Yin Yang 1 regulates the transcriptional activity of androgen receptor

Article metrics

Abstract

The multifunctional protein Yin Yang 1 (YY1) has an important role in epigenetic regulation of gene expression. YY1 is highly expressed in various types of cancers, including prostate cancer. Currently, the mechanism underlying the functional role of YY1 in prostate tumorigenesis remains unclear. In this report, we investigated the functional interplay between YY1 and androgen receptor (AR), and the effect of YY1 on AR-mediated transcription. We found that YY1 physically interacts with AR both in a cell-free system and in cultured cells. YY1 is required for the optimal transcriptional activity of AR in promoting the transcription of the prostate-specific antigen (PSA) promoter. However, ectopic YY1 expression in LNCaP cells did not further enhance the reporter driven by the PSA promoter, suggesting that an optimal level of YY1 is already established in prostate tumor cells. Consistently, YY1 depletion in LNCaP cells reduced endogenous PSA levels, but overexpressed YY1 did not significantly increase PSA expression. We also observed that YY1–AR interaction is essential to YY1-mediated transcription activity of AR and YY1 is a necessary component in the complex binding to the androgen response element. Thus, our study demonstrates that YY1 interacts with AR and regulates its transcriptional activity.

Introduction

As a transcription factor, YY1 (Yin Yang 1) either represses or activates gene expression, depending on the recruited cofactors. Many YY1-regulated genes have crucial roles in cell proliferation and differentiation (Shi et al., 1997; Thomas and Seto, 1999), whereas YY1 gene is also regulated by other factors, such as prohibitin and nuclear factor-κB (Joshi et al., 2007; Wang et al., 2007). In addition, the function of YY1 is modulated by different posttranslational modifications (Yao et al., 2001; Deng et al., 2007; Rylski et al., 2008). Several studies, including ours, demonstrated that YY1 is a negative regulator of p53 (Gronroos et al., 2004; Sui et al., 2004; Bain and Sinclair, 2005; Santiago et al., 2007). YY1 also recruits p300, HDAC1 and Ezh2 to mediate histone modifications (Yang et al., 1996; Yao et al., 2001; Caretti et al., 2004) indicating a pivotal role of YY1 in genomic imprinting and chromatin remodeling.

YY1 is overexpressed in different human cancers, including prostate cancer (PCa; Erkeland et al., 2003; Begon et al., 2005; Seligson et al., 2005; de Nigris et al., 2006; Baritaki et al., 2007). Recent studies suggest that YY1 has a role in PCa development and progression. YY1 is increased in the prostatic intraepithelial neoplasia stage (Seligson et al., 2005). Importantly, YY1 is essential to histone methylation mediated by Ezh2 (Caretti et al., 2004; Wilkinson et al., 2006) that is an oncogene in prostate tumorigenesis (Bracken et al., 2003). Moreover, p53 mutations are frequently observed in the late stage of PCa, but not in the primary tumors. Therefore, YY1 may antagonize the tumor suppression surveillance of p53 during malignant transformation of prostate cells.

Androgenic hormones are necessary for normal prostate development and also have critical roles in stimulating the proliferation and progression of PCa cells through androgen receptor (AR; Heinlein and Chang, 2004). As a ligand-regulated transcription factor, AR shares common features with other members of the nuclear receptors. AR protein contains two activation domains (AF-1 and AF-2), a DNA-binding domain (DBD) and a ligand-binding domain (LBD; Dehm and Tindall, 2005). The binding of androgens changes the conformation of AR and composition of the AR-containing complex, which leads to its translocation from cytoplasm to nucleus. Nuclear AR associates with androgen response elements (AREs) in the promoters and enhancer regions of its target genes, and stimulates their expression (Dehm and Tindall, 2005).

Due to the crucial role of AR in PCa progression, androgen deprivation is used as a therapeutic approach for patients with advanced PCa (Sharifi and Farrar, 2006). Several studies indicate that this treatment exerts selective pressure on androgen-signaling pathways (Craft et al., 1999; Taplin et al., 1999). This leads to elevated AR levels (Koivisto et al., 1997; Gregory et al., 2001), increased AR sensitivity to androgens or other steroid hormones (from gain-of-function mutations; Veldscholte et al., 1992; Gaddipati et al., 1994) and enhanced AR nuclear localization (Lapouge et al., 2007). These adaptive changes sustain the function of AR in stimulating cell proliferation and preventing apoptosis, and consequently cause recurrent cancers that are apparently androgen independent.

As YY1 has great potential in regulating cancer development and is increasingly expressed in early stages of PCa, we investigated whether there is a functional interplay between YY1 and AR. In this study, we focused on the role of YY1 in mediating the transcriptional activity of AR.

Results

YY1 directly interacts with AR

We conducted immunoprecipitation experiment in 293T cells cotransfected with pcDNA3/Flag-AR and pcDNA3/HA-YY1. When the cell lysates were immunoprecipitated with an AR antibody (N-20) and analysed by western blot using an hemagglutinin (HA) antibody (F-7), we detected HA-YY1 brought down by the AR antibody in the sample with both plasmids transfected (lane 4, Figure 1a). HA-YY1 was not detected if Flag-AR was absent in transfection or a control rabbit immunoglobulin G (IgG) replaced the AR antibody (lanes 3 and 2, Figure 1a). In the reciprocal immunoprecipitation experiment, when HA antibody (F-7) was used to bring down HA-YY1, Flag-AR could be pulled down when both HA-YY1 and Flag-AR were transfected (lane 4, Figure 1b). However, no such significant amount of Flag-AR could be brought down when the control antibody was used or when Flag-AR was replaced by an empty vector (lanes 2 and 3, Figure 1b). We further examined whether endogenous YY1 and AR interact in prostate cells. We immunoprecipitated AR from LNCaP cell lysates using AR antibody N-20 and C-19, recognizing the N- and C terminals of AR, respectively. YY1 could be coimmunoprecipitated by both AR antibodies, but not the control IgG (lanes 3, 4 vs 2, Figure 1c). AR N-20 antibody could pull down more YY1 than AR C-19, proportional to the AR amounts brought down by them (lower panel). In the reciprocal experiment, both YY1 antibodies (C-20 and H-414) brought down significant amounts of endogenous AR compared to the control rabbit IgG (lanes 3, 4 vs 2, Figure 1d). Again, the amounts of AR brought down by these antibodies were proportional to the levels of immunoprecipitated YY1 (lower panel).

Figure 1
figure1

Interaction of Yin Yang 1 (YY1) and androgen receptor (AR) in cells. (a and b) Interaction of transfected YY1 and AR in 293T cells. pcDNA3/HA-YY1 (1 μg) or an empty vector and pcDNA3/Flag-AR (1 μg) or the empty vector were cotransfected into 293T cells. One day posttransfection, cells were cultured in medium containing 10 nM R1881 for 24 h, followed by cell lysis. In coimmunoprecipitation (CoIP) experiments, AR polyclonal antibody (N-20), HA antibody or control antibodies were incubated with cell lysates. Precipitated samples were applied to western blot analysis using monoclonal antibodies against HA (a) or Flag (b). HA-YY1 and Flag-AR are indicated at left. Lower panels: western blots of the protein inputs in CoIP. (c and d) Interaction of endogenous YY1 and AR in LNCaP cells. LNCaP cells were treated with 10 nM R1881 for 24 h and 500 μg of cell lysates were used in reciprocal CoIP analysis using indicated antibodies (3 μg each). Antibodies used in western blot studies (upper and lower panels) are indicated at left.

We then asked whether the YY1-AR association occurred through direct protein–protein interaction. As the full-length AR was poorly expressed in bacteria, we produced and purified recombinant AR N terminal (1–555, containing AF-1 domain), AR C terminal (556–919, containing DBD, LBD and AF-2 domains), and full-length YY1, fused with either 6-histidine tag (His × 6) or glutathione S-transferase (GST), from Escherichia coli Rosetta cells. First, we incubated His × 6-YY1 with GST-AR-N-terminal(1–555), GST-AR-C-terminal(556–919) and GST alone, followed by the addition and incubation of glutathione-conjugated agarose (Sigma, St Louis, MO, USA). GST-AR(556–919), but not GST or GST-AR(1–555), brought down His × 6-YY1 (lanes 4 vs 2 and 3, Figure 2a), which showed a similar intensity to that brought down by GST-p53 (lane 5), as we previously demonstrated (Sui et al., 2004). We also performed a reciprocal binding assay by incubating His × 6-AR(556–919) with GST-YY1 and GST. We observed that GST-YY1, but not GST alone, could bring down His × 6-AR(556–919) (lanes 3 vs 2, Figure 2b). We mapped the interacting domain of AR on YY1 protein using three GST-YY1 mutants, and observed that YY1(198–414) and YY1(331–414), but not YY1(1–224), interacted with His × 6-AR(556–919) (lanes 5, 6 vs 4, Figure 2b), suggesting that AR-binding site is located at the C terminal of YY1. In this experiment, GST-p53 (lane 7) served as a positive control, because p53 directly interacts with AR (Shenk et al., 2001).

Figure 2
figure2

In vitro protein-binding studies to determine Yin Yang 1 (YY1) and androgen receptor (AR) interaction domains. (a) Identification of YY1-binding domain on AR. Purified GST-AR(1–555), GST-AR(556–919) and GST-p53 proteins (3 μg each) were individually incubated with His × 6-YY1 (1.2 μg). Samples brought down by glutathione agarose beads were analysed by western blot using YY1 (H-10) antibody. Lower panel: Ponceau S staining of transferred membrane to show input of glutathione S-transferase (GST) fusion proteins. (b) Identification of AR-binding domain on YY1. Top panel: diagram of GST–YY1 fusion proteins. Purified GST-YY1 proteins (3 μg each) and GST-p53 were individually incubated with purified His-AR(556–919) (1.2 μg each). Samples brought down by glutathione agarose beads were analysed by western blot using AR (N-20) antibody. Lower panel: Ponceau S staining of the transferred membrane to demonstrate input of the GST fusion proteins.

YY1 enhances the transcriptional activity of AR

After observing direct YY1–AR interaction, we asked whether YY1 has any regulatory effect on AR-regulated gene expression. We determined the response of the prostate-specific antigen (PSA) promoter to different YY1 concentrations with or without AR. We transfected 293T cells with PSA-Fluc (300 ng), pcDNA3/Flag-AR (300 ng) and increasing amounts (75, 150 and 300 ng) of pcDNA3/HA-YY1, and cultured the cells with or without the synthetic androgen R1881. Although the activation of AR to the PSA promoter was greatly stimulated by R1881 (8 vs 2, Figure 3a), HA-YY1 alone also slightly enhanced the PSA promoter regardless of the R1881 presence (3 vs 1, and 9 vs 7). When Flag-AR was cotransfected, an initial amount (75 ng) of HA-YY1 displayed increased transcription in the absence of R1881 compared to HA-YY1 or Flag-AR alone (4 vs 3 and 4 vs 2), but this increase was markedly enhanced by R1881 (10 vs 9 and 10 vs 8), exhibiting a synergistic effect of HA-YY1 and Flag-AR. Strikingly, further YY1 increases (150 and 300 ng) inversely affected this synergistic activation (5, 6 vs 4, and 11, 12 vs 10). Western blot analysis indicated that YY1 increases did not apparently alter AR levels (right, Figure 3a). We repeated this reporter assay with lower concentrations (50 and 75 ng) of YY1, but did not observe any significant difference between these two conditions (data not shown).

Figure 3
figure3

Studies of Yin Yang 1 (YY1) expression on the activity of androgen receptor (AR)-mediated prostate-specific antigen (PSA) promoter. (a) Effects of YY1 increase on PSA promoter activity with transfected AR in 293T cells. pcDNA3/HA-YY1 (75, 150 and 300 ng), pcDNA3/Flag-AR (300 ng), PSA-Fluc reporter (300 ng) and actin-SEAP (100 ng) were cotransfected into 293T cells cultured in 12-well plates with or without 10 nM R1881. Each transfection was in triplicate and total DNA was compensated to the same amount with an empty vector, if necessary. Firefly luciferase (Fluc) in cell lysates was measured at 48 h after transfection and then normalized against secreted alkaline phosphatase (SEAP) activity in the same sample. Right panel: representative western blot analysis of Flag-AR, HA-YY1 and GAPDH. (b) Effects of YY1 increase on the PSA promoter in LNCaP cells. Differing amounts of pcDNA3/HA-YY1 (50, 100 and 200 ng), PSA-Gluc reporter (80 ng) and actin-SEAP (100 ng) were cotransfected into LNCaP cells cultured in the absence or presence of R1881 in triplicates with an empty vector to compensate for each transfection if necessary. Gaussia luciferase (Gluc) in the medium was measured at 48 h after transfection and then normalized against SEAP activity in the same sample. Right panel: representative western blot analysis of HA-YY1, endogenous AR and GAPDH. (c) Effects of YY1 depletion on the PSA promoter in LNCaP cells. LNCaP cells were cotransfected with PSA-Gluc reporter (80 ng), actin-SEAP (100 ng) and the plasmid expressing the scrambled siRNA or yy1 siRNA (300 ng), as labeled, in triplicates. After cells were cultured in medium with or without 10 nM R1881 for 48 h, Gluc activity in the cell lysates was measured and normalized to SEAP activity in the same sample. Expression of YY1 and AR was analysed by western blot and shown at right.

To determine how YY1 affects endogenous AR, we studied the effects of altered YY1 expression on transcriptional activity of AR in LNCaP cells that highly express both YY1 and AR proteins. We cotransfected LNCaP cells with 80 ng of PSA-Gluc and three different amounts (50, 100 and 200 ng) of HA-YY1. HA-YY1 increases did not significantly change the PSA promoter activity with or without R1881 (2, 3, 4 vs 1 and 6, 7, 8 vs 5, Figure 3b). A representative western blot analysis of endogenous AR and transfected HA-YY1 is on the right. The results suggested that highly expressed endogenous YY1 in LNCaP cells is sufficient for the optimal activity of AR; hence, ectopically introduced YY1 would not further enhance its activity. Instead, when YY1 increased from 100–200 ng, we observed a slight decrease in luciferase activity driven by the PSA promoter, which is consistent with the effects of increased amounts of YY1 in 293T cells (Figure 3a).

To determine whether YY1 is required for AR-mediated transcription, we cotransfected LNCaP cells with U6/scrambled siRNA or U6/yy1 siRNA and 80 ng of PSA-Gluc, with or without R1881. In the presence of R1881, YY1 depletion (38% of its original level; right in Figure 3c) decreased Gaussia luciferase (Gluc) activity by 50% (4 vs 3; P<0.01). Without R1881, the reporter expressed basal levels of Gluc despite of YY1 expression (2 vs 1).

YY1 knockdown leads to decreased endogenous PSA expression

As YY1 is required for AR-mediated expression (Figure 3), we asked whether altered YY1 expression changes the endogenous PSA level. We infected LNCaP cells with lentivirus generated from an empty pSL2 vector or pSL2-YY1 that employed the cytomegalovirus promoter to drive yy1 cDNA (upper panel, Figure 4a) and then cultured the cells with or without R1881. Although endogenous PSA expression could be increased by ectopically expressed YY1 in the absence of R1881 (1.8-fold; lanes 3 vs 1, Figure 4b), there was no obvious change of PSA levels in the presence of R1881 (lanes 4 vs 2, Figure 4b). In addition, YY1 increase did not change the expression of endogenous AR. This result is consistent with our observation in the reporter assay (Figure 3b).

Figure 4
figure4

Effect of Yin Yang 1 (YY1) on expression of endogenous prostate-specific antigen (PSA) in LNCaP cells. (a) Schematic diagrams of lentiviral vectors used to express YY1 and siRNAs in LNCaP cells. Sequences between the two long terminal repeats (LTRs) in the lentiviral vectors are integrated into the genome of the infected cells. CMVp, cytomegalovirus promoter; IRES, internal ribosomal entry site; UCP, ubiquitin c promoter; EGFP, enhanced green fluorescent protein. (b) Effect of overexpressed YY1 on endogenous PSA. LNCaP cells were infected with lentivirus generated from empty pSL2 vector or pSL2/YY1. Infected cells were first cultured in phenol-red-free RPMI medium with 10% charcoal stripped serum (CSS) for 48 h, followed by another 24 h of culture with or without 10 nM R1881. Cell lysates were then analysed by western blot using antibodies against PSA (BiosPacific Inc., Emeryville, CA, USA), YY1 (H-10), androgen receptor (AR; N-20) and β-actin (MAB1501; Chemicon International Inc., Temecula, CA, USA). (c and d) Effect of YY1 depletion on endogenous PSA. LNCaP cells were infected by lentivirus generated from pLu-U6/scrambled siRNA, pLu-U6/yy1 and pLu-U6/AR. Infected cells were treated as described above and western blot was used to detect expression of PSA, YY1, AR and β-actin (c). The samples infected by the scrambled siRNA and yy1 siRNA were also analysed by real-time quantitative PCR (qPCR) for PSA expression with GAPDH as a control (d).

We further studied how YY1 knockdown affects endogenous PSA expression. LNCaP cells were individually infected with lentiviruses carrying U6/scrambled, U6/yy1 and U6/AR siRNAs (lower panel, Figure 4a) and then cultured with or without R1881 for 3 days, followed by western blot analysis. The two middle panels of Figure 4c indicate that the endogenous YY1 and AR were correspondingly knocked down. Without R1881, PSA expression was low when the scrambled siRNA was introduced, but markedly decreased when either YY1 or AR was knocked down (lanes 3, 5 vs 1, top panel of Figure 4c). Consistently, when R1881 was added, both YY1 and AR knockdown also reduced the expression of endogenous PSA compared to the scrambled siRNA control (lanes 4, 6 vs 2, Figure 4c).

To quantitatively analyse the effects of YY1 depletion on PSA expression, we determined the levels of PSA mRNA in LNCaP cells expressing the scrambled siRNA and yy1 siRNA, respectively. As shown in Figure 4d, the expression of PSA mRNA dropped to 55.6% in the YY1 depleted cells compared to the cells expressing scrambled siRNA.

YY1–AR, but not YY1–DNA, interaction is essential to YY1-enhanced AR activity

As YY1 directly binds to AR, we asked whether YY1–AR interaction determines YY1-regulated transcriptional activity of AR. As we mapped the AR-binding domain to the C terminal of YY1 (Figure 2a), we tested the interaction of AR with YY1 chimera 17 (YY1-Chi17) that has the second zinc finger of YY1 replaced by a zinc finger from growth factor independence-1 (Figure 5a) (Galvin and Shi, 1997). In the in vitro protein-binding experiments, whereas wild-type (wt) His × 6-YY1 interacted with GST-AR-C-terminal, His × 6-Chi17 lost this interaction (lanes 8 vs 4, Figure 5B). Reciprocally, GST-YY1, but not GST-Chi17, brought down His × 6-AR(556–919) (lanes 3 vs 4, Figure 5c), with GST-PIASy as a positive control (lane 5), as it interacts with AR (Gross et al., 2004). We asked if the loss of interaction between YY1-Chi17 and AR could result from a destructive distortion of YY1 conformation after the substitution of its second zinc finger. We therefore tested the binding affinity of YY1-Chi17 with Hdm2 and PIASy, which have been identified by us to interact with the spacer region and Zinc finger domain of YY1, respectively (Sui et al., 2004; Deng et al., 2007). As shown in Figures 5d and e, GST-wt YY1 and GST-YY1-Chi17 exhibited comparable affinity to His × 6-Hdm2 and His × 6-PIASy, suggesting that YY1-Chi17 likely retains a conformation similar to wt YY1. The result in Figure 5e also indicated that, although PIASy binds to the zinc finger region of YY1, the zinc finger 2 is not essential to this interaction.

Figure 5
figure5

Zinc finger 2 of Yin Yang 1 (YY1) is essential to YY1–AR interaction. (a) Schematic diagram of domain structures of wild-type YY1 and YY1-Chi17. The four zinc fingers of YY1 (ZnF1 to ZnF4) are labeled on the top. GFI, growth factor independence-1. (b and c) Substitution of zinc finger 2 of YY1 disrupts YY1–AR interaction. YY1-Chi17 mutant has the second zinc finger of YY1 replaced by a zinc finger of GFI-1 (Galvin and Shi, 1997). In (b), equal amount (3 μg) of glutathione S-transferase (GST), GST-AR-N (that is, AR 1-555) and GST-AR-C (that is, AR 556-919) were individually incubated with 1.5 μg of His × 6-YY1 or His × 6-YY1-Chi17 for 4 h at 4 °C. Samples brought down by glutathione agarose were analysed by western blot using YY1 (H-10) antibody. Lower panel: Ponceau S staining of transferred membrane to demonstrate input of GST fusion proteins. In (c), GST, GST-YY1, GST-YY1-Chi17 and GST-PIASy (3 μg each) were individually incubated with 1.5 μg of His × 6-AR(556–919) under the same conditions described above. Samples pulled down by glutathione agarose were applied to western blot to detect His × 6-AR(556–919) using androgen receptor (AR) antibody (C-20). Lower panel: protein inputs. (d and e) YY1-Chi17 retains its interaction with Hdm2 and PIASy. GST, GST-YY1 and GST-YY1-Chi17 (3 μg each) were individually incubated with 1.5 μg of His × 6-Hdm2 or His × 6-PIASy, respectively. Samples brought down by glutathione agarose were analysed by western blot using Hdm2 antibody (Smp14) or PIASy antibody (I-19). Lower panels: protein inputs stained by Ponceau S.

To determine whether YY1–AR interaction is necessary to YY1-mediated transcription activity of AR, we individually transfected pcDNA3/HA-YY1 and pcDNA3/HA-YY1-Chi17 with both PSA-Fluc and pcDNA3/Flag-AR in 293T cells and tested firefly luciferase (Fluc) activity in cell lysates. Although HA-YY1 could synergistically stimulate PSA promoter transcription (3 vs 1 and 2, Figure 6a), HA-YY1-Chi17 completely lost this function (3 vs 4). Comparable expression of HA-YY1 and HA-YY1-Chi17 was demonstrated by western blots with GAPDH as a control (right, Figure 6a).

Figure 6
figure6

YY1–AR interaction is essential to Yin Yang 1 (YY1)-enhanced androgen receptor (AR) transcriptional activity. (a) YY1-Chi17 does not activate AR-mediated transcription of the prostate-specific antigen (PSA) promoter. About 100 ng of pcDNA3/HA-YY1 or pcDNA3/HA-YY1-Chi17 were cotransfected with 300 ng of pcDNA3/Flag-AR, 300 ng of PSA-Fluc and 100 ng of actin-SEAP into 293T cells in 12-well plates cultured in medium with or without 10 nM R1881, as described above. Firefly luciferase (Fluc) activity was normalized against secreted alkaline phosphatase (SEAP) activity (left panel); expression of HA-YY1, HA-YY1-Chi17 and GAPDH is shown on the right panel. (b) Schematic diagram of reporter constructs of PSA-Gluc and m-PSA-Gluc with mutated putative YY1-binding site. Mutation of the putative YY1-binding site is indicated. (c) YY1 could stimulate the expression of the PSA promoter with mutated putative YY1-binding site. LNCaP cells were cotransfected by 300 ng of siRNA plasmid carrying the scrambled or yy1 siRNA, 80 ng of PSA-Gluc or m-PSA-Gluc, and 100 ng of actin-SEAP, in triplicates. Transfected cells were cultured in medium without or with 10 nM R1881; normalized Gaussia luciferase (Gluc) activity in each condition is presented. (d) Schematic diagram of the 3 × ARE-Fluc reporter construct. See ‘Materials and methods’ for detail. (e) Effects of YY1 depletion on 3 × ARE promoter in LNCaP cells. LNCaP cells were cotransfected with 3 × ARE-Fluc reporter (500 ng), actin-SEAP (100 ng) and plasmid (300 ng) expressing scrambled siRNA or yy1 siRNA, as labeled.

As YY1 is a transcription factor, we asked whether YY1 could potentially bind to the PSA promoter and whether YY1–DNA interaction could affect AR-mediated PSA transcription. Using an algorithm to search for binding elements of transcriptional factors (Heinemeyer et al., 1998), we identified a potential YY1-binding site (A−2040AGATGGTC−2032) in the PSA promoter. The nucleotides are numbered relative to the transcription start site of PSA mRNA and the sequence ‘ATGG’ is reversely complementary to the core sequence (‘CCAT’) of YY1-binding element (Shi et al., 1997). To study whether this potential YY1-binding site determined YY1-mediated PSA transcription, we disrupted this site by mutating ‘A−2040AGATGGTC−2032’ to ‘A−2040AacTaGTC−2032’, which deleted the YY1-binding elements and introduced a SpeI site (ACTAGT) used in DNA subcloning. Hence, we generated a reporter construct with an YY1-binding site-mutated PSA promoter driving Gluc cDNA (designated as ‘m-PSA-Gluc’; Figure 6b).

To determine whether YY1–DNA interaction is required for PSA promoter expression, we individually transfected PSA-Gluc and m-PSA-Gluc reporter plasmids into LNCaP cells with or without endogenous YY1 knockdown. As shown in Figure 6c, wt and m-PSA promoters showed no significant differences in driving Gluc expression, either with or without R1881. Similar results were obtained in 293T cells (data not shown). We also tested the effect of YY1 on an artificial promoter, 3 × ARE, which contains three concatemeric repeats of the PSA promoter ARE-I and lacks an YY1-binding element (Figure 6d). In this reporter assay using 3 × ARE-Fluc, YY1 knockdown decreased the AR-mediated transcription by 35% (P<0.01) in the presence of R1881 (4 vs 3, Figure 6e), similar to the YY1's effects on AR-mediated transcription of the endogenous PSA promoter (compare Figure 6e with Figure 3c).

YY1 is required for the AR-ARE complex

We used electrophoretic mobility shift assay (EMSA) to determine whether YY1 regulates AR-ARE association. We first analysed the 32P-labeled synthetic probe containing ARE-I of the PSA promoter without or with the incubation of LNCaP cell nuclear extract and detected a slowly migrated band when the nuclear proteins was added (lines 1 vs 2, Figure 7a). This band likely corresponds to an AR-ARE-containing complex, because its intensity increased when the cells were cultured in R1881-containing medium (lanes 3 vs 2), which enhances AR-ARE association and/or promotes AR expression, and decreased when the unlabeled probe was added to compete with the labeled probe (lanes 4 vs 3). To further validate the identity of this band, we individually incubated the 32P-labeled ARE-I-containing probe and a control probe consisting of a scrambled sequence with the LNCaP cell nuclear extracts. As shown in Figure 7a (lanes 5 vs 6), the ARE-I probe exhibited a specific band with the same migration as lane 2, but the control showed mostly nonspecific binding. To confirm the presence of AR in this band, we incubated the labeled ARE-I-containing probe with the nuclear protein extracts of LNCaP cells infected by lentiviruses expressing the scrambled siRNA or the AR siRNA that efficiently knocked down endogenous AR (data not shown). The depletion of AR markedly reduced the intensity of the detected band (lanes 7 vs 8, Figure 7a). Therefore, these gel-shift experiments unequivocally indicate that the slowly migrating band, as pointed at the left of Figure 7a, is the AR-ARE-I-containing complex.

Figure 7
figure7

Yin Yang 1 (YY1) is essential to the formation of androgen response element (ARE) complex. (a) Validation of the AR-ARE complex by electrophoretic mobility shift assay (EMSA). Lane 1: 32P-labeled ARE-I containing probe incubated with bovine serum albumin (BSA). Lanes 2–4: labeled probe was incubated with nuclear protein extracts from LNCaP cells cultured without or with R1881, as indicated. Sample in lane 4 was coincubated with unlabeled probe. Lanes 5 and 6: nuclear protein extracts from R1881-treated LNCaP cells were incubated with 32P-labeled DNA probes containing either the ARE-I sequence or a scrambled sequence, as indicated on top. Lanes 7 and 8: the extracts from R1881-treated LNCaP cells infected by lentivirus expressing either a scrambled siRNA or the androgen receptor (AR) siRNA (indicated on top) were incubated with the labeled ARE-I-containing probe. Samples were applied to a 5% non-denaturing polyacrylamide gel for EMSA. (b) YY1 antibodies attenuate the formation of AR-ARE complex. Nuclear protein extracts in lanes 1–4 were pretreated with different antibodies (indicated on top) before the incubation with labeled probe. Lane 5: labeled probe incubated with BSA. Cont, control; r, rabbit; m, mouse; YY1 (r), YY1 antibody 13G10; YY1 (m), YY1 antibody H-10; no Ab, no antibody added.

To determine the presence of YY1 in this AR-ARE-I-containing complex, we pretreated the LNCaP nuclear extracts with YY1 antibodies (13G10 generated from rabbit and H-10 from mouse) before the analysis of EMSA. Both YY1 antibodies markedly decreased the intensity of the AR-ARE-I-containing complex, compared to the samples treated with the control antibodies (lanes 2 vs 1 and 4 vs 3, Figure 7b). These results indicate that YY1 is an essential component of the AR-ARE-I complex, because the association of YY1 and YY1 antibody disrupted the complex formation.

Discussion

YY1 regulates different epigenetic processes by mediating gene expression and protein modifications. The multifunctional properties of YY1 may explain its elevated expression in different cancers, including PCa. This study provided unequivocal evidence of the functional YY1–AR interaction and unveiled another regulatory activity of YY1: acting on AR-mediated transcription through direct protein–protein interaction. Our data suggest that YY1 has an important role in prostate cell differentiation and PCa development.

We and others demonstrated that YY1 enhances Mdm2-mediated p53 ubiquitination/degradation, inhibits p300-mediated p53 acetylation and antagonizes p53's transcriptional activity (Gronroos et al., 2004; Sui et al., 2004). In these studies, we revealed that YY1 promotes AR-mediated transcription without significantly altering its expression and stability. Therefore, YY1 exhibits differential regulatory functions to different proteins. As YY1 interacts with various transcriptional cofactors, it is possible that the binding of YY1 provides an additional interface to AR-mediated gene expression and determines the components in the AR-containing complex.

Ample evidence indicates that YY1 regulates gene expression through recruiting various chromatin-remodeling proteins to target promoters. Depending on the recruited cofactors, YY1 can either activate or inhibit gene expression. In this study, we observed that a medium increase of YY1 in 293T cells could markedly enhance AR-mediated transcription of the PSA promoter. However, further increases in transfected YY1 did not add to, and actually slightly decreased, AR-mediated transcription (Figure 3a). Similarly, in LNCaP cells, although YY1 knockdown significantly reduced PSA promoter transcription in the presence of R1881, ectopically expressed YY1 did not cause any increased transcription (Figure 3b). We believe that the already high levels of YY1 in PCa cells, such as LNCaP, contribute to the elevated activity of AR. Conversely, instead of promoting AR's function, the further increase in YY1 inversely affects the activity of AR due to the ‘squelching effect’ resulted from an excessive titration of a transcription factor, as described previously by Gill and Ptashne (1988). Consequently, this leads to compromised performance of AR, as shown in the cotransfection experiments in 293T cells (Figure 3a). Based on this observation, we proposed that the regulation of AR by YY1 follows a model schematically depicted in Figure 8. In a condition with an optimal increase of YY1, it is recruited by AR and activates gene expression. In PCa cells that exhibit elevated YY1 expression (Seligson et al., 2005), we propose that this recruitment leads to stimulated expression of AR-targeted genes and consequently promotes proliferation of PCa cells. When YY1 is either depleted by siRNA or robustly overexpressed, the expression of AR-targeted genes will be significantly reduced by the absence of YY1 or attenuated due to the squelching effect caused by excessive YY1, respectively.

Figure 8
figure8

Schematic model for the effects of altered Yin Yang 1 (YY1) expression on androgen receptor (AR)-mediated gene expression. (a) At the condition of YY1 increase, such as in prostate cancer, YY1 binds to AR and recruits cofactors to facilitate the expression of AR-targeted genes. (b) When AR was depleted by siRNA, the cofactors will not be recruited and therefore the AR-mediated gene transcription is decreased. (c) When YY1 is robustly overexpressed, excessive YY1 will individually interact with AR and the cofactors, which interferes with the formation of the optimized transcriptional complex. This so-called ‘squelching effect’ (Gill and Ptashne, 1988) will inversely affect the expression of AR-targeted genes. The sizes of the arrows on AR-targeted promoter represent the strength of transcription. C1 and C2: cofactors recruited by YY1.

The manner of YY1-regulated transcriptional activity of AR is reminiscent to a reported correlation between YY1 expression and outcomes of PCa patients (Seligson et al., 2005). In that study, YY1 expression was predominantly elevated in the prostatic intraepithelial neoplasia stage and in tumors from intermediate to high morphologic grade. However, increased YY1 expression in PCa tissues was inversely associated with a higher risk to develop recurrent disease. Therefore, although YY1 could be used as a diagnostic and prognostic marker, its functional role may be either proliferative or antiproliferative, depending on the stages of PCa development and progression.

Androgen receptors are critical for the proliferation and survival of PCa cells. Current therapeutic approaches in PCa to inhibit AR function include preventing ligand synthesis and using androgen antagonists that bind to AR (Hirawat et al., 2003). Although these treatments are initially effective, they will eventually fail due to restored AR activity despite the presence of therapeutics. Therefore, understanding the molecular mechanism of AR regulation may provide fundamental support to the therapeutic treatment of PCa. In this study, we demonstrated that YY1 physically interacts with the C terminal of AR and is essential to the AR-mediated transcription of the PSA promoter. These results imply that YY1 may be an alternative therapeutic target to inhibit AR's activity and PCa development. In addition, the regulation of YY1 to AR's transcriptional activity depends on the YY1–AR interaction but not on the YY1–DNA association, suggesting that YY1 may either mediate the posttranslational modification of AR or act as a cofactor, recruiting other proteins to facilitate AR-mediated transcription. Further investigation is required to delineate whether altered YY1 expression will affect the posttranslational modification of AR and/or the association of AR with other critical transcription cofactors.

We have demonstrated that the YY1–AR interaction is important to AR-mediated transcription of the PSA promoter. Whether the direct binding of YY1 to the PSA promoter affects PSA expression is still unclear. Our results indicate that the transcription of the PSA promoter and the artificial 3 × ARE-I promoter without any YY1-binding site exhibited similar effects to altered YY1 expression (Figures 3c and 6d). Thus, it is unlikely that YY1–DNA association is critical to PSA promoter transcription. However, as increased YY1 expression inversely affected AR-mediated transcription, it is also possible that excessive YY1 in the nucleus may enhance the YY1-DNA association, in turn reducing transcription of the PSA promoter.

Both AR and YY1 are proteins with multiple functions and their regulatory roles in PCa have been indicated by many previous reports. This study revealed the functional interplay between these two proteins in PCa cells. We demonstrated that, as a novel AR-interacting protein, YY1 is an essential component of the ARE complex and crucial in maintaining the AR-mediated transcription. Future study is needed to delineate the mechanism underlying YY1-regulated function of AR.

Materials and methods

Antibodies and DNA vectors

All antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA) unless otherwise indicated. Plasmids expressing AR proteins, including pcDNA3/Flag-AR, GST-AR(1–556), GST-AR(557–919) and pHis × 6-AR(557–919), were generated in our laboratory. Plasmids expressing HA-YY1, GST-YY1 wt and mutants, and YY1 chimera 17 (YY1-Chi17, with the second zinc finger of YY1 replaced by a zinc finger from growth factor independence-1) were described previously (Galvin and Shi, 1997; Sui et al., 2004). The reporter plasmid, PSA-Fluc, with the 5.8 kb PSA promoter driving Fluc was kindly provided by Dr Weber (Gioeli et al., 2002). The PSA promoter with its putative YY1-binding site mutated from ‘AAGATGGTC’ to ‘AAACTAGTC’ was generated by PCR and confirmed by DNA sequencing analysis.

Gluc (Tannous et al., 2005) is a secreted protein and its activity can be measured in collected culture medium. To generate PSA-Gluc, we subcloned either the PSA promoter or its mutated form into the HindIII site of pGluc-Basic (New England Biolabs, Beverly, MA, USA). The lentiviral vector, pLu-Puro, was used to deliver U6/siRNAs expression cassettes. U6/scrambled and U6/yy1 siRNAs were described previously (Sui et al., 2004). The construction of U6/AR siRNA followed our previously published protocol (Sui et al., 2002; Sui and Shi, 2005), with a target sequence as ‘IndexTermGAGGCACCTCTCTCAAGAGTT’.

Cell culture and transfection

293T cells were maintained in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum (FBS). LNCaP cells were cultured in RPMI1640 medium with 10% FBS and transfected at 80% confluence in 12-well plates with Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA). After 6 h of transfection, the medium was replaced with phenol-red-free RPMI 1640 containing 10% charcoal stripped serum (CSS). If needed, R1881 was added in the medium to a final concentration of 10 nM at 24 h posttransfection. Cell lysates or medium were collected at 48 h posttransfection for further analyses.

Lentiviral production and infection

Lentivirus production followed a previous report (Rubinson et al., 2003) by transfecting 293T cells with a lentiviral plasmid and three packaging plasmids (pMDLg/pRRE, pRSV-REV and pVSV-G). To infect cells, concentrated lentivirus was added to the medium with 8 μg/ml polybrene and incubated for 6 h before changing back to normal medium.

Reporter assay

We used three reporter constructs: PSA-Gluc, PSA-Fluc and 3 × ARE-Fluc. The detailed construction procedure of the 3 × ARE-Fluc containing three concatemeric repeats of the ARE-I (Rao et al., 2003) from the PSA promoter and one PSA proximal promoter is described in the supplementary material. The actin-SEAP plasmid with β-actin promoter-driven secreted alkaline phosphatase (SEAP) was used as a transfection control and each condition was tested in triplicate and repeated over three times. For cells transfected with PSA-Gluc, 50 μl of medium, collected at 48 h posttransfection, was mixed with 100 μl of substrate solution containing 0.5 μg/ml of coelenterazine (CTZ), 200 mM NaCl, 50 mM Tris-HCl and 0.01% Triton X-100, pH 8.7. Light emission was measured at 480 nm and normalized to the SEAP expression. The assays with Fluc activity were performed as described before (Deng et al., 2007).

Protein interaction studies

The in vitro protein-binding studies and immunoprecipitation experiments for transfected or endogenous proteins have been described previously (Deng et al., 2007).

Electrophoretic mobility shift assay

The EMSA followed a previously published protocol (Cleutjens et al., 1997). We incubated nuclear protein extract with the ARE-I probe labeled by 32P-dATP (Perkin Elmer, Boston, MA, USA) and used a control double-stranded probe generated by annealing the following two oligonucleotides: 5′-IndexTermTATCAGAAGGACGTATACGTATTACAACT-3′ and 5′-IndexTermGATCAGTTGTAATACGTATACGTCCTTCT-3′. To determine the role of YY1 in the AR complex, nuclear extract was individually pretreated by YY1 antibodies (13G10; Cell Signaling, Danvers, MA, USA; and H-10) and normal antibodies before adding the probe.

Quantitative PCR analysis

This was conducted using Taqman Gene Experssion Assay kit (Applied Biosystems, Foster City, CA, USA). Briefly, RNA extraction and reverse transcription were carried out according to the manufacturer's instruction. mRNA levels of PSA and GAPDH were determined by unlabeled PCR primers and the FAM dye-labeled TaqMan MGB probes (Hs03063374 for PSA and Hs99999905 for GAPDH) using the ABI7000 Real-time PCR system. The samples were analysed in triplicates and repeated in three individual experiments. Comparative Ct method was used to calculate relative levels of PSA mRNA normalized to GAPDH.

Statistical analysis

All data in reporter assay and quantitative PCR (qPCR) are presented as mean±s.d. Comparisons between two groups on a single parameter were conducted using Student's t-test. Statistical analyses were performed using SigmaStat (Systat Software Inc., Richmond, CA, USA). The criterion for statistical significance was set at P<0.05.

Conflict of interest

The authors declare no conflict of interest.

References

  1. Bain M, Sinclair J . (2005). Targeted inhibition of the transcription factor YY1 in an embryonal carcinoma cell line results in retarded cell growth, elevated levels of p53 but no increase in apoptotic cell death. Eur J Cell Biol 84: 543–553.

  2. Baritaki S, Sifakis S, Huerta-Yepez S, Neonakis IK, Soufla G, Bonavida B et al. (2007). Overexpression of VEGF and TGF-beta1 mRNA in Pap smears correlates with progression of cervical intraepithelial neoplasia to cancer: implication of YY1 in cervical tumorigenesis and HPV infection. Int J Oncol 31: 69–79.

  3. Begon DY, Delacroix L, Vernimmen D, Jackers P, Winkler R . (2005). Yin Yang 1 cooperates with activator protein 2 to stimulate ERBB2 gene expression in mammary cancer cells. J Biol Chem 280: 24428–24434.

  4. Bracken AP, Pasini D, Capra M, Prosperini E, Colli E, Helin K . (2003). EZH2 is downstream of the pRB-E2F pathway, essential for proliferation and amplified in cancer. EMBO J 22: 5323–5335.

  5. Caretti G, Di Padova M, Micales B, Lyons GE, Sartorelli V . (2004). The Polycomb Ezh2 methyltransferase regulates muscle gene expression and skeletal muscle differentiation. Genes Dev 18: 2627–2638.

  6. Cleutjens KB, van der Korput HA, van Eekelen CC, van Rooij HC, Faber PW, Trapman J . (1997). An androgen response element in a far upstream enhancer region is essential for high, androgen-regulated activity of the prostate-specific antigen promoter. Mol Endocrinol 11: 148–161.

  7. Craft N, Chhor C, Tran C, Belldegrun A, DeKernion J, Witte ON et al. (1999). Evidence for clonal outgrowth of androgen-independent prostate cancer cells from androgen-dependent tumors through a two-step process. Cancer Res 59: 5030–5036.

  8. de Nigris F, Botti C, de Chiara A, Rossiello R, Apice G, Fazioli F et al. (2006). Expression of transcription factor Yin Yang 1 in human osteosarcomas. Eur J Cancer 42: 2420–2424.

  9. Dehm SM, Tindall DJ . (2005). Regulation of androgen receptor signaling in prostate cancer. Expert Rev Anticancer Ther 5: 63–74.

  10. Deng Z, Wan M, Sui G . (2007). PIASy-mediated sumoylation of Yin Yang 1 depends on their interaction but not the RING finger. Mol Cell Biol 27: 3780–3792.

  11. Erkeland SJ, Valkhof M, Heijmans-Antonissen C, Delwel R, Valk PJ, Hermans MH et al. (2003). The gene encoding the transcriptional regulator Yin Yang 1 (YY1) is a myeloid transforming gene interfering with neutrophilic differentiation. Blood 101: 1111–1117.

  12. Gaddipati JP, McLeod DG, Heidenberg HB, Sesterhenn IA, Finger MJ, Moul JW et al. (1994). Frequent detection of codon 877 mutation in the androgen receptor gene in advanced prostate cancers. Cancer Res 54: 2861–2864.

  13. Galvin KM, Shi Y . (1997). Multiple mechanisms of transcriptional repression by YY1. Mol Cell Biol 17: 3723–3732.

  14. Gill G, Ptashne M . (1988). Negative effect of the transcriptional activator GAL4. Nature 334: 721–724.

  15. Gioeli D, Ficarro SB, Kwiek JJ, Aaronson D, Hancock M, Catling AD et al. (2002). Androgen receptor phosphorylation. Regulation and identification of the phosphorylation sites. J Biol Chem 277: 29304–29314.

  16. Gregory CW, Johnson Jr RT, Mohler JL, French FS, Wilson EM . (2001). Androgen receptor stabilization in recurrent prostate cancer is associated with hypersensitivity to low androgen. Cancer Res 61: 2892–2898.

  17. Gronroos E, Terentiev AA, Punga T, Ericsson J . (2004). YY1 inhibits the activation of the p53 tumor suppressor in response to genotoxic stress. Proc Natl Acad Sci USA 101: 12165–12170.

  18. Gross M, Yang R, Top I, Gasper C, Shuai K . (2004). PIASy-mediated repression of the androgen receptor is independent of sumoylation. Oncogene 23: 3059–3066.

  19. Heinemeyer T, Wingender E, Reuter I, Hermjakob H, Kel AE, Kel OV et al. (1998). Databases on transcriptional regulation: TRANSFAC, TRRD and COMPEL. Nucleic Acids Res 26: 362–367.

  20. Heinlein CA, Chang C . (2004). Androgen receptor in prostate cancer. Endocr Rev 25: 276–308.

  21. Hirawat S, Budman DR, Kreis W . (2003). The androgen receptor: structure, mutations, and antiandrogens. Cancer Invest 21: 400–417.

  22. Joshi B, Rastogi S, Morris M, Carastro LM, DeCook C, Seto E et al. (2007). Differential regulation of human YY1 and caspase 7 promoters by prohibitin through E2F1 and p53 binding sites. Biochem J 401: 155–166.

  23. Koivisto P, Kononen J, Palmberg C, Tammela T, Hyytinen E, Isola J et al. (1997). Androgen receptor gene amplification: a possible molecular mechanism for androgen deprivation therapy failure in prostate cancer. Cancer Res 57: 314–319.

  24. Lapouge G, Erdmann E, Marcias G, Jagla M, Monge A, Kessler P et al. (2007). Unexpected paracrine action of prostate cancer cells harboring a new class of androgen receptor mutation--a new paradigm for cooperation among prostate tumor cells. Int J Cancer 121: 1238–1244.

  25. Rao A, Chang BL, Hawkins G, Hu JJ, Rosser CJ, Hall MC et al. (2003). Analysis of G/A polymorphism in the androgen response element I of the PSA gene and its interactions with the androgen receptor polymorphisms. Urology 61: 864–869.

  26. Rubinson DA, Dillon CP, Kwiatkowski AV, Sievers C, Yang L, Kopinja J et al. (2003). A lentivirus-based system to functionally silence genes in primary mammalian cells, stem cells and transgenic mice by RNA interference. Nat Genet 33: 401–406.

  27. Rylski M, Amborska R, Zybura K, Mioduszewska B, Michaluk P, Jaworski J et al. (2008). Yin Yang 1 is a critical repressor of matrix metalloproteinase-9 expression in brain neurons. J Biol Chem 283: 35140–35153.

  28. Santiago FS, Ishii H, Shafi S, Khurana R, Kanellakis P, Bhindi R et al. (2007). Yin Yang-1 inhibits vascular smooth muscle cell growth and intimal thickening by repressing p21WAF1/Cip1 transcription and p21WAF1/Cip1-Cdk4-cyclin D1 assembly. Circ Res 101: 146–155.

  29. Seligson D, Horvath S, Huerta-Yepez S, Hanna S, Garban H, Roberts A et al. (2005). Expression of transcription factor Yin Yang 1 in prostate cancer. Int J Oncol 27: 131–141.

  30. Sharifi N, Farrar WL . (2006). Androgen receptor as a therapeutic target for androgen independent prostate cancer. Am J Ther 13: 166–170.

  31. Shenk JL, Fisher CJ, Chen SY, Zhou XF, Tillman K, Shemshedini L . (2001). p53 represses androgen-induced transactivation of prostate-specific antigen by disrupting hAR amino- to carboxyl-terminal interaction. J Biol Chem 276: 38472–38479.

  32. Shi Y, Lee JS, Galvin KM . (1997). Everything you have ever wanted to know about Yin Yang 1. Biochim Biophys Acta 1332: F49–F66.

  33. Sui G, Affar el B, Shi Y, Brignone C, Wall NR, Yin P et al. (2004). Yin Yang 1 is a negative regulator of p53. Cell 117: 859–872.

  34. Sui G, Shi Y . (2005). Gene silencing by a DNA vector-based RNAi technology. Methods Mol Biol 309: 205–218.

  35. Sui G, Soohoo C, Affar el B, Gay F, Shi Y, Forrester WC et al. (2002). A DNA vector-based RNAi technology to suppress gene expression in mammalian cells. Proc Natl Acad Sci USA 99: 5515–5520.

  36. Tannous BA, Kim DE, Fernandez JL, Weissleder R, Breakefield XO . (2005). Codon-optimized Gaussia luciferase cDNA for mammalian gene expression in culture and in vivo. Mol Ther 11: 435–443.

  37. Taplin ME, Bubley GJ, Ko YJ, Small EJ, Upton M, Rajeshkumar B et al. (1999). Selection for androgen receptor mutations in prostate cancers treated with androgen antagonist. Cancer Res 59: 2511–2515.

  38. Thomas MJ, Seto E . (1999). Unlocking the mechanisms of transcription factor YY1: are chromatin modifying enzymes the key? Gene 236: 197–208.

  39. Veldscholte J, Berrevoets CA, Ris-Stalpers C, Kuiper GG, Jenster G, Trapman J et al. (1992). The androgen receptor in LNCaP cells contains a mutation in the ligand binding domain which affects steroid binding characteristics and response to antiandrogens. J Steroid Biochem Mol Biol 41: 665–669.

  40. Wang H, Hertlein E, Bakkar N, Sun H, Acharyya S, Wang J et al. (2007). NF-kappaB regulation of YY1 inhibits skeletal myogenesis through transcriptional silencing of myofibrillar genes. Mol Cell Biol 27: 4374–4387.

  41. Wilkinson FH, Park K, Atchison ML . (2006). Polycomb recruitment to DNA in vivo by the YY1 REPO domain. Proc Natl Acad Sci USA 103: 19296–19301.

  42. Yang WM, Inouye C, Zeng Y, Bearss D, Seto E . (1996). Transcriptional repression by YY1 is mediated by interaction with a mammalian homolog of the yeast global regulator RPD3. Proc Natl Acad Sci USA 93: 12845–12850.

  43. Yao YL, Yang WM, Seto E . (2001). Regulation of transcription factor YY1 by acetylation and deacetylation. Mol Cell Biol 21: 5979–5991.

Download references

Acknowledgements

We thank Dr Purnima Dubey and Ms Karen Klein for critical reading of the article. We also thank Dr Suzy Torti and Dr Wei Wang for providing the access to some equipment, and Dr Kazushi Inoue for the EMSA protocol. The support for the work came from the grant of American Cancer Society (RSG-09-082-01-MGO) and the startup fund from the Department of Cancer Biology and Comprehensive Cancer Center of Wake Forest University to GS. PC is supported by NCI training Grant 5T32CA079448-09.

Author information

Correspondence to G Sui.

Additional information

Supplementary Information accompanies the paper on the Oncogene website (http://www.nature.com/onc)

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Keywords

  • androgen receptor
  • Yin Yang 1
  • prostate cancer
  • PSA promoter
  • transcription

Further reading