| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
Inhibits Tumorigenicity and Represses Vascular Endothelial Growth Factor Transcription in Prostate Cancer Cells
1 Departments of Cancer Biology,
2 Urology, and
3 Surgical Oncology, University of Texas M. D. Anderson Cancer Center, Houston, Texas
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(AP-2) is a transcription factor that regulates proliferation and differentiation in mammalian cells. We have shown previously that although AP-2 is expressed highly in normal prostatic epithelium, its expression is lost in high-grade prostatic intraepithelial neoplasia and prostate cancer, suggesting that loss of AP-2 plays a role in prostate cancer development. We demonstrate that forced AP-2 expression in the prostate cancer cell line LNCaP-LN3 (AP-2 negative) inhibited dramatically tumor incidence in nude mice. To identify the genes that might have been responsible for this effect, we used microchip expression array. We found several genes known to be involved in malignancy were deregulated, including the vascular endothelial growth factor (VEGF) gene. Because VEGF was down-regulated by 14.7-fold in the AP-2-transfected cells and because it is a major angiogenic factor in prostate cancer development and progression, we chose to examine the AP-2-VEGF interaction. Our evidence suggests that AP-2 repressed transcriptionally the VEGF promoter by competing with the transcriptional activator Sp3. Loss of AP-2 in prostate cancer cells reduced the AP-2:Sp3 ratio and activated VEGF expression. AP-2 acts as a tumor-suppressor gene in prostate cancer. Elucidating the molecular events resulting from loss of AP-2 in the prostate epithelium has implications for the understanding and prevention of the onset of prostate cancer. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(AP-2) in prostate cancer clinical specimens and shown that although AP-2 is expressed highly in the luminal cells of the normal prostatic epithelium, its expression is lost in high-grade PIN and prostate cancer. These results suggest that loss of AP-2 is important in prostate cancer development, although the details remain unclear (2)
. We hypothesized that loss of AP-2 promotes the acquisition of the malignant phenotype by inducing aberrant expression patterns in the prostate epithelium. Loss of AP-2 expression has been associated with progression of melanoma, breast, and colorectal cancer. We have shown that normal melanocytes and nonmetastatic melanoma cell lines express high levels of AP-2, whereas highly metastatic melanoma cell lines do not express AP-2. Furthermore, transfection of highly metastatic melanoma cells with full-length AP-2 reduces greatly their tumorigenicity and metastatic potential in nude mice (3 , 4) . These observations have been supported by immunohistochemical studies of human melanoma, breast, and colorectal cancer specimens, in which loss of AP-2 coincides with poor prognosis (5, 6, 7, 8) .
This tumor-suppressor-like role of AP-2 in melanoma and breast cancer is linked directly to its ability to regulate the expression of genes involved in adhesion, survival, and invasion. Loss of AP-2 in metastatic melanoma causes deregulated expression of c-Kit, MUC18, thrombin receptor (PAR-1), and MMP-2 genes, inducing resistance to apoptosis, adhesion to endothelial cells, and increased angiogenesis and invasion, thus promoting the metastatic phenotype (3 , 4 , 9 , 10) . Unlike melanoma, breast, and colorectal cancer, in which loss of AP-2 is a late event, prostate cancer seems to lose AP-2 expression early, promoting prostate cancer development.
One way AP-2 expression may affect prostate cancer development is by effecting changes in microvessel density (MVD) in the stroma and a concomitant increase in expression of vascular endothelial growth factor (VEGF) in the prostatic epithelium. The stroma surrounding PIN has a higher MVD than the stroma surrounding normal epithelium, and the MVD is even higher in the stroma around prostate cancer (11) . Increased MVD is accompanied by increased expression of VEGF in the prostatic epithelium. Normal prostate epithelium expresses low levels of VEGF, whereas premalignant lesions, such as PIN, have increased VEGF expression, which is additionally increased in prostate adenocarcinoma (12) .
VEGF is a ubiquitous cytokine that regulates embryonic vasculogenesis, angiogenesis, and permeability in numerous physiologic and pathologic conditions. VEGF-A is a member of a larger family of structurally related glycoproteins of the platelet-derived growth factor superfamily (13) . The VEGF-A gene can produce six isoforms through differential splicing, consisting of 121, 141, 165, 183, 189, and 206 amino acids (14, 15, 16, 17) . The expression patterns and biological function of the 141, 183, and 206 isoforms are not understood currently. However, VEGF189, VEGF165, and VEGF121 are expressed in the prostatic epithelium and prostate cancer (18) . The human and rat VEGF promoters contain binding sites for AP-2 and for Sp1, Sp1-related transcription factors, AP-1, signal transducers and activators of transcription 3, and hypoxia-inducible factor 1, which regulate transcription of the gene during hypoxia (19 , 20) .
In this article, we explore whether loss of AP-2 expression plays a causal role in prostate cancer development. We found that re-expression of AP-2 in LNCaP-LN3 (AP-2-negative) cells abolished their tumorigenicity in nude mice. To determine which AP-2 target genes were responsible for the observed changes in tumorigenicity, we examined the expression profiles of AP-2-transfected cells versus control cells using microchip expression array analysis. Among the dominant changes in gene expression observed was reduction in VEGF expression, which might be responsible for the observed changes in tumorigenicity. We show that AP-2 mediated transcriptional repression of the VEGF gene through competition with the transcriptional activator Sp3 for binding to a GC-rich proximal region of the VEGF promoter. Loss of AP-2 in prostate cancer cells reduced the AP-2:Sp3 ratio and activated VEGF expression. We conclude that AP-2 acts as a tumor-suppressor gene in prostate cancer through this mechanism.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Stable Transfections.
LNCaP-LN3-stable AP-2 transfectants were established using the Tet-ON system (Clontech Corp., Palo Alto, CA) as outlined by the manufacturer. Tetracycline-responsive element AP-2 was constructed by cloning the AP-2 cDNA between the HindIII and XbaI sites of the TRE2 vector. Stable transfections were performed using Lipofectin reagent (Invitrogen, Carlsbad, CA) according to the manufacturer’s instructions.
Neo C1 was obtained by transfection of LNCaP-LN3 with the Tet-ON construct to provide tetracycline responsiveness. Neo C1 was transfected additionally with the TRE-AP-2 or TRE C2 constructs. Stable AP-2 and Neo transfectants were screened for AP-2 expression in the absence and presence of 1 µg/ml of doxycycline (Dox). The TRE-AP-2 transfectants showed "leaky" AP-2 expression (6–9-fold) with minimal inducibility (1.5-fold) by addition of Dox; therefore, analysis of gene expression was done in the absence of Dox.
RNA Extraction and Northern Blotting.
Total RNA was extracted using TRIzol reagent (Invitrogen) according to the manufacturer’s instructions. Briefly, cells grown to 70–80% confluence were washed with PBS and lysed in TRIzol reagent. Proteins and DNA were extracted in 0.2 volumes of chloroform. RNA was precipitated from the aqueous phase with isopropanol, washed with 75% ethanol-diethyl pyrocarbonate, air dried, and resuspended in diethyl pyrocarbonate-treated water. For the microchip expression array assay, the ethanol-washed pellet was resuspended in 1 ml of TRIzol reagent, and the extraction procedure was repeated. RNA concentration was determined by measuring the absorbance at 260 nm in a UV/visible light spectrophotometer (Ultrospec 3000 pro; Amersham Pharmacia Biotech, Cambridge, United Kingdom). Northern blot analysis for VEGF was performed as described previously (22)
.
Semiquantitative Reverse Transcription-PCR.
One µg of total RNA was reverse transcribed using the Advantage reverse transcription-PCR kit (Clontech Corp.). PCR was performed using the Advantage cDNA PCR kit (Clontech Corp.). For AP-2 quantitation, cDNA was amplified by PCR using specific primers for glyceraldehyde-3-phosphate dehydrogenase (GAPDH; sense, 5'-GAGCCACATCGCTCAGAC-3' and antisense, 5'-CTTCTCATGGTTCACACCC-3') and human AP-2 (sense, 5'-CTGCCAACGTTACCCTGC-3' and antisense, 5'-TAGTTCTGCAGGGCCGTG-3'). AP-2 and GAPDH cDNAs were amplified by PCR in the same reaction mixture as follows: initial denaturation for 2 min at 96°C, 27 cycles consisting of denaturation at 94°C for 30 s, annealing at 58°C for 30 s, extension at 72°C for 30 s, and a final elongation step at 72°C for 5 min. For VEGF quantitation, commercially available GAPDH primers (Clontech Corp.) and specific primers for VEGF were used for PCR as described previously (23)
.
Western Blotting.
Cells grown on a monolayer to 70–80% confluence were collected in a conical tube, and nuclear proteins were extracted as described previously (24)
. Fifteen µg of nuclear extract were subjected to 10% SDS-PAGE in a Bio-Rad Mini Protean III gel apparatus (Bio-Rad, Hercules, CA). The proteins were transferred electrophoretically to a polyvinylidene difluoride membrane (Millipore Corp., Bedford, MA), and the membrane was blocked in blocking solution (5% nonfat dry milk/Tris-buffered solution) and then incubated with primary antibody overnight. For AP-2 detection, a 1:3000 dilution of an anti-AP-2 polyclonal antibody (Santa Cruz Biotechnology Inc., Santa Cruz, CA) was used. The unbound primary antibody was removed by washing the membrane with 0.1% Tween/Tris-buffered solution, followed by incubation with horseradish peroxidase-conjugated secondary antibody diluted 1:2000 in 0.5% nonfat dry milk/Tween/Tris-buffered solution. Proteins were visualized using enhanced chemiluminiscence reagent (Amersham Pharmacia Biotech) and X-ray film.
Densitometric Quantitation.
Images were captured in a Gel Doc 2000 System (Bio-Rad) connected to a charge-coupled device camera. Densitometric readings of DNA fragments separated in agarose gels were quantitated using Quantity One Software Version 4 for Windows (Bio-Rad). Northern and Western blot densitometric analyses were performed in the linear range of the film.
Animals and Orthotopic Implantation of Tumor Cells.
Male athymic BALB/c nude mice were obtained from the Animal Production Area of the National Cancer Institute-Frederick Cancer Research Institute. The mice were kept under a laminar airflow cabinet under specific pathogen-free conditions. The American Association for Accreditation of Laboratory Animal Care approved the animal facilities following the standards of the United States Department of Agriculture, Department of Health and Human Services, and regulations of the institutional animal care and use committee.
For orthotopic implantations, 8–10-week-old mice were anesthetized by i.p. injection of pentobarbital. Low abdominal incisions were performed, and a suspension of 2 x 106 viable cells diluted in 40 µl of HBSS was injected into the prostate as described previously (21) . Mice underwent necropsy on day 90. Tumors were detected by gross histology. Tumors were weighed and immediately processed for H&E and immunohistochemical staining. To test the potential effects of Dox on AP-2 expression in vivo, 10 mice were given 2 µg/ml Dox + 2% sucrose in the drinking water, and 10 mice received only 2% sucrose in the drinking water. Because no statistically significant differences in tumorigenicity were produced by treatment with Dox (P = 0.6084), all 20 of the mice were grouped for statistical analysis of tumor incidence.
ELISA.
For VEGF ELISA, 5 x 104 viable cells were plated in triplicate in six-well plates and grown for 48 h in RPMI 1640 supplemented with 0.5% or 10% fetal bovine serum. After the conditioned medium was collected, the cells in each well were trypsinized, collected, and lysed. VEGF concentration in the supernatants was measured in triplicate using a commercial VEGF ELISA (R&D Systems, Minneapolis, MN) according to the manufacturer’s instructions. VEGF protein levels in the supernatant were normalized to total protein concentration in the cell lysate for each well. VEGF concentration was calculated as the average of the three wells and expressed as picograms of VEGF protein per microgram of total protein.
Luciferase Assay.
The VEGF promoter deletion constructs VEGF-EcoRI, SpeI, PstI, ApaI, and SacII were the same as those described previously (25)
. The VEGF deletion constructs P1176, P88, P66, P52, and P27 were obtained from Dr. Gilles Pages (Universite de Nice, Nice, France) and were described previously (26)
. Transfection efficiency for the luciferase assay was determined using the ß-actin Renilla construct, in which the Renilla luciferase gene is driven by the ß-actin promoter.
Cells were grown in 24-well plates to 60% confluence and transfected transiently using Lipofectin reagent with 0.6 µg of a luciferase expression construct and 20 ng of ß-actin Renilla. Cotransfection was performed by adding 1.2 µg of AP-2 expression construct or empty vector to the DNA solutions for transfection. Luciferase activity was determined 36 h after transfection using the dual-luciferase reporter assay system (Promega Corp., Madison, WI) in a microplate luminometer (Luminoskan Ascent; Labsystems Inc., Franklin, MA) as outlined in the manufacturer’s protocols. Luciferase units were calculated using the following formula: firefly luciferase units/Renilla luciferase units.
Electrophoretic Mobility Shift Assay.
Electrophoretic mobility shift assay for AP-2 was performed as described previously (3)
. VEGF88-66 oligonucleotides were end-labeled with [
-P32]ATP (Promega Corp.) and incubated with 5 µg of nuclear extract and 1x binding buffer [10 mM Tris-HCl (pH 7.8), 100 mM KCl, 5 mM MgCl2, 1 mM EDTA, 10% glycerol, and 1 mM DTT] for 30 min on ice. For competition assays, the nuclear extract was incubated for 30 min with the unlabeled oligonucleotide in binding buffer and then incubated with VEGF88-66 oligonucleotides for an additional 30 min on ice. For supershift assay, the nuclear extracts were incubated with the radiolabeled oligonucleotides in binding buffer for 30 min, followed by incubation with anti-AP-2, -Sp1, or -Sp3 polyclonal antibodies (Santa Cruz Biotechnology) for 1 h on ice. The DNA-protein complexes were resolved in a 4% polyacrylamide gel for 5 h in 0.5 x Tris-borate EDTA buffer at 4°C.
Chromatin Immunoprecipitation.
Chromatin immunoprecipitation was performed as outlined by the ChIP assay kit (Upstate Biotechnology, Lake Placid, NY). DNA binding proteins were cross-linked to DNA and lysed in SDS lysis buffer containing 1x protease inhibitors. DNA was sheared to 200–500-bp fragments by eight 10-s sonications, each using a sonic dismembranator model 60 (Fisher Scientific, Pittsburgh, PA).
The chromatin solution was precleared with salmon sperm DNA/protein A agarose-50% slurry (Upstate Biotechnology) for 30 min at 4°C. The precleared supernatant was incubated with anti-c-fos, -AP-2, -Sp1, or -Sp3 polyclonal antibodies (Santa Cruz Biotechnology) overnight at 4°C. The region between -105 and -5 of the VEGF promoter was amplified from the immunoprecipitated chromatin using the following primers: sense, 5'-GGCTGAGGCTCGCCTGTC-3' and antisense, 5'-CCCGCTACCAGCCGACTTTT-3'. The 100-bp PCR product was separated on a 3% NuSieve low-melting-point agarose gel (FMC Corporation, Philadelphia, PA), stained with ethidium bromide, and visualized under UV light. Densitometric reading of each sample was expressed as a ratio of sample:input.
Statistical Analysis.
Statistical analysis and graphical data were prepared using GraphPad Prism version 3 (GraphPad Software, Inc., San Diego, CA). Student’s t test was used to analyze statistical significance at the 95% confidence interval.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and protein (Fig. 1B)
expression levels in neo control and AP-2 transfectants. Both AP-2-transfected clones, AP-2 C1 and AP-2 C2 (Fig. 1, A and B
, Lanes 5 and 6), demonstrated 6–9-fold the AP-2 expression levels of mRNA and protein as compared with parental LNCaP-LN3 and the neo controls Neo C1 and Neo C2 (Fig. 1, A and B
, Lanes 2–4). A nonspecific band was equally prominent in all of the nuclear extracts (Fig. 1B)
and was used to assume equal loading. No morphologic changes after transfection with the AP-2 gene were observed. In addition, no changes were observed in the proliferative rate in vitro as determined by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay (data not shown).
|
. The AP-2-specific DNA-protein complex was identified by competition with 100x unlabeled AP-2 oligonucleotide (Fig. 1C
, Lane 8) but not with 100x NF-1 oligonucleotide (nonspecific competitor; Fig. 1C
, Lane 9), and by supershifting with an anti-AP-2 polyclonal antibody (Fig. 1C
, Lanes 2 and 10). Nuclear extracts from AP-2 C1 and AP-2 C2 demonstrated higher AP-2 binding activity (Fig. 1C
, Lanes 6 and 7) than nuclear extracts from LNCaP-LN3, Neo C1, or Neo C2 (Fig. 1C
, Lanes 3–5).
Clones AP-2 C1 and Neo C1 were injected into the prostates of nude mice, and blood was collected for prostate-specific antigen reading 45 days after injection and at the time mice were killed (90 days). Transfection with AP-2 abolished completely tumor development in AP-2 C1 (0 of 19), a significant difference from control Neo C1 (10 of 16; P < 0.001; Table 1
). The Neo C1 group had detectable prostate-specific antigen at day 21. At the time of death, prostate-specific antigen levels in the control group correlated directly with tumor volume but were undetectable in the AP-2 C1 group (data not shown).
|
(cutoff value = 7.0).
|
14.7-fold in the stable AP-2 transfectants (boldface in Table 2
). In addition, several cytoplasmic kinases associated with cell cycle regulation were down-regulated in the AP-2 C1 (protein kinase A, p55PIK, p21-activated kinase, Rac protein kinase ß, GTP-binding protein Rap2, RAB5, and RAB6). Interestingly, several of the differentially expressed genes have been reported previously to contain AP-2 consensus sequences in their regulatory region. Among them are the VEGF gene, tissue inhibitor of metalloproteinases-2 gene, and the c-erbB2 tyrosine kinase receptor gene (Table 2)
. These observations suggest that any one of several mechanisms may account for the differential expression profile of AP-2-transfected versus neo-transfected cells. AP-2 may alter expression directly by binding to the promoter region of its target genes and indirectly by influencing expression of signal transduction molecules that modify expression of downstream target genes.
Down-Regulation of VEGF Expression in AP-2-Transfected Cells.
Because VEGF was down-regulated by 14.7-fold in the AP-2-transfected cells and because it plays a cardinal role in prostate cancer development and progression, we decided to additionally investigate VEGF regulation by AP-2 in prostate cancer. To confirm the down-regulation of VEGF in the AP-2 C1 transfectants detected by microchip expression array, we analyzed VEGF expression in total RNA isolated from LNCaP-LN3 parental, Neo C1, and AP-2 C1. Our analysis confirmed expression of VEGF mRNA in LNCaP-LN3 parental or Neo C1 transfectants and a reduction in AP-2 C1 (Fig. 2A)
. Densitometric analysis of VEGF expression normalized to GAPDH expression revealed that the AP-2 C1 expressed approximately one-tenth of the VEGF mRNA as control Neo C1, which is compatible with the 14.7-fold difference detected by microchip expression array (Fig. 2A)
.
|
, Lanes 1 and 3 versus Lanes 2, 4, and 5). A summary of the densitometric analyses for the VEGFs normalized to GAPDH is shown in Fig. 2C
. We also analyzed VEGF secretion by ELISA and found that both AP-2-transfected clones secreted significantly lower levels of VEGF protein than parental LNCaP-LN3 or Neo C1 transfectants grown in the presence of low or high serum (Fig. 2D)
. Taken together, our results indicate that transfection of AP-2-negative prostate cancer cells with the AP-2 gene reduced VEGF expression and that this reduction was AP-2 dependent rather than a result of clonal selection.
A Proximal GC-Rich Region Is Responsible for AP-2-Mediated Repression of the VEGF Promoter.
The regulatory region of the VEGF gene contains at least three AP-2 consensus elements (illustrated in Fig. 3A
by shadow boxes), suggesting the possibility that AP-2 can exert its regulatory function through direct binding to the VEGF promoter (19)
. To identify the AP-2 binding sites that might mediate transcriptional repression, we used a series of deletion fragments of the VEGF-A promoter cloned in front of the luciferase reporter gene (Fig. 3A)
. Full-length promoter activity was reduced in the AP-2-stably transfected clone AP-2 C1 and in the Neo C1 clone transfected transiently with an AP-2 expression construct. Similar luciferase values were observed after deletion of the regions -2362 to -135. However, deletion of the region -135 to +585 relative to the transcription initiation site abolished completely basal and AP-2-mediated repression of VEGF promoter activity (Fig. 3A)
. These results suggest that the two AP-2 consensus elements located upstream of -135 do not regulate VEGF promoter activity. Therefore, the AP-2-responsive element most likely resides within the proximal region of the VEGF promoter.
|
. Deletion up to position -88 did not prevent AP-2-mediated repression but did reduce basal promoter activity when compared with full-length promoter activity. Additional deletions from -66 to -27 abolished basal promoter activity given that the luciferase readings were similar to those obtained with empty vector (Fig. 3B)
. A GC-rich region between -88 and -66 containing an AP-2 consensus sequence overlapping two Sp1 consensus sequences has been described previously (27)
. Our results define this region as mediating minimal basal promoter activity and AP-2 responsiveness in prostate cancer cells. Nonetheless, GC boxes between -135 and -88 were essential for full basal promoter activity.
A Functional Interplay between AP-2 and Sp3 Mediates Transcriptional Activation or Repression.
To identify the transcription factors that bind to the -88 and -66 region of the promoter, we reacted a DNA oligonucleotide containing the sequences -88 to -66 of the VEGF promoter (Fig. 4A)
with nuclear extracts from Neo C1 and AP-2 C1 transfectants. Three shifted complexes were observed on the electrophoretic mobility shift assay gel. Addition of an anti-AP-2 antibody did not produce a supershifted complex but did reduce the intensity of the shifted complexes in extracts from the AP-2 C1 (Fig. 4B
, Lanes 4 versus 3), whereas an antibody against Sp1 supershifted partially complex 2 (Fig. 4B
, Lanes 5 and 6). Surprisingly, an anti-Sp3 antibody supershifted completely all of the three complexes, and this supershifted complex had a lower intensity in the AP-2 C1 extract than in the Neo C1 extract (Fig. 4B
, Lanes 8 versus 7). Together, these data demonstrate that in prostate cancer cells, Sp3 bound with the highest affinity to the -88 to -66 region, although AP-2 and Sp1 also bound to this region in vitro.
|
using a chromatin immunoprecipitation assay. In the Neo C1 cells, Sp3 was the predominant Sp1 family member bound to the proximal region of the VEGF promoter (Fig. 4D
, Lane 7). In the AP-2-stable transfectant, AP-2 was bound strongly to this region (Fig. 4D
, Lane 5), whereas very low levels of bound Sp3 were detected (Fig. 4D
, Lane 7). This finding suggests that AP-2 overexpression inhibits VEGF promoter activity by competing for the binding site with Sp3, thereby blocking it from inducing VEGF promoter activity. Interestingly, similar levels of Sp1 bound to this region were observed in the AP-2-transfected and control cells (Fig. 4D
, Lane 6). Although our gel shift data showed that Sp1 did bind to the -88 to -66 region in vitro, it is possible that it binds in vivo upstream GC boxes that confer AP-2-independent full basal promoter activity.
Previous studies have demonstrated that an increase in the AP-2:Sp1 ratio modulates transcriptional repression of several genes. Therefore, we next determined whether alterations in the ratio of AP-2 to Sp3 would promote transcriptional repression or activation. To that end, we analyzed luciferase activity driven by the proximal GC-rich region of the VEGF promoter in the presence of different ratios of AP-2 and Sp3 expression constructs. Because the cytomegalovirus promoter drives both genes in these constructs, the amounts of proteins expressed after transfection should be correlated with the amounts of the transfected DNA. In the Neo C1, a high AP-2:Sp3 ratio (3:0 or 2:1) promoted transcriptional repression, whereas a low ratio (1:2 or 0:3) promoted activation (Fig. 5A)
. Similarly, reducing the AP-2:Sp3 ratio in AP-2-overexpressing cells by transfection with increasing amounts of an Sp3 expression construct increased VEGF promoter activity (Fig. 5B)
. We conclude that forced AP-2 expression in prostate cancer cells increased the AP-2:Sp3 ratio and thereby repressed VEGF transcription.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The results presented here might provide a mechanism to explain the role of VEGF in early prostate cancer development and the observed loss of tumorigenicity in the AP-2-transfected cells. Mazzucchelli et al. (12) analyzed recently VEGF expression and its correlation with the capillary architecture of PIN and prostate cancer. They observed negative to low VEGF immunoreactivity in normal prostate from the transition and nontransition zones, a higher intensity of VEGF expression in high-grade PIN, and even higher expression in moderately differentiated prostate cancer (12) . Our previous studies of AP-2 expression in radical prostatectomy specimens revealed an inverse pattern of AP-2 expression to those reported for VEGF expression, suggesting that loss of AP-2 might be a contributing factor to the increased VEGF expression in early prostate cancer development (2) . Moreover, LNCaP-LN3 cells express high levels of VEGF, which is thought to be the predominant inducer of angiogenesis in vivo: treatment of established orthotopic LNCaP-LN3 tumors with an anti-VEGF receptor-2 (flk-1) antibody (DC101) inhibits tumor incidence and MVD (28) . Besides its role in angiogenesis (paracrine effects), recent evidence suggests that VEGF also can exert mitogenic effects on the tumor cells themselves (autocrine effects). Treatment of the LNCaP prostate cancer cell line with recombinant VEGF stimulates proliferation in vitro, which can be neutralized by antisera against flk-1 (29) . Therefore, loss of the angiogenic properties and inhibition of autocrine growth-stimulatory pathways may inhibit the ability of AP-2-transfected cells to form detectable tumors.
In this report, we also demonstrate that AP-2 mediates transcriptional repression of the VEGF gene through competition with the transcriptional activator Sp3 for binding to the -88 to -66 region. This region contains the classical AP-2 consensus element 5'-GCCNNNGGC-3' (30)
, which has been shown to mediate induction of VEGF expression induced by serum withdrawal, UVA exposure (31)
, transforming growth factor (TGF) - (27)
, hepatocyte growth factor (32
, 33) , and epidermal growth factor (34)
stimulation. Although these studies show that Sp1 binds to this region to stimulate VEGF expression, mutations in the AP-2 consensus element from GCCGGGGGC to GCCTAGGGC reduced TGF- and UVA-mediated induction by 50%. We analyzed VEGF promoter activity in Neo C1 and AP-2 C1 transfectants treated with TGF-, epidermal growth factor, or hepatocyte growth factor and did not detect induction of VEGF promoter activity in control or AP-2-transfected cells.4
Similarly, in colon carcinoma cells, whereas interleukin 1ß and insulin-like growth factor I induce VEGF promoter activity, epidermal growth factor, hepatocyte growth factor, platelet-derived growth factor-BB, platelet-derived endothelial cell growth factor, interleukin 6, TGF-, and TGF-ß do not (35)
. Together, these observations suggest that although the -88 to -66 region of the VEGF promoter exerts essential basal and inducible regulatory functions, these activities are governed by cell-type-specific mechanisms.
Different transcriptional activators of the Sp1 family binding to this region may provide the cell-type specificity. In cell types in which AP-2 has been shown to have an activator function, Sp1 acts on the proximal GC-rich region to regulate basal VEGF promoter activity. In prostate cancer cells, we determined that Sp3 has higher affinity than Sp1 to the region -88 to -66 in vitro, and it is bound to the active chromatin in cells that express high VEGF levels. Our analysis demonstrated that Sp3 binding to this region of the promoter was reduced greatly in the presence of high levels of AP-2, suggesting that AP-2 mediates repression by competing for the binding site with Sp3, which is a transcriptional activator of this region in prostate cancer cells. Although a functional interplay between AP-2 and Sp1 has been described previously for several genes, such as MMP-2, MnSO dismutase, CYP11A, and keratins 3 and 14, to our knowledge, this is the first study to describe a functional interplay between AP-2 and Sp3.
Sp3 was described originally as a transcriptional repressor (36) . However, recent evidence demonstrates that it is a bifunctional transcription factor with modular independent activation and repression domains (37) . The activator and repressor functions of Sp3 depend largely on cell context: it acts as a repressor when bound through multiple DNA-binding sites and as an activator when targeted to the promoter via a single DNA-binding site (36) . Activation and repression activities are mediated through complex interactions with other transcriptional activators and members of the basal transcriptional machinery.
Our findings suggest that in prostate cancer cells, high levels of VEGF transcription are induced by binding of Sp1 to GC boxes located between -135 and -88 and by binding of Sp3 to a single site located between -88 and -66. We speculate that protein-protein interactions between Sp1 and Sp3 with the basal transcription machinery may provide a stable transcriptional complex driving high levels of transcription of the VEGF gene. On the basis of our in vitro and in vivo VEGF promoter analysis, we propose the following model. The ubiquitous transcription factor Sp1 confers basal VEGF promoter activity by binding to GC boxes upstream to -88, and in the absence of AP-2, Sp3 binds to the -88 to -66 region, resulting in high levels of VEGF expression. When AP-2 is present, it occupies the binding site between -88 and -66, thus preventing binding of Sp3 and repressing transcription of the VEGF gene (see model illustrated in Fig. 6
).
|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
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.
Requests for reprints: Menashe Bar-Eli, Department of Cancer Biology, Unit-173, M.D. Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77030. E-mail: mbareli{at}mdanderson.org
Received 9/ 2/03. Revised 10/21/03. Accepted 11/ 3/03.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
is preferentially cleaved by caspase 6 and degraded by proteasome during tumor necrosis factor á-induced apoptosis in breast cancer cells. Mol. Cell. Biol., 21: 4856-4867, 2001.This article has been cited by other articles:
|
|
 |
|
  Y. Xu, F. Fang, S. K. Dhar, A. Bosch, W. H. St. Clair, E. J. Kasarskis, and D. K. St. Clair Mutations in the SOD2 Promoter Reveal a Molecular Basis for an Activating Protein 2-Dependent Dysregulation of Manganese Superoxide Dismutase Expression in Cancer Cells Mol. Cancer Res., December 1, 2008; 6(12): 1881 - 1893. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Orso, E. Penna, D. Cimino, E. Astanina, F. Maione, D. Valdembri, E. Giraudo, G. Serini, P. Sismondi, M. De Bortoli, et al. AP-2{alpha} and AP-2{gamma} regulate tumor progression via specific genetic programs FASEB J, August 1, 2008; 22(8): 2702 - 2714. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.-J. Choi, T.-W. Chung, S.-J. Kim, S.-Y. Cho, Y.-S. Lee, Y.-C. Lee, J.-H. Ko, and C.-H. Kim The AP-2{alpha} transcription factor is required for the ganglioside GM3-stimulated transcriptional regulation of a PTEN gene Glycobiology, May 1, 2008; 18(5): 395 - 407. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W.-G. Deng, G. Jayachandran, G. Wu, K. Xu, J. A. Roth, and L. Ji Tumor-specific Activation of Human Telomerase Reverses Transcriptase Promoter Activity by Activating Enhancer-binding Protein-2beta in Human Lung Cancer Cells J. Biol. Chem., September 7, 2007; 282(36): 26460 - 26470. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. R. Qin, D. Iliopoulos, T. Nakamura, S. Costinean, S. Volinia, T. Druck, J. Sun, H. Okumura, and K. Huebner Wwox Suppresses Prostate Cancer Cell Growth through Modulation of ErbB2-Mediated Androgen Receptor Signaling Mol. Cancer Res., September 1, 2007; 5(9): 957 - 965. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A.F. Hannay, J. Liu, Q.-S. Zhu, S. V. Bolshakov, L. Li, P. W.T. Pisters, A. J.F. Lazar, D. Yu, R. E. Pollock, and D. Lev Rad51 overexpression contributes to chemoresistance in human soft tissue sarcoma cells: a role for p53/activator protein 2 transcriptional regulation Mol. Cancer Ther., May 1, 2007; 6(5): 1650 - 1660. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Lopez-Soto, A. Quinones-Lombrana, R. Lopez-Arbesu, C. Lopez-Larrea, and S. Gonzalez Transcriptional Regulation of ULBP1, a Human Ligand of the NKG2D Receptor J. Biol. Chem., October 13, 2006; 281(41): 30419 - 30430. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Wajapeyee, R. Britto, H. M. Ravishankar, and K. Somasundaram Apoptosis Induction by Activator Protein 2{alpha} Involves Transcriptional Repression of Bcl-2 J. Biol. Chem., June 16, 2006; 281(24): 16207 - 16219. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Wajapeyee, C. G. Raut, and K. Somasundaram Activator Protein 2{alpha} Status Determines the Chemosensitivity of Cancer Cells: Implications in Cancer Chemotherapy Cancer Res., October 1, 2005; 65(19): 8628 - 8634. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Egener, E. Roulet, M. Zehnder, P. Bucher, and N. Mermod Proof of concept for microarray-based detection of DNA-binding oncogenes in cell extracts Nucleic Acids Res., May 12, 2005; 33(8): e79 - e79. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Pages and J. Pouyssegur Transcriptional regulation of the Vascular Endothelial Growth Factor gene-a concert of activating factors Cardiovasc Res, February 15, 2005; 65(3): 564 - 573. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. Salah, M. Maoz, I. Cohen, G. Pizov, D. Pode, M. S. Runge, and R. Bar-Shavit Identification of a novel functional androgen response element within hPar1 promoter: implications to prostate cancer progression FASEB J, January 1, 2005; 19(1): 62 - 72. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. B. Heimberger, E. C. McGary, D. Suki, M. Ruiz, H. Wang, G. N. Fuller, and M. Bar-Eli Loss of the AP-2{alpha} Transcription Factor Is Associated with the Grade of Human Gliomas Clin. Cancer Res., January 1, 2005; 11(1): 267 - 272. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. F. McCarty Targeting Multiple Signaling Pathways as a Strategy for Managing Prostate Cancer: Multifocal Signal Modulation Therapy Integr Cancer Ther, December 1, 2004; 3(4): 349 - 380. [Abstract] [PDF]
|
|
|
|
|
|
J. M. Pellikainen, K. M. Ropponen, V. V. Kataja, J. K. Kellokoski, M. J. Eskelinen, and V.-M. Kosma Expression of Matrix Metalloproteinase (MMP)-2 and MMP-9 in Breast Cancer with a Special Reference to Activator Protein-2, HER2, and Prognosis Clin. Cancer Res., November 15, 2004; 10(22): 7621 - 7628. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Q. Li and R. H. Dashwood Activator Protein 2{alpha} Associates with Adenomatous Polyposis Coli/{beta}-Catenin and Inhibits {beta}-Catenin/T-cell Factor Transcriptional Activity in Colorectal Cancer Cells J. Biol. Chem., October 29, 2004; 279(44): 45669 - 45675. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
, and Sp1 elements are shown in 
. Neo C1 and AP-2 C1 cells were transfected with luciferase constructs and cotransfected with empty vector or an AP-2 expression construct. Mean luciferase activity in Neo C1, AP-2 C1, and Neo C1 cells cotransfected with AP-2 and measured 36 h after transfection is shown; bars ± SD. Data shown are representative of those from three independent experiments. B, schematic of deletion mutants of the VEGF promoter proximal region. Mean luciferase activity of Neo C1, AP-2 C1, and Neo C1 cells cotransfected with AP-2 was determined as described in (A); bars ± SD.
