JBC Anatrace, Inc.

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


     

Originally published In Press as doi:10.1074/jbc.M105682200 on January 4, 2002

J. Biol. Chem., Vol. 277, Issue 11, 8989-8998, March 15, 2002
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
277/11/8989    most recent
M105682200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Jiang, P.
Right arrow Articles by Xu, X.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Jiang, P.
Right arrow Articles by Xu, X.

Cloning and Characterization of the Human Heparanase-1 (HPR1) Gene Promoter

ROLE OF GA-BINDING PROTEIN AND Sp1 IN REGULATING HPR1 BASAL PROMOTER ACTIVITY*

Ping JiangDagger , Aseem Kumar§, Joseph E. Parrillo§, Laurie A. Dempsey||, Jeffrey L. Platt**Dagger Dagger , Richard A. PrinzDagger , and Xiulong XuDagger §§

From the Dagger  Department of General Surgery and the § Division of Cardiovascular Diseases and Critical Care, Department of Medicine, Rush-Presbyterian-St. Luke's Medical Center, Chicago, Illinois 60612 and the Departments of  Surgery, ** Immunology, and Dagger Dagger  Pediatrics, Mayo Clinic, Rochester, Minnesota 55905

Received for publication, June 20, 2001, and in revised form, December 21, 2001

    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

Heparanase-1 (HPR1) is an endoglycosidase that specifically degrades the heparan sulfate chains of proteoglycan, a component of blood vessel walls and the extracellular matrix. Recent studies demonstrated that HPR1 expression is increased in a variety of malignancies and may play a critical role in tumor metastases. The HPR1 gene and its genomic structure have been recently cloned and characterized. To understand the mechanisms of HPR1 gene expression and regulation, we first mapped the transcription start site of the HPR1 gene and found that HPR1 mRNA was transcribed from the nucleotide position 101 bp upstream of the ATG codon. A 3.5-kb promoter region of the HPR1 gene was cloned. Sequence analysis revealed that the TATA-less, GC-rich promoter of the HPR1 gene belongs to the family of housekeeping genes. This 3.5-kb promoter region exhibited strong promoter activity in two thyroid tumor cell lines. Truncation analysis of the HPR1 promoter identified a minimal 0.3-kb region that had strong basal promoter activity. Truncation and mutational analysis of the HPR1 promoter revealed three Sp1 sites and four Ets-relevant elements (ERE) significantly contributing to basal HPR1 promoter activity. Binding to the Sp1 sites by Sp1 and to the ERE sites by GA-binding protein (GABP) was confirmed by electrophoretic mobility shift assay and competition and supershift electrophoretic mobility shift assays. Cotransfection of Sp- and GABP-deficient Drosophila SL-2 cells with the HPR1 promoter-driven luciferase construct plus the expression vector encoding the Sp1, Sp3, or GABP gene induced luciferase gene expression. Mutation or truncation of the Sp1 or ERE sites reduced luciferase expression in both SL-2 cells and thyroid tumor cell lines. Coexpression of GABPalpha /beta and Sp1 or Sp3 further increased luciferase reporter gene expression. Our results collectively suggest that Sp1 cooperates with GABP to regulate HPR1 promoter activity.

    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

Heparanases are endo-beta -glucuronidases that specifically degrade the heparan sulfate chains of proteoglycan, one of the chief components in the cell membrane and extracellular matrix (1-3). HPR1 activity can be detected in hematopoietic cell types such as platelets (4), neutrophils (5, 6), activated T lymphocytes (7) and monocytes (8) and in many malignant tissues such as prostate carcinoma cells, melanomas, and murine B lymphoma as well as murine and human fibrosarcoma and melanoma cell lines (9-11). Increased HPR activity can be detected in the sera (12) and urine of metastatic tumor-bearing animals and cancer patients (13). Overexpression of HPR in non-metastatic human leukemia T cell lines confers the ability of these cells to metastasize (13). These observations collectively suggest that not only is HPR required for immune cells to migrate to local inflammatory sites, but that it also plays a critical role in tumor metastases.

Two heparanases, HPR1 and HPR2, have been recently characterized (13-18). The HPR1 gene encodes a protein of 543 amino acids that contains a predicted signal peptide of 35 residues, but lacks any recognizable membrane anchor sequence (13, 15, 16, 18). The HPR1 gene, spreading in an ~50-kb region covering 14 exons and 13 introns, was mapped to a single locus on chromosome 4q21 using fluorescent in situ hybridization of metaphase chromosomes (14, 19, 20). Northern blot analysis revealed that HPR1 mRNA exists mainly in two different forms, probably because of alternate splicing (15, 16, 18). A 1.7-kb HPR1 mRNA is expressed at high levels in placenta and peripheral blood leukocytes and at moderate levels in immune organs, but not in heart, brain, lung, liver, skeletal muscle, kidney, colon, stomach, or pancreas (15, 16, 18). A 5.0-kb HPR1 mRNA is present at low levels in most nonimmune and immune organs (15, 16, 18).

Given that HPR1 plays an important role in tumor metastases, it is crucial to understand the molecular mechanisms of HPR1 gene expression and regulation. In this study, we first mapped the transcription start site of the HPR1 gene. A 3.5-kb HPR1 promoter was cloned and sequenced. Functional analysis of the HPR1 promoter revealed three Sp1 sites and four Ets-relevant elements in a minimal promoter region of the HPR1 gene. Mutational and functional analysis and cotransfection studies indicated that Sp1 and GABP can cooperatively regulate HPR1 promoter activity. Thus, our study establishes a molecular basis for further understanding the mechanisms governing HPR1 gene expression.

    EXPERIMENTAL PROCEDURES
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

Materials-- Three thyroid tumor cell lines (NPA87, papillary carcinoma; KAT-4, anaplastic carcinoma; and MRO87, follicular carcinoma) were kindly provided by Dr. G. J. F. Juillard (University of California, Los Angeles, CA) and K. B. Ain (University of Kentucky Medical Center, Lexington, KY). NPA87 and MRO87 cells express HPR1 at high levels when analyzed by reverse transcription-PCR and Western blotting, whereas HPR1 expression is undetectable in KAT-4 cells when analyzed by reverse transcription-PCR, but can be detected by Western blotting.2 These cell lines were grown in complete RPMI 1640 medium containing 10% fetal bovine serum. SL-2 cells, an Sp- and GABP-deficient Drosophila cell line (kindly provided by Dr. R. L. Widow), were grown in Schneider's medium supplemented with 10% fetal bovine serum. pPac, pPac/Sp1, and pPac/Sp3 expression vectors were kindly provided by Drs. R. Tjian (University of California, Berkeley, CA) and G. Suske (Philipps-Universität Marburg, Marburg, Germany). Anti-Sp1 monoclonal antibody and goat anti-Sp3 IgG were purchased from Santa Cruz Biotechnology (San Diego, CA). Rabbit anti-GABPalpha and anti-GABPbeta antisera were kindly provided by Dr. U. A. Rapp (University of Würzburg, Würzburg, Germany). Restriction and modification enzymes were purchased from Invitrogen (Carlsbad, CA). Luciferase substrate, cell lysis buffer, and the pGL3/Basic plasmid were purchased from Promega (Madison, WI). Poly(dI-dC)·poly(dI-dC) was purchased from Amersham Biosciences, Inc.

Mapping of the Transcription Start Site of the HPR1 Gene-- The transcription start site of the HPR1 gene in MRO87 and KAT-4 cells was mapped using the RLM-RACE kit (Ambion Inc., Austin, TX) following the manufacturer's protocol. Briefly, total RNA was extracted from MRO87 and cycloheximide-treated KAT-4 cells using TRIzol. (Cycloheximide treatment dramatically induces HPR1 mRNA accumulation.2) Three antisense primers corresponding to the untranslated and encoding regions (see Fig. 1A) were used in two nested PCRs. The primary PCR was conducted using antisense gene-specific primer 1 (GSP1; 5'-GTAGTGATGCCATGTAACTGAATC-3', complementary to nucleotides +894 to +871, with the A nucleotide of ATG designated as position +1) and outer adapter primer 1 (AP1). The PCR conditions were as follows: 94 °C for 2 min; 35 cycles of 94 °C for 45 s, 60 °C for 30 s, and 72 °C for 1.5 min; and 72 °C for 7 min. The secondary PCR was conducted using antisense gene-specific primer 2 (GSP2; (5'-GCAGGCTTCGAGCGCAGCAGCATCTTG-3', complementary to nucleotides +23 to -4) and inner adapter primer 2 (AP2). The PCR conditions were as follows: 94 °C for 2 min; 35 cycles of 94 °C for 30 s, 60 °C for 30 s, and 72 °C for 30 s or 1.5 min; and 72 °C for 7 min. Me2SO (5% final concentration) was added to the primary PCR to facilitate amplification of the GC-rich region of the 5'-end region of the HPR1 cDNA.

Dong et al. (20) recently suggested that HPR1 gene transcription may be initiated from two different sites. To explore this possibility, we conducted RLM-RACE using a second set of the HPR1 gene-specific primers. The primary PCR was conducted using GSP2 and AP1. The PCR conditions were as follows: 94 °C for 2 min; 35 cycles of 94 °C for 45 s, 60 °C for 30 s, and 72 °C for 1.5 min; and 72 °C for 7 min. The secondary PCR was conducted using GSP3 (5'-GAGAGTCGAGAGCTCTAGCACTTCCTC-3' (complementary to nucleotides -45 to -71) and AP2. The same PCR conditions as described above were used. The PCR products were analyzed on a 1.5% agarose gel, and the DNA band was excised, extracted, and used as a template in the sequencing reactions.

Cloning of the HPR1 Promoter-- The HPR1 promoter was cloned using the GenomeWalker kit (CLONTECH, Palo Alto, CA). PCRs were conducted using two sense primers supplied with the kit and two synthesized antisense oligonucleotides (GSP2 and GSP3) complementary to the 5'-untranslated region of HPR1 cDNA. Me2SO (5% final concentration) was added to the primary and secondary PCRs to facilitate amplification of the GC-rich HPR1 promoter. An ~3.5-kb PCR product from the PvuII library and an ~0.7-kb PCR product from the DraI library were obtained. The PCR fragments were then treated with Pfu DNA polymerase to polish the 3'-end overhang, phosphorylated by T4 polynucleotide kinase, and ligated to SmaI-digested and calf intestine phosphatase-treated pGL3/Basic reporter construct. The ligation reaction was used to transform competent Escherichia coli DH5alpha cells. Plasmid DNA was extracted, and the orientation was identified by MluI digestion. DNA sequencing was conducted in the University of Chicago Cancer Research Center. The nucleotide sequence of the HPR1 promoter has been submitted to the GenBankTM/EBI Data Bank under accession number AF461265.

Manipulation of Plasmid DNA-- The D3 construct was generated by digesting pGL3/HPR-3.5 DNA with BamHI, followed by treatment with Pfu DNA polymerase to make a blunt end. The DNA fragment was then digested with BglII to produce a 300-bp fragment, which was then ligated to SmaI- and BglII-digested pGL3/Basic vector. The D2 con- struct was generated by digesting the D3 construct with the BglI restriction enzyme, followed by mung bean nuclease treatment to blunt the 5'-protruding end. The linearized fragment was then digested with HindIII (a HindIII site is in the pGL3/Basic vector), and an ~200-bp fragment was extracted from the agarose gel and ligated to SmaI- and HindIII-digested pGL3/Basic. To generate the D3.2, D3.1, D2.1, and D2.2 constructs, four HindIII-tagged oligonucleotides flanking the sequence starting from 268, 239, 113, and 98 bp upstream of the ATG codon were synthesized and used in the PCRs to amplify the truncated HPR1 promoters with sizes of 268, 239, 113, and 98 bp, respectively. The D2M and D2.1M constructs were similarly generated using two oligonucleotides (5'-GAAGGTACCAGGCGGTTCGGGGTTGGATTGG-3' and 5'-AAAGGTACCGTAACGGTTCGGAGGAAAGGAG-3') with mutated Sp1 sites flanking nucleotides -209 and -113, respectively. The D3M1, D3M2, D3M3, D3M4, D3M5, D3M6, and D3M7 constructs were generated using the PCR-based QuikChange kit (Stratagene, La Jolla, CA) with the ERE site-mutated primers listed in Fig. 7A following the manufacturer's instruction. The D4 construct was generated by digesting the D3 construct with the BglI restriction enzyme, followed by mung bean nuclease treatment to blunt the 5'-protruding end. The linearized fragment was then digested with MluI (an MluI site is present in the original pGL/Basic vector), and an ~100-bp fragment was extracted from the agarose gel and ligated to SmaI- and MluI- digested pGL3/Basic.

Transfection-- Thyroid tumor cell lines and SL-2 cells were grown in 24-well plates. Upon 80% confluence, the cells were transfected with pGL3/Basic or HPR1 promoter-driven luciferase plasmid using FuGENETM 6 transfection reagent (Roche Molecular Biochemicals) following the manufacturer's instructions. Briefly, 100 µl of serum-free medium was premixed with 4.5 µl of FuGENETM 6 transfection reagent, incubated at room temperature for 5 min, and then transferred to a tube containing 1 µg of luciferase reporter construct plus 0.5 µg of pCMV/SPORT as an internal control. The DNA/FuGENETM 6 mixture was aliquoted to six wells (16 µl/well) and incubated for 24 h. Cells were washed twice with ice-cold phosphate-buffered saline and harvested. Cell lysates were prepared. Luciferase activity was quantitated using a luciferase substrate kit (Promega) and read in a Packard luminometer.

Preparation of Nuclear Extracts and EMSA-- Nuclear extracts were prepared from thyroid tumor cell lines (1 × 107/ml) following a standard protocol as previously described (21). Double-stranded oligonucleotides were end-labeled with T4 polynucleotide kinase. The labeled probes were separated by NucTrap probe purification columns. The sequences of two canonical Sp sites derived from the promoter of the HPR1 gene are 5'-GGGCAGGCGGGGCGGGGTTGGGAT-3' (Sp1-B) and 5'-GGCGTAACGGGGCGGAGGAAAGG-3' (Sp1-A). The sequence of a non-canonical Sp1 site (Sp1-C) is 5'-TCCCGGCCATCTCCGCACCCTTCAAGTGGGTGTGGGTGAT-3'. The sequences of two canonical Sp1 sites with mutation of the Sp1 site are 5'-GGGCAGGCGGTTCGGGGTTGGGAT-3' (Sp1-Bm) and 5'-GGCGTAACGGTTCGGAGGAAAGG-3' (Sp1-Am) (mutated nucleotides are shown in boldface). The sequence of a consensus Sp1 probe is 5'-ATTCGATCGGGGCGGGGCGAGC-3'. An unrelated oligonucleotide (interferon-gamma activation site) was synthesized with the sequence 5'-ATTTTCCCCGAAAT-3' and used as a negative control in competition EMSA. The sequences of the oligonucleotides containing ERE sites were synthesized and used for EMSA as well as for mutation of the ERE site (see Fig. 7A). A consensus nuclear fac- tor-kappa B probe (5'-AGTTGAGGGGACTTTCCCAGGC-3') and GABP probe (5'-AGCTTGCGGAACGGAAGCGGAAACCGCCGGATCG-3') derived from the ICP4 gene of human herpes simplex virus-1 were synthesized and used as negative and positive controls in competition EMSA (see Fig. 8), respectively. Labeled oligonucleotide (20,000-50,000 cpm) was incubated at room temperature for 20 min with 2 µg of nuclear extract protein. The DNA binding buffer contained 4% glycerol, 1 mM MgCl2, 0.5 mM EDTA, 0.5 mM dithiothreitol, 50 mM NaCl, 10 mM Tris-HCl (pH 7.5), 50 µg/ml poly(dI-dC)·poly(dI-dC), and 1 µg of bovine serum albumin/reaction. For competition EMSA, nuclear extract (2-5 µg/reaction) was preincubated with 5-, 20-, 100-fold molar excesses of unlabeled oligonucleotide at 4 °C for 30 min, and then labeled probes were added and incubated at room temperature for 20 min. For supershift EMSA, the 32P-labeled probe (30,000 cpm/sample) was incubated with 2 µg of nuclear extract from KAT-4 cells (see Fig. 4B) or NPA87 cells (see Fig. 7B) at room temperature for 20 min. Antibody against Sp1, Sp3 (1 µg/sample), GABPalpha , and/or GABPbeta was added and incubated on ice for 30 min. Normal mouse IgG and normal goat IgG were included as negative controls. The reactions were separated on a 5% nondenaturing polyacrylamide gel. The dried gel was exposed to X-Omat film.

    RESULTS
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

Mapping of the Transcription Start Site-- Cloning of the HPR1 gene by several groups revealed the sequence variation at the 5'-end of the HPR1 cDNA (13-16, 18). For example, Vlodavsky et al. (13), using a traditional RACE assay, identified the 5'-end of the HPR1 gene at the nucleotide position 99 bp upstream of the translation start site, whereas a recent study by Dong et al. (20) suggested that HPR1 mRNA transcribed in two different splicing forms could result from differential transcription initiation. Here we employed a modified RLM-RACE method to map the 5'-end of the HPR1 gene. Unlike the classic RACE, which has no selection for amplification of the full-length mRNA or degraded mRNA, but rather favors amplification of the degraded mRNA, RLM-RACE selectively amplifies the DNA fragment from the capped full-length mRNA. Two sets of antisense primers complementary to the three locations of the HPR1 gene were synthesized and used in two RLM-RACE reactions (Fig. 1A). As shown in Fig. 1B, RLM-RACE using GSP1 and GSP2 produced a PCR product of ~160 bp (lane 4), whereas RLM-RACE with GSP2 and GSP3 produced a PCR product of ~90 bp (lane 2). The PCR product derived from the RLM-RACE reaction with GSP1 and GSP2 was extracted from the agarose gel and used as the template in a sequencing reaction. Sequence analysis revealed that transcription of the HPR1 gene started at the nucleotide position 101 bp upstream of the ATG site (Fig. 2), which is two nucleotides upstream of the 5'-end previously identified by Vlodavsky et al. (13) using classic RACE. Prolonging the extension time in the primary and secondary PCRs did not result in amplification of a larger PCR product (data not shown). These results indicate that the transcription start site of the HPR1 gene in both KAT-4 and MRO87 cell lines is located at the nucleotide position 101 bp upstream of the ATG site.


View larger version (60K):
[in this window]
[in a new window]
  		Fig. 1.   RLM-RACE analysis of the transcription start site of the HPR1 gene. A, schematic presentation of the primers used in RLM-RACE. Two RLM-RACE assays were conducted using two adapter primers (AP1 and AP2) and three HPR1 gene-specific antisense primers (GSP1, GSP2, and GSP3, corresponding to nucleotides +871, +23, and -45, respectively). In the first RLM-RACE, AP1-GSP1 and AP2-GSP2 primer pairs were used in the primary and secondary PCRs, respectively. In the second RLM-RACE, AP1-GSP2 and AP2-GSP3 primer pairs were used in the primary and secondary PCRs, respectively. B, analysis of the PCR products on agarose gel. Total RNA was extracted from MRO87 and KAT-4 cells using TRIzol. RLM-RACE was conducted as described under "Experimental Procedures" following the manufacturer's instructions. UTR, untranslated region.


View larger version (77K):
[in this window]
[in a new window]
  		Fig. 2.   Nucleotide sequence of the human HPR1 promoter and its 5'-flanking cDNA region. The ATG site is shown in boldface and designated as +1. Two gene-specific primers (GSP2 and GSP3) used in the PCR-based genome walking and mapping of the 5'-end of HPR1 mRNA are dash-underlined. Potential transcription factor-binding sites are solid-underlined. Two restriction enzyme sites (BamHI and BglI) are wavy-underlined. HPR1 cDNA sequence is in lowercase letters. The accession number of the HPR1 promoter in the GenBankTM/EBI Data Bank is AF461265.

Cloning of the Human HPR1 Gene Promoter-- A 3.5-kb DNA fragment of the HPR1 promoter was amplified using the GenomeWalker kit with two HPR1 gene-specific primers derived from the 5'-untranslated region of the HPR1 cDNA (Fig. 2). The final PCR product was polished by Pfu DNA polymerase and then blunt-ligated to SmaI-digested pGL3/Basic luciferase reporter vector. Computer analysis of the 3'-HPR1 promoter (Fig. 2) revealed a highly GC-rich content in its proximal promoter region. A GRAIL search identified a CpG island of 396 bp with a GC content of 58.83% in the first 700-bp region. A BLAST search of known sequences in the GenBankTM/EBI Data Bank was used to identify DNA homology. No significant homologous sequence was present. The HPR1 promoter lacks a typical TATA or CCAAT box, as seen with many GC-rich promoters. The TESS search program predicted a number of potential transcription factor-binding sites near or upstream of the major putative transcription initiation site, including E47, Max1, N-Myc, E74A, NCFI-C, Sp1, and p300.4.

Localization and Identification of the cis-Responsive Elements That Contribute to Basal HPR1 Promoter Activity-- To map the minimal promoter region of the HPR1 gene required for initiating its gene transcription, we conducted functional analysis on this 3.5-kb HPR1 promoter. Three luciferase reporter constructs in Fig. 3A, containing 3.5-, 0.7-, and 0.3-kb fragments of the HPR1 promoter, respectively, were transduced into KAT-4 and MRO87 cells. After incubation for 24 h, the cells were harvested and analyzed for their luciferase activity. As shown in Fig. 3B, KAT-4 and MRO87 cells transfected with the 3.5-kb HPR1 promoter-driven luciferase reporter construct expressed high-level luciferase activity. Deletion of the HPR1 promoter up to 0.7 kb upstream of the ATG translation initiation site did not reduce luciferase activity, but rather increased it by ~2-3-fold. Further truncation to the 0.3-kb length only slightly reduced luciferase activity compared with that observed in cells transfected with the 0.7-kb promoter-driven luciferase reporter gene. These results suggest that the cis-regulatory elements required for the basal promoter activity of the HPR1 gene are mainly located in a 0.3-kb region upstream of the ATG initiation site.


View larger version (45K):
[in this window]
[in a new window]
  		Fig. 3.   Identification of two putative Sp1 sites in the HPR1 promoter required for the basal promoter activity of the HPR1 gene. A, schematic presentation of the luciferase reporter gene driven by various lengths of the HPR1 promoter. Three luciferase reporter constructs containing 3.5 kb (pGL3/HPR-3.5), 0.7 kb (pGL3/HPR-0.7), 0.3 kb (pGL3/HPR-0.3 or D3) of the HPR1 promoter were generated as described under "Experimental Procedures." B, basal promoter activity of the HPR1 gene in two thyroid tumor cell lines. C, schematic presentation of the luciferase reporter gene driven by the HPR1 promoter containing mutated or truncated Sp1 sites. D, functional analysis of HPR1 promoter activity. E, schematic presentation of the luciferase reporter gene driven by various lengths of the HPR1 promoter. F, transfection analysis of the truncated HPR1 promoter. Thyroid tumor cell lines were transfected with the luciferase reporter constructs as described under "Experimental Procedures." Twenty-four hours later, the cells were harvested and monitored for luciferase activity. The relative light unit (RLU) in each sample was then normalized against beta -galactosidase activity measured by a colorimetric assay. The results are the means of a representative experiment performed in triplicate from two to three independent ones with similar results.

Sequence analysis revealed that several putative Sp1 sites and EREs with the core sequence GGAA are present in the proximal HPR1 promoter region (Fig. 2). We first tested whether truncation or mutation of two canonical Sp1 sites, Sp1-A and Sp1-B, affected basal HPR1 promoter activity. A panel of the luciferase reporter constructs driven by various mutated or truncated HPR1 promoters (Fig. 3C) were transduced into KAT-4 and MRO87 cells. Transient expression of the luciferase reporter gene was analyzed by quantitating luciferase activity. As shown in Fig. 3D, deletion of a 105-bp fragment from -310 to -205 (D2 construct) reduced luciferase activity by ~50% in both KAT-4 and MRO87 cells compared with the activity seen in cells transfected with the D3 construct. These results suggest that besides two putative Sp1 sites, a third cis-regulatory element is present at nucleotides -310 to -205. Mutation (D2M) or truncation (D2.1) of the Sp1-B site in the HPR1 promoter further reduced luciferase activity by 50% compared with the activity in cells transfected with D2. Mutation (D2.1M) or truncation (D2.2) of the Sp1-A site in the HPR1 promoter reduced luciferase activity by 3-5-fold compared with the activity in cells transfected with D2.1. These results collectively suggest that both the Sp1-B and Sp1-A sites are required for basal HPR1 promoter activity.

To further identify the cis-responsive elements located between nucleotides -310 to -205, a series of truncated HPR1 promoters (Fig. 3E) were generated and examined for their ability to drive luciferase reporter gene expression in three thyroid neoplastic cell lines: KAT-4, NPA87, and MRO87 cells. As shown in Fig. 3F, luciferase activity in cells transfected with the D3.2 or D3.1 construct was higher than that in cells transfected with the D2 construct, particularly in NPA87 and MRO87 cells, whereas luciferase activity in all three cell lines transfected with the D3.2 or D3.1 construct was significantly lower than that in cells transfected with the D3 construct. Sequence analysis revealed that an Sp1-binding GT box located between nucleotides -310 and -268 and a GABP-binding site consisting of two ERE repeats between nucleotides -268 and -209 may contribute to basal HPR1 promoter activity (Figs. 2 and 3E).

Identification of the Sp1 Sites in the HPR1 Promoter-- We next conducted EMSA to examine the binding to these Sp1 sites by Sp1 and its closely related homolog, Sp3. Nuclear extracts from KAT-4, NPA87, and MRO87 cells were incubated with three wild-type HPR1 Sp1 probes (Fig. 4A) or the Sp1 site-mutated Sp1-A and Sp1-B probes. A consensus Sp1 probe was included as a positive control. As shown in Fig. 4A, nuclear extracts from KAT-4 and MRO87 cells formed three complexes (C1, C2, and C3) with the consensus Sp1 probe and a fourth complex (C4) with the Sp1-A and Sp1-B probes. The C2 complex was missing when the nuclear extract from NPA87 cells was used. Mutation of the Sp1 site in the Sp1-A or Sp1-B probe eliminated or dramatically reduced the formation of all the complexes. Nuclear extracts from all three cell lines formed only a predominant C3 complex with the Sp1-C probe. These results suggest that the binding of the Sp1 and/or Sp3 transcription factors to the Sp1-A and Sp1-B sites in the HPR1 promoter is specific.


View larger version (50K):
[in this window]
[in a new window]
  		Fig. 4.   Binding activity of three putative Sp1 sites in the HPR1 promoter. A, the double-stranded oligonucleotides derived from the HPR1 promoter, each containing a putative or mutated Sp1 site, were synthesized and end-labeled with [gamma -32P]ATP. A consensus Sp1 probe was included as a positive control. The 32P-labeled probes (50,000 rpm/sample) were incubated at room temperature for 20 min with the nuclear extracts (2 µg/sample) prepared from KAT-4, MRO87, and NPA87 cells. The reactions were resolved on a 5% nondenaturing gel. The dried gel was exposed to X-Omat film overnight. B, shown are the results from supershift EMSA. The 32P-labeled consensus Sp1 and Sp1-B probes were incubated with the nuclear extracts prepared from KAT-4 cells at room temperature for 20 min. One microliter of antibody against Sp1 (alpha Sp1) or Sp3 (alpha Sp3), normal goat IgG (gIgG), or normal mouse IgG (mIgG) was added and incubated on ice for 30 min. The binding reactions were separated on a 5% nondenaturing polyacrylamide gel. The dried gel was exposed to X-Omat film overnight.

To resolve the composition of the C1, C2, C3, and C4 complexes, supershift EMSA was conducted by incubating nuclear extracts from KAT-4 cells with the radiolabeled Sp1 probes in the presence of control, anti-Sp1, and/or anti-Sp3 antibodies (Fig. 4B). Although the supershifted complex was not clearly identified, addition of anti-Sp1 antibody (but not anti-Sp3 antibody) to the binding reaction abrogated the formation of the C1 complex, but did not affect the migration of the other complexes. Combination of anti-Sp1 and anti-Sp3 antibodies had the same effect as anti-Sp1 antibody did alone. These results suggest that Sp1 is the only nuclear factor present in the C1 complex and that the binding of Sp1 to the Sp1 sites in the HPR1 promoter is specific. The nuclear factors involved in the formation of the C2, C3, and C4 complexes are currently unknown. Similar results were obtained when the radiolabeled Sp1-A probe was used in supershift EMSA (data not shown).

To assess the specificity of Sp1 binding to the Sp1 sites in the HPR1 promoter, competition EMSA was conducted using unlabeled wild-type or mutated Sp1 probes to compete with the 32P-labeled probe for binding to nuclear protein from KAT-4 cells. As shown in Fig. 5A, the Sp1-B probe was slightly less efficient than the consensus Sp1 probe, but slightly more efficient than the Sp1-A probe, in competing with the radiolabeled consensus Sp1 probe. The Sp1-C probe exhibited the weakest ability to compete with the consensus Sp1 probe. An unlabeled interferon-gamma activation site included as a negative control did not compete with the 32P-labeled Sp1 probe. Mutation of the Sp1 site in the Sp1-A and Sp1-B probes abolished competition for binding with the consensus Sp1 probe. Similar observations were made when the radiolabeled Sp1-A (Fig. 5B) and Sp1-B (Fig. 5C) probes were used in competition EMSA. These results clearly show that Sp1 was able to bind both the Sp1-A and Sp1-B sites with an affinity comparable to that of the consensus Sp1 probe and was able to bind the Sp1-C site with a very low affinity.


View larger version (65K):
[in this window]
[in a new window]
  		Fig. 5.   Competition EMSA carried out to analyze the specificity of binding of the Sp1 sites by Sp1. Nuclear extract prepared from KAT-4 cells was preincubated with 5-, 20-, and 100-fold molar excesses of the indicated unlabeled probes on ice for 20 min. The 32P-labeled consensus Sp1 (A), Sp1-A (B), and Sp1-B (C) probes (30,000 cpm/sample) were added and incubated at room temperature for another 20 min. The DNA-protein interaction was resolved on a 5% native polyacrylamide gel, and the dried gel was exposed to X-Omat film. GAS, interferon-gamma activation site.

Induction of Luciferase Reporter Gene Expression by Cotransfection with the Sp1 and Sp3 Expression Vectors in SL-2 Cells-- To further examine the role of Sp1 in regulating HPR1 promoter activity, we conducted cotransfection experiments in the Sp-deficient Drosophila SL-2 cell line using the HPR1 promoter-driven luciferase reporter gene plus the empty expression vector or the expression vector encoding the Sp1 and/or Sp3 cDNA. As shown in Fig. 6, cotransfection of SL-2 cells with the D3 construct and the Sp1, Sp3, or Sp1/Sp3 expression vector induced luciferase gene expression by 69-, 35-, and 71-fold, respectively, compared with luciferase activity in cells cotransfected with the pPac empty vector. Deletion of the Sp1-C site in the HPR1 promoter (D2 construct) resulted in a significant reduction of luciferase activity in SL-2 cells cotransfected with the Sp1, Sp3, or Sp1/Sp3 expression vector. Mutation (D2M) or truncation (D2.1) of the Sp1-B site further reduced luciferase expression in cells cotransfected with the Sp expression vectors compared with luciferase activity in cells cotransfected with the D2 construct plus the Sp expression vectors. Mutation (D2.1M construct) or truncation (D2.2 construct) of the Sp1-A site reduced luciferase activity to a level comparable to that observed in cells cotransfected with pGL3/Basic. These results collectively suggest that all three Sp1 sites in the HPR1 promoter play an important role in regulating HPR1 gene expression, although the binding affinity of Sp1 for the Sp1-C site was very low.


View larger version (25K):
[in this window]
[in a new window]
  		Fig. 6.   Induction of HPR1 promoter-driven luciferase reporter gene expression by Sp1 and Sp3. Sp-deficient SL-2 cells were transfected with the luciferase reporter gene driven by various HPR1 promoters (200 ng each) plus the pPac expression vector or the expression vector encoding Sp1 or Sp3 (600 ng each). The pCMV/SPORT plasmid DNA encoding beta -galactosidase was included as an internal control. After incubation for 24 h, the cells were harvested, and cell lysates were prepared. Luciferase activity was analyzed in a luminometer. The relative luciferase light unit was further normalized against beta -galactosidase activity measured by a colorimetric assay. The -fold induction of luciferase activity equals luciferase activity in cells transfected with the luciferase reporter gene plus Sp1, Sp3, or Sp1/Sp3 divided by luciferase activity in cells transfected with the luciferase reporter gene plus the pPac empty expression vector. The results are the means of a representative experiment performed in triplicate from two independent experiments giving similar results.

Identification of GABP-binding Sites in the HPR1 Promoter-- Functional and sequence analysis of the HPR1 promoter revealed two GABP-binding sites, each consisting of two reiterated ERE sites (Figs. 2 and 3C). To examine the role of GABP in regulating HPR1 promoter activity, we first conducted EMSA to analyze the DNA-binding activity of GABP for these EREs. A panel of oligonucleotides (Fig. 7A) containing one or two ERE sites derived from the HPR1 promoter were end-labeled with [gamma -32P]ATP and used as probes for EMSA. Nuclear extracts prepared from NPA87 and KAT-4 cells generated a dominant gel shift complex (C1) with 32P-labeled ERE-A and ERE-B and two main complexes (C1 and C2) with the ERE-C/D probe. All C1 complexes were formed at a similar position. Mutation of both ERE sites in the ERE-C/D probe (ERE-C/Dm) eliminated the formation of the C1 and C2 complexes, but resulted in a complex migrating slightly slower than the C1 complex. Mutation of the ERE site in both the ERE-B and ERE-A probes reduced the formation of the C1 complex, but did not completely abolish the formation of the C1 complex. Interestingly, mutation of the ERE site in the ERE-A probe also generated a complex migrating slightly faster than the C1 complex. It appears that the ERE-A oligonucleotide had a slightly higher affinity for GABP to form the C1 complex than the ERE-B oligonucleotide.


View larger version (48K):
[in this window]
[in a new window]
  		Fig. 7.   Binding activity of GABP for four EREs in the HPR1 promoter. A, the double-stranded oligonucleotides derived from the HPR1 promoter, each containing one or two ERE sites, were synthesized and end-labeled with [gamma -32P]ATP. The 32P-labeled probes (50,000 rpm/sample) were incubated at room temperature for 20 min with the nuclear extracts (2 µg/sample) prepared from KAT-4 or NPA87 cells. The reactions were resolved on a 5% nondenaturing gel. The dried gel was exposed to X-Omat film overnight. B, shown are the results from supershift EMSA carried out to identify GABP as a nuclear factor binding to four ERE sites in the HPR1 promoter. The 32P-labeled ERE-C/D, ERE-B, or ERE-A probe was incubated with the nuclear extracts prepared from NPA87 cells at room temperature for 20 min. One microliter of antibody against GABPalpha (alpha GABPalpha ) and/or GABPbeta (alpha GABPbeta ) or normal rabbit (NRS) was added and incubated on ice for 30 min. The binding reactions were separated on a 5% nondenaturing polyacrylamide gel. The dried gel was exposed to X-Omat film overnight. SS, supershifted complex.

To further confirm the specificity of binding of these ERE sites by GABP, supershift EMSA was conducted by incubating nuclear extract from NPA87 cells with the radiolabeled ERE probes in the presence of control, anti-GABPalpha , and/or anti-GABPbeta antibodies. As shown in Fig. 7B, no supershift was detected when normal rabbit serum was used as a negative control; addition of anti-GABPalpha and/or anti-GABPbeta antibodies resulted in the formation of a supershifted complex that migrated much slower than the C1 complex. Interestingly, addition of either of the anti-GABP antisera resulted in the formation of two extra complexes (marked by asterisks). These results suggest that the removal of the GABPs may allow other members of the Ets family to bind to the ERE sites. Nevertheless, our results suggest that GABP is able to specifically bind all four ERE sites.

To assess the specificity of GABP binding to the ERE sites in the HPR1 promoter, competition EMSA was conducted using unlabeled wild-type or mutated ERE probes to compete with the 32P-labeled probe for binding to nuclear protein from NPA87 cells. A GABP probe derived from the ICP4 gene of human herpes simplex virus-1 was used as a positive control. A consensus nuclear factor-kappa B probe was used as a negative control. As shown in Fig. 8, the wild-type GABP-C/D probe was more efficient than the ICP4 probe in competing with the radiolabeled GABP-C/D probe. Mutation of the ERE sites in the GABP-C/D probe abolished competition for binding with the wild-type GABP-C/D probe. The unlabeled nuclear factor-kappa B probe did not compete with the 32P-labeled GABP-C/D probe at all. Similar results were obtained when the radiolabeled ERE-B (data not shown) and ERE-A (Fig. 8, right panel) probes were used in competition EMSA. These results further suggest that GABP is able to bind all four ERE sites with high affinities.


View larger version (91K):
[in this window]
[in a new window]
  		Fig. 8.   Competition EMSA carried out to analyze the binding specificity of GABP for the EREs. Nuclear extract prepared from NPA87 cells was preincubated with 5-, 20-, and 100-fold molar excesses of the indicated unlabeled probes on ice for 30 min. The 32P-labeled ERE-C/D or ERE-A probe (30,000 cpm/sample) was added and incubated at room temperature for another 20 min. The DNA-protein interaction was resolved on a 5% native polyacrylamide gel, and the dried gel was exposed to X-Omat film. NF-kappa B, nuclear factor-kappa B.

Mutational Analysis of the GABP-binding Sites-- To test whether these GABP sites have any functional role in regulating HPR1 gene expression, four ERE sites were individually or simultaneously mutated (Fig. 9A) and assessed for their ability to induce the expression of the linked luciferase reporter gene. As shown in Fig. 9B, mutation of the ERE-A (D3M1), ERE-B (D3M2), ERE-A/B (D3M3), or ERE-C/D (D3M4) site reduced luciferase activity by 30, 48, 90, and 53% in KAT-4 cells, respectively. Simultaneous mutation of the ERE-C/D site with ERE-A (D3M5) or ERE-B (D3M6) reduced luciferase activity by 89 and 82%, respectively. Mutation of all ERE sites reduced luciferase activity by 95%. Similar results were obtained using MRO87 cells. These results suggest that, although both distal and proximal GABP sites are involved in regulating HPR1 promoter activity, the proximal GABP site has a more profound effect than the distal GABP site on the regulation of HPR1 promoter activity (compare the luciferase activities in D3M3- and D3M4-transfected cells).


View larger version (42K):
[in this window]
[in a new window]
  		Fig. 9.   Analysis of the effect of mutation of GABP sites on HPR1 promoter activity. A, schematic presentation of the luciferase reporter gene driven by the HPR1 promoter with mutation of the ERE sites. B, transfection analysis of the truncated HPR1 promoter. KAT-4 and MRO87 cells were transiently transfected with the luciferase reporter constructs depicted in A. A pCMV/SPORT vector encoding the beta -galactosidase gene was included as an internal control. After incubation for 24 h, the cells were harvested and analyzed for luciferase activity in a luminometer. The relative luciferase activity was normalized against beta -galactosidase activity. The results are the means of a representative experiment performed in triplicate from three independent experiments with similar results.

GABP Cooperates with Sp1 or Sp3 to Regulate HPR1 Promoter Activity-- Previous studies have demonstrated that GABP can cooperate with Sp1 to regulate the expression of several genes such as CD18 (22-24) and the matrix protein tenascin C (25). Our EMSA studies have demonstrated that GABP was able to bind four ERE sites in the HPR1 promoter (Figs. 7 and 8). To test whether GABP and Sp1 can cooperatively regulate HPR1 gene expression, we conducted cotransfection of SL-2 cells using the luciferase reporter gene plus the expression vectors encoding the GABP, Sp1, and Sp3 genes. As shown in Fig. 10, cotransfection of SL-2 cells with the D3 construct (which contains both the proximal and distal GABP sites) plus GABPalpha or GABPbeta did not increase luciferase expression compared with luciferase activity in cells cotransfected with the pPac empty vector. Cotransfection of SL-2 cells with the D3 construct plus Sp1, Sp3, or GABPalpha /beta induced luciferase gene expression by 42-, 11-, and 3-fold, respectively. Cotransfection of SL-2 cells with the D3 construct plus Sp1 and GABPalpha /beta or plus Sp3 and GABPalpha /beta further induced luciferase expression. These results suggest that GABP and Sp1 can cooperatively regulate HPR1 promoter activity.


View larger version (26K):
[in this window]
[in a new window]
  		Fig. 10.   GABP cooperates with Sp1 and Sp3 to activate the HPR1 promoter in SL-2 cells. SL-2 cells were cotransfected with the luciferase reporter gene driven by various HPR1 promoters (200 ng each) plus the pPac empty expression vector or the expression vector encoding Sp1, Sp3, GABPalpha , or GABPbeta (600 ng each); GABPalpha /beta (300 ng each); or Sp1 + GABPalpha /beta or Sp3 + GABPalpha /beta (200 ng each). After incubation for 24 h, the cells were harvested, and cell lysates were prepared. Luciferase activity was analyzed in a luminometer. The -fold induction of luciferase activity equals luciferase activity in cells transfected with the luciferase reporter gene plus GABPalpha , GABPbeta , GABPalpha /beta , Sp3, Sp3 + GABPalpha /beta , Sp1, or Sp1 + GABPalpha /beta divided by luciferase activity in cells transfected with the luciferase reporter gene plus the pPac empty expression vector. The results are the means of a representative experiment performed in triplicate from two independent experiments giving similar results.

To test whether the proximal GABP site is required to cooperate with Sp1 to regulate HPR1 promoter activity, we conducted a cotransfection experiment in SL-2 cells using the D2.1 construct (which contains the Sp1-A site and the proximal GABP site only) plus the Sp1 or Sp3 expression vector with or without GABPalpha /beta . As shown in Fig. 10, cotransfection of SL-2 cells with the D2.1 construct plus Sp1, Sp3, or GABPalpha /beta induced luciferase expression by 20-, 5-, and 9-fold, respectively. Cotransfection of SL-2 cells with the D2.1 construct plus Sp1 and GABPalpha /beta or plus Sp3 and GABPalpha /beta further induced luciferase expression by 33- and 23-fold, respectively. These results suggest that the proximal GABP-binding site, which is located in the untranslated region of the first exon, is able to cooperate with Sp1 to initiate HPR1 gene transcription.

We then examined the ability of the distal GABP site to cooperate with Sp1 to regulate HPR1 promoter activity. A luciferase reporter construct (D4) was generated by inserting a BamHI/BglI-digested DNA fragment of the HPR1 promoter (from nucleotides -209 to -306) upstream of the luciferase reporter gene. The HPR1 promoter in this construct contains the Sp1-C site and the distal GABP site only. SL-2 cells were transfected with the D4 construct plus the Sp1 or Sp3 expression vector with or without the GABPalpha /beta expression vectors. As shown in Fig. 10, cotransfection of SL-2 cells with the D4 construct plus Sp1, Sp3, or GABPalpha /beta induced luciferase expression by 10-, 4-, and 3-fold, respectively. Cotransfection of SL-2 cells with the D4 construct plus Sp1 and GABPalpha /beta or plus Sp3 and GABPalpha /beta further induced luciferase gene expression by 14- and 7-fold, respectively. In addition, SL-2 cells cotransfected with pGL3/Basic plus the expression vectors encoding the GABPalpha and GABPbeta genes also slightly increased luciferase reporter gene expression. These results suggest that GABP by itself is unable to initiate luciferase reporter gene expression driven by the distal GABP site (D4 construct) and is only able to weakly initiate luciferase reporter gene expression driven by the proximal GABP site (D2.1 construct). Nevertheless, our results suggest that both the Sp1-C site and the distal GABP-binding site are functional and that GABP is able to cooperate with Sp1 and Sp3 to initiate HPR1 gene transcription.

    DISCUSSION
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

HPR1 is expressed at high levels in a variety of malignancies such as hepatomas and breast and colon cancers (13). Although the mechanisms by which HPR1 gene expression is regulated remain unknown, it has been speculated that HPR1 expression may be due to the dysregulation of signaling molecules and/or transcription factors at a late stage of tumor progression (26). In this study, we cloned and characterized the HPR1 promoter and found that the GC-rich promoter does not contain TATA or CCAAT boxes, but does contain two GC boxes and one GT box. Three Sp1 and four ERE sites were identified in the minimal promoter region of the HPR1 gene. Further studies demonstrated that Sp1 and GABP were able to regulate HPR1 promoter activity. These observations suggest that the transcription factors of the Sp1 and Ets families play a critical role in regulating HPR1 gene expression.

Sp1 is the founding member of a growing family that binds to and acts through the GC boxes (27, 28). Sp1 is abundantly expressed in most cell types, but its level of expression changes during development and varies in different cell types (27, 28). Sp1 regulates the expression of many different types of genes such as structural proteins, metabolic enzymes, cell cycle regulators, transcription factors, growth factors, surface receptors, and others (28-31). Although it is difficult to define the function of Sp1 because of the potentially redundant or antagonistic actions of its related family members, Sp1 is essential for embryogenesis because Sp1-/- embryos display growth retardation and die early in gestation (32). We have demonstrated here that mutation and truncation of the Sp1 sites in the HPR1 promoter greatly impaired its promoter activity; cotransfection of the HPR1 promoter-driven luciferase reporter with the Sp1 expression vector dramatically increased luciferase gene expression. These observations suggest that Sp1 plays a critical role in regulating basal HPR1 promoter activity. Because most cell types constitutively express active Sp1, the lack or low level of HPR1 gene expression in normal epithelial cells may result from post-transcriptional modification of HPR1 mRNA or from differential DNA methylation of the HPR1 promoter.

Sp3 is a member of the Sp1 transcription factor family. Sp3 is homologous to Sp1. It binds to the Sp1 site with an affinity comparable to that of Sp1 and is expressed in a variety of tissues in which Sp1 is also expressed. However, Sp3 can function as either a transcription activator or repressor by binding to and competing with Sp1 for the Sp1 site (33-37). The experimental conditions under which Sp3 functions as a repressor are not fully understood. It appears that the ratio of Sp1 to Sp3 in a particular cell context determines whether Sp3 acts as an activator or a repressor (28). Our cotransfection studies demonstrated that Sp3 by itself or in combination with GABP was able to up-regulate HPR1 promoter-driven luciferase reporter gene expression in SL-2 cells. These observations suggest that Sp3 does not suppress, but instead can stimulate HPR1 gene expression. It is not clear whether Sp3 has a role in regulating HPR1 expression in thyroid tumor cell lines because supershift EMSA did not reveal Sp3 binding activity.

GABP consists of alpha - and beta -subunits. Previous studies have demonstrated that the GABP-binding site is present in many TATA-less promoters and can initiate the assembly of the preinitiation complex (22, 38-43). These promoters are found in genes encoding mitochondrial proteins involved in oxidative phosphorylation and the proteins that play important roles during embryogenesis, angiogenesis, tissue remodeling, and tumor metastases (22, 38-43). Our present study identified two GABP sites, each containing at least two repeats of the GGAA core sequence. The proximal GABP site is adjacent to the transcription initiation site, suggesting that this GABP site may have a role in directing transcription initiation. The characteristics of the HPR1 promoter further suggest that the HPR1 gene product belongs to a group of enzymes that are involved in tumor metastases and angiogenesis.

GABP regulates gene expression through the interaction with other transcription factors bound to the cognate motifs in the vicinity of the GABP site. For example, our previous study showed that GABP cooperates with AP1 to regulate fas gene expression through two ERE sites and a "sandwiched" AP1 site in the distal promoter region of the fas gene (21). GABP and Sp1 cooperate to regulate the promoter activity of CD18 (22-24), the matrix protein tenascin C (25), and the neutrophil elastase promoter (44, 45). Our present study has demonstrated that mutation of the GABP sites dramatically reduced HPR1 promoter activity and that coexpression of GABP and Sp1 or Sp3 in SL-2 cells increased HPR1 promoter-driven luciferase activity. These results collectively suggest that GABP cooperates with Sp1 or Sp3 to regulate the promoter activity of the HPR1 gene. Although the mechanisms by which GABP cooperates with the adjacent transcription factor to regulate gene expression are currently unknown, it has been proposed that GABP bound to its cognate site can provide a platform for the assembly of the transcription initiation machinery containing Sp1, transcription factor IID, and RNA polymerase II (46).

    ACKNOWLEDGEMENTS

We thank Drs. G. J. F. Juillard and K. B. Ain for kindly providing thyroid tumor cell lines; Dr. R. L. Widom (Boston University School of Medicine) for the SL-2 cell line; Drs. R. Tjian and G. Suske for pPac, pPac/Sp1, and pPac/Sp3 expression vectors; Dr. B. J. Graves (University of Utah) for pPac/GABPalpha and pPac/GABPbeta expression vectors; and Dr. U. A. Rapp for rabbit anti-GABPalpha and anti-GABPbeta antisera.

    FOOTNOTES

* This work was supported in part by grants from the Thyroid Research Advisory Council and NCI Grant CA76407 from the National Institutes of Health (to X. X.) and by the Department of General Surgery at Rush-Presbyterian-St. Luke's Medical Center.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 GenBankTM/EBI Data Bank with accession number(s) AF461265.

|| Present address: Nature Immunology, 345 Park Ave. South, New York, NY 10010-1707.

§§ To whom correspondence should be addressed: Dept. of General Surgery, Rush-Presbyterian-St. Luke's Medical Center, 1653 W. Congress Pkwy., Chicago, IL 60612. Tel.: 312-942-5000 (Ext. 21368); Fax: 312-942-2867; E-mail: xxu@rush.edu.

Published, JBC Papers in Press, January 4, 2002, DOI 10.1074/jbc.M105682200

2 X. Xu, unpublished data.

    ABBREVIATIONS

The abbreviations used are: HPR, heparanase; GABP, GA-binding protein; RLM-RACE, RNA ligase-mediated rapid amplification of cDNA ends; GSP, gene-specific primer; AP, adapter primer; ERE, Ets-relevant element; EMSA, electrophoretic mobility shift assay.

    REFERENCES
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

1. Chambers, A., and Matrisian, L. (1997) J. Natl. Cancer Inst. 89, 1260-1270[Abstract/Free Full Text]
2. Fidler, I., and Radinsky, R. (1996) J. Natl. Cancer Inst. 88, 1700-1703[Free Full Text]
3. Fidler, I. (1999) Cancer Chemother. Pharmacol. 43, S3-S10
4. Hoogewerf, A., Leone, J., Reardon, I., Howe, W., Asa, D., Heinrikson, R., and Ledbetter, S. (1995) J. Biol. Chem. 270, 3268-3277[Abstract/Free Full Text]
5. Matzner, Y., Bar-Ner, M., Yahalom, J., Ishai-Michaeli, R., Fuks, Z., and Vlodavsky, I. (1985) J. Clin. Invest. 76, 1306-1313
6. Matzner, Y., Vlodavsky, I., Bar-Ner, M., Ishai-Michaeli, R., and Tauber, A. (1992) J. Leukocyte Biol. 51, 519-524[Abstract]
7. Naparstek, Y., Cohen, I., Fuks, Z., and Vlodavsky, I. (1984) Nature 310, 241-244[CrossRef][Medline] [Order article via Infotrieve]
8. Sewell, R. F., Brenchley, P. E., and Mallick, N. P. (1989) Biochem. J. 264, 777-783[Medline] [Order article via Infotrieve]
9. Nakajima, M., Irimura, T., and Nicolson, G. (1986) Anal. Biochem. 157, 162-171[CrossRef][Medline] [Order article via Infotrieve]
10. Nakajima, M., Irimura, T., and Nicolson, G. (1986) Cancer Lett. 31, 277-283[CrossRef][Medline] [Order article via Infotrieve]
11. Nakajima, M., Irimura, T., and Nicolson, G. (1988) Cell Biochem. 36, 157-167
12. Nakajima, M., Irimura, T., DiFerrante, D., DiFerrante, N., and Nicholson, G. L. (1983) Science 220, 611-613[Abstract/Free Full Text]
13. Vlodavsky, I., Friedmann, Y., Elkin, M., Aingorn, H., Atzmon, R., Ishai-Michaeli, R., Bitan, M., Pappo, O., Peretz, T., Michal, I., Spector, L., and Pecker, I. (1999) Nat. Med. 5, 793-802[CrossRef][Medline] [Order article via Infotrieve]
14. Dempsey, L. A., Plummer, T. B., Coombes, S. L., and Platt, J. L. (2000) Glycobiology 10, 467-475[Abstract/Free Full Text]
15. Hulett, M., Freeman, C., Hamdorf, B., Baker, R., Harris, M., and Parish, C. (1999) Nat. Med. 5, 803-809[CrossRef][Medline] [Order article via Infotrieve]
16. Kussie, P. H., Hulmes, J. D., Ludwig, D. L., Patel, S., Navarro, E. C., Seddon, A. P., Giorgio, N. A., and Bohlen, P. (1999) Biochem. Biophys. Res. Commun. 261, 183-187[CrossRef][Medline] [Order article via Infotrieve]
17. McKenzie, E., Tyson, K., Stamps, A., Smith, P., Turner, P., Barry, R., Hircock, M., Patel, S., Barry, E., Stubberfield, C., Terrett, J., and Page, M. (2000) Biochem. Biophys. Res. Commun. 276, 1170-1177[CrossRef][Medline] [Order article via Infotrieve]
18. Toyoshima, M., and Nakajima, M. (1999) J. Biol. Chem. 274, 24153-24160[Abstract/Free Full Text]
19. Baker, E., Crawford, J., Sutherland, G. R., Freeman, C., Parish, C. R., and Hulett, M. D. (1999) Chromosome Res. 7, 319[CrossRef][Medline] [Order article via Infotrieve]
20. Dong, J., Kukula, A. K., Toyoshima, M., and Nakajima, M. (2000) Gene (Amst.) 253, 171-178[CrossRef][Medline] [Order article via Infotrieve]
21. Li, X. R., Chong, A. S., Wu, J., Roebuck, K. A., Kumar, A., Parrillo, J. E., Rapp, U. R., Kimberly, R. P., Williams, J. W., and Xu, X. (1999) J. Biol. Chem. 274, 35203-35210[Abstract/Free Full Text]
22. Bottinger, E. P., Shelley, C. S., Farokhzad, O. C., and Arnaout, M. A. (1994) Mol. Cell. Biol. 14, 2604-2615[Abstract/Free Full Text]
23. Rosmarin, A. G., Caprio, D. G., Kirsch, D. G., Handa, H., and Simkevich, C. P. (1995) J. Biol. Chem. 270, 23627-23633[Abstract/Free Full Text]
24. Rosmarin, A. G., Luo, M., Caprio, D. G., Shang, J., and Simkevich, C. P. (1998) J. Biol. Chem. 273, 13097-13103[Abstract/Free Full Text]
25. Shirasaki, F., Makhluf, H. A., LeRoy, C., Watson, D. K., and Trojanowska, M. (1999) Oncogene 18, 7755-7764[CrossRef][Medline] [Order article via Infotrieve]
26. Eccles, S. (1999) Nat. Med. 5, 735-736[CrossRef][Medline] [Order article via Infotrieve]
27. Lania, L., Majello, B., and De Luca, P. (1997) Int. J. Biochem. Cell Biol. 29, 1313-1323[CrossRef][Medline] [Order article via Infotrieve]
28. Suske, G. (1999) Gene (Amst.) 238, 291-300[CrossRef][Medline] [Order article via Infotrieve]
29. Braun, H., and Suske, G. (1998) J. Biol. Chem. 273, 9821-9828[Abstract/Free Full Text]
30. Courey, A. J., and Tjian, R. (1988) Cell 55, 887-898[CrossRef][Medline] [Order article via Infotrieve]
31. Philipsen, S., and Suske, G. (1999) Nucleic Acids Res. 27, 2991-3000[Abstract/Free Full Text]
32. Saffer, J. D., Jackson, S. P., and Annarella, M. B. (1991) Mol. Cell. Biol. 11, 2189-2199[Abstract/Free Full Text]
33. Conn, K. J., Rich, C. B., Jensen, D. E., Fontanilla, M. R., Bashir, M. M., Rosenbloom, J., and Foster, J. A. (1996) J. Biol. Chem. 271, 28853-28860[Abstract/Free Full Text]
34. Hagen, G., Muller, S., Beato, M., and Suske, G. (1994) EMBO J. 13, 3843-3851[Medline] [Order article via Infotrieve]
35. Ihn, H., LeRoy, E. C., and Trojanowska, M. (1997) J. Biol. Chem. 272, 24666-24672[Abstract/Free Full Text]
36. Ihn, H., and Trojanowska, M. (1997) Nucleic Acids Res. 25, 3712-3717[Abstract/Free Full Text]
37. Majello, B., De, Luca, P., and Lania, L. (1997) J. Biol. Chem. 272, 4021-4026[Abstract/Free Full Text]
38. Boone, T. C., Johnson, M. J., De, Clerck, Y. A., and Langley, K. E. (1990) Proc. Natl. Acad. Sci. U. S. A. 87, 2800-2804[Abstract/Free Full Text]
39. Campbell, C. E., Flenniken, A. M., Skup, D., and Williams, B. R. (1991) J. Biol. Chem. 266, 7199-7206[Abstract/Free Full Text]
40. Gutman, A., and Wasylyk, B. (1990) EMBO J. 9, 2241-2246[Medline] [Order article via Infotrieve]
41. Huhtala, P., Chow, L. T., and Tryggvason, K. (1990) J. Biol. Chem. 265, 11077-11082[Abstract/Free Full Text]
42. Ogawa, K., Burbelo, P. D., Sasaki, M., and Yamada, Y. (1988) J. Biol. Chem. 263, 8384-8389[Abstract/Free Full Text]
43. Villena, J. A., Martin, I., Vinas, O., Cormand, B., Iglesias, R., Mampel, T., Giralt, M., and Villarroya, F. (1994) J. Biol. Chem. 269, 32649-32654[Abstract/Free Full Text]