| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,
,||,1
* Division of Biochemistry, University of Leuven, Leuven, Belgium;
Institute of Human Genetics and Anthropology, Jena, Germany,
Regina Elena Cancer Institute, Department of Experimental Oncology, Molecular Oncogenesis Laboratory, Rome, Italy;
National Research Council, Institute of Neurobiology and Molecular Medicine, Rome, Italy; and
|| Department of Biochemistry, University of Kuopio, Kuopio, Finland
1Correspondence: Institute of Human Genetics and Anthropology, Medical Faculty, Kollegiengasse 10, 07743 Jena, Germany. E-mail: aban{at}mti.uni-jena.de
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: transrepression • gene silencing • agonist • antagonist • chromatin
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2
3)
. Interestingly, however, PCa occurs predominantly in elderly men when androgen levels are already reduced (4
, 5)
. Therefore, it has been suggested that, at least initially, androgens play a protective role in the generation of prostate hyperplasia and cancer (6
7
8
9)
. But as soon as cancer develops, androgens acquire an inducing role for cell proliferation (10)
.
The telomerase is activated in almost all cancer cells and is quiescent in almost all normal cells. Repression of telomerase activity in somatic cells and the associated replicative senescence is considered an important mechanism of tumor suppression, whereas the activation of telomerase activity appears to be one essential mechanism in carcinogenesis and growth factor independence (11
12
13)
.
In PCa, the most common malignancy in elderly men in Western countries, increased telomerase activity is already evident at the very early stages of the disease, namely prostate in situ neoplasia. Indeed, evaluation of telomerase activity in prostate biopsies has become a valuable diagnostic marker for this malignancy (14
, 15)
. Furthermore, telomerase is considered an important drug target for PCa therapy (16
17
18)
because inhibition of enzymatic activity leads to growth inhibition (19
, 20)
. The regulation of the human telomerase reverse transcriptase (hTERT) promoter is a very important mechanism to control telomerase activity. To date, several factors have been identified as activators of hTERT transcription, whereas few factors, including the tumor suppressor p53, are known to inhibit the expression of hTERT (16
, 21)
. However, the molecular mechanisms underlying the loss of telomerase repression and/or activation of telomerase during PCa development are still largely undefined.
Interestingly, and in line with the role of androgens as a protective agent, androgen ablation in animal model systems led to induction of telomerase activity (8
, 22
, 23)
. Furthermore, androgen treatment of castrated animals reversed this effect and led to down-regulation of telomerase activity (23)
. These data suggest not only that telomerase activity is repressed by androgens and that reducing androgen levels leads to its activation but also that cancer cells may overcome this repression. However, the mechanisms by which androgens inhibit the enzymatic activity of telomerase remain unknown.
AR, a ligand-activated transcription factor, is a member of the nuclear hormone receptor superfamily and consists of an amino (N)-terminal domain harboring the major transactivating function, a DNA-binding domain and the carboxyl (C)-terminal hormone-binding domain (HBD) (24)
. Classically, androgen binding induces its translocation to the nucleus and the activation of target genes.
The in vivo androgen ablation experiments that led to the induction of hTERT has prompted us to investigate whether the hTERT promoter is inhibited by androgens and whether this inhibition is abrogated in PCa cells.
Here, we report that the hTERT promoter is down-regulated by wild-type (WT) AR in an agonist-dependent manner. The N terminus of AR is essential for down-regulation, while the length of the Q-repeats does not influence AR-mediated repression. Interestingly, AR cooperates with p53 in repression of hTERT. In the human PCa cell line exhibiting hormone-dependent growth (LNCaP), however, the repression of hTERT by androgens is lost. We show here that the loss of transrepression results from the T877A point mutation of AR expressed in these cells, a finding highlighting a molecular mechanism to escape the inhibition of hTERT by androgens. On the molecular level, the loss of transrepression appears to result from less efficient recruitment of the AR T877A mutant to hTERT chromatin. Our results indicate also that this mutation, frequently found in PCa tissue (25)
, leads to not only a broader binding spectrum of other ligands and subsequent activation of AR but, in addition, to escape from the repression of hTERT. These findings also prompt the search for novel androgen antagonists that specifically inhibit AR-mediated transactivation but not the AR-mediated repression of hTERT.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
27
28)
. Additional hTERT deletion constructs were cloned as follows: pGL3–3751: pGL3–3996 was digested with Acc65I and Eco105I, both sites were treated with Klenow and religated; pGL3–439: pGL3–3996 was digested with Acc65I and BstEII, the BstEII site was filled up with Klenow and the plasmid was religated; pGL3-
1185–3551: pGL3–3996 was digested with Acc65I and Alw44I, the Alw44I site was filled in by Klenow and the plasmid was religated. The p53 expression vector (pRcCMV-p53wt) was a generous gift from Dr. Matthias Dobbelstein, University of Goettingen, Goettingen, Germany. The AR polyglutamine repeats (poly-Q) mutants were kindly provided by Dr. Angelo Poletti, University of Milan, Milan, Italy.
Cell culture and transfection experiments
CV1 (a green monkey kidney cell line lacking endogenously expressed functional androgen, glucocorticoid, and progesterone receptors), PC3 (a human prostate cancer cell line lacking endogenously expressed AR), and PC3-AR cells (PC3 cells stably expressing wild-type AR) were grown in Dulbecco modified Eagle medium (DMEM), and LNCaP cells were grown in RPMI medium; all media were supplemented with 10% FBS, and for PC3-AR cells, with 0.6 mg/ml G418 (Invitrogen, Karlsruhe, Germany). Cells were incubated at 37°C with 5% CO2. Transfections of CV1 and PC3-ARwt were done using a modified CaPO4 method as described earlier (39)
. PC3 cells were transfected with lipofectamin according to the manufacturer’s protocol. The cells, 1 x 105, were seeded out in 6-well dishes using DMEM supplemented with 10% charcoal-treated FBS and transfected 4 h later with 1 µg of reporter plasmids, 0.8 µg of AR expression constructs, 0.2 µg pCMV-LacZ for internal normalization, and, where mentioned, 0.5 µg of p53 expression vector. Cells were stimulated with cyproterone acetate (CPA; Sigma, Munich, Germany) (10–7 M), dihydrotestosterone (DHT; Sigma) (10–8 M), Casodex (Cas; Schering AG, Berlin, Germany) (10–7 M), hydroxyflutamide (OH-F; Schering AG) (10–7 M), or methyltrienolone (R1881; Perkin Elmer, Rodgau, Germany) (10–8 M) and harvested 72 h after transfection. LNCaP cells were transfected according to a method described by Fronsdal et al. (29)
, with some modifications. Three days before transfection, 3 x 105 cells were seeded out in 6-well dishes using RPMI with 10% FBS. The night before transfection with the CaPO4 method, the medium was changed to DMEM with 10% charcoal-treated FBS. At 6–8 h post-transfection, cells were glycerol shocked and refed with RPMI supplemented with 10% charcoal-treated FBS and the appropriate hormones. Then, 24 h later, cells were harvested. Luciferase activity and β-galactosidase activity for normalization were measured. Independent duplicate or triplicate experiments were performed each time. Each experiment was repeated at least 3 times. At least 2 different double CsCl gradient-purified plasmid preparations were used. Error bars in the figures represent the deviation of the mean value.
Chromatin immunoprecipitation (ChIP)
Growth of LNCaP cells was described earlier (30)
. After 3 days of cultivation, cells were treated with 10–8 M R1881 (Perkin Elmer) for 2 h. Nuclear proteins were cross-linked to DNA by adding formaldehyde directly to the medium to a final concentration of 1% and incubating at 37°C for 10 min. Cross-linking within the cells was stopped by adding glycine with a final concentration of 0.125 M and incubating at room temperature for 5 min on a rocking platform. Cells then were rinsed twice with ice-cold PBS and collected into ice-cold PBS supplemented with a protease inhibitor cocktail (Roche Diagnostics GmbH, Mannheim, Germany). After centrifugation, the cell pellets were resuspended in lysis buffer (1% SDS; 10 mM EDTA; 50 mM Tris-HCl, pH 8.0; and protease inhibitors). The lysates were sonicated on ice 10x, 10 s at 10% maximum power (Branson W-250/W), to yield DNA fragments of 1000 bp in length. After centrifugation, supernatants were collected and diluted in ChIP dilution buffer (0.01% Triton X-100, 2 mM EDTA, 150 mM NaCl, and 20 mM Tris-HCl, pH 8.0), followed by preclearing with 30 µl of salmon sperm DNA/protein A agarose 50% slurry (Upstate Biotechnology, Lake Placid, NY, USA) for 1 h at 4°C with agitation. Immunoprecipitation was performed overnight at 4°C with the rabbit anti-AR antibody (Upstate Biotechnology). The immunecomplexes were collected with 30 µl of salmon sperm DNA/protein A agarose 50% slurry for another 2 h with rotation at 4°C. Agarose beads were pelleted by centrifugation and washed sequentially for 10 min each with 1 ml of the following buffers: light salt wash buffer (0.1% SDS; 1% Triton X-100; 2 mM EDTA; 20 mM Tris-HCl, pH 8.0; and 150 mM NaCl), high salt wash buffer (0.1% SDS; 1% Triton X-100; 2 mM EDTA; 20 mM Tris-HCl, pH 8.0; and 500 mM NaCl), and LiCl wash buffer (0.25 M LiCl, 1% Nonidet P-40, 1% deoxycholate, 1 mM EDTA, and 10 mM Tris-HCl, pH 8.0). Finally, the beads were washed 2x with TE buffer (10 mM Tris-HCl and 1 mM EDTA, pH 8). The immunocomplexes were eluted twice from the beads by adding freshly prepared elution buffer (1% SDS and 0.1 M NaHCO3). Eluates were pooled, and the cross-linking was reversed by adding NaCl to a final concentration of 200 mM and heated at 65°C overnight. The remaining proteins and RNA were digested by adding proteinase K (final concentration 40 µg/ml) and RNase A (20 µg/ml), respectively, and incubating at 55°C for 3 h. The DNA fragments were purified with a DNA purification kit (QIAquick PCR Purification Kit, Qiagen, Hilden, Germany).
For polymerase chain reaction (PCR), 2 µl of 50 µl DNA extraction was used for amplification. The cycling conditions were as follows: preincubation at 94°C for 3 min, 33 cycles of denaturation at 94°C for 45 s, annealing at 60–62°C for 30 s, elongation at 72°C for 90 s, and one final incubation at 72°C for 10 min. The PCR products were separated by electrophoresis through a 2.0% agarose gel supplemented with 0.2 µg/ml ethidium bromide. The experiments were performed at least 4 times, consistently revealing the differential efficient recruitment of AR between LNCaP and PC3-AR cells. The primer sequences were as follows: PSA-ARE-for: TCTGCCTTTGTCCCCTAGAT PSA-ARE -rev: AACCTTCATTCCCCAGGACT –0.1kb TERT s: 5'-CCCCTTCACCTTCCAGCTC-3' (–200/–185) –0.1kb TERT as: 5'-GAGTTTCAGGCAGCGCTG-3' (–64/–47) –4kb TERT s: 5'-CGGTGGACAGTTCCTCACAG-3' (–3858/–3839) –4kb TERT as: 5'-AAAAGGTGCGTGCAGTAGCC-3' (–3669/–3650) –10kb TERT s: 5'-TGGCCAGTATTACCCTGATTC-3' (–10008/–9988) –10kb TERT as: 5'-CTTCCATCCCTGGAATAAATC-3' (–9822/–9802)
Reverse transcription (RT)-PCR
RNA was isolated from cells using TRIzol according to manufacturer’s instructions (Sigma). RNA transcripts were characterized by RT-PCR using Super Script One Step RT-PCR with Platinium Taq (Invitrogen) according to manufacturer’s instructions. PCR was executed under the following conditions: cDNA synthesis at 50°C for 20 min, then 94°C for 2 min, PCR amplification for 35 cycles with denaturation at 94°C for 15 s, annealing at 55°C for 30 s, and extension at 72°C for 1 min. The PCR products were visualized by ethidium bromide staining. The primer sequences were as follows:
PSA-for: ACTGCATCAGGAACAAAAGCGTGA
PSA-rev: CGCACACACGTCATTGGAAATAAC
GAPDH-for: CGGAGTCAACGGATTTGGTCGTAT
GAPDH-rev: AGCCTTCTCCATGGTGGTGAAGAC
TRAP-assay and real-time PCR experiments were performed as described previously (31)
. Cells were homogenized in Trizol (Invitrogen), and total RNA was isolated according to the manufacturer’s instructions. cDNA synthesis was performed from 2 independent RNA preparations using High Capacity cDNA Archive Kit (Applied Biosystems, Lincoln, CA, USA), and quantitive PCR was performed with the ABI Prism 7500 PCR instruments (Applied Biosystems) to amplify samples in triplicate. Predesigned TaqMan primers and probe (Applied Biosystems) specific for hTERT, 18S rRNA, and β-actin were used. The mRNA of each gene was quantified using the Standard Curve Method (5 log dilutions in triplicate) and expressed relative to the 18S rRNA or β-actin.
Telomerase activity was measured by the TRAP method described by Kim et al. (17)
. Briefly, cell extracts were prepared by lysing the cells in Nonidet P-40 Lysis buffer. After incubation on ice, the lysate was centrifuged, and the supernatant was immediately used in the assay. The reaction was carried out using 0.5–1 µg of protein extractions in 50 µl of reaction mixture, to which an internal telomerase assay standard (internal control) was added for estimation of the levels of telomerase activity and identification of any false-negative samples containing Taq polymerase inhibitors. The products were specifically amplified by PCR with the downstream primer ACX (5'-GCGCGGCTTACCCTTACCCTTACCCTAACC-3') and the upstream-labeled primer TS (5'-AATCCGTCGAGCAGAGTT-3'). The TRAP internal control, TSNT (5'-AATCCGTCGAGCAGAGTTAAAAGGCCGAGAAGCGAT-3'), was amplified by the TS primer and by its own dedicated return primer NT (5'-ATCGCTTCTCGGCCTTTT-3'), which is not a substrate for telomerase. Telomerase activity was calculated as the ratio of the intensity of the telomerase ladder to the intensity of the 36 bp internal standard. All assays were repeated at least 3x with 3 different preparations of cell lysates. For quantitative analysis of telomerase activity, the radioactive bands were scanned by the NIH Image 1.61 software (U.S National Institutes of Health, Bethesda, MD, USA). As positive and negative controls, 0.1 µg of protein from telomerase-positive HeLa cells was assayed before and after heat inactivation.
Statistical analysis
Differences among subject groups were assessed by ANOVA. Statistical significance between means of 2 paired groups was determined using Student’s t test. A 95% confidence interval (P<0.05) was considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 22
, 23)
, we wanted to determine whether AR is directly involved in repression of the hTERT promoter. For this purpose, the 4 kb hTERT reporter construct (pTERT–3996-Luc) was cotransfected with the WT human AR in CV1 cells that lack endogenous AR and other steroid receptors. Because DHT can be metabolized in cells and may then have side effects, we compared DHT with the synthetic and more stable agonist R1881 (31)
. Furthermore, we evaluated the effects on the hTERT promoter activity of androgen antagonists (antiandrogens) used in PCa therapy (24
, 32)
, such as CPA, hydroxyflutamide (OH-F), and Cas (Fig. 1
A). Androgen antagonists partially reduce or completely block human AR-mediated transactivation.
|
Treatment of cells with DHT or R1881 led to repression of hTERT promoter activity to a similar extent in the presence of expressed AR (P<0.01) (Fig. 1A
). In the absence of AR, however, treatment with R1881 had no effect, suggesting that the repression of hTERT is dependent on AR and is controlled by ligand. Treatment with the pure antagonists, either OH-F or Cas, did not result in significant repression (P<0.1), whereas the antiandrogen CPA, known to have partial agonistic activity, led to inhibition of the hTERT promoter, although significantly less pronounced than with pure agonists (P<0.03) (Fig. 1A
).
This result suggests that AR agonists mediate the transrepression through the human AR, whereas antagonists mediate only partially or lack completely this activity.
Using the same cells and experimental setup, a potent hormone induction of the well-characterized positively regulated MMTV promoter was observed with the pure agonists DHT and R1881 (Fig. 1B
), weaker induction with the partial agonist CPA, and no significant induction with the complete antagonists Cas or OH-F (P<0.003) (Fig. 1B
). These results also were obtained using the prostate specific antigen (PSA) promoter and are in line with previous observations (26)
.
To test whether hTERT repression is dependent on AR expression levels, dose-response experiments were performed with increasing amounts of AR expression vector. Inhibition by agonist-bound AR enhanced with increasing amounts of AR expression plasmid, indicating that hTERT repression is dose-dependent (data not shown). Under the same experimental conditions, hormone induction of the MMTV promoter was enhanced by increasing amounts of AR expression plasmid (data not shown). However, with the highest AR amount, the known squelching effect for positively regulated genes was observed (33)
. In contrast, the inhibitory dose response by AR as observed on the hTERT-promoter is rather typical of a transrepression mechanism without direct DNA binding (34)
. Thus, our data indicate that the AR represses the hTERT promoter activity in an agonist-dependent manner and suggest distinct mechanisms for dose-dependent activation and repression.
The N terminus but not the HBD of AR is involved in hTERT repression
To get some insights into the mechanism by which AR mediates repression of hTERT, a battery of AR mutants and deletions were employed. A known variation of AR found in patients with Kennedy’s disease is the poly-Q extension at the N terminus of AR, whereas a shorter poly-Q stretch has been correlated with higher risk of developing PCa (35)
. Repression of the hTERT promoter activity was observed independently of the length of the poly-Q stretch (P<0.03), suggesting that this motif does not influence the agonist-bound AR (Fig. 2
A).
|
Corepressors, such as silencing mediator of retinoid and thyroid hormone receptor (SMRT), were shown to bind to AR. This binding was abrogated when the small ubiquitin-like modifier (SUMO) -ylation sites of AR at positions aa 385 and aa 518 are mutated (26)
. This mutant AR was used to test whether these SUMOylation sites are involved in the AR-mediated hTERT repression. The extent of repression was similar to that with WT AR (Fig. 2A
), suggesting that the SUMOylation sites play no role in this process. In line with this, overexpression of SMRT or the corepressor Alien (30)
did not enhance AR-mediated repression of the hTERT promoter activity, indicating that corepressors are not involved (data not shown).
To define the receptor domains involved in transrepression of the hTERT promoter, an AR mutant that lacks the DBD (hAR-DBD) and the naturally occurring AR mutant A596T that has an aa-exchange in the first zinc finger of the AR DBD and causes Reifenstein syndrome (36)
were employed. This AR mutant was shown to exhibit strongly reduced DNA binding and thus transactivation on single AREs (36)
. Interestingly, both AR mutants are able to repress hTERT activity to a similar extent as the WT AR (P<0.03) (Fig. 2A
). This indicates that the DNA binding function of AR is not involved in repression of the hTERT promoter, supporting the existence of distinct mechanisms of transrepression and transactivation. Further, deletion of the AR N terminus (hAR-N) abrogated transrepression, whereas deletion of the HBD (hAR-HBD) did not (P<0.03) (Fig. 2B
). These data suggest that the AR N terminus mediates the repression of the hTERT promoter.
Both the hTERT promoter as well as the AR-mediated transactivation are known to be repressed by the tumor suppressor p53. Interestingly, p53 achieves the latter effect by preventing a direct DNA binding of AR to its response elements (37
, 38)
. To test for the influence of both factors on the hTERT promoter activity, the –3996-TERT promoter construct together with empty vectors, p53 expression vectors alone, or in combination with AR expression vectors were used. p53 expression led to inhibition of hTERT promoter activity, which is in line with previous data (37)
and, as expected, is independent of the presence of the R1881 agonist (P<0.01) (Fig. 3
A). AR repressed the hTERT promoter in a ligand-dependent manner, which was synergistically enhanced in combination with p53 expression in the presence of R1881.
|
This suggests that p53 does not abrogate AR-mediated repression of hTERT but rather cooperates with R1881-activated AR to further down-regulate hTERT expression. In contrast, expression of p53 decreased the AR-mediated induction of the positively regulated MMTV promoter (P<0.01) (Fig. 3B
), in agreement with previous findings (38)
. These results also underlie the observation that the DNA-binding function of AR is dispensable for transrepression of hTERT by AR.
Thus, the AR-mediated transrepression of the hTERT promoter is not inhibited but rather enhanced by p53. Taken together, these findings indicate that the AR N terminus mediates hTERT repression and cooperates with the tumor suppressor p53.
The LNCaP mutant AR T877A is unable to repress the hTERT promoter
It is known that in LNCaP cells, which endogenously express the AR T877A mutant, telomerase activity is not repressed but rather activated by AR agonists, presumably in an indirect manner (19
, 39)
. Therefore, we wondered whether the –3996-TERT promoter construct is androgen-regulated in PCa cells and therefore compared LNCaP with PC3-AR cells that express the WT human AR. Interestingly, in LNCaP cells, the hTERT promoter is not repressed by androgen agonists but apparently is slightly and reproducibly activated, as shown by a decrease in fold repression (P<0.05 for DHT) (Fig. 4
A), according to previous observations (19
, 39)
. As expected, the MMTV reporter exhibits a potent hormone induction in these cells (P<0.003) (Fig. 4B
). Although the MMTV reporter is also strongly induced by AR agonists in PC3-AR cells (P<0.001) (Fig. 4D
), the hTERT promoter is repressed by both DHT and R1881 (P<0.03) (Fig. 4C
). To ensure that this repression is AR-dependent, the parental PC3 cells were used, lacking endogenously expressed AR, and observed neither down-regulation of hTERT promoter activity by androgen agonists nor hormone responsiveness of the MMTV promoter (data not shown). These data indicate that endogenously expressed androgen-activated AR in LNCaP cells is unable to inhibit the hTERT promoter whereas the hTERT promoter is inhibited in PC3-AR cells.
|
To verify the functional activity and responsiveness to androgens of the endogenous telomerase, both hTERT expression and telomerase activity were evaluated vs. time on treatment with R1881, DHT, or Cas (Fig. 5
A) in PC3-AR and LNCaP cells. PC3 cells served as negative control. Quantitative real-time PCR (qRT-PCR) was used to detect the level of transcripts of the hTERT subunit. In agreement with previous findings, we detected significant (P<0.05) reduction of hTERT mRNA by R1881 and DHT in the presence of WT AR (PC3-AR). In contrast, with the AR T877A mutant (LNCaP), rather than repression of hTERT mRNA levels, a slight induction on DHT treatment was observed (Fig. 5A
). Treatment with Cas had no significant effect on hTERT mRNA levels in PC3 and PC3-AR cells but slightly reduced hTERT levels in LNCaP cells (Fig. 5A
).
|
To assess the functionality of the endogenous telomerase activity, TRAP-assays were performed in the presence of R1881, DHT, or Cas. The telomerase activity was unaffected, inhibited, or slightly increased in PC3, PC3-AR, and LNCaP in the presence of R1881 and DHT, respectively (Fig. 5B
). Treatment with Cas did not significantly affect telomerase activity in PC3 and PC3-AR cells and slightly reduced it in LNCaP cells. As control, untreated or heat inactivated HeLa cell extracts were used (Fig. 5B
), and the induction of PSA gene expression in LNCaP cells by the AR agonist R1881 was verified by qRT-PCR (data not shown).
Thus, the data suggest that both the endogenous hTERT expression and the hTERT activity are repressed by AR agonist in PC3-AR cells but not in LNCaP cells.
Since LNCaP cells express the mutant AR T877A whereas the PC3-AR cells the WT AR, we hypothesized that the mutation in AR abrogates the AR-mediated repression of the hTERT promoter. To test for the loss of transrepression, we transfected the mutant T877A or the WT AR expression plasmids in the AR-negative CV1 cells with either the hTERT or MMTV reporter. Interestingly, unlike WT AR, the AR T877A mutant was unable to repress the hTERT promoter in the presence of androgen agonist (P<0.03) (Fig. 6
A). In contrast, with both WT and mutant AR T877A, a similar hormone induction of the MMTV promoter occurred (P<0.001) (Fig. 6B
). To analyze whether WT AR is able to mediate repression in LNCaP cells, we overexpressed it and observed repression of hTERT promoter activity (data not shown), suggesting that the mechanism of transrepression is intact in these cells.
|
Thus, in contrast to WT AR, the mutant AR T877A is unable to repress hTERT expression.
The AR is recruited to the hTERT promoter in vivo
Promoter deletion analyses were performed to test for androgen-dependent inhibition of the hTERT promoter and to compare the responsiveness of WT and mutant AR. The results indicate that WT AR is able to inhibit all used promoter deletions in an AR- and agonist-dependent manner, whereas the AR T877A mutant has strongly reduced or absent transrepression activity (Fig. 7
). The promoter deletions also indicate that AR-mediated repression operates through very proximal promoter sites but does not exclude involvement of AR on the distal promoter sequences as well.
|
To investigate whether AR is recruited to the endogenous hTERT promoter in vivo, we used ChIP analyses with anti-AR antibodies. Both LNCaP and PC3-AR cells were used to assess potential differences in AR recruitment. In the absence of ligand, the AR is weakly recruited on the positively regulated PSA gene promoter (Fig. 8
A), which is in agreement with previous observations (30
, 40)
, and this recruitment is strongly enhanced by addition of ligand in both PCa cell types. Analyzing the hTERT promoter region, the AR was found to be recruited to the proximal promoter region, hTERT –0.1 kb, as well as to the –4 kb region, but not to the –10 kb region in PC3-AR cells (Fig. 8A
). The recruitment of AR to the hTERT promoter in PC3-AR cell regions was strongly enhanced by treatment with androgen agonists. However, using LNCaP cells, no significant recruitment of AR to the hTERT promoter was observed, albeit AR recruitment is detected under the same conditions to the PSA promoter (Fig. 8A
). This result indicates that within cells exhibiting hTERT repression, the AR is stronger recruited to the promoter compared to LNCaP cells lacking hTERT transrepression by AR.
|
Thus, these data strongly suggest a hormone-regulated recruitment of AR at specific promoter regions of hTERT and less efficient recruitment of the AR T877A mutant, which may account for the loss of transrepression.
Further, we investigated whether the AR displays a timely dynamic recruitment. For that purpose, different time points of ligand treatment were analyzed (Fig. 8B
). An agonist-enhanced recruitment was observed within 30 min of treatment in PC3-AR cells, being more efficient between 30 and 40 min of ligand treatment. Interestingly, also in LNCaP cells, we observed recruitment of AR to the hTERT promoter, albeit compared to the input, the recruitment was much weaker in these cells, suggesting that the AR is recruited more efficiently to the hTERT promoter in PC3-AR cells. This result was observed although the endogenous protein level of the AR was significantly higher in LNCaP cells than in PC3-AR (data not shown). This difference in recruitment could be one underlying mechanism of the differential hormone-regulated hTERT expression in these cells.
Thus, endogenous WT AR in PC3-AR cells is recruited in vivo onto the hTERT promoter, whereas in LNCaP cells the mutant receptor is recruited to a much lesser extent, supporting the notion that the strongly reduced recruitment of the AR mutant might be the basis for the loss of androgen-regulated transrepression.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
7
8
9)
. There are several potential mechanisms for this escape. We provide novel evidence that one such mechanism involves mutation of the AR HBD in Helix 11. A single amino acid exchange in the T877A AR mutant abrogates androgen-mediated inhibition of hTERT. Similarly, we have tested another naturally occurring mutation of AR in helix 11, AR H874Y, and observed a reduced transrepression of the TERT promoter (Supplemental Data). The T877A mutation seems to occur frequently in PCa and represents a hot spot mutation (25)
. To date, this mutation has been thought to allow AR to bind only to ligands to which it normally does not bind, including androgen antagonists such as OH-F, which, instead of inhibiting AR-mediated transactivation, activates the receptor (41
, 42)
. Our findings strongly suggest that, in addition to these functions, the T877A mutation has the important role of abrogating hTERT transrepression. The obtained data indicate that the lack of hTERT repression in LNCaP cells could result from the weaker recruitment of AR onto the hTERT promoter, compared to PC3-AR cells. Also, a timely distinct recruitment of AR was observed when comparing LNCaP and PC3-AR cells. A change of the dynamics in the recruitment of transcriptional regulators was postulated to change the gene expression (43)
. Therefore, a different time pattern of recruitment also could account for differences in hTERT gene expression. Thus, our findings suggest that, in addition to broadening of ligand spectrum binding, the T877A mutation has the important role of abrogating hTERT transrepression, which represents a mechanism for AR mutations to escape a protective role.
An N- to C-terminal interaction of AR has been shown to be involved in full AR-mediated transactivation (44)
. Concerning transrepression, although the AR N terminus is sufficient to repress hTERT, the T877A receptor mutant suggests a possible interaction between this region and Helix 11, at least for transcriptional control through transrepression.
Based on the results presented here, we propose a model whereby AR represses telomerase activity in the normal prostate in a hormone-dependent manner. This action would be a tumor-suppressing mechanism mediated by AR. However, during the establishment of PCa, there could be a selection for those AR mutants that have lost the ability to repress hTERT but retain the ability to activate those genes that promote tumor proliferation. Cancer cells expressing such AR mutation would have a clear growth advantage. Taken together, our data suggest a protective mechanism of androgens that may not be functional in PCa.
Telomerase inhibitors are being considered useful for treatment of PCa (16
17
18)
and, therefore, knowing precisely how AR-mediated hTERT inhibition occurs and also how to maintain it would be helpful. The underlying mechanisms of transrepression of hTERT by AR are currently unclear and might involve the inhibition of the transcriptional activators AP1, Ets, and SP1 (45
, 46)
for which putative binding sites were identified in the hTERT promoter regions to which AR is recruited in vivo.
The findings shown here, that androgen receptor antagonists (OH-F, Cas, and, to a certain extent, CPA) do not induce hTERT repression by AR, suggest that novel types of androgen antagonists should be developed that, on one hand, inhibit AR-mediated transactivation but, on the other hand, do not influence AR-mediated inhibition of hTERT expression.
|
ACKNOWLEDGMENTS |
|---|
Received for publication July 19, 2007. Accepted for publication October 11, 2007.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Eisold, M. Asim, H. Eskelinen, T. Linke, and A. Baniahmad Inhibition of MAPK-signaling pathway promotes the interaction of the corepressor SMRT with the human androgen receptor and mediates repression of prostate cancer cell growth in the presence of antiandrogens J. Mol. Endocrinol., May 1, 2009; 42(5): 429 - 435. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Neuwirt, M. Puhr, F. R. Santer, M. Susani, W. Doppler, G. Marcias, V. Rauch, M. Brugger, A. Hobisch, L. Kenner, et al. Suppressor of Cytokine Signaling (SOCS)-1 Is Expressed in Human Prostate Cancer and Exerts Growth-Inhibitory Function through Down-Regulation of Cyclins and Cyclin-Dependent Kinases Am. J. Pathol., May 1, 2009; 174(5): 1921 - 1930. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
