Background
Ovarian cancer is one of the most common cancers among women and the leading cause of death from gynecological malignancies in the world [
1,
2]. The underlying causes of ovarian cancer are poorly understood and largely untested, but estrogen, as a major steroidal product of the ovary, has been shown to be associated with increased ovarian cancer risk in estrogen receptor (ER)-expressing cells [
3‐
5]. Hormone replacement therapy (HRT) has been widely used in women with estrogen-withdrawal syndromes. Estrogen has been associated with an increased ovarian cancer risk and it can promote tumor growth and cell proliferation in ER-expressing cell lines. The main biological functions of estrogen are manifested through transcriptional activation of the ligand-dependent ERs, ERα and ERβ [
3,
6]. More than two-thirds of ovarian cancer patients are positive for ERα [
6‐
8]. The activation of ER results in an altered expression of its direct transcriptional targets, thereby affecting a series of downstream secondary biological activities. However, the regulatory effects of estrogen on the behavior of ovarian tumor cells involves a complex signaling network and the underlying mechanisms are still not fully understood. Therefore, identification of novel targets regulated by estrogen will be very important to clarify the specific impact of estrogen on ovarian tumor growth and facilitate the development of new diagnostic and therapeutic markers [
4,
5,
9].
In recent years, a number of studies have shown that numerous estrogen-responsive genes, including insulin-like growth factor binding protein (IGFBP) family members (
IGFBP3,
IGFBP4 and
IGFBP5), trefoil factor (
TFF) family members (
TFF1 and
TFF3), and TNF receptor-associated protein 1 (
TRAP1) among others, affect the growth and development of ovarian cancer [
10,
11]. Genomic analysis of ovarian cancer identified gene amplification of the
WFDC2 gene (that encodes the human epididymis protein 4 (
HE4)) and the whey acidic protein gene loci in a large proportion of epithelial ovarian cancers [
12,
13]. HE4 exhibits a tumor-restricted, upregulated pattern of expression in ovarian cancer, making it a potential marker [
12,
14,
15]. The previous work has shown a direct linkage between
HE4 expression and cell proliferation; however the molecular mechanisms are still unclear [
12,
16]. To date, the majority of studies have focused on the potential value of HE4 as a diagnostic using various serologic tests, but very little attention has been paid to the role of HE4 in tumor development of ovarian cancer [
12,
14,
17].
The
WFDC2 gene is located on human chromosome 20q12-13.1, a region that includes 14 genes that encode proteins with a WAP-type four-disulfide core (WFDC) domain [
14,
17]. Two of the best-studied members of the WAP gene family are
SLPI and
PI3 (that encodes for elafin), both having antiproteinase activity. They are co-expressed with
WFDC2 and involved in cancer development or progression in various carcinomas affected by sex hormones [
9,
14,
18]. So we could not help to speculate that WFDC2 might also play some role in the estrogen-sensitive ovarian cancers.
As a cancer-specific gene, several hormone-response elements were found within the
WFDC2 promoter, including an estrogen response element (ERE) and RORA, which may be attributed to HE4 upregulation in ovarian cancer and ovarian cancer specificity [
19]. The amount of HE4 in blood samples was significantly different between follicular (FP) and ovulatory (OP) phases of the hormonal cycle, being lower in the FP than in the OP [
20]. The menstrual cycle phase-dependent variability indicated that
WFDC2 expression might be affected by the menstrual cycle of women. These results suggested that
WFDC2 might be an estrogen response gene, and play important roles in the cell proliferation and malignant transformation of ovarian cancer.
In this study, we investigated the regulatory effects of estrogen and estrogen antagonist on WFDC2 gene expression in estrogen sensitive HO8910 cells and estrogen insensitive SKOV3 cells, with the aim to determine whether WFDC2 is an estrogen-responsive gene. And then, we transfected these cells with short hairpin RNA (shRNA) against WFDC2, and investigated the effect of WFDC2 silencing on cell proliferation, its interaction with ER and its effect on ER-mediated signaling.
Methods
Cells and treatments
The cell bank of the Chinese Academy of Sciences (Shanghai, China) supplied the human ovarian cancer cell lines, HO8910 and SKOV3 (American Type Culture Collection (ATCC), Manassas, VA, USA). Cells were maintained in minimal essential medium supplemented with 10 % (v/v) fetal bovine serum (FBS) at 37 °C in an atmosphere of 95 % air, 5 % CO2. The ligand 17β-estradiol (E2) and the selective ER modulator (SERM), tamoxifen (TAM), were purchased from Sigma-Aldrich (St Louis, MO, USA). Before the cells were treated with the ligands, the medium was replaced with minimal essential medium supplemented with 0.5 % FBS. Cells were treated with different concentrations of E2 (5, 25, 125, 625 and 1250 ng/ml), and TAM (100 ng/ml), for 24 h prior to quantitative real-time PCR (QPCR) and 48 h prior to western blotting and protein array analysis.
RNA extraction and QPCR
Total RNA was isolated following the manufacturer’s instructions (PrimeScript 1st Strand cDNA Synthesis Kit, TAKARA). Reverse transcription was also performed following the manufacturer’s instructions in a total volume of 20 μl using an oligo-dT primer and 1 μg of total RNA. Each primer set was designed using Primer Express software v3.0 to flank an intron to prevent the amplification of genomic DNA. β-actin was used to evaluate the efficiency and variability of the reverse transcription step. CDNA samples (0.1 μg) were amplified using the SYBR Green PCR Master Mix (TAKARA) under conditions recommended by the manufacturer: (a) pre-incubation at 95 °C for 30 s; (b) 40 PCR cycles of 95 °C for 5 s, 55 °C for 30 s, and 72 °C for 34 s. Samples were assayed in duplicate using the ABI Prism 7500 detection system (Perkin Elmer Applied Biosystems). The relative quantification number was then calculated by subtracting the average CT from the corresponding average CT for β-actin.
Western blotting
Total protein was extracted using sonication in radio-immunoprecipitation assay (RIPA) buffer (50 mM Tris–HCl pH 7.5, 150 mM NaCl, 5 mM EDTA, 0.5 % Nonidet P-40, 5 mM dithiothreitol, 10 mM NaF, and protease inhibitor cocktail). One hundred micrograms of denatured protein was separated on an SDS-polyacrylamide gel and transferred to a Hybond membrane (Amersham, Germany), which was then blocked overnight in 5 % skim milk in Tris-buffered saline with Tween 20 (TTBS, 10 mM Tris–HCl, 150 mM NaCl, 0.1 % Tween 20). For immunoblotting, the membrane was incubated for 15 min with the WFDC2 antibody. The membrane was rinsed with TTBS and incubated with anti-rabbit IgG conjugated to horseradish peroxidase (DAKO, USA, 1:1000) for 15 min. All incubations were performed in a microwave oven to allow intermittent irradiation. Bands were visualized on an ImageQuant LAS4010 (GE Healthcare Life Science, USA) using ECL-Plus detection reagents (Santa Cruz, USA). Densitometric quantification of protein bands with GAPDH as an internal control was performed using Image J (NIH, USA).
Gene silencing
The WFDC2-specific shRNA sequence (5′-GCTCTCTGCCCAATGATAAGG-3′) (based on the Gene Bank Accession No. NM_0006103.3) and its control sequence (5′-GTTCTCCGAACGTGTCACGT-3′) were chemically synthesized and cloned into the pGLV-U6-GFP vector by Shanghai GenePharma Co. Ltd (Camsonne et al.). Lentiviruses, purchased from the same company, were transduced into the HO8910 cell line according to the manufacturer’s instructions. For stable silencing of WFDC2, the transduced HO8910 cell line, named HO8910-209, was selected using puromycin. Puromycin-resistant colonies were then picked, expanded and analyzed separately.
Cell proliferation assays
The in vitro proliferation assay was performed using the 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (MTT) assay following the manufacturer’s instructions (Sigma). Briefly, 500 cells per well were plated in 96-well plates in triplicate. After a 24 h incubation, cells were serum-starved for 24 h and then treated with different concentrations of estrogen (0, 5, 25, 125, 625 and 1250 ng/ml) or TAM for 6 days. At the indicated time, cell proliferation was determined by measurement of the absorbance values using the MTT assay method, and cell growth curves were then plotted.
Flow cytometry
Cells were trypsinized and washed three times with phosphate-buffered saline (PBS). Cells were then digested with 1 % RNase at 37 °C for 30 min and stained with annexin V-FITC for 30 min, followed by staining with propidium iodide (PI, Sigma, Shanghai, China) for 5 min at 4 °C for cell apoptotic analysis. The results were analyzed using WinMDI software with 10,000 events collected for each sample. Cell suspensions were also incubated with antibodies to ERα and ERß (1 μg/1 × 106 cells) for 30 min at 37 °C. Cells were then washed three times with PBS, and incubated with PE and an FITC-labeled secondary antibody for 30 min at 37 °C. Cell suspensions were washed three times with PBS. Expression of ERα and ERß per 10000 cells was determined using flow cytometry.
Annexin V-FITC/PI staining
Apoptosis was evaluated using the Annexin V-FITC/PI Apoptosis Detection kit (BestBio, Shanghai, China) using fluorescence microscopy according to the manufacturer’s instructions. Briefly, cells were grown onto a cover slide and incubated with E2 (625 ng/ml) for 24 h. The adherent cells were washed twice with ice-cold PBS and stained with annexin V-FITC and PI for 15 min and examined under a light microscope equipped with appropriate filters. Apoptotic cells stained with annexin V-FITC showed green fluorescence, and necrotic cells stained with both annexin V-FITC and PI produced red fluorescence as well as green fluorescence.
Human antibody array for apoptotic-related proteins
Apoptotic-related proteins induced by E2 directly and indirectly were determined using the RayBio® Label-Based Human Antibody Array kit (RayBio® Human Apoptosis Antibody array) (RayBiotech, Norcross, GA). Densitometric analysis was performed on a Kodak ImageStation 4000 M (Eastman Kodak Company, Rochester, NY) with background subtraction from spot edges following the manufacturer’s instructions. Spot data were normalized to a positive control spot on each array.
Statistics
Microsoft Office Excel 2007 and the statistical software SPSS13.0 were used in data processing and the t-test was used in analyzing significance. P values < 0.05 were considered statistically significant. Data were expressed as the mean ± SD from at least three independent experiments.
Discussion
Estrogen plays a crucial role in the control of development, sexual behavior and reproductive functions. Its effects have been linked to the progression of the majority of human ovarian cancers and acts as a potent mitogen for many ovarian cancer cell lines [
3,
23]. In this study, we undertook to prove the hypothesis that
WFDC2 is regulated by estrogen and that it might play a role in tumorigenesis induced by estrogen in ovarian cancer.
In this study, we demonstrated that high-dose E2 induced the upregulation of WFDC2 gene expression in estrogen-sensitive HO8910 cells, while no induction effect was observed in estrogen-insensitive SKOV3 cells. However, the estrogen selective inhibitor TAM could not block estrogen-induced WFDC2 expression. These results provide evidence of a positive relationship between WFDC2 expression and estrogen action; however, the regulatory effect might not be through the ERα pathway only, which will require further experiments to prove definitively.
This study also provides the first evidence for a functional role of the
WFDC2 gene in the estrogen-dependent proliferation of ovarian cancer cells. It is generally considered that the proliferative effects of estrogen in cell culture occur at picomolar or nanomolar concentrations, typical levels in serum [
1]. However, the ovary is an estrogen secretion organ, and thus micromolar concentrations of estrogen occur within the ovary and play a role in cell biology or transformation of normal ovarian surface epithelium and ovarian cancer cells, which has been largely overlooked [
24,
25]. Proliferation in response to physiological concentrations of E2 has been reported in cultured ovarian cancer cells expressing ER, but currently there is no direct evidence in support of the hypothesis that high E2 levels present within the microenvironment of the ovary following ovulation contribute to the induction of ovarian cancer. Although physiological concentrations of estrogen can stimulate breast cancer cell proliferation, high-dose estrogens cause regression of some ER-positive human breast tumors.
In this study, we observed significant inhibition of cell growth induced by high-dose estrogen in estrogen-sensitive HO8910 cells, and a slight inhibition of cell growth in estrogen-insensitive SKOV3 cells. The suppression of the endogenous WFDC2 in ovarian cancer cells not only inhibited cell growth, but also significantly strengthened the response of estrogen-insensitive SKOV3 cells to estrogen. These data suggest the involvement of WFDC2 in estrogen signaling and in estrogen-responsiveness of ovarian cancer cells. We also found that WFDC2 knockdown could affect the hormone-dependent proliferation not only in HO8910-sensitive but also in SKOV3-insensitive cells. WFDC2 knockdown decreased proliferation of SKOV3 ovarian cancer cells and, more strikingly, switched on a strong estrogen response in this estrogen-unresponsive cell line.
This transformation of ovarian cancer cells from hormone-independent to an estrogen-responsive phenotype by knockdown of a single gene was further validated in proliferation experiments combining E2 with TAM. As shown in the results, WFDC2-knockdown increased the sensitivity of cells to TAM not only in estrogen-sensitive, but also in estrogen-insensitive cells. The suppression of the endogenous WFDC2 in ovarian cancer cells was more likely suppressed by TAM. Thus, the specific role of WFDC2 in cell growth in the microenvironment with high-dose estrogen may be associated with ovarian epithelial cancer cell growth and transformation, and may also be involved in TAM resistance.
Previously, we demonstrated that the expression of
WFDC2 promoted cell proliferation as well as G1/S transition, while stimulating cyclin D1 expression [
12]. Cyclin D1 has been confirmed as one of the estrogen target genes, and confers the mitogenic role of estrogen [
26]. These results led us to postulate that
WFDC2 participated in estrogen-dependent proliferation through regulating expression of cell cycle checkpoint proteins. While, no previous evidence has been presented,
WFDC2 has been shown to be associated with cell apoptosis [
12]. We further analyzed whether apoptosis, regulated by estrogen, could be affected by
WFDC2 expression. Using FACS analysis, we showed that high concentrations of estrogen could significantly increase the level of cell apoptosis, and that loss of
WFDC2 expression resulted in a significant increase in the rate of apoptosis under high-dose estrogen both in estrogen-sensitive and insensitive cells. Our results demonstrated that pharmacological concentrations of estrogen induced apoptosis in human ovarian cancer cells that involved
WFDC2 expression.
ER has been considered to be an important regulator of estrogen-sensitive cell behavior. The response of neurons to estrogen depends on the ER subtype expressed in the cell. ERα is known to stimulate proliferation in response to estrogens by increasing expression of genes connected with cell cycle progression like cyclin D1 or growth factors, and by downregulating antiproliferative and proapoptotic genes [
26,
27]. Lokich et al. demonstrated that
WFDC2 interacted with ERα, and that
WFDC2 overexpression resulted in ERα downregulation in ovarian cancer cells [
19]. In our study, we did not observe upregulation of ERα in HO8910-209 cells; indeed, expression of ERα was almost unchanged by
WFDC2 knockdown. While
WFDC2 knockdown resulted in upregulation of ERβ in both HO8910 and SKOV3 cells. ERβ has been described to act as an antagonist of ERα in certain settings and to act as a tumor suppressor with proapoptotic and antiproliferative properties. Furthermore, loss of ERβ in ovarian epithelial cells has been linked to tumorigenesis and has been shown to increase proliferation of ovarian cancer cells [
6,
28]. A clear upregulation in the level of ERβ was noted in
WFDC2-knockdown cells, which could partially explain why high-dose estrogen caused more apoptosis in those cells. Thus, we hypothesize that upregulation of ERβ caused by suppression of
WFDC2 is part of the mechanism underlying the decreased growth of these cells.
Because of these results, we further studied the correlation between
WFDC2 and a series of genes that were related to cell proliferation and apoptosis using an apoptosis antibody array. In estrogen-sensitive HO8910 cells, insulin-like growth factor-1 (IGF-1) and insulin-like growth factor binding protein-1 and 2 (IGFBP-1 and IGFBP-2), which have been considered to be estrogen-responsive genes and related to the malignancy and metastasis of ovarian cancer [
29‐
31], were downregulated in
WFDC2-knockdown cells in the protein array.
WFDC2 knockdown has also been shown to decrease HSP27 expression, which has also been identified as an estrogen-binding protein, playing a complex role in cell proliferation, migration and differentiation of estrogen-responsive tumors [
32,
33]. X-linked inhibitor of apoptosis protein (XIAP) and survivin, apoptotic suppressors [
34‐
36], were also found to be downregulated in
WFDC2-knockdown cells. These results are consistent with our previous hypothesis and explain the role of
WFDC2 in estrogen-dependent cell proliferation and apoptosis induced by estrogen in ovarian cancer cells. However, not all the results were in line with our expectations. We expected, that the expression of tumor suppressor genes P21 [
37] and HTRA [
38,
39] should be increased by
WFDC2 knockdown. But in fact, expression of these two genes in
WFDC2-209 cells was significantly downregulated.
The results we observed in SKOV3 cells were also not entirely consistent with what we observed in the HO8910 cells. There was no evidence to confirm the relationship between
WFDC2 and the insulin-like growth factor pathway, and we did not observe any change in IGFBP-A, IGF-1, HSP27 or XIAP in SKOV3 cells, although the data were of interest. It is well known that Bcl-2 family proteins are key regulators of apoptosis [
40,
41], in particular, where involvement of the mitochondrial apoptotic pathway is concerned. In estrogen-insensitive SKOV3 cells, proapoptotic genes belonging to the Bcl2 family, such as Bad, Bax, and Bim were upregulated by
WFDC2 knockdown. These data led us to suppose that
WFDC2 could take part in cell apoptosis through the mitochondrial apoptotic pathway. Accordingly, proapoptotic genes, Fasl, Hsp60, caspase 3 and caspase 8 were also upregulated in
WFDC2-knockdown cells, which also revealed the proapoptotic effects of
WFDC2 knockdown. Additionally, the tumor suppressor gene P21 was downregulated in
WFDC2-knockdown SKOV3-209 cells, and as seen in HO8910-209 cells. The contradictory results suggest that
WFDC2 might be involved in a complex apoptotic net that will need further study to unravel.
The inconsistent results obtained using SKOV3 and HO8910 cells suggested that WFDC2 regulated cell apoptosis through different pathways in estrogen-sensitive and estrogen-insensitive cells. The focus of our upcoming research, therefore, will be to gain a better understanding of the relationship between WFDC2 and the insulin-like growth factor pathway to ascertain the role of WFDC2 on the physiological and pathological behavior of ovarian cancer. However, it was the downregulation of the tumor suppressor genes P21 and HTRA, which led us to question the pathophysiological role of WFDC2 in tumor initiation and progression. Understanding the detailed molecular mechanisms of action of WFDC2 in cell apoptosis induced with high-dose E2 will require further study.
Competing interests
The authors have no conflicts of interest to declare. All authors have approved the manuscript for publication.
Authors’ contributions
Conceived and designed the experiments: YC, LTC, SHW, YSW, JLL and ML. Performed the experiments: YC, SHW, and LTC. Analyzed the data: YC, YSW, JLL and ML. Contributed reagents/materials/analysis tools: YC, LTC and SHW. Wrote the paper: YC, YSW, JLL and ML. All authors read and approved the final manuscript.