This article is mentioned in:

Introduction

In western countries, ovarian cancer can be attributed to the primary cause of cancer-associated mortality. Based on the statistical data of the American Cancer Society in 2013, 70–90% of women with ovarian cancer are susceptible to relapse or metastasis, and ~30% of patients with advanced ovarian cancer develop peritoneal metastasis (1). The importance of long non-coding RNAs (lncRNAs) has been increasingly recognized, as they are widely involved in carcinogenesis, metastasis and drug resistance. Thus, it is particularly important to examine the association between ovarian carcinogenesis and the abnormal expression of lncRNAs, thereby elucidating the complexity of tumor biology.

At present, only a relatively small number of lncRNAs have been identified, and they have been found to be widely involved in gene expression at transcriptional and post-transcriptional levels (2,3). Previous studies have suggested that lncRNAs serve as competing endogenous RNAs (ceRNAs) to sponge micro (mi)RNAs, thereby regulating gene expression (4). Linc-MD1 has been reported to function as an important ceRNA, which acts as an effective decoy to protect MyoD mRNA from miRNA-mediated degradation (5). Furthermore, lnc-RoR may serve as an important ceRNA to regulate the expression of pluripotency-associated genes. The liver specific lncRNA, HULC, has also been found to act as an endogenous 'sponge', thereby repressing the activities of certain miRNAs, including miR-372 (6). These previous studies suggest that lncRNAs function as ceRNAs and are involved in ovarian carcinogenesis.

As a transcript from the HOXC locus, Hox transcript antisense intergenic RNA (HOTAIR) significantly inhibits the transcription of HOXD in foreskin fibroblasts (7). In primary breast tumors and breast cancer metastases, HOTAIR has been found to be significantly upregulated, which is then involved in the invasiveness and metastasis of the cancer (8). In addition, in colorectal cancer, hepatocellular carcinoma and other types of cancer, HOTAIR is positively associated with malignant processes and poor outcome (9–12). However, the specific role of HOTAIR in ovarian carcinoma remains to be fully elucidated.

In the present study, the level of HOTAIR in ovarian carcinoma was examined, and the role of HOTAIR in ovarian carcinogenesis was investigated. The results suggested that HOTAIR may act as a ceRNA to regulate the expression of Rab22a through the competitive binding of miR-373. The present study also indicated a positive correlation between HOTAIR and Rab22a, as well as the interaction with miR-373, which may assist in elucidating the molecular mechanism of ovarian carcinogenesis.

Materials and methods

Tissue collection

Ovarian cancer tissues and adjacent non-tumorous ovarian samples were obtained from 30 Chinese patients (age, 45.8±10.5 years) at Tongji University Hospital affiliated to Shanghai Tongji University (Shanghai, China) between 2008 and 2010. Based on histopathological evaluation, all cases were carefully reviewed as ovarian cancer. None of the patients received adjuvant therapy and the tissues were collected during biopsy. The present study was approved by the Research Ethics Committee of Tongji University Hospital affiliated to Shanghai Tongji University. Informed consent was obtained from each patient.

Cell lines and culture conditions

The human epithelial ovarian cancer (EOC) cells (SKOV3, A2780 and CP70), the human ovarian immortalized nontumorigenic ovarian surface epithelial (IOSE) cells, EOC cells (HeyC2) and human embryonic kidney (HEK)293T cells were purchased from American Type Culture Collection (Manassas, VA, USA). The cells were cultured in RPMI-1640 medium containing 10% fetal bovine serum (FBS). HEK293T cells were cultured in Dulbecco's modified Eagle's medium (Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) with 10% FBS and 1% penicillin-streptomycin and used for the dual luciferase reporter assay. All cell lines were maintained at 37°C in a humidified atmosphere with 5% CO2.

RNA isolation and reverse-transcription-quantitative polymerase chain reaction (RT-qPCR) analysis

Total RNA from the tissues and cells was extracted using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.). RNA quality and concentration were determined using a Nanodrop 2000 system (Thermo Fisher Scientific, Inc.). To analyze the levels of HOTAIR, a reverse transcription kit (Applied Biosystems, Foster City, CA, USA) was used, with GAPDH used as the reference. To measure the level of HOTAIR, a SYBR Premix Ex Taq™ kit (Takara Bio, Dalian, China) was used, and the expression of GAPDH was used as an endogenous control. The reaction system contained 2X TaqMan Universal PCR Master Mix, No AmpErase UNG (Applied Biosystems; Thermo Fisher Scientific, Inc.), 20X TaqMan MicroRNA Assay mix (Applied Biosystems; Thermo Fisher Scientific, Inc.) and template cDNA. The primers were as follows: Forward, 5′-GCG CTG CAA GTG CTT ACT GTGCA-3′ and reverse, 5′-CCG AGG TAT TCG CAC TGG ATAC-3′ for HOTAIR; and forward, 5′-GTC GGT GTG AAC GGA TTTG-3′ and reverse, 5′-AAG ATG GTG ATG GGC TTCC-3′ for GAPDH. The thermal cycling conditions were as follows: 95°C for 10 min; 40 cycles at 95°C for 15 sec and 60°C for 1 min. RT-qPCR was performed using the Applied Biosystems 7900 Fast Real-Time PCR system (Applied Biosystems). The data were analyzed using the 2−ΔΔcq method (13).

Plasmid construction

The cDNA was cloned into a pcDNA3.1 vector (Invitrogen; Thermo Fisher Scientific, Inc.), and miRNA precursors were cloned into the modified pLL3.7 vector (Invitrogen; Thermo Fisher Scientific, Inc.). In addition, the 3′-untranslated region (UTR) of Rab22a and HOTAIR with the predicted potential miRNA binding sites were cloned into a pLUC luciferase vector (Ambion, Carlsbad, CA, USA).

Oligonucleotide transfection

The cells (1×106 cells/well) were plated in a six-well plate 1 day prior to transfection. Using Lipofectamine 2000 (Invitrogen; Thermo Fisher Scientific, Inc.), the cells were transfected with the plasmid vectors (50 nmol/l). The HOTAIR small interfering (si)RNA (si-HOTAIR) and negative control siRNA (si-NC) were purchased from Shanghai Jima Technology Co., Ltd. The siRNAs were mixed with HiPerFect transfection reagent (Qiagen GmbH, Hilden, Germany) and incubated at room temperature for 10 min. The complex was transfected into cells for 48 h.

Dimethyl thiazolyl diphenyl tetrazolium (MTT) assay

A total of 5,000 cells per well in 100 μl of medium were seeded in 96-well plates. The cells were then transfected with pcHOTAIR/pcDNA or miR-373 mimics/NC at a final concentration of 50 nM. Following transfection, 20 μl of MTT reagent (Solarbio, Beijing, China) was added into the wells and incubated with the cells for 4 h at 37°C. The medium was removed and washed with phosphate-buffered saline (PBS) three times. The blue formazan was dissolved in 200 μl of dimethyl sulfoxide (DMSO; Solarbio), and was measured at 550 nm. Wells with cells only served as a blank control.

Hoechst 33258 staining

Following transfection with pcHOTAIR/pcDNA or miR-373 mimics/NC for 48 h, the cells were washed with PBS three times. Subsequently, the cells were incubated with Hoechst 33258 (10 μg/ml; Solarbio) for 5 min. The cells were then washed with cold PBS three times and observed under a fluorescent microscope.

Quantification of apoptotic cells

An Annexin V-fluorescein-5-isothiocyanate (FITC) apoptosis detection kit (BioVision, Inc., Paroo, CA, USA) was applied to detect apoptotic cells. Following transfection with miR-373 mimics or NC for 48 h, the cells were placed in a 5 ml tube. The cells were then washed with cold PBS and resuspended in 1X Annexin V binding buffer containing 10 mM HEPES/NaOH (pH 7.4) (1×106 cells/ml). The cells were mixed with FITC-Annexin V (5 μl) and propidium iodide for 15 min at room temperature. Following washing with PBS three times, the samples were analyzed using flow cytometry.

Bioinformatics methods

The potential miRNA binding sites of HOTAIR were predicted using the following websites: 132.77.150.113/pubs/mir07/mir07_prediction.html, regrna.mbc.nctu.edu.tw/html/prediction.html and www.microRNA.org.

Cell migration assay

The cells were grown to confluence as a monolayer in 6-well plates. To initiate migration, the cell layer was scratched using a pipette tip. The cells were subsequently transfected with miR-373 mimics or negative control. Cell migration was captured under a microscope (CX21; Olympus Corporation, Tokyo, Japan).

Statistical analysis

The data are expressed as the mean ± standard error of the mean and statistical significance was analyzed using Student's t-test (two-tailed). All statistical analyses were performed using SPSS 13.0 (SPSS, Inc., Chicago, IL, USA) or the GraphPad Prism 5.0 (GraphPad Software, Inc., La Jolla CA, USA) software packages. P<0.05 was considered to indicate a statistically significant difference.

Results

Expression of HOTAIR is upregulated in human gastric cancer tissues

RT-qPCR was used to examine the expression of HOTAIR in 30 paired ovarian cancer samples and adjacent, histologically normal tissues. Compared with the adjacent, histologically normal tissues, the expression of HOTAIR was significantly enhanced (Fig. 1A). Furthermore, the upregulation of HOTAIR was positively correlated with increased tumor size and distant metastasis (Fig. 1B and C). These data indicated that the upregulation of HOTAIR was involved in ovarian carcinogenesis or distant metastasis.

Figure 1 Expression levels of HOTAIR in ovarian cancer tissues. (A) Levels of HOTAIR in ovarian cancer tissues were higher, compared with those in non-tumor normal tissues. (B) Upregulated expression of HOTAIR was correlated with increased tumor size. (C) Expression of HOTAIR was significantly increased in distant metastatic tumors, compared with non-distant metastatic tumors. The data are presented as the mean ± standard error of the mean. *P<0.05 and **P<0.01, vs. control. HOTAIR, HOX transcript antisense intergenic RNA.

Manipulation of HOTAIR levels in gastric cancer cells

The expression level of HOTAIR was also examined in different ovarian cancer cells, including SKOV3, A2780 and CP70 cells, as well as IOSE and EOC cells (HeyC2). As shown in Fig. 2A, the level of HOTAIR in the ovarian cancer cells was significantly increased, compared with the normal human EOCs. To examine the role of HOTAIR on cell viability, a pCDNA/HOTAIR vector was transfected into HeyC2 cells. As shown in Fig. 2B, the levels of HOTAIR were increased by >108-fold in the HeyC2 cells transfected with the pHOTAIR vector, compared with the pcDNA vector. siRNAs targeting HOTAIR were also transfected into HeyC2 cells. At 48 h post-transfection, the level of HOTAIR was inhibited by 82% in the si-HOTAIR2 group, which was identified as the most effective siRNA targeting HOTAIR (Fig. 2C).

Figure 2 Expression levels of HOTAIR in ovarian cancer cells in vitro. (A) Results from RT-qPCR demonstrating the expression levels of HOTAIR in ovarian cancer cell lines (SKOV3, A2780 and CP70), compared with human ovarian immortalized nontumorigenic ovarian surface epithelial cells (HeyC2). RT-qPCR analyses were performed to determine the expression levels of HOTAIR following treatment of HeyC2 cells with (B) pHOTAIR/pcDNA or (C) siRNAs targeting HOTAIR. Data are presented as the mean ± standard error of the mean (n=3). *P<0.05, **P<0.01, ***P<0.001 vs. control. HOTAIR, HOX transcript antisense intergenic RNA; si, small intefering; NC, negative control.

Knockdown of HOTAIR reduces cell viability and promotes cell apoptosis

In order to determine the effect of HOTAIR on cell viability, HeyC2 cells were transfected with pHOTAIR or pcDNA for 24, 48 or 72 h. As shown in Fig. 3A, the upregulation of HOTAIR enhanced cell viability in the HeyC2 cells by 20 and 30% at 48 and 72 h, respectively, whereas the downregulation of HOTAIR decreased cell viability by 25 and 30% at 48 and 72 h, respectively (Fig. 3B). The effect of HOTAIR on cell apoptosis was also examined. The results showed that the inhibition of HOTAIR enhanced cell apoptosis by almost 4-fold, compared with the NC in HeyC2 cells (14.32±0.78, vs. 3.56±0.32%; Fig. 3C). Cell morphology was also examined using Hoechst 33342 staining. As shown in Fig. 3D, the number of apoptotic cells increased in the HeyC2 cells transfected with siRNA targeting HOTAIR. These data suggested that HOTAIR modulated HeyC2 cell viability and that the downregulation of HOTAIR induced HeyC2 cell apoptosis.

Figure 3 Knockdown of HOTAIR reduces cell growth and promotes cell apoptosis. The HeyC2 cells were transfected with pHOTAIR or pcDNA for 24, 48 or 72 h. (A) Upregulation of HOTAIR enhanced cell viability in HeyC2 cells by 20 and 30% at 48 and 72 h, respectively. (B) Downregulation of HOTAIR decreased cell viability by 25 and 30% at 48 and 72 h, respectively. (C) Inhibition of HOTAIR enhanced cell apoptosis by almost 4-fold, compared with the NC in HeyC2 cells, determined using an Annexin V and PI kit. (D) Numbers of apoptotic cells increased in the HeyC2 cells transfected with siRNA targeting HOTAIR, as examined by Hoechst 33342 staining (magnification, ×40). The white arrow indicate apoptotic cells. Data are presented as the mean ± standard error of the mean (n=3). *P<0.05; and **P<0.01, vs. control. HOTAIR, HOX transcript antisense intergenic RNA; si, small interfering; NC, negative control; PI, propidium iodide; OD, optical density.

HOTAIR is a target of miR-373 and positively regulates the expression of Rab22a

Previous studies have indicated that miR-331-3p and miR-124 can directly bind to the 3′UTR of HOTAIR (14). According to bioinformatics analysis, a novel miRNA binding site, miR-373 was identified in the present study. The 3′UTR of HOTAIR was cloned downstream of the luciferase gene and termed RLuc-HOTAIR (Fig. 4A), which was transfected into cells together with the miRNA-373-coding plasmid. As shown in Fig. 4B, cotransfection of RLuc-HOTAIR and pllmiR-373 significantly reduced luciferase activity, suggesting that HOTAIR was a target gene of miR-373. miR-373 has been reported to inhibit the expression of Rab22a, a member of the Rab family of small GTPases (15). Thus, the present study further examined the effects of miR-373 on the protein expression of Rab22a. The HeyC2 cells were transfected with miR-373 mimic or NC. Overexpression of miR-373 significantly suppressed the expression of Rab22a, while inhibition of miR-373 markedly enhanced the protein expression of Rab22a (Fig. 4C and 4D).

Figure 4 HOTAIR is a target of miR-373 and enhances protein levels of Rab22a. (A) A luciferase reporter assay demonstrated that HOTAIR was a direct target of miR-373. (B) Transfection with pHOTAIR significantly increased the RLU of Rab22a. (C) Transfection with miR-373 mimics significantly suppressed the protein level of Rab22a. (D) Transfection of miR-373 inhibitor enhanced the protein expression of Rab22a. Data are presented as the mean ± standard error of the mean (n=3). **P<0.01, vs. control. HOTAIR, HOX transcript antisense intergenic RNA; miR, microRNA; NC, negative control. RLU, relative luciferase units.

miR-373 suppresses ovarian cancer cell proliferation

To investigate whether HOTAIR served as an endogenous sink for target miRNAs, the level of HOTAIR was compared with the level of miR-373. RT-qPCR analysis demonstrated that the level of miR-373 was significantly reduced in the 30 pairs of advanced ovarian cancer tissue (Fig. 5A). To determine the effect of miR-373 on ovarian cancer cell proliferation, miR-373 was overexpressed in HeyC2 cells. MTT and colony-formation assays were then performed. As shown in Fig. 5B, cell viability was decreased in the miR-373-overexpressing cells. In addition, the colony-formation capacity was significantly reduced following transfection with the pllmiR-373 vector (Fig. 5C). It was also found that transfection with the miR-373 mimics significantly inhibited HeyC2 cell migration (Fig. 5D). These data indicated that the upregulation of miR-373 significantly reduced ovarian cancer cell proliferation and migration, which were in accordance with the results obtained following HOTAIR knockdown in the HeyC2 cells.

Figure 5 Expression levels of HOTAIR and Rab22a are positively correlated, and linked to the expression of miR-373. (A) Reverse transcription-quantitative polymerase chain reaction analysis demonstrated that the levels of miR-373 were significantly reduced in the 30 pairs of advanced ovarian cancer tissue samples. (B) Dimethyl thiazolyl diphenyl tetrazolium, (C) colony forming (magnification, ×10) and (D) cell migration assays were performed to analyze miR-373-transfected HeyC2 cell proliferation. *P<0.05 and **P<0.01. HOTAIR, HOX transcript antisense intergenic RNA; miR, microRNA; NC, negative control; OD, optical density.

Discussion

LncRNAs are >200 nucleotides in length with limited protein-coding capacity. In disease or developmental stages, lncRNAs are often differentially expressed, indicating the importance of lncRNAs in disease pathology (16–18). The importance of the dysregulation of lncRNAs has been increasingly reported. For example, through regulating gene expression, lncRNAs are widely involved in uncontrolled tumor growth (19–21). Thus, elucidating the role of lncRNAs may improve current understanding of tumor regulatory networks.

HOTAIR is an lnRNA, which is defined at the HOXD locus. RNA fluorescence in situ hybridization assays have demonstrated that HOTAIR locates to the nucleus and the cytoplasm (9). A previous study suggested that, through suppressing polycomb repressive complex 2 (PRC2), HOTAIR in the nucleus can widely alter gene expression in the whole genome (9). For example, through the trimethylation of histone H3 lysine-27 of the HOXC locus, HOTAIR retargets PRC2 and inhibits several metastasis suppressor genes, thereby promoting the metastasis of breast cancer cells (8). In the cytoplasm, HOTAIR can act as a scaffold to induce ubiquitin-mediated proteolysis (22). In the present study, the levels of HOTAIR in ovarian cancer tissue samples and adjacent non-tumor tissue samples were analyzed. It was found that HOTAIR was markedly increased in ovarian cancer. In accordance with the above-mentioned studies, the upregulation of HOTAIR was positively correlated with ovarian tumor size and metastasis. In addition, the results of the present study revealed that the upregulation of HOTAIR significantly enhanced ovarian epithelial cell proliferation, whereas the inhibition of HOTAIR repressed cell viability and promoted cell apoptosis. These data indicated that HOTAIR was important in controlling ovarian carcinogenesis.

Previous studies have indicated that lncRNAs may be involved in the ceRNA regulatory network (4). In this case, lncRNAs function as ceRNAs to sponge miRNAs, thereby de-repressing miRNA targets and affecting post-transcriptional regulation. In human gastric cancer tissues, HOTAIR is also significantly upregulated; as a sink for miR-331-3p, HOTAIR positively regulates the expression of human epithelial growth factor receptor 2 and facilitates the malignant phenotype of the gastric tumor (14). In the present study, whether HOTAIR functions as a ceRNA by directly interacting with miRNAs was investigated. It was found that miR-373 was able to bind with the 3′UTR of HOTAIR, and the RLuc-HOTAIR reporter assay suggested that miR-373 had a translational repressive role on the expression of HOTAIR. As an miRNA sponge, the level of HOTAIR is expected to be higher or comparable to the expression level of miRNA. A previous study suggested the reduced expression of miR-373 in ovarian cancer (15), therefore, a negative correlation was identified between the levels of HOTAIR and miR-373. In addition, the present study found that the overexpression of miR-373 significantly repressed ovarian cancer proliferation, which mimicked the effects of HOTAIR inhibition in ovarian cancer cells. These results indicated that HOTAIR may be correlated with miRNA to be involved in ovarian carcinogenesis.

Rab22a is a small GTPase, which belongs to the Rab family endocytic pathway (23). Through the recruitment of Rabex-5, Rab22 interacts with Rabex-5 and Rab5, which further modulates cancer cell proliferation through the integrin-mediated signaling pathway (24–26). The activation of Rab5 has also been found to enhance transforming growth factor-β signaling (27), leading to the epithelial-mesenchymal transition process in the progression of cancer. The overexpression of miR-373 reduced the level of Rab22a, whereas the reduction in miR-373 increased the expression of Rab22a. The present study also identified a positive correlation between HOTAIR and Rab22a, indicating the interactive role of miR-373 and HOTAIR in ovarian cancer.

In the present study, specific crosstalk between the lncRNA HOTAIR and miR-373 was identified. Consistent with HOTAIR sequestration of miR-373, it was found that the inhibition of HOTAIR reduced the expression of Rab22a, whereas the overexpression of HOTAIR enhanced the protein level of Rab22a. These data suggested that the HOTAIR lncRNA positively regulated the protein expression of Rab22a through competition for miR-373 binding. In conclusion, the positive regulation between HOTAIR and Rab22a may be partially attributed to the ceRNA regulatory network via miR-373, indicating a potential therapeutic target for ovarian cancer.

References