JBC Origene Your Gene Company

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


     

Originally published In Press as doi:10.1074/jbc.M309159200 on September 15, 2003

J. Biol. Chem., Vol. 278, Issue 47, 46632-46642, November 21, 2003
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
278/47/46632    most recent
M309159200v1
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 Tellez, C.
Right arrow Articles by Bar-Eli, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tellez, C.
Right arrow Articles by Bar-Eli, M.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Loss of Activator Protein-2{alpha} Results in Overexpression of Protease-activated Receptor-1 and Correlates with the Malignant Phenotype of Human Melanoma*

Carmen Tellez, Marya McCarty, Maribelis Ruiz, and Menashe Bar-Eli{ddagger}

From the Department of Cancer Biology, the University of Texas M. D. Anderson Cancer Center, Houston, Texas 77030

Received for publication, August 18, 2003 , and in revised form, September 10, 2003.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Increasing evidence implicates the protease-activated receptor-1 (PAR-1) as a contributor to tumor invasion and metastasis of human melanoma. Here we demonstrate that the metastatic potential of human melanoma cells correlates with overexpression of PAR-1. We also provide evidence that an inverse correlation exists between the expression of activator protein-2{alpha} (AP-2) and the expression of PAR-1 in human melanoma cells. Reexpression of AP-2 in WM266-4 melanoma cells, which are AP-2-negative, resulted in decreased mRNA and protein expression of PAR-1. The promoter of the PAR-1 gene contains multiple putative consensus elements for the transcription factors AP-2 and specificity protein 1 (Sp1). Chromatin immunoprecipitation analysis of the PAR-1 promoter regions bp –365 to –329 (complex 1) and bp –206 to –180 (complex 2) demonstrated that Sp1 was predominantly bound to the PAR-1 promoter in metastatic cells, whereas AP-2 was bound to the PAR-1 promoter in nonmetastatic cells. In vitro analysis of complex 1 demonstrated that AP-2 and Sp1 bound to this region in a mutually exclusive manner. Transfection experiments with full-length and progressive deletions of the PAR-1 promoter luciferase constructs demonstrated that metastatic melanoma cells had increased PAR-1 promoter activity compared with low and nonmetastatic melanoma cells. Our data show that exogenous AP-2 expression decreased promoter activity, whereas transient expression of Sp1 further increased expression of the reporter gene. Mutational analysis of complex 1 within PAR-1 luciferase constructs further demonstrated that the regulation of PAR-1 was mediated through interactions with AP-2 and Sp1. Our data suggest that loss of AP-2 in metastatic cells alters the AP-2/Sp1 ratio, resulting in overexpression of PAR-1. Taken together, our results provide strong evidence that loss of AP-2 correlates with overexpression of PAR-1, which in turn contributes to the acquisition of the malignant phenotype of human melanoma.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
We have previously reported that loss of the transcription factor activator protein-2{alpha} (AP-2)1 is associated with the transition of melanoma cells from radial growth phase to vertical growth phase. Our studies demonstrated that nonmetastatic melanoma cell lines expressed high levels of AP-2, whereas highly metastatic melanoma cell lines did not express AP-2 (1–3). Furthermore, we showed that transfection of highly metastatic melanoma cells with full-length AP-2 significantly reduced the tumorigenicity and metastatic potential of these cells in vivo (2, 3). Similarly, inactivation of AP-2 by stable transfection with a dominant negative AP-2 gene (AP-2B) into AP-2-positive primary cutaneous melanoma cells increased tumor growth in vivo (4). These observations have been supported by several studies in human melanoma clinical specimens.

We have previously demonstrated that loss of AP-2 is an important molecular event in melanoma progression, which results in deregulation of AP-2 target genes involved in tumor growth and metastasis. For example, loss of AP-2 expression in metastatic melanoma cells resulted in overexpression of the melanoma cell adhesion molecule (MCAM/MUC18) (2) and lack of expression of the tyrosine kinase receptor c-KIT (3). Increased expression of MUC18 allowed the metastatic cells to adhere to the endothelial cells in blood vessels and supported their migration to the metastatic site (5), whereas low c-KIT expression rendered the cells resistant to apoptosis (6). In addition, inactivation of AP-2 in AP-2-positive primary cutaneous melanoma cells by means of transfection with a dominant negative AP-2 gene (AP-2B) led to deregulation of the matrix metalloproteinase-2 gene (4). Furthermore, immunohistochemical analyses of advanced primary and metastatic melanoma clinical specimens demonstrated that the loss of AP-2 expression correlated with low p21/WAF1, E-cadherin, and c-KIT expression and poor prognosis (7, 8). Functional AP-2-binding elements have been identified in other genes involved in the progression of human melanoma, including p21/WAF1 (9), intercellular adhesion molecule (10), c-erbB-2/HER-2/neu (11–13), plasminogen activator inhibitor type I (14), insulin-like growth factor-binding protein-5 (15), transforming growth factor-{alpha} (16), vascular endothelial growth factor (17, 18), E-cadherin (19), and hepatocyte growth factor (20).

The protease-activated receptor-1 (PAR-1) is a unique G-coupled protein receptor, which belongs to the protease-activated receptor family. Activation of PAR-1 involves proteolytic cleavage of the extracellular amino-terminal domain by thrombin to unmask a new amino terminus capable of serving as a tethered ligand for the receptor, which leads to downstream cell signaling events that evoke a variety of cellular responses (21). Overexpression of PAR-1 has been detected in numerous human cancers, including colon (22), laryngeal (23), breast (24), pancreatic (25), and oral squamous cell carcinomas (26). Recent evidence suggests that PAR-1 plays a central role in tumorigenesis and metastasis (24, 27, 28). PAR-1 was reported to be a rate-limiting factor in thrombin-enhanced experimental pulmonary metastasis of a murine melanoma cell line, demonstrating a potential role for PAR-1 in metastasis of melanoma (28). Furthermore, thrombin acts as a growth factor in human melanoma cells, which suggests that PAR-1 signaling is involved in the biological responses of these cells (29).

PAR-1 expression has been reported to be up-regulated in the human metastatic melanoma cell line A375SM (30). However, the mechanism for up-regulation of PAR-1 expression in malignant melanoma is unknown. The regulatory region of the PAR-1 gene has been cloned, and DNA sequence analysis indicates the presence of multiple putative AP-2 and specificity protein 1 (Sp1) regulatory elements at the proximal 3' region of the promoter (31, 32). The genomic organization of the promoter shares remarkable similarities with other AP-2 target genes, such as a G + C-rich sequence, lack of conventional TATA and CAAT sequences, and multiple AP-2-binding elements. Previously, it has been shown that Sp1 transactivates the PAR-1 promoter in human endothelial cells (33). We hypothesized that PAR-1 expression is regulated by AP-2 and that loss of AP-2 may contribute to PAR-1 up-regulation in malignant melanoma.

Here we provide further evidence that AP-2 is involved in the etiology of malignant human melanoma. We observed a direct correlation between the expression of PAR-1 and the metastatic potential of human melanoma cell lines. Thus, we demonstrated that an inverse correlation exists between the levels of expression of AP-2 and PAR-1 in human melanoma cell lines. Furthermore, we found that enforced expression of AP-2 in the human metastatic melanoma cell line WM266-4, which is AP-2-negative, resulted in decreased expression of PAR-1. In determining the mechanisms of PAR-1 gene regulation, we analyzed specific DNA elements of its promoter. The nuclear proteins AP-2 and Sp1 were found to bind to elements of the PAR-1 promoter both in vivo and in vitro. Analysis of the DNA-binding and protein expression of AP-2 and Sp1 in a panel of melanoma cell lines revealed a marked decrease in the ratio of AP-2 to Sp1 expression, which correlated with an overexpression of PAR-1 in metastatic melanoma cells. These results provide strong evidence for an additional mechanism by which loss of AP-2 expression and overexpression of PAR-1, contributes to the malignant phenotype of human melanoma.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Cell Lines and Culture Conditions—The human melanoma cell line WM266-4, purchased from American Type Culture Collection (Manassas, VA), is tumorigenic and metastatic in nude mice (34). The highly metastatic A375SM human melanoma cell line was established from pooled lung metastases produced by A375-P cells injected intravenously into nude mice (35). The SB2 cell line was isolated from a primary human cutaneous melanoma and was a gift from Dr. B. Giovanella (St. Joseph's Hospital, Houston, TX). In nude mice, SB2 cells are poorly tumorigenic and are nonmetastatic (36). The human melanoma MeWo cell line was established in culture from a lymph node metastasis of a melanoma patient and was kindly provided by Dr. S. Ferrone (New York Medical College, New York, NY). In nude mice, MeWo cells are tumorigenic and have low metastatic potential (37). The human breast adenocarcinoma cell lines MDA-MB-231 and MCF-7 were provided by Dr. J. E. Price (University of Texas M. D. Anderson Cancer Center, Houston, TX). MDA-MB-231 cells are highly tumorigenic when injected into the mammary fat pad and are metastatic in nude mice (38). MCF-7 cells require estradiol supplementation for growth in vivo and are tumorigenic when injected into the mammary fat pad. However, sites such as the lung rarely support growth of MCF-7 tumors (39). The melanoma and breast cancer cell lines were maintained as adherent monolayers in Eagle's minimal essential medium supplemented with 10% fetal bovine serum, as previously described (3). The stably transfected cell lines were maintained in the same medium, which contained 0.5 µg/ml G418 (Invitrogen).

Preparation of Total and Nuclear Cell Extracts—For total extracts, the cells were lysed in 500 µl of Triton lysis buffer (25 mM Tris-HCl, pH 7.5, containing 150 mM NaCl, 1% Triton X-100, 5 mM EDTA, and protease inhibitors: 1 mM phenylmethylsulfonyl fluoride, 20 µM leupeptin, 0.15 units/ml aprotinin, 1 mM Na3VO4, and 10 mM NaF). For nuclear extracts, the cells were incubated on ice in 400 µl of 10 mM HEPES buffer, pH 7.9, containing 1.5 mM MgCl2, 10 mM KCl, 0.5 mM dithiothreitol, and protease inhibitors, and then they were lysed with a Dounce tissue grinder until over 80% of the nuclei were released, as determined by trypan blue staining. The cytoplasmic fraction was separated by centrifugation. The nuclear pellet was resuspended in 50 µlof 20 mM HEPES buffer, pH 7.9, containing 25% glycerol, 450 mM NaCl, 1.5 mM MgCl, 0.2 mM EDTA, 0.5 mM dithiothreitol, and protease inhibitors. For both extracts, the soluble proteins in the lysates were separated by centrifugation. The protein content was quantified using the Bio-Rad protein assay.

Antibodies—Antibodies to AP-2 (C18), Sp1 (PEP2), and Sp3 (D-20), which were used in Western blot and gel shift analyses, were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). The antibody for {beta}-actin was purchased from Sigma, and the monoclonal antibody for PAR-1 (WEDE15) was purchased from Immunotech-Coulter (Miami, FL). The polyclonal antibodies used in chromatin immunoprecipitation (ChIP) assay for AP-2 and Sp1 were obtained from Geneka Biotechnology, Inc. (Montreal, Canada).

Western Blot Analysis—PAR-1, AP-2, Sp1, and {beta}-actin were detected in cell extracts by Western blot, as previously described (3). Briefly, proteins of total cell extracts (40 µg) and nuclear extracts (20 µg) were separated by SDS-polyacrylamide gel electrophoresis and transferred to Immobilon P transfer membrane (Millipore Corp., Billenica, MA). The membranes were washed in Tris-HCl-buffered saline (10 mM Tris-HCl, pH 8, containing 150 mM NaCl) and blocked with 5% nonfat milk in TBS for 1 h at room temperature. The blots were then probed overnight in Tris-HCl-buffered saline with the relevant antibodies at a dilution of 1:2000.

Electrophoretic Mobility Shift Assay (EMSA)—The EMSA probes consisted of annealed synthetic complementary oligonucleotides corresponding to bp –365 to –329 and bp –206 to –180 of the published PAR-1 promoter sequence (31). 32P-end-labeled oligonucleotides (20,000 cpm) were incubated for 30 min on ice with 5 µg of nuclear extract in 20 µl of binding buffer containing 25 mM HEPES, pH 7.9, 0.5 mM EDTA, 0.5 mM dithiothreitol, 4% glycerol, 50 mM NaCl, and 0.5 µg of poly(dI-dC). The binding reaction was incubated on ice for 30 min. Competition reactions were performed with a 100-fold molar excess of unlabeled double-stranded AP-2 and Sp1 competitor DNA (Promega Corp., Madison, WI). For supershift analysis, the nuclear extracts were incubated with polyclonal anti-human Sp1, Sp3, and AP-2 antibodies for 1 h following the binding reaction with labeled probe. The DNA-protein complexes were separated on a 4% native polyacrylamide gel in 0.5x TBE buffer.

RNase Protection Assay—PAR-1 mRNA levels were analyzed by RNase protection assay using the RiboQuant multiprobe set (BD Biosciences, San Jose, CA) according to the manufacturer's instructions. In brief, 5 µg of RNA was hybridized overnight to the 32P-labeled RNA probe, which had been previously synthesized from the supplied template set (hAngio-1). Single-stranded RNA and the free probe were digested by RNase A and T1. Subsequently, protected RNA was isolated by phenol extraction, ethanol-precipitated, and analyzed on a 6% denaturing polyacrylamide gel. The PAR-1 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) transcripts were identified by the lengths of the respective fragments. For quantitation, the protected RNA was determined using QuantityOne software (Bio-Rad), and PAR-1 values were expressed as a percentage of the mean values of GAPDH expressed for each cell line.

ChIP—The ChIP assay was performed as described by Yan et al. (40). Briefly, the cells were treated with 1% formaldehyde, followed by the addition of 0.125 M glycine. The cells were pelleted and resuspended in 200 µl of SDS lysis buffer (1% SDS, 10 mM EDTA, and 50 mM Tris-HCl, pH 8.1, containing protease inhibitors). The cell lysates were then sonicated with a sonic dismembrator (Fisher) at 30% maximum power for six 20-s pulses on ice. Cell lysates were diluted 10-fold in immunoprecipitation buffer (0.01% SDS, 1.1% Triton X-100, 1.2 mM EDTA, 16.7 mM Tris-HCl, pH 8.1, 167 mM NaCl, and protease inhibitors). Chromatin solutions were incubated overnight at 4 °C with 5 µg of anti-AP-2 and anti-Sp1 antibodies. The immune complexes were then mixed with 60 µl of a 50% protein A-agarose slurry containing 20 µg of single-stranded DNA and 1 mg/ml bovine serum albumin (Upstate Biotechnology, Inc., Lake Placid, NY). The immune complexes were eluted by adding 250 µl of 1% SDS in 0.1 M NaHCO3 to the pelleted beads and then incubated at room temperature for 15 min. Then 20 µl of 5 M NaCl was added, and the complexes were incubated at 65 °C for 4 h. The DNA was recovered by phenol/chloroform extraction and ethanol precipitation using 20 µg of glycogen as a carrier. The precipitated DNA was then dissolved in 20 µl of TE buffer and analyzed by PCR. PAR-1 primer sequences (5'-ACT TCT AGG CCC GGC AGT G-3' and 5'-GGT AAG ATC AGG GTC CAA GC-3') were used. The PCR was subjected to an initial denaturation step (2 min at 96 °C), followed by 30 cycles of denaturation (1 min at 94 °C), annealing (1 min at 60 °C), and extension (1 min at 72 °C). Then reaction was subjected to a final extension time of 5 min at 72 °C. PCR products were analyzed on a 3% agarose gel containing ethidium bromide.

Semiquantitative Reverse Transcriptase-PCR—One microgram of total RNA was reverse-primed with an oligo(dT) primer and extended with Maloney murine leukemia virus reverse transcriptase (Clontech, Palo Alto, CA). The PCR was performed, using the Clontech Advantage cDNA PCR kit, in a 50-µl reaction mixture containing 1x PCR buffer, 5 µl of cDNA, 0.2 mM dNTP, and 2.5 units of Taq polymerase. For quantitation of AP-2, cDNA was amplified by PCR using specific primers for human AP-2 (sense, 5'-CTG CCA ACG TTA CCC TGC-3'; antisense, 5'-TAG TTC TGC AGG GCC GTG-3') and the housekeeping gene GAPDH (sense 5'-GAG CCA CAT CGC TCA GAC-3'; antisense, 5'-CTT CTC ATG GTT CAC ACC C-3'). AP-2 and GAPDH cDNAs were amplified by PCR in the same reaction mixture as follows: an initial denaturation for 2 min at 94 °C, followed by 27 cycles of denaturation at 94 °C for 1 min, annealing at 58 °C for 1 min, and extension at 72 °C for 1 min, with a final elongation step at 72 °C for 5 min. For PAR-1 quantitation, specific primers (5'-GCA GAG CCC GGG ACA ATG GGG-3' and 5'-AGA TGG CCA GAC AAG TGA AGG-3') were used. The PCR was carried out by an initial denaturation at 94 °C for 5 min, followed by 30 cycles of denaturation at 94 °C for 1 min, annealing at 63 °C for 1 min, and extension at 72 °C for 1 min. A final elongation step was carried out at 72 °C for 5 min.

Expression Constructs—pcDNA3.1 (Invitrogen) is a mammalian cell expression construct in which expression is driven by the cytomegalovirus promoter and provides neomycin resistance. The pcDNA3.1-AP-2 construct was used for stable and transient transfections and was created by cloning the AP-2 cDNA into the EcoRI site of pcDNA3.1. The pcDNA3.1-Sp1 construct that contained the cDNA encoding for Sp1 was used for transient transfections.

Construction of PAR-1 Luciferase Plasmids—DNA from human melanoma WM266-4 cells was prepared with a Qiagen genomic tip system (Valencia, CA) according to the manufacturer's instructions. The nucleotide upstream of the translation start site was numbered –1. The 1400-, 890-, and 500-bp genomic fragments upstream of the translation start site were generated by PCR from genomic DNA using the Qiagen PCR kit. The underlined restriction sites in each primer were added to the 5' ends to facilitate subcloning. PCR was used to generate the 1400-bp product with primers complementary to the regulatory region of the PAR-1 gene (sense 5'-GGT ACC GCC AGT GGC AAA GCA ACT TA-3' and antisense 5'-GCT AGC CTC TCT CCT GAC TTC TGC GG-3'). The 890-bp product of the PAR-1 promoter was generated with the following primers: sense, 5'-GGT ACC CGC TCT TCC TAT TCC ACT C-3'; antisense, 5'-GCT AGC CTC TCT CCT GAC TTC TGC GG-3'. Similarly, the 500-bp region of the PAR-1 promoter was generated with the primers sense 5'-GGT ACC CGG TCC CAT TCC AAG GAC-3' and antisense 5'-GCT AGC CTC TCT CCT GAC TTC TGC GG-3'. The PCR products were subcloned in the promoterless pGL3 basic vector (Promega) and digested with NheI and KpnI, generating the constructs –1400/PAR1-Luc, –890/PAR1-Luc, and –500/PAR1-Luc. The constructs were confirmed by sequence analysis.

Reporter Constructs—The 3xAP-2-Luc luciferase reporter construct contains three AP-2 consensus elements from the human metallothionein gene IIA ligated in front of a minimal thymidine kinase promoter, as previously described (3). The luciferase reporter plasmid 3xSP1-Luc contains three consensus Sp1-binding sites from the SV40 promoter.

Site-directed Mutagenesis—Mutagenesis was performed with the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA) in accordance with the manufacturer's instructions. The following oligonucleotides (single-stranded sense and antisense) were used as primers to introduce mutations within the AP-2- and Sp1-binding elements between bp –365 and –329 of the PAR-1 promoter (mutated nucleotides are indicated by underlined letters): M1 sense, 5'-CCA GTA GGG CAG TTC GGT TCG GGG CGG GGC GCA CAG A-3'; M1 antisense, 5'-GCG CCC CGC CCC GAA CCG AAC TGC CCT ACT GG-3'; M2 sense, 5'-CCA GTA GGG CAG GGC GGT TCG GGG CTT GGC GCA CAG A-3'; M2 antisense, 5'-CTG TGC GCC AAG CCC CGA ACC GCC CTG CCC-3'; M3 sense, 5'-CCA GTA GGG CAG TTC GGG GCG GGG CTT GGC GCA CAG A-3'; M3 antisense, 5'-CTG TGC GCC AAG CCC CGC CCC GAA CTG CCC TAC TGG-3'; M4 sense, 5'-CCA GTA GGG CAG TTC GGT TCG GGG CTT GGC GCA CAG A-3'; M4 antisense, 5'-CTG TGC GCC AAG CCC CGA ACC GAA CTG CCC TAC TGG-3'. PCRs were performed with the wild-type –500/PAR1-Luc reporter vector as a template. The newly synthesized PCR products were digested with DpnI and used for transformation. The presence of mutations was verified by sequencing. Reporter vectors containing mutations were selected for large scale DNA preparations and used in transfection experiments

Transient Transfection and Luciferase Reporter Assay—The cells were grown in 12-well plates to 60% confluence for at least 18 h and then were transiently transfected using Lipofectin reagent (Invitrogen) with 1.5 µg of a firefly luciferase reporter gene and 50 ng of the Renilla luciferase reporter gene, driven by the {beta}-actin promoter (Promega, Madison, WI). Cotransfection was performed by adding 1 µg of expression constructs to the DNA solutions. Luciferase activity was determined using the dual luciferase reporter assay system (Promega) in a microplate Luminoskan Ascent luminometer (Thermo Labsystems Inc., Franklin, MA) according to the manufacturer's instructions. Normalization of transfection efficiency was based on cotransfected {beta}-actin Renilla luciferase activities.

Densitometric Quantitation—Images were captured in a Gel Doc 2000 system (Bio-Rad) connected to a CCD camera. Densitometric readings of DNA fragments separated in agarose gels were quantitated using QuantityOne software. RNase protection assay and Western blot densitometric analysis were performed in the linear range of the film using QuantityOne software.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
PAR-1 Expression in Human Melanoma Cell Lines Correlates with Metastatic Potential and Is Inversely Correlated with AP-2 Expression—The cell lines in this investigation included primary and metastatic human melanoma cells with different tumorigenic and metastatic potentials in nude mice (34). Since the regulatory region of the PAR-1 gene contains several putative AP-2 binding sites, a panel of AP-2-negative and -positive melanoma cell lines was surveyed by RNase protection assay for differential expression of PAR-1. To further investigate the contribution of AP-2 in human melanoma, we reexpressed AP-2 in the human metastatic melanoma cell line WM266-4, which expressed negligible levels of endogenous AP-2. AP-2 expression, DNA binding activity, and promoter activation by AP-2 in the transfected cells were verified by reverse transcriptase-PCR, Western blot analysis, EMSA, and luciferase reporter assay (data not shown).

We observed a correlation between the level of PAR-1 expression and the metastatic potential of human melanoma cell lines. As shown in Fig. 1, high levels of PAR-1 mRNA were found in the metastatic melanoma cell lines A375SM and WM266-4 (lanes 1 and 2); on the other hand, low levels were found in the low and nonmetastatic melanoma MeWo and SB2 cells, respectively (lanes 8 and 9). The expression of PAR-1 was significantly lower in the WM266-4-AP-2-transfected cell lines C8 and C12A (lanes 4 and 5) than in the WM266-4 parental and neotransfected cell line (lanes 2 and 3). Similar results were observed in the metastatic breast cancer cell line MDA-MB-231, which expressed high levels of PAR-1 (lane 7), whereas the nonmetastatic breast cancer cell line MCF-7 expressed undetectable levels of PAR-1 (lane 6). The PAR-1 mRNA amounts determined by densitometry, which was normalized to GAPDH expression, were ~4.5-fold higher in the metastatic melanoma cell lines (A375SM and WM266-4) than in the low and nonmetastatic melanoma cell lines (MeWo and SB2). In summary, PAR-1 was highly expressed in metastatic melanoma and breast cancer cell lines and expressed at significantly lower levels in nonmetastatic cell lines.



View larger version (44K):
[in this window]
[in a new window]
  		FIG. 1.
RNase protection assay for the expression of PAR-1 in human carcinoma cell lines. Metastatic melanoma cell lines (WM266-4 and A375SM) expressed high levels of PAR-1 transcripts, whereas low and nonmetastatic melanoma cell lines (MeWo and SB2) expressed low levels of PAR-1. Down-regulation of PAR-1 gene expression was observed in AP-2-transfected cells in comparison with the parental and neotransfected (Neo C5) cells. The metastatic breast cancer cell line MDA-MB-231 expressed high levels of PAR-1, whereas the nonmetastatic breast cancer cell line MCF7 did not express PAR-1. Metastatic potential in nude mice was as follows: H, high; L, low; N, nonmetastatic.

 
The RNase protection assay indicated that reexpression of AP-2 in WM266-4 cells resulted in decreased expression of PAR-1, which led us to further investigate the relationship between the expression of AP-2 and PAR-1. We evaluated by reverse transcriptase-PCR and Western blot analysis the expression of AP-2 and PAR-1 in a panel of melanoma cell lines (Fig. 2). The metastatic human melanoma cell lines WM266-4 and A375SM (Fig. 2, A and B, lanes 1 and 7) expressed negligible levels of AP-2 and high levels of PAR-1. In contrast, low and nonmetastatic melanoma cell lines MeWo and SB2 (Fig. 2, A and B, lanes 5 and 6) expressed high levels of AP-2 and low levels of PAR-1. Reexpression of AP-2 in WM266-4 cells resulted in down-regulation of the PAR-1 mRNA and protein (Figs. 2, A and B, lanes 3 and 4).



View larger version (57K):
[in this window]
[in a new window]
  		FIG. 2.
Expression of AP-2 and PAR-1 in human melanoma cell lines. Shown are reverse transcriptase-PCR (A) and Western blot analysis (B) for the expression of PAR-1 and AP-2 in human melanoma cell lines. High levels of PAR-1 expression were observed in WM266-4, neotransfected (Neo C5), and A375SM cell lines, which expressed negligible levels of AP-2. AP-2-transfected C8 and C12A, MeWo, and SB2 melanoma cell lines expressed high levels of AP-2 and low levels of PAR-1. The AP-2-negative breast cancer cell line MDA-MB-231 expressed high levels of PAR-1, whereas the AP-2-positive breast cancer cell line MCF7 lacked expression of PAR-1.

 
Similar results were obtained in breast carcinoma cell lines. The metastatic breast cancer cell line MDA-MB-231 expressed high levels of PAR-1 and undetectable levels of AP-2 (Fig. 2, A and B, lane 9), whereas the nonmetastatic breast cancer cell line MCF-7 expressed high levels of AP-2 and did not express PAR-1 (Fig. 2, A and B, lane 8). Collectively, these data demonstrate an inverse correlation between the expression of AP-2 and the expression of PAR-1 in both melanoma and breast carcinoma cell lines.

In Vivo ChIP Analysis of the PAR-1 Promoter Demonstrates AP-2 and Sp1 Binding—The regulatory region of the PAR-1 gene has multiple putative AP-2- and Sp1-binding motifs, suggesting that these transcription factors probably mediate the promoter activity of PAR-1 (31, 32). To determine the roles of AP-2 and Sp1 in regulating the expression of PAR-1, we investigated whether these nuclear proteins were associated with the PAR-1 promoter in vivo, using the ChIP assay. Chromatin fragments from cultured cells were immunoprecipitated with an antibody to either AP-2 or Sp1, and DNA from the immunoprecipitates was isolated. From this DNA, a 276-bp fragment of the PAR-1 promoter region was amplified by PCR. This region of the promoter contained two AP-2·Sp1 complexes with multiple overlapping binding motifs for these transcription factors (Fig. 3A) (31). In cells that did not express AP-2 (WM266-4 and MDA-MB-231), Sp1 was predominantly bound to the PAR-1 promoter (Fig. 3B). In contrast, in cells that expressed AP-2 (WM266-4-AP-2-C8 and MCF-7), 9–10-fold more AP-2 was bound to the same region of the promoter (Fig. 3B). These data demonstrate that both AP-2 and Sp1 bind to the 3' regulatory region of the PAR-1 promoter in a mutually exclusive manner.



View larger version (31K):
[in this window]
[in a new window]
  		FIG. 3.
ChIP assay to determine transcription factors binding to the PAR-1 promoter in vivo. A, 5'-flanking sequence of the PAR-1 gene and potential AP-2 and Sp1 regulatory motifs. AP-2-like elements are noted as open small ovals, AP-2-binding motifs are noted as large dashed ovals, and the Sp1-binding motifs are noted as black circles. B, the immunoprecipitated (IP) DNA was purified, and the regions containing complex 1 and complex 2 were amplified by PCR. The signal ratio for AP-2 binding to its motif in the promoter was 0.06 in WM266-4 cells and 0.07 in MDA-MB-231 cells. The ratio was increased in cells expressing AP-2 (WM266-4-AP-2-C8 cells, 0.57; MCF-7 cells, 0.74). The signal ratio for Sp1 binding was 0.59 in WM266-4 cells and 0.36 in MDA-MB-231. The ratio was decreased in cells expressing AP-2 (WM266-4-AP-2-C8, 0.17; MCF-7, 0.07).

 
AP-2 and Sp1 Bind to Complex 1 of the PAR-1 Promoter, Forming Distinct DNA-Protein Complexes—The in vivo analysis of the promoter by ChIP led us to further examine the two AP-2·Sp1 complexes within the PAR-1 promoter (Fig. 3A). To determine which nuclear proteins were bound to the regulatory region, we performed EMSA with oligonucleotide probes that corresponded to bp –365 to –329 (complex 1) and bp –206 to –180 (complex 2) of the 5'-flanking sequence of the PAR-1 gene (Table I) (31). To confirm the DNA binding activity of the nuclear proteins AP-2 and Sp1, nuclear extracts from WM266-4-AP-2-C8 cells were used in a gel shift assay that contained end-labeled double-stranded oligonucleotides of AP-2- and Sp1-binding motifs. Extracts from the WM266-4-AP-2-C8 cells revealed both AP-2 and Sp1 DNA binding to the AP-2 and Sp1 probes (data not shown). Incubation of the radiolabeled complex 1 (C1) probe with nuclear extracts from the WM266-4-AP-2-C8 cells resulted in the formation of four distinct shifted bands (labeled A, B, C, and D in Fig. 4A, lane 1). We also observed DNA-protein complexes migrating faster than these bands, as seen in Fig. 4A (lanes 5–7). The complexes appeared intermittently in our reactions but most often were present in the supershift analysis and, thus, might have represented complexes stabilized by the antibody or nonspecific interactions. The observed bands A, B, C, and D were identified in part by competition with a 100-fold excess of unlabeled double-stranded oligonucleotides consisting of Sp1 and AP-2 consensus sequences. The binding of unlabeled Sp1 oligonucleotides inhibited the formation bands B, C, and D (Fig. 4A, lane 3), whereas the binding of unlabeled AP-2 oligonucleotides inhibited the formation of band A (Fig. 4A, lane 4).


View this table:
[in this window]
[in a new window]
  		TABLE I
Sequence of PAR-1 promoter oligonucleotides used as EMSA probes

Sequences of electrophoretic mobility shift assay putative binding elements are shown. The AP-2 site is depicted as a solid line, and the Sp1 is site depicted as a broken line. Oligonucleotide sequences are written as the sense strand in the 5' -> 3' direction, and mutations against the wild-type sequence are noted in boldface type.

 



View larger version (58K):
[in this window]
[in a new window]
  		FIG. 4.
EMSA. A, to determine AP-2 and Sp1 binding to complex 1 of the PAR-1 promoter, nuclear extracts from WM266-4-AP-2-transfected C8 cells were incubated with a radiolabeled complex 1 (C1) probe (lane 1). The protein-DNA complexes were confirmed by competition with unlabeled Sp1 and AP-2 oligonucleotides (lanes 3 and 4). Supershift assays with specific antibodies for Sp1, Sp3, and AP-2 are depicted in lanes 5–7. To determine the nucleotide sequences required for the binding of AP-2 and Sp1 to complex 1, we synthesized a series of mutated complex 1 oligonucleotides (see Table I). Mutated complex 1 oligonucleotides were incubated with nuclear extracts from WM266-4-AP-2-transfected C8 cells. Competition and supershift assays were used to confirm the identity of these complexes. B, EMSA with the mutant probe M1. C, EMSA with the mutant probe M3. D, to determine whether AP-2 and Sp1 bind to complex 2 of the PAR-1 promoter, nuclear extracts from WM266-4-AP-2-transfected C8 cells were incubated with a radiolabeled complex 2 (C2) probe (lane 1). The protein-DNA complexes were confirmed by competition with unlabeled Sp1 and AP-2 oligonucleotides (lanes 3 and 4). Supershift assays with specific antibodies for Sp1, Sp3, and AP-2 are shown as lanes 5–7.

 
To further define the nuclear proteins associated with complex 1, we performed supershift assays. When nuclear extracts were incubated with a specific anti-Sp1 antibody, bands B and C were diminished, and retarded (supershifted) complexes were formed, which indicates that these bands contained immunoreactive Sp1 protein (Fig. 4A, lane 5). Incubation of the nuclear extracts with a specific anti-Sp3 antibody abolished bands C and D and resulted in supershifted bands (Fig. 4A, lane 6). Similarly, incubation with a specific anti-AP-2 antibody resulted in the loss of band A and the formation of a supershifted band (Fig. 4A, lane 7).

To identify the nucleotide sequences required for AP-2 and Sp1 to bind to complex 1, we synthesized a series of mutated complex 1 oligonucleotides (Table I). The mutations were introduced to the overlapping AP-2 and Sp1 motifs and thus altered the nucleotides important for AP-2 and Sp1 to bind to the native sequence. We incubated radiolabeled mutant probes with nuclear extracts from WM266-4-AP-2-C8 cells. EMSA indicated that the mutations introduced in mutant probes eliminated band A and greatly affected the ability of AP-2 to bind to complex 1 (Fig. 4, B and C). The bindings of the nuclear factors Sp1 and Sp3 were also affected by mutations within the sequence of complex 1. The mutated probes M1 and M3 did not alter the formation of bands B, C, and D (Fig. 4, B and C), which coincided precisely with the DNA-protein complexes of complex 1 of the promoter (Fig. 4A). However, the mutant probes M2 and M4 completely abolished the formation of all DNA-protein complexes (data not shown), which suggests that AP-2, Sp1, and Sp3 bound to specific sequences within the PAR-1 promoter.

To further characterize the regulatory regions of the PAR-1 promoter, we performed EMSA using complex 2 (C2) as a probe and nuclear extracts from WM266-4-AP-2-C8 cells. Three major bands A, B, and C appeared (Fig. 4D). Competition and supershift assays confirmed the identity of distinct Sp1 and Sp3 DNA-protein complexes (Fig. 4D). In contrast, the formation of DNA-protein complexes with the sequence corresponding to complex 2 of the promoter did not involve AP-2 binding, which demonstrated binding specificity of AP-2 with complex 1 of the PAR-1 promoter. Taken together, the results presented in Fig. 4 demonstrate that AP-2, Sp1, and Sp3 bound to overlapping motifs in the complex 1 element of the human PAR-1 promoter in a mutually exclusive manner.

The 5' GC Box (Complex 1) Is a Functional Element of the PAR-1 Promoter: Repression by AP-2 and Transactivation by Sp1—To define the functional elements responsible for PAR-1 gene regulation, we constructed a series of luciferase reporter plasmids containing serial 5' deletions (Fig. 5A). These plasmid constructs were transfected into a panel of melanoma cell lines that differentially express PAR-1. Using the {beta}-actin-Renilla luciferase plasmid as a control, to normalize for transfection efficiency, we found that the promoter activity driven by the –1400/PAR-1-Luc construct was increased in A375SM and WM266-4 cell lines and decreased in WM266-4-AP-2-C8, SB2, and MeWo cell lines (Fig. 5B). Deletion of bp –1400 to –891 to create the –890/PAR1-Luc construct did not produce any marked differences in PAR-1 promoter activity (Fig. 5B), which implied that the elements in this region were not required for basal activity of the PAR-1 promoter. Further deletions of sequences between bp –890 to –501, which left the remaining 500-bp region of the promoter (–500/PAR1-Luc), increased PAR-1 promoter activity by 2-fold in A375SM and WM266-4 cell lines, whereas only a minimal effect was observed in WM266-4-AP-2-C8, SB2, and MeWo cell lines (Fig. 5B). This indicated that this 500-bp segment of the promoter is driven by a strong transcriptional activator. Furthermore, this 500-bp region of the promoter contains both complex 1 (bp –365 to –329) and complex 2 (bp –206 to –180), implicating these regulatory motifs in the regulation of this gene.



View larger version (41K):
[in this window]
[in a new window]
  		FIG. 5.
Luciferase reporter assay. A, schematic of the PAR-1 promoter luciferase constructs. Full-length PAR-1 promoter, bp –1400 to +1, and promoter deletions, bp –890 to +1 and bp –500 to +1, were cloned into the luciferase reporter construct pGL3. AP-2-like elements are depicted as small open ovals. Sp1 sites are shown as black circles. Complex 1 contains three AP-2 sites and five Sp1 sites and is depicted as a large gray oval. Complex 2 contains a single AP-2 site and two Sp1 sites and is shown as a striped oval. B, PAR-1 promoter luciferase activity in human melanoma cells. The human melanoma cell lines A375SM, WM266-4, WM266-4-AP-2-C8, SB2, and MeWo were transfected with full-length PAR-1 reporter constructs and deletion constructs. Luciferase activity was measured 36 h after transfection. Data shown are cumulative of three independent experiments performed in triplicate. C, WM266-4 and WM266-4-AP-2-C8 cells were co-transfected with –500/PAR1-Luc reporter construct and expression plasmids for AP-2, Sp1, or an empty vector (pcDNA3.1). Luciferase activity was measured 36 h after transfection. Data shown are cumulative of three independent experiments performed in triplicate. D, schematic representation of PAR-1 deletion mutant luciferase constructs. The base pair substitutions within the AP-2 and Sp1 binding sites of complex 1 are noted in boldface type and underlined. E, WM266-4 and WM266-4-AP-2-C8 cells were cotransfected with PAR-1 deletion mutant reporter constructs and expression plasmids for AP-2, Sp1, or an empty vector (pcDNA3.1). Luciferase activity was measured 36 h after transfection. Data shown are cumulative of three independent experiments performed in triplicate.

 
To further determine the effects of AP-2 and Sp1 on PAR-1 promoter activity, expression constructs containing cDNA encoding for AP-2 and Sp1 or an empty vector (pcDNA3.1-Neo) were transiently cotransfected with –500/PAR1-Luc construct in WM266-4 and WM266-4-AP-2-C8 cell lines (Fig. 5C). In comparison with –500/PAR1-Luc construct cotransfected with pcDNA3.1-Neo, transfection with –500/PAR1-Luc construct and pcDNA3.1-AP-2 resulted in a marked decrease in reporter gene activity in WM266-4 cells and a further decrease (1.5–2-fold) in luciferase activity in WM266-4-AP-2-C8 cells (Fig. 5C). In contrast, cotransfection of –500/PAR1-Luc construct with pcDNA3.1-Sp1 resulted in a marked increase in promoter activity by 2–2.5-fold. These results demonstrate that AP-2 functioned as a transcription repressor of PAR-1, whereas Sp1 was a strong transcriptional activator of PAR-1.

To assess the functional role of complex 1 (bp –365 to –329) in PAR-1 gene regulation, we performed site-directed mutagenesis within the Sp1 and AP-2 sites, similar to the mutations used in the gel shift analysis (Fig. 5D). Mutant luciferase reporter constructs were transiently transfected into the WM266-4 and WM266-4-AP-2-C8 melanoma cell lines, and their activity was compared with that of the –500/PAR1-Luc, which contained the native PAR-1 promoter sequences (Fig. 5E). Disruption of complex 1 with 2-bp mutations at the distal and middle regions (M1/PAR1-Luc) increased promoter activity in the WM266-4-AP-2-C8 cells by 1.4-fold, but no changes in luciferase activity were observed in WM266-4 cells compared with –500/PAR1-Luc activity. Cotransfection of M1/PAR1-Luc with AP-2 in WM266-4 and WM266-4-AP-2-C8 cell lines had no effect on promoter activity. Transfection with Sp1 resulted in a 2-fold increase in luciferase activity. Similarly, 2-bp mutations in the middle and proximal regions of complex 1 (M2/PAR1-Luc) decreased promoter activity by 0.5–2-fold in WM266-4 and WM266-4-AP-2-C8 cell lines. Introduction of AP-2 had no effect on promoter activity, whereas Sp1 had a minimal increase in the promoter activity of M2/PAR1-Luc. When 2-bp mutations were introduced in the distal and proximal region of complex 1 (M3/PAR1-Luc), we observed a ~2-fold increase in promoter activity in WM266-4 and WM266-4-AP-2-C8 cell lines compared with –500/PAR1-Luc. Transfection with AP-2 had little effect on promoter activity, whereas Sp1 transactivated M3/PAR1-Luc. To our surprise, when 2-bp mutations were introduced simultaneously at the distal, middle, and proximal region of complex 1 (M4/PAR1-Luc), promoter activity was abolished in WM266-4 and WM266-4-AP-2-C8 cell lines, thus inhibiting the positive effect of Sp1 transactivation on complex 1 (bp –365 to –329).

The results of this targeted mutation analysis of the PAR-1 promoter, in conjunction with the deletion analysis and cotransfection with AP-2 and Sp1 expression plasmids, demonstrated that complex 1 was necessary for activation by Sp1 and repression by AP-2. Furthermore, when this region was significantly mutated, the promoter activity was lost, which suggests that complex 2 (bp –206 to –180) may have a minimal role in promoter activation.

The Ratio of AP-2 to Sp1 DNA Binding Activity and Expression Are Lower in Metastatic Melanoma Cell Lines—The presence of overlapping AP-2- and Sp1-binding motifs within the PAR-1 promoter suggests that AP-2 probably repressed expression of this gene by interfering with activated or basal transcription. Therefore, the levels of AP-2 and Sp1 DNA binding activity and expression in relation to the level of PAR-1 expression were examined by luciferase reporter assay and Western blot analysis. To determine the DNA binding activity of AP-2 and Sp1, we performed luciferase assays using either a reporter construct that contained three AP-2-binding sites (3xAP-2-Luc) or a vector that contained three Sp1-binding motifs (3xSp1-Luc) in a panel of human melanoma cell lines (Fig. 6A). The data in Fig. 6B demonstrate that metastatic melanoma cell lines A375SM and WM266-4 had very low AP-2 binding activity in comparison with the low and nonmetastatic cell lines MeWo and SB2, which had significantly higher AP-2 activity. The AP-2 binding activity was 2.7-fold higher in WM266-4-AP-2-C8 cell line than in the parental WM266-4 cell line. Sp1 DNA binding activity was similar in the metastatic cell lines A375SM and WM266-4. The Sp1 activity was only slightly lower in SB2 cell line than in to A375SM and WM266-4 cell lines. However, we observed a marked decrease in Sp1 DNA binding activity in the MeWo cell line. To calculate the ratio of AP-2 to Sp1 DNA binding activity, we used their relative luciferase activities. We determined that the metastatic cell lines A375SM and WM266-4 had a low AP-2/Sp1 ratio (~0.23), and the low and nonmetastatic MeWo and SB2 cell lines had significantly higher ratios of 3.1 and 1.4, respectively. The WM266-4-AP-2-C8 cells, in which AP-2 activity was restored, had a ratio equal to 0.68, which suggests that the balance of the transcriptional regulators determines the level of PAR-1 expression in melanoma cells. Similarly, the AP-2/Sp1 ratio was determined by Western blot and densitometry analyses (normalized to {beta}-actin expression) (Fig. 6C). The metastatic melanoma cell lines A375SM and WM266-4 had an AP-2/Sp1 ratio of less than or approximately equal to 1 and expressed high levels of PAR-1 (Fig. 6C). In contrast, the low and nonmetastatic melanoma cell lines MeWo and SB2 had a ratio of AP-2/Sp1 greater than 1 and expressed minimal levels of PAR-1 (Fig. 6C). The WM266-4-AP-2-C8 cells displayed an increase in the ratio of AP-2/Sp1, which coincided with decreased PAR-1 expression. We conclude that the loss of AP-2 in melanoma cells resulted in a noticeable decrease in the ratio of AP-2 to Sp1 and correlated with high expression levels of PAR-1 and metastatic potential. Furthermore, reexpression of AP-2 restored the ratio of AP-2 expression to Sp1 expression in melanoma cells and resulted in down-regulation of PAR-1 (Fig. 7).



View larger version (30K):
[in this window]
[in a new window]
  		FIG. 6.
AP-2 and Sp1 DNA binding activity and expression in human melanoma cell lines. A, illustrations of the luciferase reporter gene 3xAP-2 driven by a trimer of AP-2-binding motifs and minimal thymidine kinase promoter and the luciferase reporter gene 3xSp1 driven by a trimer of Sp1-binding elements. B, luciferase activity of 3xAP-2 and 3xSp1 reporter plasmids in A375SM, WM266-4, WM266-4-AP-2-C8, SB2, and MeWo cell lines. The ratio of AP-2 to Sp1 DNA binding activity was determined from the luciferase activity of 3xAP-2 and 3xSp1. Luciferase activity was measured 36 h after transfection. Data shown are cumulative of three independent experiments performed in triplicate. C, Western blot analysis for the expression of Sp1, AP-2, {beta}-actin, and PAR-1 in human melanoma cell lines. Relative Sp1 and AP-2 expression levels were quantified by densitometry to determine the AP-2/Sp1 ratio. Melanoma cells with an AP-2/Sp1 ratio approximately equal to 1 (A375SM, WM266-4, and neotransfected cell line) expressed high levels of PAR-1, whereas cells with ratios greater than 1 (MeWo, SB2, and AP-2-transfected cell line) expressed low levels of PAR-1.

 



View larger version (19K):
[in this window]
[in a new window]
  		FIG. 7.
Schematic of the proposed mechanism for PAR-1 promoter regulation by the transcriptional regulators AP-2 and Sp1. We hypothesize that the low levels of AP-2 expressed in malignant melanoma cells enable Sp1 to bind to its motif in the PAR-1 promoter, resulting in activation of the gene. AP-2-positive nonmetastatic melanoma cells and restoration of AP-2 expression in metastatic melanoma cells result in AP-2 binding to its element in the promoter and suppressing PAR-1 gene expression. We conclude that loss of AP-2 expression in malignant human melanoma cells results in overexpression of PAR-1.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
This investigation provides evidence that loss of AP-2 results in increased expression of PAR-1 in metastatic human melanoma cells, suggesting that PAR-1 overexpression contributes to the metastatic phenotype of melanoma. However, it should be noted that PAR-1 alone does not induce a metastatic phenotype; other malignant and invasive properties are also required. For example, Nierodzik et al. (28) reported that NIH/3T3 fibroblast cells, which expressed PAR-1 at low levels, and NIH/3T3 cells transfected to overexpress PAR-1 did not form pulmonary metastases after being intravenously injected into syngeneic mice. In a similar study conducted using B16F10 melanoma cells transfected with PAR-1, the metastatic potential increased 5–6-fold, demonstrating the need for already established metastatic properties of malignant cells (28). EvenRam et al. (30) demonstrated that PAR-1 increased the invasive properties of human melanoma cells in vitro, primarily by promoting adhesion to extracellular matrix components. In their study, Even-Ram et al. (30) used the noninvasive human melanoma cell line SB2 transfected with PAR-1 cDNA and compared its properties to those of the highly invasive melanoma cell line A375SM. Indeed, SB2 clones expressing high levels of PAR-1 had an increased ability to invade, as was determined by Matrigel invasion assays. To that end, the invasiveness of the PAR-1-transfected SB2 cells was further enhanced when they were activated by either thrombin or the thrombin receptor-activating peptide that corresponds to the PAR-1 internal ligand SFFLRN. The invasive properties of the PAR-1-transfected SB2 cell line were explained, in part, by its preferential adhesion and specific recruitment of integrin {alpha}v{beta}5 to focal contact sites accompanied by the cytoskeletal reorganization of F-actin toward migration-favoring morphology. Furthermore, PAR-1 probably contributed to metastasis by augmenting already established invasive properties of malignant cells. For example, the urokinase gene was induced by the activation of PAR-1 in human PC-3 prostatic cancer cell line (41), resulting in the production of an enzyme required for tumor cell invasion (42, 43). These observations further implicate PAR-1 in the metastatic process of tumor progression and invasion.

The formation of new blood vessels is a critical determinant of tumor progression. Tumors are limited in size by their access to oxygen and nutrients from nearby blood vessels. Hence, solid tumors cannot exceed a diameter of 2 mm without the innermost cells undergoing necrosis (44). Thrombin is a potent promoter of angiogenesis via activation of PAR-1. Indeed, activation of PAR-1 in a variety of cell types can elicit a range of cellular responses and expression of thrombin-responsive genes. Many of the gene products are precisely those required for tumor angiogenesis and invasion, including interleukin-8 (45), vascular endothelial growth factor (46), basic fibroblast growth factor (47), platelet-derived growth factor (48), matrix metalloproteinase-2 (49), urokinase-type plasminogen activator (41), and {alpha}IIb{beta}3 (50), {alpha}v{beta}3 (51), and {alpha}v{beta}5 integrins (30). This suggests that activation of PAR-1 may facilitate tumor invasion and metastasis by inducing the expression of cell adhesion molecules and matrix-degrading proteases and by stimulating the secretion of angiogenic factors. Taken together, these observations indicate a strong link between the cellular effects resulting from PAR-1 activation and the invasive properties of malignant melanoma.

Analysis of the PAR-1 promoter region revealed putative binding sites for many sequence-specific transcription factors, including Sp1, which is ubiquitously expressed, and AP-2, which is a cell type-specific transcriptional regulator. We noted with special interest the presence of multiple AP-2 consensus sequences in the PAR-1 promoter. AP-2 is a developmentally regulated and tissue-specific transcription factor expressed primarily in the neural crest (from which melanocytes are derived) and epidermal cell lineages. Although AP-2 was initially described as a transcriptional activator, several recent studies have demonstrated its ability to behave as a repressor. In both the human acetylcholinesterase gene and the K3 keratin gene, there are overlapping Sp1 and AP-2 sites in the promoters, with Sp1 activating and AP-2 repressing gene transcription (52, 53). Indeed, our findings implicate Sp1 as a transcriptional activator and AP-2 as a transcriptional repressor of PAR-1 gene transcription. Although Sp1 is ubiquitously expressed, it is involved in the transcriptional regulation of a number of tissue-specific and differentiation-dependent genes, including genes that regulate neural, epithelial, and hematopoietic cell differentiation (52–54). A cooperation between ubiquitous and cell type-specific transcription factors is required to achieve precise gene expression. In fact, transactivation by Sp1 is due, in part, to its synergistic interaction with many cell type-specific and ubiquitous transcription factors. Examples of these factors include Ets (55, 56), E2F (57), HNF3 (58), STAT1 (59), Octamer (60), and GATA-1 (61). Interestingly, putative binding sites for Ets and Octamer have been found within the PAR-1 promoter region (32), raising the possibility of a cooperation between Sp1 and cell type-specific transcription factors in the regulation of PAR-1.

In our current investigation, we showed that Sp1 and AP-2 are important regulators of PAR-1 gene expression based on the following evidence. First, ChIP assays demonstrated an in vivo association between Sp1 and AP-2 and the PAR-1 promoter. We demonstrated that AP-2 is primarily bound to the PAR-1 promoter in low and nonmetastatic melanoma cell lines and that Sp1 is associated with the promoter in metastatic cell lines. Second, we have shown that AP-2 and Sp1 to bind bp –365 to –329 (complex 1) in vitro in a mutually exclusive manner. Third, functional promoter studies demonstrated an increase in promoter activity in cells lacking expression of AP-2 and a noticeable decrease in promoter activity in cells expressing high levels of AP-2. Furthermore, an increase in promoter activity was observed upon deletion of sequences just upstream of complex 1. Fourth, PAR-1 reporter gene assays, along with cotransfection of AP-2, resulted in decreased PAR-1 promoter activity, whereas cotransfection with Sp1 increased PAR-1 promoter activity. Finally, mutation analysis of complex 1 within the promoter indicated that PAR-1 is regulated via interactions with AP-2 and Sp1 at this functional regulatory element. In summary, we show that both AP-2 and Sp1 bind to the regulatory elements bp –365 to –329 of the PAR-1 promoter and that these transcriptional regulators have differential effects on promoter activity. Here we demonstrate that stably and transiently expression of AP-2 resulted in diminished PAR-1 promoter activity, whereas overexpression of Sp1 increased the reporter gene's activation in human melanoma cells. We hypothesize that the ubiquitous transcription factor Sp1 confers basal PAR-1 promoter activity by binding to GC boxes within bp –365 to –329 in the absence of a transcriptional competitor (most likely AP-2), resulting in very high levels of PAR-1 expression. On the other hand, a high level of AP-2 expression results in AP-2 occupying the region between bp –365 and –329, thus preventing Sp1 from binding and down-regulating PAR-1 expression. A schematic presentation of the interplay between Sp1 and AP-2 in regulating PAR-1 promoter activity in melanoma cells is illustrated in Fig. 7.

Alteration of transcription factor function has been established as a frequent cause of neoplastic transformation. However, little is known about how the dysfunction of transcriptional activators and repressors and the consequent alteration of specific transcriptomes lead to specific tumor phenotypes. Our present observations raise the interesting possibility that high levels of AP-2 in nonmetastatic cells hinder the binding of Sp1 to the complex 1 element of the PAR-1 promoter and thus suppression of PAR-1 gene expression. To that end, we investigated the DNA binding activity and expression levels of AP-2 and Sp1 in metastatic and nonmetastatic cell lines. We report here that the DNA binding activity and expression level of Sp1 was similar in metastatic and nonmetastatic cells. However, metastatic melanoma cells had a marked decrease in AP-2 DNA binding activity and expression; thus, the ratio of AP-2 to Sp1 activity and expression decreased. Furthermore, a decrease of AP-2 binding activity in relation to Sp1 binding activity correlated with an overexpression of PAR-1 in metastatic cell lines. We theorize that the noticeable increase in Sp1 over AP-2 DNA binding activity may have led to a greater occupation of the complex 1 element by Sp1 and, hence, activation of the PAR-1 gene in metastatic cells. In addition, we showed that restoring the balance of AP-2 and Sp1 by transfection with the AP-2 gene reduced expression of PAR-1. Collectively, our results provide evidence that a decrease of AP-2 expression in relation to Sp1 expression in metastatic melanoma cell lines contributes to the overexpression of PAR-1, thus probably enhancing the metastatic phenotype of these cells.

In summary, the data presented here add further weight to the notion that loss of AP-2 is a crucial event in the progression of human melanoma (62). Previously, we determined that c-KIT (3), MCAM/MUC18 (2), and matrix metalloproteinase-2 (4) genes, all of which contribute to the malignant phenotype of human melanoma, are regulated by AP-2. Here we have identified PAR-1 as an additional AP-2 target gene that can play an active role in melanoma progression; overexpression and activation of PAR-1 in melanoma cells allow these cells to modulate the expression of genes involved in invasion and metastasis.


   FOOTNOTES
 
* This work was supported by National Institutes of Health Grant CA76098. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

{ddagger} To whom correspondence should be addressed: The University of Texas M.D. Anderson Cancer Center, Dept. of Cancer Biology, 173, 1515 Holcombe Blvd., Houston TX 77030. Tel.: 713-794-4004; Fax: 713-792-8747; E-mail: mbareli{at}mdanderson.org.

1 The abbreviations used are: AP-2, activator protein-2; PAR-1, protease-activated receptor-1; Sp1, specificity protein 1; ChIP, chromatin immunoprecipitation; EMSA, electrophoretic mobility shift assay; GAPDH, glyceraldehyde-3-phosphate dehydrogenase. Back


   ACKNOWLEDGMENTS
 
We thank Dr. Boyd and Dr. Yan for technical assistance with the ChIP analysis and Walter Pagel and Katie Matias for expert scientific editing of the manuscript.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Bar-Eli, M. (1999) Cancer Metastasis Rev. 18, 377–385[CrossRef][Medline] [Order article via Infotrieve]
  2. Jean, D., Gershenwald, J. E., Huang, S., Luca, M., Hudson, M. J., Tainsky, M. A., and Bar-Eli, M. (1998) J. Biol. Chem. 273, 16501–16508[Abstract/Free Full Text]
  3. Huang, S., Jean, D., Luca, M., Tainsky, M. A., and Bar-Eli, M. (1998) EMBO J. 17, 4358–4369[CrossRef][Medline] [Order article via Infotrieve]
  4. Gershenwald, J. E., Sumner, W., Calderone, T., Wang, Z., Huang, S., and Bar-Eli, M. (2001) Oncogene 20, 3363–3375[CrossRef][Medline] [Order article via Infotrieve]
  5. Johnson, J. P., Bar-Eli, M., Jansen, B., and Markhof, E. (1997) Int. J. Cancer 73, 769–774[CrossRef][Medline] [Order article via Infotrieve]
  6. Huang, S., Luca, M., Gutman, M., McConkey, D. J., Langley, K. E., Lyman, S. D., and Bar-Eli, M. (1996) Oncogene 13, 2339–2347[Medline] [Order article via Infotrieve]
  7. Baldi, A., Santini, D., Battista, T., Dragonetti, E., Ferranti, G., Petitti, T., Groeger, A. M., Angelini, A., Rossiello, R., Baldi, F., Natali, P. G., and Paggi, M. G. (2001) J. Cell. Biochem. 83, 364–372[CrossRef][Medline] [Order article via Infotrieve]
  8. Karjalainen, J. M., Kellokoski, J. K., Eskelinen, M. J., Alhava, E. M., and Kosma, V. M. (1998) J. Clin. Oncol. 16, 3584–3591[Abstract]
  9. Zeng, Y. X., Somasundaram, K., and El-Deiry, W. S. (1997) Nat. Genet. 15, 78–82[CrossRef][Medline] [Order article via Infotrieve]
  10. Grether-Beck, S., Olaizola-Horn, S., Schmitt, H., Grewe, M., Jahnke, A., Johnson, J. P., Briviba, K., Sies, H., and Krutmann, J. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 14586–14591[Abstract/Free Full Text]
  11. Bosher, J. M., Williams, T., and Hurst, H. C. (1995) Proc. Natl. Acad. Sci. U. S. A. 92, 744–747[Abstract/Free Full Text]
  12. Hollywood, D. P., and Hurst, H. C. (1993) EMBO J. 12, 2369–2375[Medline] [Order article via Infotrieve]
  13. Turner, B. C., Zhang, J., Gumbs, A. A., Maher, M. G., Kaplan, L., Carter, D., Glazer, P. M., Hurst, H. C., Haffty, B. G., and Williams, T. (1998) Cancer Res. 58, 5466–5472[Abstract/Free Full Text]
  14. Descheemaeker, K. A., Wyns, S., Nelles, L., Auwerx, J., Ny, T., and Collen, D. (1992) J. Biol. Chem. 267, 15086–15091[Abstract/Free Full Text]
  15. Duan, C., and Clemmons, D. R. (1995) J. Biol. Chem. 270, 24844–24851[Abstract/Free Full Text]
  16. Wang, D., Shin, T. H., and Kudlow, J. E. (1997) J. Biol. Chem. 272, 14244–14250[Abstract/Free Full Text]
  17. Gille, J., Swerlick, R. A., and Caughman, S. W. (1997) EMBO J. 16, 750–759[CrossRef][Medline] [Order article via Infotrieve]
  18. Gille, J., Reisinger, K., Asbe-Vollkopf, A., Hardt-Weinelt, K., and Kaufmann, R. (2000) J. Invest. Dermatol. 115, 30–36[CrossRef][Medline] [Order article via Infotrieve]
  19. Batsche, E., Muchardt, C., Behrens, J., Hurst, H. C., and Cremisi, C. (1998) Mol. Cell. Biol. 18, 3647–3658[Abstract/Free Full Text]
  20. Jiang, J. G., DeFrances, M. C., Machen, J., Johnson, C., and Zarnegar, R. (2000) Biochem. Biophys. Res. Commun. 272, 882–886[CrossRef][Medline] [Order article via Infotrieve]
  21. Grand, R. J., Turnell, A. S., and Grabham, P. W. (1996) Biochem. J. 313, 353–368[Medline] [Order article via Infotrieve]
  22. Wojtukiewicz, M. Z., Tang, D. G., Ben-Josef, E., Renaud, C., Walz, D. A., and Honn, K. V. (1995) Cancer Res. 55, 698–704[Abstract/Free Full Text]
  23. Kaufmann, R., Schafberg, H., Rudroff, C., and Nowak, G. (1997) Cancer 80, 2068–2074[CrossRef][Medline] [Order article via Infotrieve]
  24. Henrikson, K. P., Salazar, S. L., Fenton, J. W., II, and Pentecost, B. T. (1999) Br. J. Cancer 79, 401–406[CrossRef][Medline] [Order article via Infotrieve]
  25. Rudroff, C., Schafberg, H., Nowak, G., Weinel, R., Scheele, J., and Kaufmann, R. (1998) Pancreas 16, 189–194[CrossRef][Medline] [Order article via Infotrieve]
  26. Liu, Y., Gilcrease, M. Z., Henderson, Y., Yuan, X. H., Clayman, G. L., and Chen, Z. (2001) Cancer Lett. 169, 173–180[CrossRef][Medline] [Order article via Infotrieve]
  27. Even-Ram, S., Uziely, B., Cohen, P., Grisaru-Granovsky, S., Maoz, M., Ginzburg, Y., Reich, R., Vlodavsky, I., and Bar-Shavit, R. (1998) Nat. Med. 4, 909–914[CrossRef][Medline] [Order article via Infotrieve]
  28. Nierodzik, M. L., Chen, K., Takeshita, K., Li, J. J., Huang, Y. Q., Feng, X. S., D'Andrea, M. R., Andrade-Gordon, P., and Karpatkin, S. (1998) Blood 92, 3694–3700[Abstract/Free Full Text]
  29. Fischer, E. G., Ruf, W., and Mueller, B. M. (1995) Cancer Res. 55, 1629–1632[Abstract/Free Full Text]
  30. Even-Ram, S. C., Maoz, M., Pokroy, E., Reich, R., Katz, B. Z., Gutwein, P., Altevogt, P., and Bar-Shavit, R. (2001) J. Biol. Chem. 276, 10952–10962[Abstract/Free Full Text]
  31. Li, F., Baykal, D., Horaist, C., Yan, C. N., Carr, B. N., Rao, G. N., and Runge, M. S. (1996) J. Biol. Chem. 271, 26320–26328[Abstract/Free Full Text]
  32. Schmidt, V. A., Vitale, E., and Bahou, W. F. (1996) J. Biol. Chem. 271, 9307–9312[Abstract/Free Full Text]
  33. Wu, Y., Ruef, J., Rao, G. N., Patterson, C., and Runge, M. S. (1998) Biochem. J. 330, 1469–1474[Medline] [Order article via Infotrieve]
  34. Luca, M., Xie, S., Gutman, M., Huang, S., and Bar-Eli, M. (1995) Oncogene 11, 1399–1402