PES1 differentially regulates the expression of ERα and ERβ in ovarian cancer

Authors

  • Jieping Li,

    Corresponding author
    1. Department of Clinic Medical Laboratory, General Hospital of Fujian Corps of CAPF, Fuzhou, China
    • Address correspondence to: Jieping Li, Department of Clinic Medical Lab, General Hospital of Fujian Corps of CAPF, Fuzhou 350003, China. Tel: +86 591 83128221, Fax: +86 59183129106. E-mail: liejieping@hotmail.com or Xiaopeng Lan, Institute of Clinic Lab Medicine, Fuzhou General Hospital of Nanjing Military Command, PLA, Fuzhou 350003, China. Tel & Fax: +86 591 83732529. E-mail: lanxp@sina.com

    Search for more papers by this author
  • Qinren Zhuang,

    1. Department of Clinic Medical Laboratory, General Hospital of Fujian Corps of CAPF, Fuzhou, China
    Search for more papers by this author
  • Xiaopeng Lan,

    Corresponding author
    1. Institute of Clinic Laboratory Medicine, Fuzhou General Hospital of Nanjing Military Command, PLA, Fuzhou, China
    • Address correspondence to: Jieping Li, Department of Clinic Medical Lab, General Hospital of Fujian Corps of CAPF, Fuzhou 350003, China. Tel: +86 591 83128221, Fax: +86 59183129106. E-mail: liejieping@hotmail.com or Xiaopeng Lan, Institute of Clinic Lab Medicine, Fuzhou General Hospital of Nanjing Military Command, PLA, Fuzhou 350003, China. Tel & Fax: +86 591 83732529. E-mail: lanxp@sina.com

    Search for more papers by this author
  • Guobin Zeng,

    1. Department of Clinic Medical Laboratory, General Hospital of Fujian Corps of CAPF, Fuzhou, China
    Search for more papers by this author
  • Xuping Jiang,

    1. Department of gynaecology and obstetrics, General Hospital of Fujian Corps of CAPF, Fuzhou, China
    Search for more papers by this author
  • Zongming Huang

    1. Department of Pathology, General Hospital of Fujian Corps of CAPF, Fuzhou, China
    Search for more papers by this author

Abstract

Estrogen exhibits mitogenic activity in early ovarian carcinogenesis and plays an important role in ovarian tumorigenesis. Due to the increased expression of ERα and decreased expression of the ERβ, the ratio of ERα and ERβ is markedly increased in ovarian cancer. We have recently reported that PES1 regulates the balance of ERα and ERβ at the post-transcriptional level in breast cancer. Here, we report that PES1 inversely regulates the expression of ERα and ERβ in addition to their transcriptional activities in epithelial ovarian cancer. We found that the ablation of PES1 resulted in the significant downregulation of ERα and estrogen-responsive genes such as cylin D1, HIF-1α and VEGF and the up-regulation of ERβ and p21WAF1. Cell proliferation in both tested ovarian cell lines was markedly inhibited and cells were arrested in G2 after PES1 was ablated. Further analysis of clinical samples showed that expression of PES1 correlated positively with ERα expression and negatively with ERβ expression. Our results demonstrate that PES1 may play important role in the progression of ovarian cancer by inversely regulating the ERα and ERβ expression. PES1 may be a new target for ovarian cancer therapy. © 2013 IUBMB Life, 65(12):1017–1025, 2013.

Introduction

Ovarian cancer is one of most common gynecological malignancies, with the highest mortality rate. Approximately 90% of ovarian cancers arise from ovarian surface epithelial cells (OSE) [1]. Incessant ovulatory cycles have been proposed as the most common hypothesis for ovarian carcinogenesis, but the precise pathogenesis and etiological factors involved in ovarian epithelial carcinogenesis have not been defined. The ovary is the main source of estrogen and one of the major target organs, where this hormone plays an important role in the growth and function of reproductive tissues [2]. Although the results have not always been in concordance, major studies have suggested overall that estrogen plays an important role in promoting the progression of ovarian carcinoma [1, 3-6].

The functions of estrogen are mediated by the estrogen receptor (ER). Estrogen combines with ER to activate genes that are responsible for cell proliferation, differentiation and angiogenesis, including c-myc, Bcl-2, cathepsin D and, HIF-1α amongst others [3]. The relative expression levels of these genes are important for the progression of carcinoma. There are two isoforms ER (ERα and ERβ), which are both members of the steroid hormone superfamily of nuclear receptors that act as ligand-activated transcription factors [7-9]. ERα and ERβ share a similar structure that is characterized by several functional domains. The transcriptional activity of both receptors depends on two distinct activation function (AF) domains. The N-terminus and C-terminus of both ERs contain the AF-1 and AF-2 domain, respectively. AF-1 is a ligand-independent transactivation domain, whereas AF-2, which overlaps the ligand-binding domain is ligand-dependent. The ligand binding domain and the N terminus containing the AF-1 domain of both ERs shows 58% and 28% homology between ERα and ERβ. These differences suggest that ERα and ERβ may have different functions. It has been shown that the expression of ERβ significantly decreased and the ratio of ERα to ERβ is markedly increased in ovarian cancer compared with normal OSE [1, 3, 5]. ERα and ERβ have antagonistic effect to each other. ERα promotes the proliferation of ovarian epithelial cells and tumor growth, whereas ERβ may play a protective role against ERα-mediated cell growth. Therefore, the imbalance of ERα and ERβ expression may be related to ovarian tumorigenesis. To date, the mechanism that underlies the imbalance of the ERα and ERβ expression in epithelial ovarian cancer (EOC) remains unclear.

Pescadillo was originally identified as a gene that is essential for the normal development of the zebrafish [10]. Pescadillo is well conserved in terms of both sequence and function between species, with homologues identified in the human (PES1), yeast (YPH1, Nop7p), and mouse (Pes1) genomes [10-12]. Pescadillo contains a BRCA1 C-terminal (BRCT)-domain and a conserved site for SUMOylation [11, 12]. The BR-CT domain has been found in several proteins that are involved in DNA repair, cell cycle control, and/or recombination [13-15]. Previous studies have shown that PES1 is essential for ribosome biogenesis and nucleologenesis, and regulates cell size, which are all important components that determine the cell proliferation rate [16, 17]. PES1 has also been shown to be overexpressed in stomach cancer [18], prostatic cancer [19, 20], breast cancer [21-23], head and neck squamous cell cancer [24], colon cancer [25], malignant astrocytomas and glioblastomas [11]. The increased expression of Pes1 transforms both mouse and human fibroblasts [26], while the repression of PES1 inhibits proliferation and tumorigenicity of breast cancer cells [22]. These results suggest that PES1 is related to proliferation, transformation, and malignant conversion of cells and may contribute to tumor progression.

We have recently reported the novel function of PES1 as a new ER coregulator that regulates the balance of ERα/ERβ expression levels through the ubiquitin-proteasome pathway and hence contributes to breast cancer growth [23]. As some ERs coregulators have been found to have tissue specificity that modulates their transcriptional activities [27-29], in this study we investigated the effect of PES1 on the transcription of ERα and ERβ and their relative expression levels in ovarian cancer cells in vitro and in vivo. We provided further evidence that PES1 increased the protein level of ERα and decreased that of ERβ, inversely regulating their transcriptional activities. Meanwhile, the ablation of PES1 differentially regulated the expression of ER target genes. The ablation of PES1 expression significantly inhibited ovarian cancer growth with cell cycle arrest in the G2 phase. We further demonstrated that PES1 expression levels were positively correlated with ERα expression and negatively correlated ERβ expression in ovarian cancer tissues. Our results suggest that PES1 regulates the balance of ERα/ERβ expression and may play a crucial role in ovarian tumorigenesis.

Materials and Methods

Reagents and Plasmids

Antibodies against ERα and ERβ were from ABCOM (Cambridge, MA). The antibody against PES1 was purchased from Bethyl Laboratories (Montgomery, TX). Antibodies against cyclin D1, HIF-1α, p21 and VEGF were from Bioworld Biotechnology (Louis Park, MN); 17β-estradiol (E2), propylpyrazole triol (PPT), diarylpropionitrile (DPN), and antibodies against FLAG and GAPDH were purchased from Sigma-Aldrich (Louis, MO). The following plasmids have been described previously: pERE-LUC (estrogen-responsive element-containing luciferase reporter) and FLAG-tagged PES1 expression vectors [23], and both were a kind gift from Dr. Qinong Ye (Beijing Institute of Biotechnology, China). Pescadillo shRNA plasmid (h) (Santa Cruz Biotechnology; Santa Cruz, CA) was packaged into pLVX-shRNA2 (Clontech; Mountain View, CA).

Cell Culture and Transfection

The ovarian cancer cell lines, CAOV3 and ES2, the breast cancer cell line MCF7 and the embryonic kidney cell HEK293T were purchased from the Chinese Academy of Sciences Cell Bank of Type Culture Collection (Shanghai, China).CAOV3 cells were cultured in DMEM (Invitrogen) with glutamine and, 10% fetal bovine serum (FBS) in a humidified atmosphere with of 5% CO2 at 37 °C. ES2 cells were cultured in Mc'Coy 5a (Gibco) supplemented with 10% FBS. HEK293T cells and MCF7 cells were routinely cultured in DMEM (Invitrogen) containing 10% FBS (Hyclone). For the hormone treatment experiments, cells were cultured in medium containing phenol red–free DMEM or Mc'Coy-5a supplemented with 10% charcoal/dextran-treated FBS (Hyclone). Lipofectamine 2000 reagent was used for the transfections in accordance with the manufacturer's protocol (Invitrogen). Stable cell lines were selected in 1 μg/mL puromycin for approximately 2 months after the lentiviruses were collected and added to the culture medium. Pooled clones or individual clones were screened by standard immunoblotting protocols and produced similar results. PES1 knockdown stable cell lines grew very slowly and could only be passaged several times.

Inhibition of PES1 by RNA Interference

To knock down endogenous PES1 expression in a stable manner, lentiviruses were produced by the cotransfection of HEK293T cells with a Pescadillo shRNA plasmid and the pPACK Packaging Plasmid Mix (Clontech). Lentiviruses were collected 48 h after transfection and added to the medium of target cells with 8 μg/mL polybrene (Sigma-Aldrich).

Western Bolt Analysis

Cultured cells were washed with cold phosphatebuffered saline and lysed on ice in RIPA buffer supplemented with protease inhibitors. Following the removal of insoluble debris by centrifugation, protein concentration was determined in the Bradford procedure. Protein samples were separated by SDS–polyacrylamide gel electrophoresis and blotted to a nitrocellulose membrane. Blotted membranes were blocked overnight at 4 °C in TBST containing 5% non-fat milk. Blots were probed with primary antibodies diluted in TBST containing 5% non-fat milk for 1 h at room temperature. After washing extensively with TBST, membranes were incubated with the appropriate horseradish peroxidase-conjugated secondary antibody, followed by chemiluminescent detection according to the manufacturer's instructions (Pierce).

Luciferase Assay

The cells were transfected using Lipofectamine 2000 (Invitrogen) with 0.2 μg of ERE-LUC, 1.0 μg of the expression vector for PES1, and 0.1 μg of β-galactosidase reporter as an internal control. After treatment with 10 nM E2, 1 nM PPT, or 1 nM DPN for 24 h, the cells were harvested. Cell extracts were analyzed for luciferase and β-galactosidase activities as described previously [29, 30]. All experiments were repeated at least three times with similar results.

Anchorage-Dependent Growth Assay

Anchorage-dependent growth was analyzed as described [25]. Cells were plated on 24-well plates (2 × 102 cells per well) and cultured for 2 weeks. The colonies were stained with 0.5% crystal violet for 30 min after fixation with methanol for 30 min at room temperature.

Cell Growth Assay

Cell growth was analyzed by crystal violet assay as described previously [30]. Briefly, cells were fixed by 1% glutaraldehyde for 15 min and stained with 0.5% crystal violet for 15 min at room temperature. Plates were washed with distilled water several times and air-dried. The dye was eluted by Sorenson's solution for 30 min at room temperature with constant shaking. A microplate reader was used to read aliquots of eluant at 590 nm.

Cell Cycle Analysis

The cell cycle was analyzed by FACS [25]. Briefly, cells were harvested by centrifugation, washed twice with ice-cold PBS and fixed in 75% ethanol (in PBS) at 4 °C overnight. After they were washed twice with cold PBS, cells were resuspended in PBS containing 0.1 mg/mL RNase (Sigma). After 30 min at 37 °C, the cells were re-suspended in PBS containing 50 mg/mL propidium iodide (PI) and then analyzed using a flow cytometer (BD, CALIBUR). The cell cycle distribution was calculated using Cell Quest and Mod-fit software.

Immunohistochemistry

Ovarian cancer samples were provided by the Fuzhou General Hospital of Nanjing Military Command with the informed consent of patients. Institutional approval for the experiments was obtained from the ethics committeeof Fuzhou General Hospital of Nanjing Military Command and the General Hospital of Fujian Corps of CAPF. IHC staining was performed as described previously [23, 30] using rabbit anti-ERα, rabbit anti-ERβ, and rabbit anti-PES1 as primary antibodies.

Statistical Analysis

Differences between variables were assessed by χ2 analysis, Mann-whitney U test, or a two-tailed Student's t-test. Statistical calculations were performed using SPSS13.0. P values of less than 0.05 were considered statistically significant.

Results

PES1 Differentially Regulates the Transcriptional Activities of ERα and ERβ

Epithelial ovarian cancer cell CAOV3 and ES2 were co-transfected with the ERE-containing reporter ERE-LUC and PES1. E2 was used as the ER activator, and PPT or DPN were used as the selective activator for endogenous ERα or ERβ respectively. As shown in Fig. 1A, in the presence of E2 or PPT, PES1 enhanced the transcriptional activity of ERE reporter by approximately 1.8- or 3.3-fold, respectively, in CAOV3 cells. Surprisingly, PES1 almost completely attenuated the transcriptional activity of the ERE reporter activated with the DPN, whereas, in the absence of activators, PES1 had no effects on the transcriptional activities of the ERE reporter. These findings suggested that these effects of PES1 were ligand-dependent. In another ERα-positive ERβ-negative ovarian cancer cell, ES2 [31], DPN had no effect on transcriptional activity. However, a similar result for the ERα was obtained as in CAOV3 cells although to a different extent (Fig. 1B). These data suggested that the ectopic expression of PES1 increased the transcriptional activity of ERα and decreased the transcriptional activity of ERβ.

Figure 1.

PES1 differentially regulates the transcriptional activity of ERα and ERβ. (A and B) Luciferase reporter assays of ERα and ERβ transcriptional activity in CAOV3 (A) or ES2 (B) cells transiently transfected with ERE-LUC and FLAG-PES1 and treated for 24-hours with 10 nM E2, 1 nM PPT, or 1 nM DPN. Results shown are the mean ± SD of three independent experiments. @P < 0.01, *P < 0.01, &P < 0.01 vs. an empty vector with E2, PPT, and DPN, respectively. (C) Western blotting analysis of lysates from CAOV3 and ES2 cells that have undergone the lentiviral-mediated stable ablation of PES1. (D and E) Luciferase reporter assays of ERα and ERβ transcriptional activity with ERE-LUC, CAOV3 (D) or ES2 (E) cells that underwent the lentiviral-mediated stable ablation of PES1and were treated with 10 nM E2, 1 nM PPT, or 1 nM DPN for 24 h. @P < 0.01, *P < 0.01, and #P < 0.01 vs. the shRNA control treated with E2, PPT, and DPN, respectively. Results shown are the mean ± SD of three independent experiments. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

To investigate the role of endogenous PES1 in ER-mediated transcriptional activity, the lentiviral-mediated stable ablation of PES1 was performed in CAOV3 and ES2 cells using shRNA. As shown in Fig. 1C, PES1 shRNA obviously repressed the endogenous PES1 expression in both CAOV3 and ES2 cells. In the CAOV3 cells, the ablation of the normal expression of PES1 significantly decreased the transcriptional activity of the ERE reporter for approximately 31% with the presence of E2, or almost completely attenuated transcriptional activity activated in the presence of PPT. However, the transcriptional activity of the ERE reporter increased by 1.6-fold with DPN (Fig. 1D). In the ES2 cells, the ablation of PES1 significantly decreased ERE reporter transcriptional activity in the presence of E2 or PPT (Fig. 1E). The results were consistent with those regarding the effect of PES1 overexpression, showing that the ablation of endogenous PES1 decreases the transcriptional activity of ERα and increases that of ERβ.

PES1 Changes the Expression Levels of the Different ERs and Estrogen-Responsive Genes

We have recently reported that PES1 modulated ERα and ERβ expression at the post-transcriptional level through the ubiquitin-proteasome pathway in breast cancer cells, mediated by the carboxyl terminus of the Hsc70-interacting protein [23]. To identify whether changes in the transcriptional activity of ERs caused by luciferase reporter assays in ovarian cancer cells were due to alterations of relative ER' expression level, lentiviral-mediated stable ablation of PES1 was performed in CAOV3 and ES2 ovarian cancer cells with shRNA. As shown by the Western blot results (Fig. 2), the ablation of PES1 expression decreased the expression of ERα but increased ERβ expression. Cyclin D1, is well-known to be regulated by E2 in breast cancer relative to cell cycle and has been reported to be activated by ERα and to be repressed by ERβ [32, 33]. In this study, cyclin D1 was down-regulated by the ablation of PES1. Meanwhile, it was not surprising that p21WAF1, one of the inhibitors of cyclin-dependent kinases, was up-regulated when PES1 was silenced. Angiogenesis plays an important role in ovarian carcinogenesis. VEGF expression depends on HIF-1α, which is the target of estrogen in ovarian cancer cells [3, 34]. VEGF and HIF-1α expression levels both decreased after PES1 ablation. These data suggest that PES1 differently regulates the expression of ERs and estrogen-responsive genes.

Figure 2.

PES1 regulates the expression of ERα and ERβ and estrogen-responsive genes. (A and B) Immunoblot analysis of the expression of ERα and ERβ and estrogen-responsive genes in CAOV3 (A) and ES2 (B) cells stably transduced with shRNA control or PES1 shRNA in the presence or absence of 10 nM E2 for 24 h.

Reduction of Endogenous PES1 Inhibited the Cell Proliferation and Growth of Ovarian Cancers Cell In Vitro

To investigate the biological functions of PES1 in the pathogenesis of ovarian cancer, we determined the effect of PES1 on the growth and proliferation of ovarian carcinoma cells in vitro. PES1 was stably ablated in CAOV3 and ES2 cells treated with PES1 shRNA. As shown in Fig. 3A, the PES1 shRNA-expressing clone displayed a reduced rate of E2-mediated cell proliferation compared with the control shRNA in both CAOV3 and ES2 cells. Ablation of endogenous PES1 resulted in significant inhibition of anchorage-dependent growth (Fig. 3B). Cell cycle analysis by FACS further demonstrated that the reduction of endogenous PES1 expression caused a G2 phase delay in both of the absence or presence of E2 (Fig. 3C). These results confirmed that PES1 is very important for cell proliferation and cell cycle progression.

Figure 3.

Effect of PES1 on the growth and proliferation of ovarian cancer cells. Growth curves of CAOV3 (A) and ES2 (B) cells stably transduced with PES1 shRNA. Cells were treated with 10 nM E2 and analyzed by the MTT assay. *P < 0.001 vs. shRNA control without E2 on day 4. #P < 0.001 vs. shRNA control with E2 on day 4. Values represent means ± SD from three independent experiments. (C and D) Silencing of PES1 decreased colony formation in CAOV3 (C) and ES2 (D) cells. The photos demonstrate the results of the colony-formation assay of cells in the plate (left panel). The colony numbers (right panel) were obtained from three independent experiments. *P < 0.01 vs. shRNA control without E2. #P < 0.01 vs. shRNA control with E2. (E) Silencing of PES1 resulted in cell cycle arrest in CAOV3 (E) and ES2 cells (F). Quantification of the cell cycle distribution was derived from three independent experiments. Values represent the means ± SD. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Correlation of PES1 with ERα and ERβ in Ovarian Cancer Patients

PES1 was shown to differentially regulate the expression levels of ERα and ERβ, and, IHC staining was performed to determine the expression of PES1 in clinical samples of ovarian cancer and its relationship with ERα and ERβ expression. Twenty non-cancerous human ovarian samples and 54 samples of ovarian cancer were used for IHC staining with PES1, ERα, and ERβ antibodies. As shown in Fig. 4, PES1 exhibited strong nuclear staining in cancer cells, along with ERα and ERβ. Out of the 54 ovarian cancer tissues, 43 (79.63%) were positive for PES1 expression at an intermediate or high level, whereas only 10% (2/20) of non-cancerous tissues showed low levels PES1 staining. The positive rate of PES1 expression in ovarian cancer tissues was significantly higher than that of the non-cancerous tissues (P = 0.000, χ2 = 29.692). The positive rate of ERα was 30% (6/20) or 70.37% (38/54) in noncancerous tissues or ovarian cancer tissues respectively, whereas, the positive rate of ER β was 45% (9/20) or 5.56% (3/54) in noncancerous tissues or ovarian cancer tissues, respectively. Intriguingly, consistent with the finding that PES1 up-regulates ERα and down-regulates ERβ expression, the results of IHC staining showed that PES1 expression correlated positively with ERα expression (P = 0.004, U = 121.00) and negatively with ERβ expression (P = 0.04, U = 175.5) (Table 1). These results strongly indicate important pathological roles of PES1 in ovarian cancer.

Table 1. Summary of 54 tissue biopsies of ovarian cancer and 20 noncancerous tissues
  The expression of ER α (IHC scores)a The expression of ER β (IHC scores)a 
GroupCases (n)1234P value1234P value
  1. a

    Staining was assessed on a four-tiered scale: negative, no staining; 1+, weak staining; 2+, moderate staining; and 3+, strong staining. The IHC scores were graded as follows: 0, no staining or 1+ staining regardless of the percentage of positive cells; 1, 2+ staining in ≤30% of cells; 2, 2+ staining in >30% of cells; 3, 3+ staining in ≤ 50% of cells; and 4, 3+ staining in >50% of cells. The P values were generated using the Mann-whitney U test.

PES1 high expression43891790.004384100.04
PES1 low expression118130 7211 
Figure 4.

Association of PES1 with ERα and ERβ in clinical ovarian cancer samples. Representative immunohistochemical staining patterns are shown of PES1, ERα, and ERβ. Expression of ERα, ERβ and PES1 in human ovarian cancer tissues (A, B, C) and noncancerous tissues (D, E, F). Original magnification, ×40. Scale bar: 100 μm.

Discussion

Estrogen exhibits mitogenic activity in early ovarian carcinogenesis and plays an important role in ovarian tumorigenesis. Estrogen signaling depends on the balance between ERα and ERβ expression [35, 36]. ERβ is highly represented in normal ovarian epithelial cells or benign tumors, whereas ERα is the main form expressed in malignant tumors [3, 37]. ERβ mRNA decreases in an inversely correlated way with tumor progression compared with ERα mRNA, and the ER α/β mRNA ratio is markedly increased in ovarian cancer [3, 38]. In IHC studies, compared with ERα, ERβ was expressed at significantly lower levels in ovarian cancer tissues; the loss of ERβ expression in ovarian tumors maybe a feature of malignant transformation and correlates with shorter overall durations [39-41]. Studies have shown that reintroduction of ERβ directly inhibited ERα activity and the proliferation of ovarian cancer cells [32, 42]. These suggest that the imbalance of ERα and ERβ may play an important role during ovarian carcinoma progression.

Here, we reported that PES1 increased ERα transcriptional activity and decreased that of ERβ, while the ablation of PES1 expression decreased the level of ER α protein and increased that of ER β in ovarian cancer cells in vitro. IHC results of clinical ovarian cancer tissues also showed that PES1 expression correlated positively with ERα expression and negatively with ERβ expression, although studies with larger sample sizes should be performed to confirm the relationship between PES1 and the ERs. These data suggest the PES1 can regulate the balance of ERα and ERβ expression in ovarian cancer cells. ERα mediates cell growth, whereas ERβ acts as an ERα antagonist, thereby inhibiting the cell growth promoted by ERα. It has been reported that the estrogen-stimulated growth of epithelial ovarian carcinoma is mediated by ERα rather than ERβ [43]. The E2-triggered metastatic potential of ovarian cancer cells also exclusively depends on ERα pathway, which could be promoted by activating the PIK3/AKT pathway [44, 45]. The present results show that the ablation of endogenous PES1 inhibited the proliferation and growth of ovarian cancer cells. The expression of PES1 increased the ratio of ERα/ERβ in ovarian cancers, and might attenuate the antagonistic effect of ERβ on the function of ERα. E2 has been shown to induce the expression of PES1 [21, 46], which might further change the balance between ERα and ERβ and hence play a crucial role in ovarian tumorigenesis.

Previous studies have shown that the loss of ERβ expression was correlated with increased DNA methylation of its promoter [47, 48]. As approximately 20% of breast cancers have amplification of the gene that encodes ERα, this may be one of the important mechanisms for the regulation of ERα expression in this tumor type [49], ERα amplification occurs only rarely in ovarian cancer [50]. Therefore, the regulation of ERα and ERβ expression by PES1 may be a new factor that controls the level of the different ERs in ovarian cancer.

PES1 has been demonstrated to play important roles in embryonic development [10], cell cycle regulation [12] and ribosome biogenesis [51-53], although the molecular mechanisms that underlie these processes remain largely unknown. PES1 modulate many estrogen-responsive genes in breast cancer cells and is well known to have important functions in DNA replication and cell cycle regulation [23]. The altered regulation of the cell cycle results in uncontrolled cell proliferation in ovarian cancer [54]. Cyclins and their associated CDKs form the central machinery that governs cell cycle progression. The overexpression of cyclin D1, the major regulatory subunit for CDK4, is common in human cancers with an epithelial cell origin [55]. The ablation of PES1 in ovarian cancer cells resulted in the significant down-regulation of cyclin D1, up-regulation of p21WAF1, and G2 phase delay. These current results suggest that PES1 plays an important role in ovarian cancer cell proliferation.

Estrogen directly regulates VEGF gene transcription though a variant ERE in the VEGF promoter via both ERα and ERβ [56]. PES1 differently regulates the expression of ERα and ERβ in ovarian cancer cells, and the ablation of PES1 resulted in the significant suppression of HIF-1α and VEGF expression. Thereby, it will be very interesting to investigate how PES1 regulates VEGF expression and the relationship between PES1 and VEGF in ovarian cancer tissues.

In summary, we have demonstrated that PES1 inversely regulated the expression of ERα and ERβ at protein level, and alters their transcriptional activity on target genes. The ablation of PES1 resulted in the repression of ovarian cell proliferation and G2 arrest. Our results suggest that PES1 may have important role in the progression of ovarian cancer and may be a new target for ovarian cancer therapy.

Acknowledgements

The authors thank Professor Xu Lin (Key Laboratory of Ministry of Education for Gastrointestinal Cancer, Research Center for Molecular Medicine, Fujian Medical University, Fuzhou, Fujian, People's Republic of China) for helpful discussion. This work was supported by Social Development Key Project in Fujian Province (2010Y0048) and Natural Science Foundation of Fujian Province (2010J05082).

Ancillary