNPP2.2 software (Fig. 1A).

It is worth mentioning that the 5'-RACE-predicted TSSs reported here (-1189, -1064, -956, and -726) correspond to the longest and more frequent products obtained by this method. The multiplicity of lengths detected by 5'-RACE could be due to amplification of degraded RNA or reflect a true property of mSlo transcription. We support the second explanation as 5'-RACE was performed with high quality RNA and used capped mRNAs. Furthermore, it is now known that the majority of human genes have a broad distribution of TSSs (40) as is apparent in mSlo. Therefore, it is possible that mSlo transcription could also initiate from additional TSSs as the ones reported here resulting in different lengths of 5'-UTR. It will be interesting to find out which TSS is preferred in vivo and whether there is a difference in TSS selection in basal and activated transcription.

One could speculate that variable lengths of 5'-UTR generated by multiple TSS could direct tissue-specific translation factors contributing to variation in expression/efficiency patterns. Furthermore, transcription from -186/-184 TSS would preferentially result in a protein translated from ATG of M3 (with perfect Kozak sequence).

Multiplicity of Core Elements—mSlo gene displays a variety of core promoter elements or sequences that would aid transcription from the various TSSs, e.g. Inr, pyrimidine-purine dinucleotide (-1, +1), GC box, and imperfect ATG-desert region. In particular, Inrs are common upstream of mSlo GC-rich region (supplemental Fig. 1). Our results indicate that mSlo gene can use perfect and imperfect Inrs for the initiation of transcription. Interestingly, -1187 TSS found by primer extension (-1189 RACE-predicted product) is within two contiguous perfect Inrs that seem to serve as a sticky surface to ensure transcription from this region. RACE products of -1189, -1182, and -1180 lengths (asterisks in supplemental Fig. 1) support this view. Nonconserved Inrs could explain the -826 TSS and the faint longer extended product in Fig. 2A (arrowhead at -1209; supplemental Fig. 1). In this case, the strength of the signal coincided with the degree of Inr consensus accuracy. The Inrs encompassing -1187 TSS is associated to a DPE, which although not in an optimum position (+16 to +20 nt versus consensus +28 to +32) could further support transcription from this TSS. Inr sequences can be associated with BRE or MTE, but these motifs seem not to be involved in mSlo transcription as MTE is absent and BRE could not be located at consensus positions.

Pyrimidine-purine dinucleotides supported the -826 and -186 TSSs, whereas imperfect ATG-desert supported TSSs at -186, -184, -804, and -826. The imperfect ATG-desert (34) contains two ATGs upstream of the one encoding M1. Interestingly, the distal ATG (-278 to -276) has a perfect Kozak sequence that could result in a small peptide of 27 amino acids. It will be interesting to determine whether this potential leader peptide has a functional role on mSlo.

GC boxes (Sp1-binding sites) are common in CpG island-associated promoters, as in mSlo. A GC box (-245 to -232 nt; supplemental Fig. 1) predicted by TESS software comparing current human, mouse, and rat models may support transcription of TSSs at -186 and -184. As regions upstream -804 and -826 are also GC-rich, we cannot exclude that these TSSs are also supported by a GC box and could undergo transcription aided by Sp1 (31).

Although experimental TSSs could not be assigned to any TATA box (TATA(T/A(AA(G/A)), we found a TATA-like sequence located at -392 to -387 nt, which is conserved in mouse, rat, and human Slo genes. Probably, based on the presence of this motif, Eponine software that uses a TATA-box and GC-rich domain-based algorithm predicted a TSS around nucleotide -361 (Fig. 1A) in mSlo. The corresponding TATA-like sequence in hSlo gene has been correlated with a TSS (41). We also found three MED1 (multiple start element down-stream) motifs located at positions -515, -318, and -234 (supplemental Fig. 1). This type of sequences have been proposed to play a role in TATA-less and multiple start-site class of promoters when located downstream of TSS up to +145 (42); the potential role of these MED1 motifs in mSlo expression has yet to be established.

Multiple Promoters—We have demonstrated that the mSlo 5'-flanking sequence immediately upstream of M3 possesses promoter activity in transfected cell lines. Basal promoter activity assay showed that the maximal promoter activity could be achieved with a -1262/+7 construct and that bigger constructs yielded less promoter activity. This effect is observed in many mammalian genes (41, 43), and it has been attributed to the presence of inhibitory domains of the active promoters in upstream sequences (43, 44). A similar pattern of basal activity displayed by the various size constructs in different cell lines suggests that basal promoter activity is driven by a similar pol II complex regardless of the cell line.

Deletion analysis from 5'- or 3'-ends of the flanking sequence identified at least two promoter regions utilized by the mSlo gene. 5' deletion construct identified a proximal promoter in HeLa, A7r5 smooth muscle cells (Fig. 4B), and NIH-3T3 cells (data not shown). The basal activity of the proximal promoter in multiple cell lines indicates that this region may direct the constitutive expression of mSlo in different tissues.

Activity of 3'-deletion constructs as a function of nt deleted upstream of ATG encoding M3 (Fig. 4D) clearly showed two breaks in the curve confirming the presence of the proximal promoter and highlighted the activity of a distal promoter. The distal promoter required for its activity 186 nt from -966 to -781 ({Delta}-780 versus {Delta}-966 construct, Fig. 4, C and D). Based on the -1187 TSS mapped with primer extension, we would have expected that construct {Delta}-1171 (with extra 17 nt) presented promoter activity, but this was not the case. Because many mammalian promoters are known to require sequences down-stream of TSS for proper expression (31), we designed constructs with additional nucleotides. However, addition of 117 nt ({Delta}-1070) or 220 nt ({Delta}-966) did not result in any detectable promoter activity in the three cell lines tested. Several nonexclusive mechanisms could explain these intriguing results as follows: (i) a promoter that is strictly tissue-specific and needs other factors for its expression that are absent in the cell lines tested but present in uterus, intestine, bladder, aorta (Fig. 2A), and brain (not shown); (ii) a promoter that requires downstream sequences that were not included in the 220-nt downstream sequences of {Delta}-966 construct; and (iii) a small 26-amino acid peptide could be synthesized from an ATG (at -1170 nt, with Kozak consensus sequence) preventing luciferase expression. Thus, the distal promoter could drive transcription from -1187, from -826/-804 TSSs, or from both TSSs.

Homology search of the mSlo UTR region in the EST data base detected two clones (BY715389 [GenBank] and AK019684 [GenBank] ) whose three exons are 100% homologous to the sequences complementary to -1907 to -1060 (dashed arrow in Fig. 1A marks exon 1). Existence of these ESTs suggests that the GC-rich sequences around -826/-804-nt TSSs harbor a bi-directional promoter. It is now known that many mammalian genes can produce this kind of sense-antisense transcripts and that they can regulate each other (45). It will be interesting to know whether expression of this antisense RNA has any role on mSlo expression regulation.

Accumulating evidence also suggests that a high number of mammalian genes are regulated by multiple promoters and transcribed from multiple TSSs (33, 46-48). In many instances these multiple promoters behave differentially under different cellular conditions. For example, Drosophila Slo gene (dSlo) has multiple promoters whose expression is differentially regulated in muscle or the central nervous system (16, 18). However, the role of multiple TSSs or exactly how they are regulated in mammals is not clear. Because mSlo transcript expression is high in various tissues (Fig. 2), mSlo could serve as a prototype gene to study the regulation of mammalian transcription from multiple TSSs.

Our analysis showed that a cluster of muscle-specific transcription factor-binding sites is present within -2431 and -1269 nt that include the following: a consensus and three semi-perfect CArG boxes, a consensus MEF2-binding site, and four consensus E-boxes (Fig. 2E and supplemental Fig. 1). These observations correlate with a high expression of Slo in smooth muscle. In line with this view, our basal promoter activity assay showed that the -2924/+7 construct, which include these sites, had relatively more activity in a smooth muscle cell line (Fig. 3B).

Genomic Regulation of mSlo by Estrogen—We have demonstrated that mSlo promoters embedded into different lengths of 5'-flanking sequence respond differentially upon stimulation with estrogen. This stimulation was dependent both on the availability of ER{alpha} and on physiological concentrations of estrogen. Stimulation was observed with as low as 0.1 nM estrogen and may involve several regions of the promoter as depicted by dose-response, activity versus sequence content curves, and mutagenesis experiments (Figs. 5 and 6). Two almost perfect EREs, ERE1 and ERE2, having a fundamental role were identified along with a partial role of Sp factors mainly Sp1 (Figs. 6 and 7). Inhibition of estrogen responsiveness with its antagonist ICI 182,780 proves the involvement of estrogen in mSlo promoter activation, whereas direct binding of ER{alpha} to ERE2 demonstrates a classical genomic mechanism at this site (Figs. 5, 8, and 9).

It is worth mentioning that mSlo has another ERE-like sequence (-2072GGTgtGGCTGACC-2060), although this element has two mismatches in contrast to quasi-perfect ERE1 and ERE2. The role of this ERE-like sequence if any is minimal as its mutation (not shown) or deletion (inclusive in {Delta}ERE1-ERE2mut construct) does not lead to significant loss of estrogen responsiveness when compared with control (compare ERE1 and ERE2mut versus {Delta}ERE1-ERE2mut constructs; Fig. 6B).

Mutational analysis suggested that a classical mechanism of estrogen action might be involved in mSlo gene regulation by estrogen. Final proof for the involvement of this type of mechanism came from our assays with DBDM-ER{alpha} and direct DNA binding analysis with EMSA (Figs. 8 and 9). EMSA showed that ER{alpha} can directly bind to mSlo ERE2 irrespective of the source of the ER{alpha}, and competition experiments ruled out the possibility of the interaction in that region of DNA via other transcription factors, including Sp1.

The fact that the ERE1 and ERE2mut construct showed a reduction in estrogen responsiveness in DBDM-ER{alpha} compared with WT-ER{alpha}-expressing cells indicates that, in addition to ERE1 and ERE2, other ERE-like sequences (e.g. 1/2EREs) or alternative unknown mechanism(s) that require an intact DNA-binding domain of ER{alpha} may also participate in estrogen-mediated up-regulation of mSlo gene.

The full consensus ER-binding motif GGTCANNNTGACC (ERE) is found only in a few genes. Most of the estrogen-responsive genes identified so far contain an imperfect repeat (36, 49) or multiple 1/2ERE sites sometimes in a GC-rich region (50). In vitro studies have shown that the ER{alpha} binding to the perfect palindrome is strongest, and any deviation from the palindromic sequence causes reduced binding affinity for ER{alpha} (51). Recent studies also demonstrated that ERE sequence, flanking sequence, and promoter context determine the binding affinity as well as the rate of transcriptional activation (47, 52, 53). ER{alpha} binding to 1/2EREs in combination with Sp1 binding to a GC-rich site has also been reported (51). Alternative pathways of ER{alpha} interaction with promoters are indirect (nonclassical genomic mechanisms), in which ER{alpha} influences transcription by interaction with another transcription factor bound to the promoter-regulatory sequences. For example, ER{alpha} can interact with Sp1/Sp3/Sp4 proteins bound to the GC-rich sequence in the promoter (54-58) or activation can occur through interaction with transcription factors bound at AP-1 site (49, 59). ER{alpha} can also interact with transcription factors like nuclear factor {kappa}beta (NF-{kappa}B), STAT5, GATA-1, CCAAT/enhancer-binding protein beta (59), and CCAAT/enhancer-binding protein {alpha} (60). In this regard, mSlo gene has plenty of Sp factor-binding sites (supplemental Fig. 1) that could play a regulatory role in mSlo transcription. Activation of estrogen-mediated transcription via an Sp1 site has been reported for a small conductance Ca2+-activated K+ channel (56). Our results also showed that Sp factors can contribute to the estrogen-mediated mSlo promoter activation, especially Sp1 (Fig. 7). Further mutational studies may confirm a specific role of the multiple 1/2EREs or Sp1 sites in mSlo.

Comparison of mouse 600-nt 5'-flanking sequences with rat and human genes showed a high degree of homology (~90%) among species. This region is also highly GC-rich in rat and human. Between mouse and rat sequences the 90% homology extended to -1187 nt, and there was 80% homology within -4.5 kb (supplemental Fig. 2). Multiple 1/2EREs are found in the three species that are not positionally conserved. In addition, ERE1 and ERE2 are not conserved in human; however, rat has a fairly related sequence (GGTCGCGAAGACC) as mSlo ERE2 at a similar position. Based on these observations, we suggest that the proximal promoter is evolutionarily conserved and responsible for most of the basal activity of mSlo promoter. However, estrogen-activated transcription may be controlled by different mechanisms in a tissue- and species-specific manner. This prediction is supported by the fact that mRNA levels are decreased by estrogen in rat myometrium instead of increased as in mouse (13, 61). The presence of multiple estrogen-responsive sequences could produce additive or synergistic stimulatory effects and safeguard the response of mSlo gene to estrogen.

In summary, mSlo is a TATA-less multiple transcription start type of gene with more than one promoter that is regulated by estrogen. The multiplicity of TSSs and estrogen-responsive sequences in the promoter region of mSlo seems designed to ensure its expression and its responsiveness to estrogen via classical and nonclassical genomic mechanisms. These findings also provide an underlying mechanism for the estrogen-mediated increase in mSlo transcript levels in vivo.


   FOOTNOTES
 
* This work was supported by National Institutes of Health Grants HL54970 (to L. T.) and HD046510 (to E. S.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

Formula The on-line version of this article (available at http://www.jbc.org) contains supplemental Figs. 1 and 2. Back

1 To whom correspondence should be addressed: Dept. of Anesthesiology, UCLA, BH-509A CHS, P.O. Box 957115, Los Angeles, CA 90095-7115. Tel.: 310-794-7809; Fax: 310-825-5379; E-mail: pallob{at}ucla.edu.

2 The abbreviations used are: MaxiK, large conductance, voltage-, and Ca2+-activated K+ channel; Slo, KCNMA1, pore-forming {alpha}-subunit of MaxiK channel; mSlo, mouse Slo;dSlo, Drosophila Slo, slowpoke; ERE, estrogen-responsive element; AP-1, activator protein 1; Sp1, specificity protein 1; estrogen, 17-beta-estradiol; RT, reverse transcriptase; M3, methionine 3 encoded by the third possible translation initiation codon in mSlo; RLuc, Renilla luciferase; Luc, firefly luciferase; ER{alpha}, estrogen receptor {alpha}; TSSs, transcription start sites; Inr, initiator; BRE, TFIIB recognition element; DPE, downstream promoter element; MTE, motif 10 element; DBDM, DNA-binding domain mutant; h, human; nt, nucleotide; RACE, rapid amplification of cDNA 5'-ends; EMSA, electrophoretic mobility shift assay; WT, wild type; PIPES, 1,4-piperazinediethanesulfonic acid; AMV, avian myeloblastosis virus; siRNA, short interfering RNA; UTR, untranslated region. Back

3 P. Kundu, A. Alioua, E. Stefani, and L. Toro, unpublished data. Back


   ACKNOWLEDGMENTS
 
Human ER{alpha} clone was a kind gift from Dr. Bert W. O'Malley (Baylor College of Medicine).



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Brayden, J. E., and Nelson, M. T. (1992) Science 256, 532-535[Abstract/Free Full Text]
  2. Burdyga, T., and Wray, S. (2005) Nature 437, 290
  3. Gu, N., Vervaeke, K., and Storm, J. F. (2007) J. Physiol. 580, 859-882[Abstract/Free Full Text]
  4. Lu, R., Alioua, A., Kumar, Y., Eghbali, M., Stefani, E., and Toro, L. (2006) J. Physiol. (Lond.) 570, 65-72[Abstract/Free Full Text]
  5. Sausbier, M., Arntz, C., Bucurenciu, I., Zhao, H., Zhou, X. B., Sausbier, U., Feil, S., Kamm, S., Essin, K., Sailer, C. A., Abdullah, U., Krippeit-Drews, P., Feil, R., Hofmann, F., Knaus, H. G., Kenyon, C., Shipston, M. J., Storm, J. F., Neuhuber, W., Korth, M., Schubert, R., Gollasch, M., and Ruth, P. (2005) Circulation 112, 60-68[Abstract/Free Full Text]
  6. Ruttiger, L., Sausbier, M., Zimmermann, U., Winter, H., Braig, C., Engel, J., Knirsch, M., Arntz, C., Langer, P., Hirt, B., Muller, M., Kopschall, I., Pfister, M., Munkner, S., Rohbock, K., Pfaff, I., Rusch, A., Ruth, P., and Knipper, M. (2004) Proc. Natl. Acad. Sci. U. S. A. 101, 12922-12927[Abstract/Free Full Text]
  7. Meredith, A. L., Thorneloe, K. S., Werner, M. E., Nelson, M. T., and Aldrich, R. W. (2004) J. Biol. Chem. 279, 36746-36752[Abstract/Free Full Text]
  8. Werner, M. E., Zvara, P., Meredith, A. L., Aldrich, R. W., and Nelson, M. T. (2005) J. Physiol. (Lond.) 567, 545-556[Abstract/Free Full Text]
  9. Sausbier, M., Hu, H., Arntz, C., Feil, S., Kamm, S., Adelsberger, H., Sausbier, U., Sailer, C. A., Feil, R., Hofmann, F., Korth, M., Shipston, M. J., Knaus, H. G., Wolfer, D. P., Pedroarena, C. M., Storm, J. F., and Ruth, P. (2004) Proc. Natl. Acad. Sci. U. S. A. 101, 9474-9478[Abstract/Free Full Text]
  10. Du, W., Bautista, J. F., Yang, H. H., Diez-Sampedro, A., You, S. A., Wang, L. J., Kotagal, P., Luders, H. O., Shi, J. Y., Cui, J. M., Richerson, G. B., and Wang, Q. K. (2005) Nat. Genet. 37, 733-738[CrossRef][Medline] [Order article via Infotrieve]
  11. Benkusky, N. A., Fergus, D. J., Zucchero, T. M., and England, S. K. (2000) J. Biol. Chem. 275, 27712-27719[Abstract/Free Full Text]
  12. Eghbali, M., Toro, L., and Stefani, E. (2003) J. Biol. Chem. 278, 45311-45317[Abstract/Free Full Text]
  13. Holdiman, A. J., Fergus, D. J., and England, S. K. (2002) Mol. Cell. Endocrinol. 192, 1-6[CrossRef][Medline] [Order article via Infotrieve]
  14. Jamali, K., Naylor, B. R., Kelly, M. J., and Ronnekleiv, O. K. (2003) Endocrine 20, 227-237[CrossRef][Medline] [Order article via Infotrieve]
  15. Becker, M. N., Brenner, R., and Atkinson, N. S. (1995) J. Neurosci. 15, 6250-6259[Abstract]
  16. Bohm, R. A., Wang, B., Brenner, R., and Atkinson, N. S. (2000) J. Exp. Biol. 203, 693-704[Abstract]
  17. Brenner, R., Thomas, T. O., Becker, M. N., and Atkinson, N. S. (1996) J. Neurosci. 16, 1827-1835[Abstract/Free Full Text]
  18. Chang, W. M., Bohm, R. A., Strauss, J. C., Kwan, T., Thomas, T., Cowmeadow, R. B., and Atkinson, N. S. (2000) J. Biol. Chem. 275, 3991-3998[Abstract/Free Full Text]
  19. Brenner, R., and Atkinson, N. (1996) Dev. Biol. 177, 536-543[CrossRef][Medline] [Order article via Infotrieve]
  20. Wallner, M., Meera, P., Ottolia, M., Kaczorowski, G. J., Latorre, R., Garcia, M. L., Stefani, E., and Toro, L. (1995) Receptors Channels 3, 185-199[Medline] [Order article via Infotrieve]
  21. Livak, K. J., and Schmittgen, T. D. (2001) Methods (San Diego) 25, 402-408
  22. Dignam, J. D., Lebovitz, R. M., and Roeder, R. G. (1983) Nucleic Acids Res. 11, 1475-1489[Abstract/Free Full Text]
  23. Knudsen, S. (1999) Bioinformatics (Oxf.) 15, 356-361[Abstract/Free Full Text]
  24. Down, T. A., and Hubbard, T. J. (2002) Genome Res. 12, 458-461[Abstract/Free Full Text]
  25. Davuluri, R. V., Grosse, I., and Zhang, M. Q. (2001) Nat. Genet. 29, 412-417[CrossRef][Medline] [Order article via Infotrieve]
  26. Frazer, K. A., Pachter, L., Poliakov, A., Rubin, E. M., and Dubchak, I. (2004) Nucleic Acids Res. 32, W273-W279[Abstract/Free Full Text]
  27. Schug, J. (2003) in Current Protocols in Bioinformatics (Baxevanis, A. D., Petsko, G. A., Stein, L. D., and Stormo, G. D., eds) pp. 1-16, Wiley Interscience, New York
  28. Bajic, V. B., Tan, S. L., Suzuki, Y., and Sugano, S. (2004) Nat. Biotechnol. 22, 1467-1473[CrossRef][Medline] [Order article via Infotrieve]
  29. Fukue, Y., Sumida, N., Tanase, J., and Ohyama, T. (2005) Nucleic Acids Res. 33, 3821-3827[Abstract/Free Full Text]
  30. Maston, G. A., Evans, S. K., and Green, M. R. (2006) Annu. Rev. Genomics Hum. Genet. 7, 29-59[CrossRef][Medline] [Order article via Infotrieve]
  31. Smale, S. T., and Kadonaga, J. T. (2003) Annu. Rev. Biochem. 72, 449-479[CrossRef][Medline] [Order article via Infotrieve]
  32. Lim, C. Y., Santoso, B., Boulay, T., Dong, E., Ohler, U., and Kadonaga, J. T. (2004) Genes Dev. 18, 1606-1617[Abstract/Free Full Text]
  33. Carninci, P., Sandelin, A., Lenhard, B., Katayama, S., Shimokawa, K., Ponjavic, J., Semple, C. A. M., Taylor, M. S., Engstrom, P. G., Frith, M. C., Forrest, A. R. R., Alkema, W. B., Tan, S. L., Plessy, C., Kodzius, R., Ravasi, T., Kasukawa, T., Fukuda, S., Kanamori-Katayama, M., Kitazume, Y., Kawaji, H., Kai, C., Nakamura, M., Konno, H., Nakano, K., Mottagui-Tabar, S., Arner, P., Chesi, A., Gustincich, S., Persichetti, F., Suzuki, H., Grimmond, S. M., Wells, C. A., Orlando, V., Wahlestedt, C., Liu, E. T., Harbers, M., Kawai, J., Bajic, V. B., Hume, D. A., and Hayashizaki, Y. (2006) Nat. Genet. 38, 626-635[CrossRef][Medline] [Order article via Infotrieve]
  34. Lee, M. P., Howcroft, K., Kotekar, A., Yang, H. H., Buetow, K. H., and Singer, D. S. (2005) Genome Res. 15, 1189-1197[Abstract/Free Full Text]
  35. Kumar, M. S., and Owens, G. K. (2003) Arterioscler. Thromb. Vasc. Biol. 23, 737-747[Abstract/Free Full Text]
  36. O'Lone, R., Frith, M. C., Karlsson, E. K., and Hansen, U. (2004) Mol. Endocrinol. 18, 1859-1875[Abstract/Free Full Text]
  37. Jakacka, M., Ito, M., Weiss, J., Chien, P. Y., Gehm, B. D., and Jameson, J. L. (2001) J. Biol. Chem. 276, 13615-13621[Abstract/Free Full Text]
  38. Takai, D., and Jones, P. A. (2003) In Silico Biol. 3, 235-240[Medline] [Order article via Infotrieve]
  39. Larsen, F., Gundersen, G., Lopez, R., and Prydz, H. (1992) Genomics 13, 1095-1107[CrossRef][Medline] [Order article via Infotrieve]
  40. Suzuki, Y., Taira, H., Tsunoda, T., Mizushima-Sugano, J., Sese, J., Hata, H., Ota, T., Isogai, T., Tanaka, T., Morishita, S., Okubo, K., Sakaki, Y., Nakamura, Y., Suyama, A., and Sugano, S. (2001) EMBO Rep. 2, 388-393[Medline] [Order article via Infotrieve]
  41. Dhulipala, P. D., and Kotlikoff, M. I. (1999) Biochim. Biophys. Acta 1444, 254-262[Medline] [Order article via Infotrieve]
  42. Ince, T. A., and Scotto, K. W. (1995) J. Biol. Chem. 270, 30249-30252[Abstract/Free Full Text]
  43. Yang, F., and Bleich, D. (2004) J. Biol. Chem. 279, 35403-35411[Abstract/Free Full Text]
  44. Vernier-Magnin, S., Nemos, C., Mansuy, V., Tolle, F., Guichard, L., Delage-Mourroux, R., Jouvenot, M., and Fraichard, A. (2005) Biochim. Biophys. Acta 1731, 23-31[Medline] [Order article via Infotrieve]
  45. Katayama, S., Tomaru, Y., Kasukawa, T., Waki, K., Nakanishi, M., Nakamura, M., Nishida, H., Yap, C. C., Suzuki, M., Kawai, J., Suzuki, H., Carninci, P., Hayashizaki, Y., Wells, C., Frith, M., Ravasi, T., Pang, K. C., Hallinan, J., Mattick, J., Hume, D. A., Lipovich, L., Batalov, S., Engstrom, P. G., Mizuno, Y., Faghihi, M. A., Sandelin, A., Chalk, A. M., Mottagui-Tabar, S., Liang, Z., Lenhard, B., and Wahlestedt, C. (2005) Science 309, 1564-1566[Abstract/Free Full Text]
  46. Schmid, C. D., Perier, R., Praz, V., and Bucher, P. (2006) Nucleic Acids Res. 34, D82-D85[Abstract/Free Full Text]
  47. Klinge, C. M., Jernigan, S. C., Smith, S. L., Tyulmenkov, V. V., and Kulakosky, P. C. (2001) Mol. Cell. Endocrinol. 174, 151-166[CrossRef][Medline] [Order article via Infotrieve]
  48. Hall, J. M., McDonnell, D. P., and Korach, K. S. (2002) Mol. Endocrinol. 16, 469-486[Abstract/Free Full Text]
  49. Sanchez, R., Nguyen, D., Rocha, W., White, J. H., and Mader, S. (2002) BioEssays 24, 244-254[CrossRef][Medline] [Order article via Infotrieve]
  50. Watanabe, T., Inoue, S., Hiroi, H., Orimo, A., Kawashima, H., and Muramatsu, M. (1998) Mol. Cell. Biol. 18, 442-449[Abstract/Free Full Text]
  51. Klinge, C. M. (2001) Nucleic Acids Res. 29, 2905-2919[Abstract/Free Full Text]
  52. Stokes, K., Alston-Mills, B., and Teng, C. (2004) J. Mol. Endocrinol. 33, 315-334[Abstract/Free Full Text]
  53. Krieg, A. J., Krieg, S. A., Ahn, B. S., and Shapiro, D. J. (2004) J. Biol. Chem. 279, 5025-5034[Abstract/Free Full Text]
  54. Dong, J., Tsai-Morris, C. H., and Dufau, M. L. (2006) J. Biol. Chem. 281, 18825-18836[Abstract/Free Full Text]
  55. Fleming, J. G. W., Spencer, T. E., Safe, S. H., and Bazer, F. W. (2006) Endocrinology 147, 899-911[CrossRef][Medline] [Order article via Infotrieve]
  56. Jacobson, D., Pribnow, D., Herson, P. S., Maylie, J., and Adelman, J. P. (2003) Biochem. Biophys. Res. Commun. 303, 660-668[CrossRef][Medline] [Order article via Infotrieve]
  57. Koos, R. D., Kazi, A. A., Roberson, M. S., and Jones, J. M. (2005) Ann. N.Y. Acad. Sci. 1041, 233-247[CrossRef][Medline] [Order article via Infotrieve]
  58. Higgins, K. J., Liu, S. G., Abdelrahim, M., Yoon, K., Vanderlaag, K., Porter, W., Metz, R. P., and Safe, S. (2006) Endocrinology 147, 3285-3295[CrossRef][Medline] [Order article via Infotrieve]
  59. Bjornstrom, L., and Sjoberg, M. (2005) Mol. Endocrinol. 19, 833-842[Abstract/Free Full Text]
  60. Chen, Y. H., Lee, M. J., Chang, H. H., Hung, P. F., and Kao, Y. H. (2006) Endocrinology 147, 4496-4504[Abstract/Free Full Text]
  61. Zhu, N., Eghbali, M., Helguera, G., Song, M., Stefani, E., and Toro, L. (2005) FEBS Lett. 579, 4856-4860[Medline] [Order article via Infotrieve]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 Am. J. Physiol. Cell Physiol.Home page
L. D. Ridgway, E. Y. Kim, and S. E. Dryer
MAGI-1 interacts with Slo1 channel proteins and suppresses Slo1 expression on the cell surface
Am J Physiol Cell Physiol, July 1, 2009; 297(1): C55 - C65.
[Abstract] [Full Text] [PDF]

Home page
 Mol Biol EvolHome page
K. N. Rohmann, D. L. Deitcher, and A. H. Bass
Calcium-Activated Potassium (BK) Channels Are Encoded by Duplicate slo1 Genes in Teleost Fishes
Mol. Biol. Evol., July 1, 2009; 26(7): 1509 - 1521.
[Abstract] [Full Text] [PDF]

Home page
 PhysiologyHome page
S. Hou, S. H. Heinemann, and T. Hoshi
Modulation of BKCa Channel Gating by Endogenous Signaling Molecules
Physiology, February 1, 2009; 24(1): 26 - 35.
[Abstract] [Full Text] [PDF]

Home page
 J Mol EndocrinolHome page
S. Safe and K. Kim
Non-classical genomic estrogen receptor (ER)/specificity protein and ER/activating protein-1 signaling pathways
J. Mol. Endocrinol., November 1, 2008; 41(5): 263 - 275.
[Abstract] [Full Text] [PDF]

Home page
 Hum Mol GenetHome page
M. A. Seibold, B. Wang, C. Eng, G. Kumar, K. B. Beckman, S. Sen, S. Choudhry, K. Meade, M. Lenoir, H. G. Watson, et al.
An african-specific functional polymorphism in KCNMB1 shows sex-specific association with asthma severity
Hum. Mol. Genet., September 1, 2008; 17(17): 2681 - 2690.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
282/37/27478    most recent
M704777200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
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 arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kundu, P.
Right arrow Articles by Toro, L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kundu, P.
Right arrow Articles by Toro, L.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 2007 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement