Long non-coding RNA HOXA-AS2 promotes proliferation and invasion of breast cancer by acting as a miR-520c-3p sponge

The long non-coding RNA (lncRNA) HOXA cluster antisense RNA2 (HOXA-AS2) has recently been shown to be dysregulated and involved in the progression of several cancers. However, the biological role and clinical significance of HOXA-AS2 in the carcinogenesis of breast cancer are still unclear. In the present study, we found that HOXA-AS2 was up-regulated in human breast cancer tissues and cell lines and associated with clinicopathological characteristics. Silencing of HOXA-AS2 inhibited the progression of breast cancer cells in vitro and in vivo. Furthermore, microarray profiling indicated that HOXA-AS2 serves as an endogenous sponge by directly binding to miR-520c-3p and down-regulating miR-520c-3p expression. We demonstrated that HOXA-AS2 controls the expression of miR-520c-3p target genes, TGFBR2 and RELA, in breast cancer cells. Therefore, our study may provide a better understanding of the pathogenesis of breast cancer and suggests that HOXA-AS2 may be a potential prognostic and therapeutic target in breast cancer.


INTRODUCTION
Breast cancer is a malignant tumor originating from breast tissue and is the most common cancer in women worldwide [1,2]. Recently, the development of early diagnoses, radical surgery and adjuvant therapy has decreased the mortality rates of breast cancer [3][4][5]; however, it is still the most common gynecologic cancer. In China, the growing incidence of breast cancer threatens women's health and has gradually become a social problem [6]. Research has shown that many molecular triggers play significant roles in breast cancer development and chemotherapy resistance [7]. However, the molecular regulatory network of breast cancer is very complex, and the molecular mechanisms of breast cancer progression needed to be further explored.
In the human genome, only 1-2% of genes encode proteins [8], and the rest of the genome encodes a large amount of non-coding RNAs (ncRNA) [9].
LncRNAs are a class of RNA molecules that are longer than 200 nucleotides and cannot be translated into proteins [10]. Research has shown that lncRNAs are involved in a variety of biological processes, such as development [11], apoptosis [12,13], proliferation and differentiation [14], and carcinogenesis [15][16][17]. Recent studies have also found that lncRNAs are abnormally expressed in a variety of cancers, including breast cancer [18], lung cancer [19], hepatocellular carcinoma [20], and gastric cancer [21]. Therefore, lncRNAs are regarded as potential biomarkers and may lead to a new treatment era for various cancers [22,23]. HOXA cluster antisense RNA 2 (HOXA-AS2) is an lncRNA located between the HOXA3 and HOXA4 genes in the HOXA cluster. Recent studies indicated that HOXA-AS2 represses apoptosis in NB4 promyelocytic leukemia cells treated with trans retinoic acid [24], promotes the proliferation of gastric cancer via P21/PLK3/DDIT3 [25] and accelerates tumorigenesis of hepatocellular carcinoma [26]. However, the role of HOXA-AS2 in breast cancer development, invasion and metastasis remains unknown.
Many studies have shown that lncRNAs can serve as a competing endogenous RNAs (ceRNAs) to regulate microRNAs (miRNAs) [27,28]. CeRNAs play a posttranscriptional regulatory role in miRNA distribution on www.impactjournals.com/oncotarget/ Oncotarget, 2017, Vol. 8, (No. 28), pp: 46090-46103 Research Paper their targets. MiRNAs are a class of non-coding RNA of approximately 20 nucleotides and regulate gene and protein expression by targeting the 3' untranslated region (UTR) of mRNAs [29,30]. Previous studies have shown that miRNAs are involved in post-transcriptional regulation of various biological processes [31][32][33]. In addition, miRNAs serve as tumor suppressor genes or oncogenes and play important roles in the occurrence and development of human cancers [31,34]. In this study, we aimed to explore the ceRNA mechanism of HOXA-AS2 though miR-520c-3p and revealed the functional relevance of miR-520c-3p and HOXA-AS2 in breast cancer.

LncRNA-HOXA-AS2 was up-regulated in human breast cancer tissues and cell lines
The expression level of HOXA-AS2 was detected by qRT-PCR in 38 pairs of breast cancer tissues (Tumor) compared with corresponding adjacent normal tissues (Normal), and we found that HOXA-AS2 expression was significantly up-regulated in tumor tissues compared with normal tissues ( Figure 1A). Next, we found that high expression of HOXA-AS2 was corrected with distant metastasis (n = 16) (*P < 0.05, Figure 1B). In addition, the relationships between HOXA-AS2 expression and clinical characteristics in 38 patients with breast cancer were analyzed, and we found that expression was significantly associated with invasion (P = 0.0005), lymphatic metastasis (P < 0.0001), distant metastasis (P < 0.0001), and TNM stage (P < 0.0001) but not related to age (Table 1). Further analysis showed that HOXA-AS2 was significantly higher in breast cancer cell (MDA-MB-231, MDA-MB-453, and MCF-7) lines than that of the non-metastatic human mammary epithelial cell line MCF-10A (P < 0.05, Figure 1C). Moreover, high HOXA-AS2 expression was associated with poor overall survival compared to that of patients with low HOXA-AS2 (P < 0.05, Figure 1D). These results suggested that HOXA-AS2 might be an oncogene in breast cancer.

Silencing of HOXA-AS2 expression suppresses tumor growth and inhibits Ki-67 expression in vivo
To investigate the effects of siRNA-HOXA-AS2 on the growth of breast cancer in vivo, MCF-7 cells were transfected with control or HOXA-AS2 siRNAs and were subcutaneously injected into athymic mice. Our results showed that tumors from the siRNA-HOXA-AS2transfected MCF-7 cells grew slower than that of the control group during the whole tumor growth period (P < 0.05, Figure 4A). Five weeks after inoculation, the tumors were excised, and photographs of the excised tumors were obtained ( Figure 4B). The average weight of tumors derived from siRNA-HOXA-AS2-transfected MCF-7 cells was significantly smaller than those of control groups (P < 0.05, Figure 4C). Furthermore, the HOXA-AS2 and Ki-67 levels were detected by qRT-PCR and immunostaining. The results indicated that HOXA-AS2 expression in tumors of the siRNA-HOXA-AS2-treated group was significantly lower than the control group (P < 0.05, Figure 4D). Additionally, Ki-67 expression was significantly decreased in tumors of the siRNA-HOXA-AS2-treated group compared with the control group (P < 0.05, Figure 4E, 4F). Therefore, our results demonstrated that silencing of HOXA-AS2 expression can inhibit proliferation of breast cancer cells in vivo.

MiR-520c-3p was regulated by HOXA-AS2 in breast cancer
To investigate the effect of HOXA-AS2 on the expression of miRNAs, MCF-7 and MDA-MB-453 cells were transfected with HOXA-AS2 siRNAs and controls for 48 hrs. Hierarchical clustering of differentially expressed miRNAs is shown in Figure 5A. Then, starBase v2.0 software was used to predict the miRNAs that interacted with HOXA-AS2 ( Figure 5B). MiR-520c-3p was identified in both array results and bioinformatics analysis results; thus, it was chosen for further study. To explore the regulatory relationship between HOXA-AS2 and miR-520c-3p, MCF-7 and MDA-MB-453 cells were transfected with HOXA-AS2 siRNAs and controls, and qRT-PCR was used to detect the expression level of miR-520c-3p. The results showed that silencing of HOXA-AS2 increased the expression level of miR-520c-3p in MCF-7 and MDA-MB-453 cells (P < 0.05, Figure 5C). Meanwhile, miR-520c-3p also inhibited the expression of HOXA-AS2 in MCF-7 and MDA-MB-453 cells (P < 0.05, Figure 5D). These results indicate that HOXA-AS2 and miR-520c-3p inhibit each other in breast cancer. As shown in Figure 5E, there was a putative miR-520c-3p target site in the HOXA-AS2 sequence. To confirm this interaction, the wild type sequence of HOXA-AS2 or its mutant sequence was subcloned into the pMIR luciferase reporter. MCF-7 and MDA-MB-453 cells were co-transfected with the reporter plasmid (or the corresponding mutant reporter) and miR-520c-3p. Our results indicated that the relative luciferase activity of the pMIR-HOXA-AS2-wt construct was significantly decreased but was abolished in HOXA-AS2-mt-transfected MCF-7 (P < 0.05, Figure  5F) and MDA-MB-453 cells (P < 0.05, Figure 5G). In addition, the relationships between miR-520c-3p expression and clinical characteristics of 38 patients with breast cancer were analyzed, and we found that miR-520c-3p was significantly associated with invasion (P = 0.018), lymphatic metastasis (P = 0.0003), distant metastasis (P = 0.0016), and TNM stage (P = 0.0003), which has a negative relationship with HOXA-AS2 (Table 2).

DISCUSSION
As the most common gynecological cancer, which has a low 5-year survival rate in patients with metastasis, breast cancer research is urgently needed to improve early diagnosis and treatment in human populations [35]. Non-coding RNA (ncRNA) has been regarded as a mechanism potentially driving breast cancer progression and metastasis and may serve as a new and more effective therapy in breast cancer [36,37]. Among these ncRNAs, lncRNAs, which have extensive functions in cancer development [38,39], have emerged as a new hot topic in early diagnosis and molecular targeted therapy [40]. At present, the relationship between lncRNA HOXA-AS2 and breast cancer is unclear. In this study, our results showed that HOXA-AS2 was up-regulated in breast cancer and associated with invasion, lymphatic metastasis, distant metastasis, TNM stage, and postoperative survival. Functional studies indicated a role of HOXA-AS2 in cell proliferation, apoptosis, migration and invasion of breast cancer cells. We also found that HOXA-AS2 silencing dramatically suppressed cell proliferation in vivo. These results indicated that HOXA-AS2 may be a good early diagnosis biomarker and therapeutic target in breast cancer. Ki-67, a marker of proliferation, is present in all proliferating tumor cells and is expressed in all stages of the cell cycle except G0 phase [41]. Ki-67 measurement is a standard method to evaluate the proliferative activity of tumor cells [42]. Several studies have shown that highly expressed Ki-67 is related to the development and clinicopathological features of breast cancer [43,44]. In our study, we found that silencing of HOXA-AS2 inhibited Ki-67 protein expression in breast cancer cells, suggesting that HOXA-AS2 is closely involved in the proliferation of breast cancer and may be a critical marker along with Ki-67 in breast cancer diagnosis.
Previous studies have indicated that miRNAs have important biological functions in breast cancer, such as cell proliferation, apoptosis, differentiation, inflammation, and metastasis [45,46]. In our study, we found that miR-520c-3p showed low expression and plays important roles in human breast cancer, which indicates it may be a potential biomarker in the diagnosis and treatment of breast cancer. Our findings enhanced the knowledge of miRNAs in this field. Studies have also indicated that lncRNAs potentially affect target gene levels as ceRNAs through miRNA response elements (MREs) [27]. At present, a number of lncRNAs that serve as ceRNAs have been shown to be involved in the occurrence and development of various diseases [47,48]. In our study, we performed a microarray to screen the HOXA-AS2-regulated miRNAs in breast cancer. StarBase v2.0 was also used to predict the binding sites between HOXA-AS2 and the regulated miRNAs. Then, dual-luciferase reporter assays confirmed that miR-520c-3p, screened by microarray and bioinformatics analyses, was a HOXA-AS2-regulated miRNA in breast cancer cells. Functional studies demonstrated that HOXA-AS2 alters proliferation and invasion of breast cancer by regulating miR-520c-3p.
In the present study, we also explored the mechanism of the HOXA-AS2/miR-520c-3p axis in breast cancer and detected the influences of HOXA-AS2 on the targets (TGFBR2 and RELA) of miR-520c-3p. We found that HOXA-AS2 controls the expression of TGFBR2 and RELA by acting as a miR-520c-3p sponge. A previous study indicated that TGF-β has a dual effect (inhibiting early-stage tumor growth but promoting tumor metastasis) in cancer development [49]. TGF-β2, which belongs to the TGF-β family, contributes to the recruitment and activation of SMAD family members that regulate gene expression [50]. Our research showed that HOXA-AS2/miR-520c-3p may participate in breast cancer tumorigenesis through the TGF-β signaling pathway. Another target of miR-520c-3p, RELA (p65, NF-κB3), is a member of the transcription factor nuclear factor kappa B (NF-κB) family. RELA was shown to regulate the growth and malignancy of various tumors by regulating pro-inflammatory and pro-survival genes [51]. Our results demonstrated that the HOXA-AS2/miR-520c-3p axis controls the expression of RELA, suggesting a role for the HOXA-AS2/miR-520c-3p in breast cancer through the NF-κB signaling pathway.
In conclusion, our study found the following: a) HOXA-AS2 was up-regulated in breast tissues and cell lines. b) HOXA-AS2 expression was associated with invasion, lymphatic metastasis, distant metastasis, TNM stage, and postoperative survival of breast cancer. c) Silencing of HOXA-AS2 significantly suppressed carcinogenesis of breast cancer. d) HOXA-AS2 acts as a molecular sponge for miR-520c-3p and regulates its targets. The HOXA-AS2/miR-520c-3p regulatory network may shed light on tumorigenesis in breast cancer and may contribute to the development of novel diagnostic and treatment approaches for breast cancer.

Clinical specimens
A total of 38 patients who were diagnosed with breast cancer and had undergone a mammectomy between 2009 and 2015 in the Affiliated Tumor Hospital of Harbin Medical University were enrolled in the study. No patients received radiotherapy or chemotherapy before surgery. The study was approved by the Ethics Committee of the Affiliated Tumor Hospital of Harbin Medical University. Written informed consent was obtained from all patients in this study. The histological diagnosis of breast cancer was based on the World Health Organization (WHO) criteria. All samples (cancer and paired noncancerous tissues) were stored in liquid nitrogen with RNALater (Qiagen) immediately after resection and saved until RNA extraction.

Quantitative real-time reverse transcription PCR (qRT-PCR)
Total RNA was isolated from tissues and cells using TRIzol reagent (Invitrogen, CA, USA) according to the manufacturer's instructions. First-strand cDNAs were reverse transcribed from total RNA using the RevertAid First Strand cDNA Synthesis kit (Thermo Fisher). Quantitative PCR was performed using the SYBR-Green

Western blot assay
Western blot assays were performed to determine the protein levels of TGFBR2 and RELA. The treated MCF-7 and MDA-MB-453 cells were lysed by Radio Immunoprecipitation Assay (RIPA) buffer (Thermo Scientific, Rockford, IL, USA) containing a protease inhibitor cocktail (Roche Diagnostics). The concentrations were detected using a BCA Protein Assay kit (Thermo Fisher Scientific., Rockford, IL, USA). Proteins were separated by 10% SDS/PAGE and transferred onto a polyvinylidene fluoride membrane (PVDF, Millipore, Billerica, MA). The PVDF membranes were blocked and incubated with primary antibody overnight at 4°C. Then, the PVDF membranes were incubated with HRPconjugated secondary antibody. The results were obtained using an ECL detection kit (Millipore Corp.

Luciferase assay
MCF-7 and MDA-MB-453 cells (5 × 10 4 cells/well) were seeded in 24-well plates and co-transfected with miR-520c-3p and wild type or mutant HOXA-AS2, along with a Renilla plasmid (RL-SV40) using Lipofectamine 3000 (Invitrogen). The pRL-CMV Renilla vector served as an internal control. Cell extracts were harvested 48 hrs post-transfection, and a Dual-Luciferase Reporter Assay System (Promega) was used to measure the reporter gene activities according to the manufacturer's instructions. All experiments were performed in triplicate and repeated twice.

Proliferation assay
Cell proliferation was measured by cholecystokinin octapeptide (CCK-8) assays. The treated MCF-7 and MDA-MB-453 cells (3 × 10 3 cells/well) were seeded in a 96-well plates and incubated at 37°C for 24, 48, and 72 hrs. Then, 10 μL CCK-8 (Dojindo Laboratories, Kumamoto, Japan) was added to each well at the indicated time point. After 4 hrs, an Elx800 Reader (Bio-Tek Instruments Inc., Winooski, VT, USA) was used to record the absorbance value at 570 nm. For colony formation assays, the treated MCF-7 and MDA-MB-453 cells (200 cells/well) were seeded in six-well plates and incubated at 37°C in complete medium for 14 days (culture medium was changed every three days). Cells were fixed and stained with 0.1% crystal violet. The number of colonies was counted.

Flow cytometric analysis
For cell cycle assays, the treated MCF-7 and MDA-MB-453 cells were obtained, washed with PBS three times, fixed with 80% ethanol, subsequently incubated with RNase A (0.25 mg/ml, Sigma) for 30 min at 37°C, and then incubated with 20 μg/ml propidium iodide (KeyGen, Nanjing, China) for 20 min at room temperature. The cell cycle was analyzed by flow cytometry using a FACSCalibur Flow Cytometer (BD Biosciences, San Jose, CA, USA). The experiments were performed in triplicate. For cell apoptosis assays, the extent of apoptosis was measured using the Annexin V/7-AAD Apoptosis Detection Kit (Southern Biotechnology, Birmingham, Al, USA) according to the manufacturer's instructions. Cells (1 × 10 6 cells/mL) were digested, centrifuged, washed with cold PBS twice, and re-suspended in 100 μL of 1× binding buffer. Then, they were double stained with Annexin V and 7-AAD and incubated at room temperature in the dark for 15 mins. The stained cells were analyzed by flow cytometry (BD Biosciences) using a FACSCalibur Flow Cytometer (BD Biosciences, San Jose, CA, USA). The experiments were performed in triplicate.

Wound-healing assay
The wound healing assay was used to study cell migration. The treated MCF-7 and MDA-MB-453 cells (1 × 10 6 cells) were seeded in 6-well plates and cultured in complete medium for 12 hrs at 37°C in a 5% CO 2 incubator. A pipette tip was used to make a straight scratch, and the images of the scratch were obtained as baseline. Cells were cultured at 37°C for 48 hrs. The suspended cells were gently washed off twice with PBS. Photographs were obtained every 12 hrs for the next 36 hrs.

Transwell assay
The cell invasive and migratory abilities were evaluated using transwell assays. For migration assays, 200 mL of treated MCF-7 and MDA-MB-453 cells (5 ×10 5 ) with serum-free medium was seeded in the top 24well Transwell chamber (8-μm pore; Corning, Painted Post, NY, USA) of the inserts, and 600 µL of complete medium was added to the lower chamber. After 24 hrs of incubation, the migratory cells were fixed, stained in 0.1% crystal violet solution, and counted under a confocal microscope. For invasion assays, the invasive cells were detected by Transwell Matrigel (BD Biosciences, San Diego, CA, USA) invasion assays. Matrigel (BD Biosciences, San Diego, CA, USA) was pre-coated onto the upper chambers. Other experimental procedures were similar to the migration assay.

Tumor formation in nude mice
This study was approved by the Institutional Committee for Animal Research and performed according to the national guidelines for the care and use of laboratory animals. MCF-7 cells were transfected with HOXA-AS2 siRNAs and controls, and cells (1×10 6 cells/mouse, 0.2 ml) were subcutaneously injected into the armpit of 6-week-old BALB/c athymic nude mice. Tumor volume and weight were measured at 0, 7, 14, 21, 28, and 35 days. The mice were euthanized 5 weeks after injection, and photographs of excised tumors were obtained. The excised tumors were then used into further analyses.

Immunohistochemistry
Immunohistochemistry was performed on paraformaldehyde-fixed paraffin sections. The TMA blocks were then cut into 4 µM sections for immunostaining. The sections were incubated with Ki-67 (#9027; Cell Signaling) antibody overnight at 4°C. Histostain-Plus 3rd Gen IHC Detection Kit (Invitrogen Co., San Diego, CA) was applied for 30 mins to visualize the positive signals.

MiRNA microarrays
MCF-7 and MDA-MB-453 cells were transfected with HOXA-AS2 siRNAs and control for 48 hrs. An Agilent human miRNA microarray (V2) was used to determine the expression levels of miRNAs according to the manufacturer's instructions. First, total RNA was extracted using TRIzol (Invitrogen) and dephosphorylated with calf intestine alkaline phosphatase (Amersham). The dephosphorylated RNA was denatured by DMSO (Sigma). Second, the labeled miRNAs were re-suspended in hybridization buffer (Agilent) and hybridized with a microarray (Agilent Technologies). Finally, images were analyzed using Agilent Feature Extraction software (version A.9.5.3; Agilent Technologies).

Statistical analysis
Differences between groups were analyzed by Student's t-test using SPSS 21.0 software (SPSS Inc., Chicago, IL, USA). Survival curves were evaluated by the Kaplan-Meier method and compared by the log-rank test. P < 0.05 was considered statistically significantly. Data are shown as the mean ± standard deviation (SD), and all experiments were performed in triplicate.