HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Pengue, G. Articles by Durmowicz, M. C. Search for Related Content PubMed PubMed Citation Articles by Pengue, G. Articles by Durmowicz, M. C. Social Bookmarking                      What's this?

J Biol Chem, Vol. 274, Issue 37, 26477-26484, September 10, 1999

Functional Characterization of the Promoter of the X-linked Ectodermal Dysplasia Gene^*

Gina Pengue, Anand K. Srivastava^§, Juha Kere^¶, David Schlessinger, and Meredith C. Durmowicz^**

From the  Department of Internal Medicine, Washington University School of Medicine, St. Louis, Missouri 63110, the ^§ J. C. Self Research Institute of Human Genetics, Greenwood Genetic Center, Greenwood, South Carolina 29646, the ^¶ Department of Medical Genetics, Haartman Institute, University of Helsinki, 00014 Helsinki, Finland, and the  Laboratory of Genetics, National Institute on Aging, Baltimore, Maryland 21224

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Anhidrotic ectodermal dysplasia (EDA) is a disorder characterized by poor development of hair, teeth, and sweat glands, and results from lesions in the X-linked EDA gene. We have cloned a 1.6-kilobase 5'-flanking region of the human EDA gene and used it to analyze features of transcriptional regulation. Primer extension analysis located a single transcription initiation site 264 base pairs (bp) upstream of the translation start site. When the intact cloned fragment or truncated derivatives were placed upstream of a reporter luciferase gene and transfected into a series of cultured cells, expression comparable with that conferred by an SV40 promoter-enhancer was observed. The region lacks a TATA box sequence, and basal transcription from the unique start site is dependent on two binding sites for the Sp1 transcription factor. One site lies 38 bp 5' to the transcription start site, in a 71-bp sequence that is sufficient to support up to 35% of maximal transcription. The functional importance of the Sp1 sites was demonstrated when cotransfection of an Sp1 expression vector transactivated the EDA promoter in the SL2 Drosophila cell line that otherwise lacks endogenous Sp1. Also, both Sp1 binding sites were active in footprinting and gel shift assays in the presence of either crude HeLa cell nuclear extract or purified Sp1 and lost activity when the binding sites were mutated. A second region involved in positive control was localized to a 40-bp sequence between 673 and 633 bp. This region activated an SV40 minimal promoter 4- to 5-fold in an orientation-independent manner and is thus inferred to contain an enhancer region.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Anhidrotic ectodermal dysplasia (EDA)¹ is an X-linked recessive disorder that affects the development of ectodermal structures (1). The gene responsible for the disorder was originally isolated by positional cloning (2). Additional exons of the EDA gene have recently been identified, bringing the total number of exons in the gene to twelve (3). Mutations in affected individuals have been characterized (2-4), and interruption of the orthologous gene in mouse leads to the Tabby phenotype (5, 6). Because affected individuals have sparse hair, rudimentary teeth, and no sweat glands, and Tabby mice show similar defects, the gene is believed to function at an early stage in ectodermal development, possibly at a branch point.

Some hints as to the function of the EDA protein have been gained by findings that it associates with the cell membrane and may participate in the regulation of cell-cell or cell-matrix interactions (3, 7). Consistent with a role in such interactions, exons of the gene encode collagen-like repeat motifs (2) that have been shown to form collagenous trimers in the extracellular domain of the EDA protein.²

Studies of EDA gene expression and protein function have been complicated by the fact that the EDA transcript undergoes alternative splicing and is capable of forming eight distinct isoforms, many of which can be detected by reverse transcriptase-polymerase chain reaction (PCR) in a variety of tissues (3). In addition, in situ hybridization and immunohistochemical analysis of various human embryonic, fetal, and adult tissues have demonstrated that the EDA gene and protein are expressed at low levels in several tissues unaffected in EDA as well as in the ectodermal tissues that develop abnormally (2, 9).

In this work, we have initiated studies to analyze the regulation of EDA gene expression. The minimal promoter region, including two Sp1 sites important for promoter function, has been defined. In addition, an enhancer region centered at 653 bp upstream of the transcription start site has been identified. The enhancer segment includes putative binding sites for two transcription factors known to regulate tissue- and developmental stage-specific expression of certain cardiac genes.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Cell Culture and DNA Transfection-- HeLa, 293, and HaCaT cell lines were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum. Transfections were done with 10 µg of reporter plasmids. The 293 and HeLa cell lines were transfected in subconfluent cultures by the calcium phosphate method. The HaCaT cell line was transfected using liposomes (DOTAP, from Roche Molecular Biochemicals) with 7 µl of DOTAP per microgram of DNA. To normalize transfection efficiencies, a plasmid expressing -galactosidase (pSV--Gal plasmid, Promega) was cotransfected with the test plasmid in each experiment. Promoter activity was normalized to protein concentration and -galactosidase activity. Drosophila melanogaster SL2 cells were grown in Schneider's medium supplemented with 10% heat-inactivated fetal calf serum and transfected using calcium phosphate precipitation. Cells were harvested 48 h after transfection, and extracts were assayed for luciferase activity according to the Promega protocol.

Plasmid Constructions-- To create the p-1625 plasmid, the p2A5 genomic clone (10) was digested with EcoRI (position 1608) and SmaI (position +258), and the single-stranded ends of the released fragment were made double-stranded using the Klenow fragment of DNA polymerase I. The fragment was then blunt-end ligated into the SmaI site of the promoterless pGL2-Basic vector (Promega). Other constructs included: plasmid p-785, constructed by digestion of p2A5 with AccI (position 768) and SmaI (position +258). The AccI site was filled in as above, and the resulting fragment was cloned into the SmaI site of pGL2-Basic.

p-338 was constructed by digestion of p-785 with PstI (positions 321 and +168); the fragment was rendered blunt-ended with T4 DNA polymerase and cloned into the SmaI site of pGL2-Basic.

p-88 was created using PCR. The promoter fragment (nt 71 to nt +63) was amplified using the upstream primer 5'-TCCCCCGGGTGGAGGCCCGGCT-3' and the downstream primer 5'-GAAGATCTCCCGCCGAGGGAAT, with SmaI and BglII sites (underlined), respectively, incorporated into the primers. The PCR product was then cloned between the SmaI and BglII sites of pGL2-Basic.

p-53 was constructed by digestion of p-785 with NarI (position 35) and SmaI (position +258), filled in with Klenow fragment, and cloned into the SmaI site of plasmid pGL2-Basic.

p125(5'3'), p103(5'3'), p83(5'3'), and p63(5'3') were constructed by PCR, using the primers listed below. A SmaI site (underlined) was included in each of the 5'-primers, and a BglII site was included in each of the 3'-primers. The amplified DNA fragments were subcloned into the SmaI/BglII sites of the plasmid pGL2-Promoter (Promega). p125(5'3') and p103(5'3') were constructed using the 5'-primers 5'-TCCCCCGGGTACAGGGATCGATAG-3' and 5'-TCCCCCGGGTGTTGAATTAATTAA-3' and the same 3'-primer, 5'-GAAGATCTAGTAACAGAGAAGC-3'. p83(5'3') and p63(5'3') were constructed using the 5'-primer 5'-TCCCCCGGGCAAGAAATCCTAGGA-3' and the 3'-primers 5'-GAAGATCTAGTAACAGAGAAGC-3' and 5'-GAAGATCTATGCCAAGCGGAACTG-3', respectively.

p125(3'5'), p103(3'5'), p83(3'5'), and p63(3'5') were similarly constructed by PCR, using the primers listed below. For these constructs, the BglII site was included in each of the 5'-primers, and the SmaI site was included in each of the 3'-primers. The DNA fragments were again subcloned into the SmaI/BglII sites of the plasmid pGL2-Promoter. p125(3'5'), p103(3'5'), and p83(3'5') were made using the 3'-primer 5'-TCCCCCGGGAGTAACAGAGAAGC-3' and the 5'-primers, 5'-GAAGATCTTACAGGGATCGATAG-3', 5'-GAAGATCTTGTTGAATTAATTAA-3', and 5'-GAAGATCTCAAGAAATCCTAGGA-3', respectively. p63(3'5') was constructed using the 3'-primer, 5'TCCCCCGGGATGCCAAGCGGAACTG-3', and the 5'-primer, 5'-GAAGATCTCAAGAAATCCTAGGA-3'.

pmBox1 and pmBox2 mutant constructs were created by PCR-mediated site-directed mutagenesis (11). The mutant plasmids contain GG to TT substitutions in both of the putative Sp1 boxes as follows: GGTTCGGGG (Box B1) and GGTTCGGAC (Box B2). The mutated bases are underlined. All plasmid constructs were analyzed by DNA sequencing to confirm that the constructions were correct. Plasmids pPac and pPac-Sp1 were kindly provided by Dr. Luigi Lania.

Primer Extension Analysis-- HeLa cells were transfected as described, and RNA was isolated from these cells using the SV Total RNA isolation system from Promega. An EDA-specific primer, PE1625-22 (5'-GCAGCTCTACTCCGAGGGGTGG-3'), was end-labeled with ³²P using T4 polynucleotide kinase, and primer extension reactions were carried out as described by Ordahl, et al. (12). The same primer was used in sequencing reactions that were done with the 33P Thermo Sequenase radiolabeled cycle sequencing terminator kit (Amersham Pharmacia Biotech). All samples were electrophoresed on a 6% polyacrylamide, 8 M urea sequencing gel in 1× TBE (0.1 M Tris-HCl, 90 mM boric acid, 1 mM EDTA, pH 8.0), dried, and exposed to x-ray film with an intensifying screen.

Electrophoretic Mobility Shift Assay-- Gel shift assays were performed with radiolabeled double-stranded oligonucleotides from the EDA gene, nt 52 to nt 22 (Box B1) and nt 197 to nt 167 (Box B2). The oligonucleotides were labeled with ³²P using T4 polynucleotide kinase. Nuclear extracts were prepared from 293, HeLa, and HaCaT cells, and DNA binding assays were performed as described (13). In competition experiments, unlabeled competitor was included in the preincubation reaction with the nuclear extract. Oligonucleotides containing the consensus sequence and mutated sequence for Sp1 binding (5'-ATTCGATCGGGGCGGGGCGAGC-3' and 5'-ATTCGATCGGTTCGGGGGAC-3', respectively) were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz).

For antibody "supershift" experiments, 0.5 and 1µg of polyclonal antisera specific for Sp1 and AP2 (Santa Cruz Biotechnology, Inc.) were added to the binding assay reactions and incubated on ice for 30 min before the radiolabeled oligonucleotides were added.

DNase I Footprinting-- Plasmid F1 (containing FragmentI) was constructed using PCR amplification of the EDA sequence from nt 102 to nt +63 with primers containing a SmaI site in the 5'-primer (5'-TCCCCCGGGGGCGAACCC-3') and a BglII site in the 3'-primer (5'-GAAGATCTCCCGCCGAGGGAAT-3'). The amplified fragment was again inserted into the SmaI/BglII sites of pGL2-Basic. Plasmid F1 was cleaved with BglII restriction endonuclease, labeled by filling in ends with [-³²P]dCTP using the Klenow fragment of DNA polymerase I, and digested with SmaI. Plasmid p-785 (FragmentII), described above, was digested with NarI, labeled by end-filling with [-³²P]dCTP and Klenow DNA polymerase, and then digested with PstI. Each labeled fragment was gel purified. Binding reactions were carried out using 30-50 µg of nuclear extract or 1-2 footprinting units of purified Sp1 fragment (Promega) and 30,000 cpm of probe. After DNase digestion, DNA fragments were analyzed on 6% polyacrylamide sequencing gels containing 8 M urea. The sequences of the binding sites were confirmed by Maxam-Gilbert G + A sequencing reactions performed on each fragment.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The primary findings are that, although the level of EDA mRNA in tissues is low (2, 7, 9), the promoter contains basal Sp1 elements and an enhancer region that sustain RNA transcription at a high level from a single initiation site. Because endogenous levels of the EDA transcript are extremely low (2, 7, 9), the studies here have been carried out with cells transiently transfected with EDA constructs to increase signal strength. Although EDA is very widely expressed, epithelial-derived cell lines have been used as they are likely to be more relevant for a gene involved in skin appendage formation.

A Single Transcription Initiation Site for EDA-- The transcription start site was determined by primer extension analysis carried out on RNA preparations from HeLa epithelial cells transfected with plasmid p-1625. This plasmid contains a 1.6-kilobase genomic fragment that includes 5'-upstream sequences and part of the known cDNA sequence and has been shown to direct high levels of transcription (see below). As shown in Fig. 1, lane 2, a single transcription start site was observed at nucleotide 3453 (G) of the EDA genomic sequence. DNA sequencing has revealed that the EDA gene lacks basal elements like the TATA box or an initiator sequence. The absence of such sites and the presence of Sp1 sites, including one in the basal promoter (see below), might have been expected to result in initiation of transcription at several locations. However, a single initiation site was consistently detected in these assays and in RNase protection assays (data not shown). This analysis extends the 5'-end of the EDA transcript 47-bp further than the published cDNA sequence.

View larger version (60K): [in this window] [in a new window]   Fig. 1.   Primer extension analysis to assess the transcription start site of EDA. Lane 1, 25 µg of tRNA; lane 2, 25 µg of total RNA from HeLa cells transfected with plasmid p-1625. Primer extension reactions were carried out on both RNA samples using the EDA-specific primer PE1625-22. Sequencing reactions were performed with the same primer on the noncoding strand of plasmid p-1625. The sequence listed to the right represents the sequence of the coding strand. The transcription initiation site (G) is indicated by an asterisk.

Deletion Analysis Defines Several Regulatory Sites in a Strong Promoter-- To test for promoter activity, a genomic fragment of 1.6 kilobases, including 5'-upstream sequences and part of the known cDNA sequence, was placed 5' to a luciferase reporter gene, and its capacity to direct luciferase synthesis in transfection experiments was compared with a series of deletion mutant constructs. Each truncated segment was cloned into the promoterless reporter plasmid pGL2-Basic and transfected into human epithelia-derived cell lines. Each of these cell lines (HeLa, 293, and HaCat) have been shown to exhibit low levels of endogenous EDA gene expression, as detected by reverse transcriptase-PCR, similar to levels found in other cell types and tissues (2, 3, 9). Cells were transfected as described under "Experimental Procedures," and after 48 h, cell extracts were prepared and luciferase activity was measured (Fig. 2). This index of promoter strength was normalized to the activity of a cotransfected -galactosidase reporter gene under SV40 promoter control.

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Deletion analysis of the EDA promoter region detected by a reporter gene. Top left, schematic of the 5'-flanking region of the gene from nt 1608 to nt +258. Some restriction enzyme sites used in the preparation of some constructs are indicated. An arrow indicates the transcription start site. Closed circles, consensus binding sequences Box B1 and Box B2 for Sp1; closed rectangle, enhancer binding region (E); below left, extent of DNA upstream of luciferase reporter gene in each chimeric construct; right, luciferase activity of 293, HeLa, and HaCaT cell lines transfected with each construct, along with transcriptional activities of the positive control (pGL2 SV40 promoter-enhancer) and the negative control (pGL2-Basic vector alone). The mean and standard deviation of at least four independent experiments, with each transfection done in duplicate, are shown, with values determined after normalization for an internal control of -galactosidase activity. , 293; , HeLa; , HaCat.

Promoter constructs retaining 1.6 kilobases to 35 bp of DNA upstream of the transcription initiation site varied in potency by more than 90%. Promoter activity as strong as an SV40 control promoter-enhancer (Fig. 2) was maintained in constructs deleted up to 767 bp. Removal of the region between 767 and 321 bp, however, decreased promoter activity in HeLa cells and HaCaT cells by 50%, and in 293 cells by 40%. A further drop of the remaining activity was obtained after removing sequence extending from 321 to 71 bp (by 50% in HeLa and in HaCaT, and by 40% in 293 cells).

As is often seen in transient expression assays of promoter activity, each of the promoter constructs exhibits a rather higher level of activity in one cell type rather than another (here, 293 cells compared with HeLa or HaCat cells). However, in all cell types, the subfragment of 71 bp immediately proximal to the transcription start site provides enough DNA sequence for basal transcription of the human EDA gene. It contains a single Sp1 site. In contrast, constructs further shortened to include only 35 bp upstream of the transcription initiation site, which eliminates the last Sp1 site, show no activity over the background from the promoterless luciferase vector pGL2 in HeLa and HaCaT cells. (The 35-bp construct transfected into 293 cells retained activity about 10% higher than the promoterless pGL2, probably because of a higher transfection efficiency.) Because the Sp1 binding sites ("GC boxes") bind transcription factor Sp1 and other members of the Sp1 multigene family (14), these results provide an indication that Sp1 or similar proteins may be important in the regulation of EDA expression.

DNase I Footprinting Analysis of the Putative Promoter Region-- Based on the transfection experiments of deletion mutant constructs (Fig. 2), the region between +63 and 321 bp was chosen for further analysis. DNase I footprinting analysis was used to map binding sites of nuclear factors, using nuclear extracts prepared from HeLa cells. Similar results were obtained from the HaCaT and 293 cell lines (data not shown). Each of two fragments of the promoter was 5'-end-labeled to generate a single-strand-labeled fragment for the assays. Fragment I (102 to +63 bp) showed one protected region (45 to 31 bp; Fig. 3, lane 2), containing a canonical GC box (Box B1), i.e. a putative binding site for Sp1.

View larger version (20K): [in this window] [in a new window]   Fig. 3.   DNase I footprinting analysis of the EDA promoter. 5'-End-labeled fragment, including the sequence from +63 to 102 bp (Fragment I) was incubated without extract (lane 1) or with 30 micrograms of HeLa nuclear extract protein (lane 2), or with 1 footprinting unit of purified Sp1 (lane 3) and partially digested with DNase I. To the right, the putative Sp1 binding site (BOX 1) is indicated by an open box.

It seemed likely that Sp1 might indeed bind to this region because it functions as an essential factor for several viral and cellular promoters (15-19). To establish whether this GC box is protected by Sp1, DNase I footprinting analysis was carried out with purified Sp1 protein. As shown in Fig. 3, lane 3, recombinant Sp1 protected the same GC-rich sequence noted in the experiment with HeLa cell nuclear extract, suggesting that Box B1 of the EDA promoter region can specifically interact with the transcription factor Sp1.

In repeated comparable experiments using the second probe (-37 to 321 bp), a second potential binding site for Sp1 was detected between nt 177 and nt 189 (Box B2). However, a higher background was observed, possibly because of nonspecific binding, which resulted in a significantly weaker signal (data not shown). Stronger evidence that Box B2 also specifically interacts with Sp1 was derived from further experiments using gel shift assays and transactivation studies with an Sp1 expression vector (see below).

Characterization of Two Functional Sp1 Sites-- To characterize further the functionality of the Sp1 sites, we performed electrophoretic mobility shift assays. We used 30-bp double-stranded oligonucleotides from positions 52 to 22 bp (Box B1) and 197 to 167 bp (Box B2).

Incubation of the Box B1 probe with the HeLa nuclear extract resulted in the formation of three complexes (complex I, complex II, and a less consistently observed very weak complex III (Fig. 4A, lane 1)). Similar binding patterns were observed for the Sp1 binding sites in comparable assays with crude nuclear extract (14). Complexes I and II were competed away by unlabeled oligonucleotide, verifying the specificity of the binding to DNA. The third complex was constitutively expressed and was not competed away by unlabeled probe; it is inferred to be a nonspecific binding site.

View larger version (40K): [in this window] [in a new window]   Fig. 4.   Gel mobility shift with DNA from 45 to 31 bp. An end-labeled double-stranded oligonucleotide covering the Box B1 region protected from DNase I in footprinting experiments (Fig. 3) was used as probe. Panel A, nuclear extract from HeLa cells was incubated with labeled oligonucleotide in the absence (lane 1) or after preincubation in the presence of increasing amounts (10-, 50-, 100-, and 500-fold molar excess) of unlabeled double-stranded competitor oligonucleotide (lanes 2-5). Specific complexes I and II are indicated to the right. An asterisk indicates a weak nonspecific complex (see text). Panel B, lane 1, reaction as in lane 1, panel A. Lanes 2 and 3, a polyclonal rabbit anti-Sp1 peptide antibody (lanes 2 and 3) or a control polyclonal anti-AP2 rabbit antibody (lanes 4 and 5) was incubated for 30 min with the nuclear extract from HeLa cells before the double-stranded oligonucleotide was added. The specifically shifted protein complex is indicated at the right. Panel C, nuclear extract from HeLa cells was incubated in the absence or in the presence of cold double-stranded competitor oligonucleotide which included the consensus binding site for Sp1 or a mutated consensus binding site for Sp1 (see "Experimental Procedures"). The double-stranded competitor oligonucleotide was used in 10-, 50-, 100-, and 500-fold molar excess, respectively, in lanes 2-5 and 6-9.

To establish that binding was specific for Sp1, increasing amounts of an antibody to Sp1 were added to the binding reaction, resulting in a "supershift" (a shift to lower mobility) of Complex I (Fig. 4B, lanes 2 and 3). The inability of Sp1 to supershift the complex fully may be because of the comigration of a complex with another member of the Sp1-family, e.g. Sp3 (14, 20).

A control antibody to the transcription factor AP-2, which is expressed in HeLa cells and can also bind some GC-rich sequences (20), had no effect on the mobility of the bound complex (Fig. 4B, lanes 4 and 5). To substantiate further that Sp1 bound to Box B1, competition experiments were performed in which unlabeled Box B1 oligonucleotides and double-stranded oligonucleotides containing the consensus Sp1 binding site or containing mutated Sp1 consensus sequence were added to the reactions and gel mobility shifts were again assessed. Complexes I and II were both competed for in a dose-dependent manner by Box B1 and Sp1 consensus double-stranded oligonucleotides (Fig. 4C, lanes 2-5) but not by increasing amounts of double-stranded oligonucleotides containing a mutated Sp1 binding site (Fig. 4C, lanes 6-9). Similar results were obtained using as a probe the putative Box B2 Sp1 site (Fig. 5). As shown in Fig. 5A, lane 1, two other minor nonspecific complexes were also observed.

View larger version (67K): [in this window] [in a new window]   Fig. 5.   Gel mobility shift with DNA from 177 to 189 bp. The end-labeled double-stranded oligonucleotide covering the Box B2 region protected from DNase I in footprinting assays was incubated with HeLa nuclear extract. Protocols as in Fig. 4, but with Box B2 oligonucleotide instead of Box B1. Panel A, binding reaction in the absence of unlabeled double-stranded oligonucleotide (lane 1) or in the presence of increasing amounts (10-, 50-, 100-, and 500-fold molar excess) of unlabeled double-stranded competitor oligonucleotide (lanes 2-5) before the labeled probe was added. Specific complexes I and II are indicated to the right. Three nonspecific complexes are indicated by asterisks. Panels B and C, as in Fig. 4, but with Box B2 oligonucleotide instead of Box B1.

Taken together, the results further demonstrate directly that Sp1 binds to both the Box B1 and Box B2 GC sequences.

Sp1 Transactivates the EDA Promoter in SL2 Cells-- As another confirmation of Sp1 function, we tested the extent of its transactivation of the EDA promoter in the SL2 cell line. D. melanogaster Schneider cell line SL2 is known to be devoid of endogenous Sp1-like activity and thus serves as a useful cell line to test Sp1 effects in vivo (16, 17). When reporter plasmids p-1625 and p-88, which contain two and one Sp1 binding sites, respectively, were cotransfected with increasing amounts of the eukaryotic expression vector pPAC-Sp1 (16), we observed a strong enhancement of transcription compared with the reporter vector pGL2-Basic (Fig. 6). Note that higher levels of added pPAC-Sp1 resulted in some decrease of transcription activity, perhaps because Sp1 cofactors were squelched; but the results confirm the ability of co-expressed Sp1 to induce expression from the EDA promoter, further supporting the functionality of the Sp1 sites.

View larger version (16K): [in this window] [in a new window]   Fig. 6.   Sp1 mediated-activation of EDA promoter in the Drosophila SL2 cell line. As indicated at the bottom, 2 µg of reporter plasmids p-1625, p-88, or vector alone (pGL2-Basic) were transfected along with increasing amounts of the effector plasmid pPac-Sp1 to provide Sp1 in trans. The results are presented as a mean of three independent transfection experiments. Duplicate transfections were performed in each experiment.

Mutagenesis Analysis of Sp1 Binding Sites-- To assess the importance of each Sp1 site for EDA gene expression, we assayed the effects of point mutations in each of them. Reporter constructs were created that differed by 2 bp from the wild type promoter (548 to +63 bp; see "Experimental Procedures") and were assayed for luciferase activity in HeLa cells. The promoter activity from the mutant constructs pmBox1 and pmBox2 was decreased to 40 and 10% of the wild type promoter, respectively (Fig. 7). These results suggest that Box B2 may be functionally more important than Box B1; however, both sites were required for maximal promoter activity.

View larger version (17K): [in this window] [in a new window]   Fig. 7.   Loss of Sp1 activation at mutated binding sites. Wild-type and mutant promoter constructs were placed upstream of the luciferase gene in pGL2-Basic. 10 µg of each construct or pGL2-Basic vector was transfected into HeLa cells as indicated, along with 2 µg of -galactosidase expression vector cotransfected in each case to normalize for transcription efficiency. All transfections were done in duplicate, and all values shown are the average of at least three different experiments.

Putative Enhancer Region-- Another positive regulatory sequence extending from 673 to 550 bp was detected using constructs with this fragment cloned in both orientations into the pGL2-Promoter vector. This vector contains an SV40 minimal promoter that drives the expression of the luciferase reporter gene. When constructs were transfected into HeLa cells, the EDA promoter fragment induced the production of luciferase by 4- to 5-fold compared with the level observed for the vector alone. Consistent with the notion that this DNA segment contains an enhancer binding site, the activity of the cloned region was very similar in both cloned orientations (Fig. 8A).

View larger version (13K): [in this window] [in a new window]   Fig. 8.   Transcriptional activity of the EDA enhancer region. Panel A, plasmids truncated to localize the EDA enhancer were constructed by insertion, in both orientations, of the indicated region upstream of the luciferase gene in pGL2-Promoter vector (see "Experimental Procedures"). Each enhancer construct was transfected into HeLa cells. Luciferase activity is given as the -fold induction over the background pGL2-Promoter activity (that is, without insert). All transfections were done in duplicate, and values shown are the average of at least three different experiments after normalization for the internal control -galactosidase activity. Panel B, the DNA sequence of the fragment in plasmid p-125 was searched against the MATInspector data base of transcription factor binding sites (21). A potential GATA factor binding site is underlined as is a potential Nkx-2 factor binding site.

To define the core region that still gave enhancer activity, smaller fragments of the region were tested by repeated transfection experiments. As shown in Fig. 8A, enhancer activity was slightly reduced when the region between 673 and 653 bp was removed and completely abolished when the fragment was truncated further to 633 bp upstream of the transcription start site. From these results, we concluded that the 40-bp fragment from nt 673 to nt 633 contains regulatory sequence(s) sufficient to enhance the transcriptional activity of the SV40 minimal promoter. Possible binding sites for additional transcription factors are present in this sequence (Fig. 8B) and can guide further analysis (see "Discussion").

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The initial characterization of the 5'-flanking region of the EDA gene provides some insight into the minimal promoter structure and the regulation of its expression but also raises some questions. In particular, 1) how is transcription initiation limited to a single site in the absence of standard initiating signals; 2) how can the strength of the promoter be reconciled with the very low steady-state levels of EDA mRNA (2); and 3) what is the relevance of the apparent enhancer locus and a putative neighboring LEF-1 binding site to the regulation of the gene?

Minimal Promoter Structure and Transcription Initiation-- The results clearly define a minimal promoter region for EDA with activation of transcription from the human EDA promoter that is highly dependent upon transcription factor Sp1. Two "GC" boxes are functionally identical to Sp1 binding sites based on the following observations. From assays of shifts in electrophoretic mobility, the major Box B1 and Box B2 binding activity is recognized by Sp1 antibodies. In competition experiments, both boxes are competed for by an oligonucleotide carrying a consensus Sp1 binding site. The other, uncharacterized complex observed in the experiments may involve another member of the Sp1 multigene family because binding was effectively competed by an Sp1 consensus binding site oligonucleotide.

Consistent with a role for Sp1 in the regulation of expression of the EDA gene, DNase I footprinting experiments detect the binding of Sp1 recombinant protein to the two "GC" boxes. Furthermore, the EDA promoter is activated by Sp1 in the D. melanogaster Schneider cell line. Finally, comparison of the expression of wild-type and Sp1-mutant EDA-luciferase vectors in several cell lines shows that the EDA promoter activity is reduced when two point mutations are introduced into each Sp1 binding site.

Sp1 plays an important role in a number of other promoters, including those for SV40 (16, 22, 23), rat transforming growth factor- (TGF-) (24), cell cycle-regulated genes like thymidine kinase (TK) (25), dihydrofolate reductase (DHFR) (26) and b-myb (27). In all of these promoters, Sp1 elements are required for efficient transcription. The Sp1 binding site is also critical for the maintenance of the methylation-free CpG island of the adenine phosphoribosyltransferase gene (APRT) (28, 29) and may prevent methylation of at least a subset of CpG islands in the genome (30). For many TATA-less promoters, activation requires the multisubunit TFIID complex (16, 31), with Sp1 binding at GC boxes functioning to stabilize the interaction of TFIID with the transcription start site (32, 33). However, only part of the EDA minimal promoter function can be explained by such a model. It remains unexplained that the EDA gene has a single transcription start site (Fig. 1). Possibly, alternative sequence(s) within this promoter functions to regulate transcriptional initiation. It is also unclear how cells limit the steady state amount of EDA mRNA at a very low level (2, 9) in the face of strong promoter activity. At present, open possibilities include 1) negative transcriptional regulators elsewhere in the sequence and 2) instability of the mRNA.

Enhancer and Other Regulatory Sequences-- Two other promoter region sequences that are likely to be active in the transcriptional regulation of EDA have been detected. One of them is an enhancer element (nt 673 to nt 633) that is essential for the full activity of the EDA promoter and can boost the activity of an SV40 promoter by 4-fold or more when cloned in either orientation.

The 40-bp sequence includes a potential binding site for members of the GATA family of transcription factors (34, 35) as well as a potential weak affinity binding site for members of the Nkx-2 family of homeobox proteins (36) (Fig. 8B). GATA-4 and Nkx-2.5, members of each of these families, have been shown to interact and coactivate the transcription of cardiac-specific promoters (37). This interaction is proposed to play a role in the regulation of transcriptional activity in the embryo during early cardiogenesis. Interestingly, abundant expression of the EDA protein is seen in muscle cells of the developing heart (7). The potential role of either GATA or Nkx-2 family members in the transcriptional regulation of the EDA gene is currently under investigation.

Synergistic interactions between multiprotein complexes are known to affect both specificity and levels of transcription (38). For example, Sp1 is known to interact with both members of the GATA family (39-41) and members of the Nkx-2 family (42) to affect tissue-specific and developmentally restricted expression of target promoters. Also, the C-terminal domain of Sp1 interacts with other transcription factors (43). Speculatively, Sp1-TAF interactions at each of the Sp1 binding sites might, for example, facilitate enhancer interaction with the assembled machinery.

It is suggestive that the enhancer element inferred here lies next to a putative binding site for the transcription factor LEF-1 (nt 372 to nt 366; see Fig. 9) that is conserved between the human and mouse genomic sequences (5). LEF-1 belongs to a class of DNA-binding transcriptional regulatory proteins, referred to as "architectural factors", that induce directed bends in DNA and promote interactions between proteins bound at nonadjacent sites within complex regulatory DNA elements (38). In addition, LEF-1 can form a functional transcription factor complex with -catenin as a result of Wnt/wingless signaling (44-46). It could be relevant to the function of the EDA promoter and enhancer, because recent studies have implicated the Wnt pathway and especially LEF-1 and -catenin in the regulation of hair follicle formation (47, 48). Mice that overexpress a stabilized form of -catenin undergo de novo hair follicle morphogenesis (49), and mice in which the LEF-1 gene is disrupted show features similar to those observed in ectodermal dysplasia (8). The function of LEF-1 in the regulation of EDA transcription is the subject of ongoing studies.

View larger version (52K): [in this window] [in a new window]   Fig. 9.   Sequence of the EDA promoter region. GenBankTM accession number AF040628 (10). All regulatory elements described in the text are indicated. The core 40-bp enhancer region (nt 673 to nt 633) is underlined, and a potential LEF-1 binding site between nt 372 and nt 366 is boxed. Box B1 and Box B2 (from nt 189 to nt 177 and from nt 45 to nt 31, respectively), which contain the two functional Sp1 sites (in bold type), are double underlined.

In summary, current data suggest that the EDA promoter can be functionally divided into at least two regions. The first is a region that contains Sp1 binding sites, exhibits promoter activity, and binds the basal transcription machinery. The second region, which requires further investigation to define binding protein(s), enhances EDA promoter activity and also contains a putative element that may exert effects by binding to transcription factor LEF-1.

	ACKNOWLEDGEMENTS

We thank Dr. Luigi Lania for the pPac-Sp1 plasmid and D. melanogaster Schneider cell line. We are also grateful to Dr. Maurizio D'Esposito and Dr. Richard Mazzarella for helpful discussion and Dr. Michael Iademarco for reading the manuscript.

	FOOTNOTES

^* The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The nucleotide sequence(s) reported in this paper has been submitted to the GenBank^TM/EMBL Data Bank with accession number(s) AF040628.

^** To whom correspondence should be addressed: Laboratory of Genetics, National Institute on Aging, Triad Technology Center, Suite 4000, 333 Cassell Dr., Baltimore, MD 21224. Tel.: 410-558-8300, ext. 7087; Fax: 410-558-8331; E-mail: durmowiczm@grc.nia.nih.gov.

² S. Ezer, M. Bayes, O. Elomaa, D. Schlessinger, and J. Kere, submitted for publication.

	ABBREVIATIONS

The abbreviations used are: EDA, anhidrotic ectodermal dysplasia; nt, nucleotide(s); bp, base pair(s); PCR, polymerase chain reaction.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				C. Drogemuller, O. Distl, and T. Leeb Partial Deletion of the Bovine ED1 Gene Causes Anhidrotic Ectodermal Dysplasia in Cattle Genome Res., October 1, 2001; 11(10): 1699 - 1705. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Pengue, G. Articles by Durmowicz, M. C. Search for Related Content