| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Biological Sciences, University of Toledo, Toledo, Ohio 43606; Unité Mixte de Recherche 8161, Institut de Biologie de Lille, Centre National de la Recherche Scientifique, Universities of Lille 1 and Lille 2, Pasteur Institute of Lille, Institut Fédératif de Recherche 142, 59021 Lille Cedex, France
Address all correspondence and requests for reprints to: Lirim Shemshedini, Department of Biological Sciences, University of Toledo, Toledo, Ohio 43606. E-mail: lshemsh{at}utnet.utoledo.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
The Ets proteins are transcription factors involved in multiple processes, including cell proliferation and cancer cell invasion (reviewed in Ref. 6). All Ets proteins contain the ETS domain, which mediates binding to a central 5'-GGAA/T-3' motif (reviewed in Ref. 7). Among the multiple Ets proteins (reviewed in Ref. 8), the PEA3 group consists of ETV1 (Ets variant gene 1; also called ER81), ETV4 (also called PEA3) and ETV5 (also called ERM). All three members are 95% identical in the ETS domain and more than 85% in the acidic transactivation domain (reviewed in Ref. 7). Several studies suggest that the PEA3 group proteins are involved in intestinal tumors (9), gastric cancer (10), and breast cancer metastasis (11, 12). In nearly all Ewing’s sarcoma tumors, EWS, which encodes a RNA-binding protein, is fused by chromosomal translocation to an Ets gene, including FLI, ERG, ETV4, and ETV1. This results in the expression of chimeric proteins that may be important in tumor cell transformation (reviewed in Ref. 13). Recently, it was reported that TMPRSS2, an AR-regulated gene, is fused by translocation to the ETV1 (14), ERG (14), or ETV4 gene (15) in a subset of prostate cancers. These findings suggest an important role for PEA3 proteins in prostate cancer.
We report in this study a novel pathway by which AR and ETV1 signaling interact in prostate cancer cells to regulate cell invasion. Our results demonstrate that the ETV1 promoter is a direct AR target. Diminution of ETV1 expression disrupts the invasion ability of both androgen-dependent and androgen-independent prostate cancer cells. Lastly and importantly, ETV1 is overexpressed in prostate cancer tissues independent of TMPRSS2 translocation, suggesting that the intact ETV1 may also play an important role in prostate cancer.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
1 (sGC1) mediating prostate cancer cell proliferation (20) and multidrug resistance protein 4 (MRP4) (21). This microarray analysis also led us to identify ETV1, which was expressed in an androgen-induced manner in C14 cells (LNCaP cells transfected with empty vector), but not in A103 cells (Table 1
). This androgen-induced ETV1 expression in C14 cells is similar to what was obtained in untransfected LNCaP cells (see Fig. 4A). Importantly, the expressions of other novel (20, 21) and known androgen-regulated genes are similar in C14 and untransfected LNCaP cells (data not shown), demonstrating that the gene expression pattern is not altered in C14 cells as compared with unstransfected LNCaP cells. In addition, C14 and untransfected LNCaP cells are similar in androgen-induced proliferation (data not shown). Of the other ETs genes found on the micrarray chip, ERG, Ets-1, and MEF are not expressed in C14 cells and FLI1, Ets-2, and ERF exhibited no DHT regulation (Table 1).
|
|
, ETV1 mRNA expression was markedly induced by DHT only in LNCaP cells, and this induction was similar to that of prostate-specific antigen (PSA), a hallmark androgen-regulated gene (22). There was only weak expression in PC-3 cells unaffected by DHT and no detectable expression in PrEC cells. Interestingly, LNCaP cells exhibited no detectable expression of ETV4 and ETV5 (Fig. 1A). There was significant expression of both genes in PC-3 cells and weaker expression in PrEC cells. These results demonstrate androgen-induced ETV1 mRNA expression in LNCaP cells that is markedly higher than in PC-3 cells, confirming the gene microarray data.
|
). A dose-dependent experiment showed that 100 nM DHT was optimal for ETV1 induction, and this concentration of DHT was nearly as potent as 1 nM R1881 (Fig. 1C).
Protein expression was measured by Western blotting using an anti-ETV1 antibody, revealing a maximal DHT induction of ETV1 protein in LNCaP cells after 2 d of hormone treatment (Fig. 1D). A 2-d incubation also resulted in increased ETV1 protein expression in the absence of DHT (Fig. 1D), suggesting a time-dependent increase in ETV1 expression that does not depend on DHT. The positive DHT effect was also observed in immunocytochemistry, which showed nuclear ETV1 expression (data not shown). There was markedly weaker ETV1 protein expression in PC-3 cells (Fig. 1D). These results demonstrate that DHT enhances ETV1 protein expression in LNCaP cells.
AR Acts on ETV1 Promoter
To determine whether ETV1 is an AR target gene, the ETV1 promoter was amplified from LNCaP genomic DNA by PCR. This generated a PCR fragment starting from the ETV1 exon 1 and extending to 1 kb upstream of this exon. There is a consensus TATA box sequence (5-TATAAA-3') just upstream of exon 1 and a near consensus androgen-responsive element (ARE) (5'-AAGCCATCTTGTTC-3') at approximately 0.15 kb upstream of the exon. The PCR product was inserted into the pGL3 reporter vector just upstream of the Luciferase gene to generate the reporter plasmid ETV1-Luc. Promoter activity from the 1-kb fragment was significantly induced by DHT (3-fold), whereas activity from the promoter-less parental plasmid, pGL3, was substantially weaker and not affected by DHT (Fig. 2A). Hence, the 1-kb genomic fragment has DHT-inducible promoter activity, suggesting that ETV1 may be a direct target of AR transactivation.
|
, both concentrations of Casodex significantly repressed DHT-induced expression of ETV1 mRNA, suggesting that the androgen effect on ETV1 expression is mediated by AR.
Chromatin immunoprecipitation (ChIP) was performed to determine whether AR binds to the region around the putative ARE found in the 1-kb ETV1 genomic fragment. As shown in Fig. 2C, AR is recruited to the ETV1 1-kb promoter in a DHT-dependent manner, with a stronger DHT effect after 24 h than 8 h. This finding is comparable to AR recruitment to the PSA promoter (Fig. 2C). Importantly, another part of the ETV1 promoter (2-kb) that contain consensus ARE sequences failed to recruit AR (Fig. 2C), demonstrating the specificity of the ARE within the 1-kb fragment in AR recruitment. These results demonstrate that ligand-activated AR can bind to the ETV1 promoter in LNCaP cells.
ETV1 transcriptional activity was measured using Fes3xWT-Luc, a Luciferase reporter plasmid containing a promoter driven by three copies of an Ets-responsive element from the fes promoter (24). This promoter was strongly induced (nearly 12-fold) by transfected ETV1 in LNCaP cells (Fig. 2D). When it was tested for DHT regulation, the Fes promoter exhibited a smaller (2.5-fold) but significant DHT-induced increase in activity (Fig. 2D). These data suggest that DHT-activated AR can induce the activity of an ETV1-regulated reporter plasmid, supporting the gene expression data above showing that AR can induce the expression of ETV1.
ETV1 Induces Matrix Metalloproteinase (MMP) Gene Expression in Prostate Cancer Cells
To determine whether androgen induction of ETV1 expression affects ETV1 activity, ETV1-regulated gene expression was monitored by semiquantitative RT-PCR. ETV1 and related proteins regulate the expression of MMP genes (reviewed in Ref. 8), which encode proteins that mediate degradation of the extracellular matrix and basement membrane (reviewed in Ref. 25). MMP genes were measured for androgen regulation in C14 (LNCaP) and A103 (PC-3) cells. This analysis showed that only MMP-7, MMP-10, and MMP-13 were expressed to detectable levels in LNCaP cells. DHT had a positive effect on MMP-7 and MMP-13 and negative effect on MMP-10 (Fig. 3A). MMP-1, MMP-3, MMP-9, MMP-10, and MMP-13 mRNAs exhibited substantially higher expression in A103 than C14 cells (Fig. 3A). Interestingly, MMP-10 expression in both LNCaP and A103 cells is down-regulated by DHT (Fig. 3A).
|
) and protein expression (see Fig. 5A) in LNCaP cells. This ETV1 reduction resulted in substantially decreased expression of MMP-7 and MMP-13 (Fig. 3B). These results show that endogenous ETV1 is required for expression of MMP-7 and MMP-13 and suggest that their increased expression in response to androgen treatment is likely due to androgen-induced expression of ETV1.
|
). To determine whether exogenous overexpression of ETV1 can induce MMP gene expression in LNCaP cells, ETV1 was exogenously expressed in these cells. As shown in Fig. 3C, overexpression of ETV1 led to enhanced levels of MMP-1 and MMP-9 mRNA, but not MMP-3. These results suggest that MMP-1 and MMP-9 expression in LNCaP cells requires high levels of ETV1 protein, which is provided by transient transfection (Fig. 3C) but not by DHT treatment (Fig. 3A).
In view of the stronger ability of transiently expressed ETV1 to induced MMP gene expressed as compared with the androgen-induced ETV1 (compare Fig. 1C with Fig. 1A), we compared ETV1 mRNA expression resulting from these two treatments. As shown in Fig. 3D, transient transfection of an ETV1 expression plasmid yielded 10-fold more ETV1 mRNA than treatment with either DHT or R1881. This enhanced expression may be responsible for increased ability of transfected ETV1 to activate MMP gene expression as compared with androgen-induced ETV1.
ETV1 Expression Is Androgen-Unresponsive in Androgen-Independent Prostate Cancer Cells
LNCaP cells are not only a model cell line for studying androgen-dependent prostate cancer, but they also can be cultured to become androgen-unresponsive and therefore mimic the androgen-independent stage of prostate cancer. The C81 cells provide a good example of this, exhibiting in culture androgen-independent cell proliferation and more aggressive tumorigenesis in nude mice studies than the androgen-dependent parental cells (26, 27). C81 cells express AR protein levels that are similar to androgen-dependent LNCaP cells (26, 27). As shown in Fig. 4A, in the presence of DHT, ETV1 mRNA expression is slightly elevated in C81 cells when compared with LNCaP cells and C33 cells, which are the androgen-dependent parental cells. Interestingly, however, in the absence of DHT, C81 cells express significantly more ETV1 than do LNCaP or C33 cells (Fig. 4A), showing that ETV1 mRNA is significantly expressed in hormone-refractory cells without DHT treatment. Mimicking these mRNA results, ETV1 protein expression is not significantly affected by DHT in C81 cells (Fig. 4B), in contrast to the DHT induction that was observed in androgen-dependent LNCaP cells (see Fig. 1, A–D). Interestingly, androgen-unresponsive expression of ETV1 was also observed in CWR22-Rv1 cells (Fig. 4C), a prostate cancer cell line that is derived from the CWR22 xenograft and cultured for hormone-refractory cell growth (28). In contrast, the expression of TMPRSS2 (29) is androgen induced in C81 cells, whereas in CWR22-Rv1 cells the expression is weak and unresponsive to DHT (Fig. 4C). These results demonstrate that androgen-independent LNCaP cells can express high levels of ETV1 via a mechanism that is independent of DHT.
ETV1 activity in C81 cells was monitored by measuring the expression of MMP-7 and MMP-13, which were earlier shown to be DHT induced in androgen-dependent LNCaP cells (see Fig. 3A). Interestingly, MMP-7 mRNA expression is significantly higher in C81 cells than in androgen-dependent LNCaP cells, and it is independent of DHT (Fig. 4D). In contrast, MMP-13 was similarly expressed in the two LNCaP cell lines (Fig. 4D), whereas the expression of MMP-1 and MMP-9 was not detectable. These data show a gene-specific effect of DHT on MMP genes in androgen-independent LNCaP cells.
ETV1 Mediates Invasion of Prostate Cancer Cells
Because MMP proteins catalyze the degradation of the extracellular matrix and basement membrane and MMP genes are transcriptionally regulated by ETV1 and other Ets proteins, we were interested to study a potential function of ETV1 in metastasis of prostate cancer cells. Therefore, ETV1 endogenous levels were diminished in LNCaP and C81 cells by siRNA transfection (Fig. 5A) and in vitro cell invasion was measured. As shown in Fig. 5B, reducing ETV1 protein levels markedly attenuates the invasion of androgen-dependent LNCaP cells, both in the absence and presence of DHT. Similar observations were made in androgen-independent C81 cells (Fig. 5C), implicating a role for ETV1 in hormone-refractory prostate cancer. Together, these data demonstrate a direct relationship between ETV1 protein expression and activity and cell invasion, suggesting an important role for ETV1 in prostate cancer cell metastasis.
The Ets proteins are involved in not only cell invasion, but also cell proliferation (reviewed in Ref. 30). Therefore, LNCaP and C81 cells transfected with ETV1 siRNA were studied for effects on cellular proliferation. Cell proliferation was measured using the MTT assay, which has been used previously (20, 31, 32) and has been shown to be as accurate as direct cell counting (see supplemental Fig. S1, which is published as supplemental data on The Endocrine Society’s Journals Online web site at http://mend.endojournals.org). Reduction of endogenous ETV1 protein levels inhibited the androgen-dependent proliferation of LNCaP cells (Fig. 5D) and androgen-independent proliferation of C81 cells (Fig. 5D). However, the magnitude of this effect on proliferation was substantially smaller than on cell invasion, suggesting that AR regulation of ETV1 is more important for prostate cancer cell invasion than proliferation.
ETV1 Expression Is Highly Elevated in Prostate Cancer Tissues, and This Does Not Depend on Translocation of TMPRSS2
Semiquantitative RT-PCR and Western blot analyses were used to measure the expression of ETV1 in prostate tissues purchased from the Cooperative Human Tissue Network (CHTN). CHTN identifies these tissues as normal (N), benign prostatic hyperplasia (BPH), or malignant prostate cancer (MPC). ETV1 mRNA is highly expressed in the two MPC samples and one BPH (Fig. 6A), which express significant levels of AR mRNA; no ETV1 was detected in the one normal tissue (Fig. 6A). The absence of measurable ETV1 expression in normal tissue is in agreement with its absence in PrEC cells (see Fig. 1A). Although ETV5 expression is similar to ETV1, ETV4 expression is high in normal tissue and progressively decreases in the BPH and MPC tissues (Fig. 6A). Interestingly, in contrast to ETV1 expression, MMP-1 expression is high in normal and decreases markedly in BPH until there is no detectable MMP-1 mRNA in the MPC tissues (Fig. 6A). To better characterize these tissues, the expression of several other genes was measured. PSA expression is high in both MPC and several BPH tissues (Fig. 6A). Notably, the expression pattern of ETV1 parallels that of EZH2 (20), a gene marker for prostate cancer (33). E-cadherin expression is significantly reduced in the BPH and MPC tissues as compared with normal (20), consistent with the literature (34). These data confirm that our tissue samples represent different stages of prostate cancer and demonstrate that ETV1 mRNA is highly expressed in MPC. As expected (16), all tissues examined express AR (Fig. 6A).
|
. Three primer pairs were used spanning ETV1 exons 2 and 3, 3 and 4, or 4 and 5. Because the TMPRSS2 translocation occurs at ETV1 exon 4, the PCR product measuring exons 4 and 5 is greatly increased, in comparison to the other two exon pair PCR products, in tissues that have the translocation (14). As shown in Fig. 6B, only three of 15 MPC tissues express disproportionately high levels of ETV1 exons 4 and 5 and therefore probably have the translocation. This 20% incidence of translocation corresponds closely to the previous study reporting this phenomenon (14). Importantly, although the other 11 MPC samples do not apparently have the TMPRSS2 translocation reported by Tomlins et al. (14), these tissues nevertheless express higher levels of ETV1 mRNA than the BPH and normal tissues (Fig. 6B, inset). Analysis of variance indicates a statistical difference in mean expression between N and either BPH (P = 0.01) or MPC tissues (P = 0.03). To measure the relative proportion of epithelial vs. stromal cells in our tissues, RT-PCR was used to measure the expression of cytokeratin 18 (K18), a maker of epithelial cells (35), and keratinocyte growth factor (KGF), a marker gene for prostate stromal cells that is not expressed in prostate epithelial cells (36). K18 expression was observed in all the tissues of roughly the same magnitude (supplemental Fig. S2A), indicating that the three classes of tissues used in Fig. 6B had the roughly equal amounts of epithelial cells. KGF expression was detected in only two BPH and one carcinoma samples (supplemental Fig. S2A), suggesting insignificant stromal contamination of our prostate tissues. The ETV1 expression levels shown in Fig. 6B were recalibrated based on K18, and this analysis demonstrated that ETV1 expression is elevated in MPC tissues as compared with normal or BPH (supplemental Fig. S2B), supporting the same analysis that was done using glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as a standard (Fig. 6B). Thus, these data confirm that the ETV1 expression differences we measured reflect true differences in epithelial cell expression.
To determine whether ETV1 protein is also elevated in prostate cancer, Western blotting was performed with those tissues for which sufficient material was available. Figure 6C shows that the ETV1 protein level is increased in all MPC tissues that represent invasive tumor as compared with normal or BPH tissues. Thus, the expression of ETV1 protein, like mRNA, increases with increasing stage of cancer (Fig. 6C, inset). This is not true for only tissue 16, which, based on the exon-specific PCR analysis of Fig. 6B, would be expected to have the TMPRSS2 translocation, suggesting that the TMPRSS2:ETV1 fusion gene may produce a protein product that our antibody does not recognize. To better analyze the expression of ETV1 protein in prostate tumors, we compared normal and cancer tissues from the same patient. With the two sets of samples available, ETV1 protein is markedly higher in cancer vs. normal prostate (Fig. 6D). Our exon-walking assay with these tissues indicates the absence of translocation with TMPRSS2 (data not shown), showing that increased ETV1 expression in cancer compared with normal is not due to translocation. Collectively, all these results demonstrate that ETV1 is significantly overexpressed in prostate cancer tissues, and this overexpression does not depend on TMPRSS2 translocation.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Several experiments collectively demonstrate that ETV1 is an AR-regulated gene in prostate cancer cells. First, ETV1 expression is induced by DHT in LNCaP cells to about the same extent as PSA, a hallmark androgen-regulated gene (22). To demonstrate that the AR effect is transcriptional, we cloned into a reporter plasmid a genomic fragment spanning the ETV1 region upstream of exon 1. This 1-kb ETV1 genomic fragment has promoter activity that is activated by androgen-induced AR. Interestingly, the DHT effects on the cloned EVT1 promoter and on endogenous ETV1 mRNA expression are similar in magnitude, suggesting that our 1-kb genomic fragment is sufficient to respond to DHT-activated AR. This is supported by the presence of a near consensus ARE found within the 1-kb fragment. This ARE was shown by ChIP assay to bind ligand-activated AR in LNCaP cells. The ChIP assay was performed with two other AREs found by nucleotide sequence analysis of upstream genomic DNA. Interestingly, neither of the two consensus AREs found about 2 and 4 kb upstream of ETV1 exon 1 showed any AR-binding ability in the ChIP assay (data not shown). These results strongly suggest that the proximal ARE found within our cloned genomic fragment is the major element that recruits AR to the ETV1 promoter.
Tomlins et al. (14) recently reported that in a subset of prostate tumors the ETV1 gene is a target of translocation with TMPRSS2, an androgen-regulated gene. The results here support the finding of Tomlins et al. (14) that about 20% of prostate tumors have a TMPRSS2:ETV1 translocation that results in elevated ETV1 expression. However, our study substantially expands their work by demonstrating that neither androgen regulation of ETV1 nor its overexpression in prostate cancer depends upon this translocation. Indeed, in the 80% of prostate tumors not having the TMPRSS2:ETV1 translocation, ETV1 expression is significantly elevated as compared with what is expressed in normal prostate. Additionally, two sets of prostate tissues, normal and cancer, from the same patients reveal higher ETV1 protein expression in the cancer tissues. These data collectively argue that ETV1 overexpression is a common feature of prostate cancer, irrespective of translocation with TMPRSS2. In contrast to ETV1, ETV4 mRNA expression is high in normal tissue and progressively decreases in prostate cancer to barely detectable levels. This is interesting in view of the recent report that ETV4 is also a target of TMPRSS2 translocation (15), which may provide a mechanism for overexpression of ETV4 in prostate cancer. Thus, the expression of ETV4 is clearly different from ETV1, which exhibits significant overexpression even in prostate tumors that have not undergone the translocation, and this is likely due to direct AR induction.
It has been suggested that the TMPRSS2:ETV1 translocation would make ETV1 androgen inducible and thereby leads to aberrant overexpression in prostate tumors (14). Interestingly, the TMPRSS2:ETV1 translocation results in the synthesis of two small fusion mRNAs (129 and 200 nucleotides; accession nos. DQ204770 and DQ204771, respectively), neither of which contains a start codon that would encode any part of the ETV1 protein. On the other hand, it is possible that a reciprocal translocation between TMPRSS2 and ETV1 would result in not only a TMPRSS2:ETV1 hybrid gene, but also an ETV1:TMPRSS2 hybrid gene, which would place some TMPRSS2 exons under the regulation of the ETV1 promoter. In view of our data showing that the ETV1 promoter becomes androgen independent in hormone-refractory prostate cancer cells, whereas the TMPRSS2 promoter remains androgen-inducible, the expression of TMPRSS2 exons found in a possible ETV1:TMPRSS2 hybrid gene would be expressed in hormone-refractory prostate cancer cells independent of androgens. This would provide a novel mechanism of TMPRSS2 overexpression in androgen-refractory prostate cancer, which is a common feature of prostate cancer (29).
The PEA3 proteins are important regulators of metastasis (reviewed in Ref. 8). This prompted us to investigate a potential role for ETV1 in prostate cancer cell invasion. Down-regulation of ETV1 expression by siRNA strongly compromised the invasive capacity of LNCaP cells. This was observed not only in androgen-dependent cells, but also in androgen-independent LNCaP cells. Interestingly, the C81 cells express a higher level of ETV1 that is not responsive to androgen stimulation. Although previous work implicates PEA3 proteins in colorectal invasiveness (9) and the non-PEA3 protein Ets2 in prostate cancer invasiveness (37), this study is the first to demonstrate a link between ETV1 expression and prostate cancer invasiveness. In view of its elevated expression in prostate tumors and constitutive expression in androgen-independent prostate cancer cells, ETV1 likely represents an important protein in metastatic androgen-independent prostate tumors.
To further study the effect of AR on ETV1, we measured the expression in prostate cancer cells of known or potential ETV1 target genes, the MMP genes. Among these genes, MMP-7 and MMP-13 exhibited DHT-inducible expression in LNCaP cells, but not PC-3 cells. Surprisingly, the expression of MMP-1, a well-known ETV1 target gene (38), is very low in LNCaP cells, substantially lower than in PC-3 cells, in which there is barely any detectable ETV1 expression. The expression in LNCaP cells of MMP-7 and MMP-13, as well as MMP-1 and MMP-9 (data not shown), depends on the presence of endogenous ETV1, suggesting that MMP-9 and MMP-13, like MMP-1 (38) and MMP-7 (9), are ETV1 target genes. MMP-13 has been shown to be androgen-induced in prostate cancer cells (39). Our data suggest that this androgen effect may be mediated by androgen-induced expression of ETV1. Interestingly, MMP-9 expression, which in LNCaP cells is substantially weaker than either MMP-7 or MMP-13, also requires endogenous ETV1. Several studies have reported an association of increased expression of MMP-7, MMP-9, and MMP-13 and prostate cancer progression. MMP-7 and MMP-9 are overexpressed in BPH and prostate cancer (40, 41). Expression levels of MMP-9 and MMP-13 are elevated in metastatic prostate cancer (40). MMP-9 can increase the metastatic potential of LNCaP and other prostate cancer cells (42). By contrast, MMP-1 mRNA expression is high in normal prostate but severely diminished in prostate cancer, opposite to the expression of ETV1. Thus, it is possible that MMP-7, MMP-9, or MMP-13 may be responsible for mediating the ETV1 effect on prostate cancer cell invasiveness, something that can be addressed in future work.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
For reporter gene assays, cells were transfected and Luciferase assays performed as previously described (32). In all cases, 0.5 µg of reporter plasmid, 1 µg pCH110 (for standardizing transfection efficiency) (32), and 0.5 µg of all expression plasmids were transfected. For all transfections, empty vector (empty expression plasmid or promoter-less reporter plasmid) was used to ensure equal amounts of each kind of vector. All Luciferase values represent the average of three independent transfections plus standard deviations.
Cell Culture and siRNA Transfection
C14 cells represent LNCaP cells stably transfected with an empty expression vector, pCI-Neo (32). A103 cells are PC-3 cells stably transfected with an AR expression plasmid (18, 32). PrEC cells are primary prostate epithelial human cells that express no detectable levels of endogenous AR, as measured by both Western blotting and reporter gene assay (data not shown). C33 are parental androgen-dependent LNCaP cells that gave rise to the androgen-independent C81 cells, which have been previously characterized (26, 27). All these cells were cultured as previously described (32). CWR22-Rv1 (28) cells were grown in RPMI1640 medium with 10% fetal bovine serum (FBS). Commercial siRNAs were obtained for ETV1 and a negative control (both from Ambion, Austin, TX). X-tremeGENE siRNA transfection reagent was used to transfect siRNA into cells following the manufacturer’s protocol (Roche, Indianapolis, IN). Lipofectamine 2000 was used to transfect plasmid DNA following the manufacturer’s protocol (Invitrogen, Carlsbad, CA).
For androgen treatment, cells were grown to 60–70% confluency in 10% FBS-containing medium and then changed to serum-free medium. After 48 h of incubation, ethanol, 100 nM DHT, or 1 nM R1881 was added to the cells. Casodex (10 or 100 nM) was added at the same time as androgen. After an additional 48-h incubation, the cells were subjected to semiquantitative RT-PCR, real-time quantitative-PCR, or Western blotting.
Semiquantitative RT-PCR and Real-Time Quantitative-PCR Analyses
RNA isolation was performed using the Trizol reagent (Invitrogen) and subjected to either semiquantitative RT-PCR as previously described (32) or real-time quantitative PCR using Sybr Green (iSCRIPT from Bio-Rad, Hercules, CA). The upstream and downstream primers, respectively, used for each gene were: PSA 5'-GCAGCATTGAACCAGAGGAG-3' and 5'-CCCATGACGTGATACCTTGA-3'; AR 5'-CAATGAGTACCGCATGCAC-3' and 5'-GCC CATCCACTGGAATAATG-3; ETV1 5'-TACCCCATGGACCACAGATT-3' and 5'-CACTGGGTCGTGGTACTCCT-3'; PEA3 5'-CCGGTTTGTCAGTTCTTGGT-3' and 5'-AGATGTGGTGGAGGTGGAAG-3'; ERM 5'-ACCATGGACGGGTTTTATGA-3' and 5'-GGCATGAAGCACCAGGTTAT-3'; MMP-1 5'-ATGCTGAAACCCTGAAGGTG-3' and 5'-CTGCTTGACCCTCAGAGACC-3'; MMP-2 5'-ATGACAGCTGCACCACTGAG-3' and 5'-ATTTGTTGCCCAGGAAAGTG-3'; MMP-3 5'-GCAGTTTGCTCAGCCTATCC-3' and 5'-GAGTGTCGGAGTCCAGCTTC-3'; MMP-7 5'-GAGTGCCAGATGTTGCAGAA-3' and 5'-AAATGCAGGGGGATCTCTTT-3'; MMP-8 5'-TCTGCAAGGTTATCCCAAGG-3' and 5'-CTTGCTGGAAAACTGCATCA-3'; MMP-9 5'-GCCATTCACGTCGTCCTTAT-3' and 5'-TTGACAGCGACAAGAAGTGG-3'; MMP-10 5'-TCCCGAAGGAACAGATTTTG-3' and 5'-GGCTCTTTCACTCAGCCAAC-3'; MMP-12 5'-CCTTCAGCCAGAAGAACCTG-3' and 5'-ACACATTTCGCCTCTCTGCT-3'; MMP-13 5'-GGAGCCTCTCAGTCATGGAG-3' and 5'-TTGAGCTGGACTCATTGTCG-3'; TMPRSS2 5'-CACTGTGCATCACCTTGACC-3" and 5'-ACACACCGATTCTCGTCCTC-3'; EZH2 5'-CCTCTGAAGCAAATTCTCGG-3'and 5'-CACAACCGGTGTTTCCTCTT-3'; E-cadherin, 5'-GAACGCATTGCCACATACAC-3' and 5'-GTGGTCAGCGGAA ACTTGAT-3'; KGF 5'-AGCTTGCAATGACATGACTCCA-3'and 5'-CCATAGGAAGAAAGTGGGCTGT-3'; K18 5'-CACAGTCTGAGGTTGGA-3' and 5'-GAGCTGCTCCATCTGTAGGG-3'; and GAPDH 5'-CGACCACTTTGTCAAGCTCA-3' and 5'-AGGGGAGATTCAGTGTGGTG-3'. GAPDH was used as a control for mRNA amount. Real-time quantitative-PCR measurements of ETV1 expression are given relative to GAPDH expression.
Exon-Walking RT-PCR
The RNA from different tissues was isolated using the Trizol Reagent and cDNA was synthesized using Moloney murine leukemia virus reverse transcriptase as described above. The PCR performed to detect different ETV1 exons used primer pairs as previously described (14).
SDS-PAGE and Western Blot
SDS-PAGE and Western blotting were carried out as described (32). The antibodies used were anti-ETV1 antibody AD2 for Western blotting (12) (directed against the ETV1 acidic domain amino acids 42–73), or anti-ß-actin antibody (Abcam). ß-Actin was used as a control for protein amount.
ChIP
LNCaP cells were grown to 70% confluence in RPM1640 containing 10% FBS. The medium was changed to RPMI1640 containing 2% dextran-coated charcoal FBS, and cells were incubated for 2 d. Then, the cells were treated with either ethanol (–) or 100 nM DHT (+) for 8 or 24 h incubation, after which cells were collected and subjected to ChIP assay as described (43). The anti-AR antibody PA1–110 (ABR) and Protein A-Sepharose (Amersham Biosciences, Piscataway, NJ) were used to perform immunoprecipitation. The primers used to detect ETV1 promoter (–1 kb) are upstream 5'-TTTTGTGAATGGGACTGTCG-3', and downstream 5'-AGGGGAACAAGATGGCTTTT-3'. The primers used to detect ETV1 promoter (–2 kb) are upstream 5'-CTGTTGGACACTG GCTCCTT-3' and downstream 5'-TTAAGCAGT GAGGGCTGCAT-3'. The primers used to detect PSA promoter are upstream 5'-GCCTGGATCTGAGAGAGATATCATC-3', and downstream 5'-ACACCTTTTTTTTTCTGGATTGTTG-3'.
Proliferation Assay
The cell proliferation experiments, using the MTT assay (Sigma, St. Louis, MO), were performed as described previously using 2% charcoal-stripped serum (32). This assay provides a measure of LNCaP proliferation that is as accurate as direct cell counting (supplemental Fig. S1). Note that 100 nM DHT was used, which we have previously shown to be as active in LNCaP cell proliferation as 1 nM R1881 (32). Direct cell counting was performed using a hemacytometer.
Cell Invasion Assay
Cell invasion was measured using the Cell Invasive Assay Kit from Chemicon experiments followed the manufacturer’s protocol. Briefly, cell suspensions containing 800,000 cells/ml (in serum-free medium) and treated with or without 100 nM DHT were used to monitor cell invasion into a lower chamber containing RPMI1640 medium with 10% charcoal-stripped serum. After 72 h of incubation at 37 C, cells were stained and quantified.
Prostate Cancer Tissues
Prostate cancer tissues were purchased from the CHTN as frozen samples. The CHTN represents a group of hospitals that obtain tissues from either surgeries or autopsies and make these tissues available for research. The information provided with the tissues indicates a diagnosis of the stage of cancer, which may include a Gleason score, and type of therapy used (radiation and/or chemotherapy). No other information is provided. We extracted the tissues for RNA by using the Trizol reagent and for protein by lysing tissues in Laemmli buffer.
Affymetrix Gene Chip Assay
C14 and A103 cells were grown to 60–70% confluency in 10% FBS and then changed to FBS-free medium. After 24 h of incubation, cells were treated with either ethanol or 100 nM DHT. After 48 h of incubation, total mRNA was isolated and subjected to gene chip analysis using chips purchased from Affymetrix (GeneChip Human Genome U95Av2 Array) according to the manufacturer’s protocol. Briefly, biotinylated cRNA from ethanol or DHT-treated cells was hybridized to the gene chip on a rotisserie box, followed by washing on a fluidics station (Affymetrix). The arrays were then stained with a streptavidin-phycoerythrin conjugage (Molecular Probes, Eugene, OR). After washing, the arrays were scanned and subjected to analysis by Affymetrix Microarray Suite 5.0, which indicated a significant change in gene expression with P value below 0.002. The experiment with C14 cells was repeated twice, yielding a false discovery rate of 2.8% (three genes out of a total of 108) between the two experiments.
Statistical Analysis
The Student’s t test was used to compare ETV1 expression under different androgen conditions and different prostate cancer cell lines, as well the proliferation and invasive capacity of prostate cancer cells. At least three repeats were used for each condition. The threshold for significance was set at P < 0.05 (two-tailed).
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Disclosure Summary: The authors have nothing to disclose.
First Published Online May 15, 2007
Abbreviations: AR, Androgen receptor; ARE, androgen-responsive element; BPH, benign prostatic hyperplasia; ChIP, chromatin immunoprecipitation; DHT, dihydrotestosterone; ETV1, Ets Variant Gene 1; FBS, fetal bovine serum; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; KGF, keratinocyte growth factor; MMP, matrix metalloproteinase; MPC, malignant prostate cancer; N, normal; PSA, prostate-specific androgen; siRNA, small interfering RNA.
Received for publication November 20, 2006. Accepted for publication May 8, 2007.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
1 mediates prostate cancer cell proliferation. Oncogene 26:1606–1615[CrossRef][Medline]
gene expression in LNCaP and Tramp-C1 prostate cancer cell lines is associated with methyl-CpG-binding protein MBD2. Oncogene 25:4559–4572[CrossRef][Medline]NURSA Molecule Pages Link:
This article has been cited by other articles:
 |
|
 |
|
  S. Shin, T.-D. Kim, F. Jin, J. M. van Deursen, S. M. Dehm, D. J. Tindall, J. P. Grande, J.-M. Munz, G. Vasmatzis, and R. Janknecht Induction of Prostatic Intraepithelial Neoplasia and Modulation of Androgen Receptor by ETS Variant 1/ETS-Related Protein 81 Cancer Res., October 15, 2009; 69(20): 8102 - 8110. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C.-L. Hsieh, C. Cai, A. Giwa, A. Bivins, S.-Y. Chen, D. Sabry, K. Govardhan, and L. Shemshedini Expression of a hyperactive androgen receptor leads to androgen-independent growth of prostate cancer cells J. Mol. Endocrinol., July 1, 2008; 41(1): 13 - 23. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Hara, H. Miyazaki, A. Lee, C. P. Tran, and R. E. Reiter Androgen Receptor and Invasion in Prostate Cancer Cancer Res., February 15, 2008; 68(4): 1128 - 1135. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. E. Helgeson, S. A. Tomlins, N. Shah, B. Laxman, Q. Cao, J. R. Prensner, X. Cao, N. Singla, J. E. Montie, S. Varambally, et al. Characterization of TMPRSS2:ETV5 and SLC45A3:ETV5 Gene Fusions in Prostate Cancer Cancer Res., January 1, 2008; 68(1): 73 - 80. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
