This article is mentioned in:

Introduction

Breast cancer (BC) is a common cancer and leading cause of cancer-associated mortality in women and is the second most lethal cancer worldwide, following lung adenocarcinoma (1,2). In the last decades, approaches used for the treatment of BC have advanced, including surgery, radiotherapy, chemotherapy and endocrine therapy, leading to a significant reduction of the rates of premature mortality (3). However, the molecular mechanisms responsible for BC development and progression remain unclear. There are diverse BC subtypes, and clear and definitive targets are limited, resulting in ambiguity of the pathophysiology of BC (4). It was demonstrated that, in response to therapy, distinct signaling pathways may be activated in different subtypes of BC (5,6). The features of tumorigenesis include uncontrolled cell proliferation, and high rates of metastasis and stemness (7,8). Therefore, determining the mechanisms of dysregulation of cell proliferation, metastasis and stemness is urgently required to identify key regulators in the development of BC and improve BC therapy.

Long non-coding RNAs (lncRNAs) are a family of transcripts, with lengths >200 nucleotides, which contain no open reading frames, and these RNAs are less abundant and more variable among tissues compared with mRNA expression (9). lncRNAs have been identified to serve key roles in the regulation of a broad array of cancer processes, including proliferation (10), apoptosis (11), metastasis (12) and drug resistance (13). lncRNAs may function through a wide range of mechanisms (14), including serving as signals, decoys, guides and scaffolds to modulate the transcriptional or post-transcriptional regulation of gene expression in multiple cancer types. Serving as guides or molecular scaffolds, lncRNAs may recruit co-regulators of transcription to a specific DNA region or raise enzymes with chromatin-modifying activity to form ribonucleoprotein complexes, leading to the bridging of regulatory proteins and regulation of transcription (15). In addition, functioning as decoys, lncRNAs may bind directly with microRNAs (miRNAs) or proteins and thus modulate the functions of key regulators (16). Therefore, lncRNAs may serve an oncogenic or a tumor-suppressive role, making them good prognostic biomarkers and therapeutic targets.

lncRNA FOXD2 adjacent opposite strand RNA 1 (FOXD2-AS1; accession no. NR_026878), is located on chromosome 1p33 and consists of 2,527 nucleotides (17). It was firstly identified that FOXD2-AS1 expression is increased in gastric cancer (18,19). Previous studies demonstrated that FOXD2-AS1 is critical for cell proliferation, apoptotic cell death, invasion, migration and drug-resistance in non-small cell lung cancer (20), esophageal squamous cell carcinoma (21), nasopharyngeal carcinoma carcinogenesis (17), colorectal cancer (22) and bladder cancer (23). However, whether FOXD2-AS1 is dysregulated in BC and its underlying mechanisms remain unclear.

The present study aimed to identify the possible role of FOXD2-AS1 in the proliferation, migration, invasion and stemness in BC, and to clarify the clinical features of FOXD2-AS1 in BC. In the present study, an increased expression of FOXD2-AS1 was identified in BC tissues and cells. Critical roles of FOXD2-AS1 in the regulation of proliferation, epithelial-mesenchymal transition and stemness were identified. The results suggested that FOXD2-AS1 promoted BC malignancy and tumorigenesis by targeting the miRNA (miR)-150-5p/profilin 2 (PFN2) axis.

Materials and methods

Patients and tissue specimens

In total, 34 pairs of BC tissues and paired adjacent normal tissues were collected from the Department of Breast Surgery, Affiliated Cancer Hospital and Institute of Guangzhou Medical University (Guangzhou, China) between Jan 2012 and May 2012. The present study was approved by the Ethics Committee of the Affiliated Cancer Hospital and Institute of Guangzhou Medical University (approval no. ACHIGMU-2012-1-3-05). The specimens were immediately frozen in liquid nitrogen and subsequently stored at −80°C for further determination. The female patients did not receive any radiation or chemotherapy prior to operation, and patients with hypertension and diabetes mellitus were excluded in the present study. All tissue sections were reviewed by at least two experienced pathologists. The tumor, node, metastasis (TNM) stage was evaluated according to the American Joint Committee on Cancer (24). Written informed consent was obtained from all the enrolled patients. A 5-year follow-up was performed and overall survival was defined as the length of time between surgery and mortality or the last follow-up (if mortality did not occur).

Cell lines, culture and transfection

A human normal breast epithelial cell line (MCF-10A) and human breast cancer cell lines (MCF-7, MDA-MB-231, MDA-MB-453 and MDA-MB-468) were purchased from The American Type Culture Collection (Manassas, VA, USA). The cell lines were incubated in RPMI-1640 medium (Corning, Inc., Corning, NY, USA) supplemented with 10% fetal bovine serum (FBS; Corning, Inc.) and cultured at 37°C with 5% CO2. Cells in the exponential phase were used in the experiments. Small interfering RNA (siRNA) targeting FOXD2-AS1 (cat. no. 4390771; 100 nM) and its scrambled control siRNA (cat. no. AM4636; 100 nM) (Thermo Fisher Scientific, Inc., Waltham, MA, USA) were employed to downregulate FOXD2-AS1. Short hairpin RNA (shRNA) targeting FOXD2-AS1 (100 nM) was prepared based on those siRNAs using pSilencer 3.1-H1 puro plasmids (Shanghai GenePharma Co. Ltd., Shanghai, China). Lentivirus carrying shRNA targeting FOXD2-AS1 (1×107 IFU/ml) was constructed using pre-packaged lentivirus (Shanghai GenePharma Co. Ltd.). The miR-150-5p mimics (cat. no. 4464084; 100 nM), inhibitors (cat. no. 4464066; 100 nM) and scrambled control oligonucleotides (cat. nos. 4464059 and 4464076; 100 nM) were purchased from Thermo Fisher Scientific, Inc. Oligonucleotide and lentivirus transfection into cells were conducted using Lipofectamine® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. After 48 h of transfection, subsequent experiments were performed.

Cell proliferation

Cell proliferation was determined by Cell Counting Kit-8 (CCK-8; Beyotime Institute of Biotechnology, Haimen, China), according to the manufacturer's protocol. Following the treatment, cells were incubated in 10 µl CCK-8 solution for 1 h at 37°C. Finally, the absorbance at 450 nm was measured following the plate incubation at 37°C for 2 h (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Triplicate experiments were performed.

RNA extraction and reverse transcription-quantitative poly- merase chain reaction (RT-qPCR)

Tissues and cells were lysed and total RNA was extracted using TRIzol® (Thermo Fisher Scientific, Inc.) reagent, according to the manufacturer's protocol. RNA quality was measured using Nanodrop equipment (Thermo Fisher Scientific, Inc., Wilmington, DE, USA). For the detection of mRNA and lncRNA, 500 ng RNA was reverse transcribed into cDNA using a cDNA synthesis kit (Thermo Fisher Scientific, Inc.) according to the following conditions: 65°C for 5 min, 25°C for 10 min, 50°C for 15 min and 85°C for 5 min. The sequences of specific primers used for RT-qPCR were as follows: GAPDH, forward, 5′-GCGAGATC GCACTCATCATCT-3′ and reverse, 5′-TCAGTGGTGGACCT GACC-3′; FOXD2-AS1, forward, 5′-TGGACCTAGCTGCAGC TCCA-3′ and reverse, 5′-AGTTGAAGGTGCACACACTG-3′; E-cadherin, forward, 5′-CTGCTGCAGGTCTCCTCTTG-3′ and reverse, 5′-TGTCGACCGGTGCAATCTTC-3; Vimentin, forward, 5′-AAGGCGAGGAGAGCAGGATT-3′ and reverse, 5′-GGTCATCGTGATGCTGAGAAG-3′; N-cadherin, forward, 5′-GTGCCATTAGCCAAGGGAATTCAGC-3′ and reverse, 5′-GCGTTCCTGTTCCACTCATAGGAG-3′; Nanog, forward, 5′-GGTCCCAGTCAAGAAACAGA-3′ and reverse, 5′-GAG GTTCAGGATGTTGGAGA-3′; Oct4, forward, 5′-CCCGAAAGAGAAAGCGAACC-3′ and reverse, 5′-GCAGCCTCAAAATCCTCTCG-3′; and Sox2, forward, 5′-CATGTCCCAGCACTA CCAGA-3′ and reverse, 5′-TTTGAGCGTACCGGGTTTTC-3′. Reactions of RT-qPCR were conducted on an ABI 7500 real-time PCR System (Applied Biosystems; Thermo Fisher Scientific, Inc.). GAPDH was used as an internal control for normalization.

For the detection of miRNA, the qScript miRNA cDNA Synthesis kit (Quantabio, Beverly, MA, USA) was employed to synthesize cDNA, according to the following conditions: 50°C for 60 min and 85°C for 5 sec. miRNA RT-qPCR was performed using a miScript SYBR Green PCR kit (Qiagen GmbH, Hilden, Germany) according to the manufacturer's protocol. The sequences of specific primers used for RT-qPCR were as follows: U6, forward, 5′-CTCGCTT CGGCAGCACA-3′; reverse, 5′-AACGCTTCACGAATTTGCGT-3′; miR-150-5p, forward, 5′-TCTCCCAACCCTTGTACCAGTG-3′; reverse, 5′-CTCAACTGGTGTCGTGGTA-3′. U6 was used as an internal control for normalization. The level of target gene was calculated relative to internal control using the 2−∆∆Cq method (25). The PCR conditions were as follows: 95°C for 5 min, 40 cycles of 95°C for 20 sec and 62°C for 30 sec, followed by 72°C for 3 min.

Cytoplasmic and nuclear isolation of RNA

The location of lncRNA FOXD2-AS1 was measured using the Cytoplasmic and Nuclear RNA Purification kit (Norgen Biotek Corp., Thorold, ON, Canada) according to the manufacturer's protocol.

Luciferase reporter assay

The fragments of wild-type (WT) or mutated (MUT) 3′untranslated region (UTR) of FOXD2-AS1 and PFN2 were synthesized, inserted into the vector pGL3 (Shanghai GeneChem Co. Ltd., Shanghai, China), and subsequently miR-150-5p mimics and their respective control were transfected into 293T cells (American Type Culture Collection) using Lipofectamine® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.). The Renilla luciferase plasmid was used for normalization. The luciferase activity was measured using the Dual-Luciferase Reporter Assay System (Promega Corporation, Madison, WI, USA) according to the manufacturer's protocol. After 48 h of transfection relative luciferase activity was calculated by comparison with Renilla luciferase activity.

Sphere-formation assay

The formation of spheres was evaluated as previously described (26). BC cells were suspended and 5×104 cells/well were seeded into 6-well plates (Corning, Inc.). In total, 2 ml serum-free Dulbecco's modified Eagle's medium (DMEM; Gibco; Thermo Fisher Scientific, Inc.) was added to each well to re-suspend cells. The medium contained 4 U/l insulin, 20 mg/l epidermal growth factor and 20 mg/l human fibroblast growth factor. Sphere-formation was observed using a light microscope (magnification, ×400) and the number of spheroids was counted (Olympus Corporation, Tokyo, Japan).

Flow cytometry analysis

CD44 antigen (CD44)+/signal transducer CD24 (CD24)− cells percentage was determined using a Breast Cancer Stem Cell Isolation kit (R&D Systems Inc., Minneapolis, MN, USA) using flow cytometer analysis. Cells were trypsinized and resuspended to a density of 2×105 cells/well. Cells were incubated in PBS containing 2% FBS and human CD24 biotinylated antibody, human CD44 biotinylated antibody, human CD24 detection antibody and human CD44 detection antibody provided in the kit for 15 min at 2-8°C. A flow cytometer was used (BD Biosciences, Franklin Lakes, NJ, USA) and CD44+/CD24− cells percentage was measured (Kaluza 1.2 software; Beckman Coulter, Inc., Brea, CA, USA).

Bioinformatics analysis

The lncRNA-miRNA interaction prediction was performed using StarBase v 2.0 (http://starbase.sysu.edu.cn/starbase2). The miRNA-mRNA interaction prediction was conducted using TargetScanHuman 7.1 (http://www.targetscan.org/vert_71/).

Migration and invasion assays

A wound healing assay was conducted to evaluate the migration of the cells. 5×105 cells were plated in 6-well plates and when cells were grown to 90% confluence, a scratch was made through the monolayer using a sterile pipette tip. The floating cells were removed with PBS and subsequently, cells were cultured in serum-free medium. Images were captured at 0 and 24 h. The width of the scratches at 0 and 24 h was measured. The migration distance was calculated following the formula: Migration rate = migration distance/original distance. Relative migration distance was expressed as fold-change of the control.

A Matrigel assay was conducted to evaluate the invasion of the cells. 1×105 cells in 200 µl serum-free medium were added to the upper chamber of Transwell chambers (Corning, Inc.) were coated with DMEM-diluted Matrigel (BD Biosciences). In total, 800 µl medium containing 10% FBS was added to the lower chamber. Following incubation at 37°C for 48 h, the upper membrane surface was scraped using a cotton tip to remove non-invaded cells. Finally, the membrane was fixed with 4% paraformaldehyde for 30 min at room temperature and stained with 1% crystal violet for 15 min at room temperature. Staining was observed using a light microscope (magnification, ×400; Olympus Corporation) and the number of invaded cells was counted and expressed as the fold-change of the control.

Nude mice experiments

Xenograft mice were established as per the experimental protocols and the study was approved by the Affiliated Cancer Hospital and Institute of Guangzhou Medical University. The mice were housed in a specific pathogen-free animal laboratory under temperature (23±2°C) and humidity (55±5%) condition with a standard light cycle (12 h light/dark) and free access to food and water. In total, 12 female BALB/c nu/nu mice (4-weeks; weighing 16±2 g; purchased from The Animal Centre of Guangzhou Medical University) were randomly divided into two groups with six mice in each group. In total, 1×107 MCF-7 cells were suspended in physiological saline. An equal volume of Matrigel (Corning, Inc.) was added to the cell suspension. Each mouse was subcutaneously injected with 150 µl mixture to generate a tumor. 17b-Estradiol pellets (0.72 mg; 60 day release; Innovative Research of America, Sarasota, FL, USA) were implanted subcutaneously using a precision trochar (10 gages) at the time of cell injection. The tumor size was measured using a caliper every 3 days and the volume of the tumor was calculated using the following formula: Volume = (length/2) × width2. The experimental period was 21 days. Subsequently, the mice were sacrificed, and tumors were resected and weighed. Tumor tissues were stored at −80°C for further analysis.

Statistical analysis

The data are presented as the mean ± standard deviation. A total of three independent experiments were performed. Differences between two groups were analyzed with Student's t-test. Differences among more than two groups were analyzed with analysis of variance followed by Tukey's test. Overall survival was analyzed using the Kaplan-Meier method and log-rank test. P<0.05 was considered to indicate a statistically significant difference. GraphPad Prism 6.01 software (GraphPad Software, Inc., La Jolla, CA, USA) was used to perform statistical analyses.

Results

lncRNA FOXD2-AS1 expression is increased in BC tissues and is associated with poor prognosis

To examine the pattern of lncRNA FOXD2-AS1 expression in BC tissues and non-tumor tissues, the mRNA expression of FOXD2-AS1 in 34 paired BC tissues and adjacent normal tissues was determined. The results demonstrated that mRNA expression of FOXD2-AS1 was significantly increased in BC tissue samples, compared with adjacent normal tissues (Fig. 1A and B; P<0.01). According to the median expression (6.9) of FOXD2-AS1, the BC tissues specimens were divided into two groups: A higher FOXD2-AS1 expression group and the other group with lower expression of FOXD2-AS1 (Fig. 1C). The differences of overall survival between groups with either higher or lower expression of FOXD2-AS1 were compared, using Kaplan-Meier curves and a log-rank test. The results demonstrated that the overall survival was poor in patients with BC with high expression levels of FOXD2-AS1, compared with patients with low expression levels of FOXD2-AS1 (Fig. 1D). The clinicopathologic features of the patients are presented in Table I. It was demonstrated that higher FOXD2-AS1 expression was associated the positive expression of estrogen receptor, human epidermal growth factor receptor 2, distant metastases, lymphatic metastasis and tumor, node, metastasis stage. Overall, the present results suggested that FOXD2-AS1 expression was increased in BC tissues and upregulation of FOXD2-AS1 was associated with a poor prognosis BC. The results demonstrated that FOXD2-AS1 may serve an oncogenic role in the tumorigenesis of BC.

Figure 1 FOXD2-AS1 expression is increased in BC tissues and was associated with poor prognosis of BC. (A) Expression of FOXD2-AS1 in 34 pairs of BC tissues and paired adjacent non-tumor tissue was determined by reverse transcription-quantitative polymerase chain reaction. (B) Respective association of FOXD2-AS1 expression between each BC tissue specimen and paired adjacent normal tissue. (C) In total, 34 tissues samples were divided into two groups: Higher FOXD2-AS1 expression group and lower FOXD2-AS1 expression group, according to the median expression of FOXD2-AS1. (D) Overall survival in patients with BC with high/low FOXD2-AS1 expression levels. BC, breast cancer; FOXD2-AS1, FOXD2 adjacent opposite strand RNA 1.

Table I Association between FOXD2-AS1 expression and clinicopathological characteristics of patients with breast cancer.

Table I

Association between FOXD2-AS1 expression and clinicopathological characteristics of patients with breast cancer.

Clinicopathological characteristics	Number of patients	FOXD2-AS1 expression		P-value
		Low, n=15	High, n=19
Age
<50	18	8	9	0.327
≥50	16	7	10
ER
Positive	23	8	14	0.006a
Negative	11	7	5
PR
Positive	19	9	9	0.623
Negative	15	6	10
HER2
Positive	20	9	13	0.025a
Negative	14	6	6
EGFR
Positive	18	8	9	0.712
Negative	16	7	10
Distant metastases
Yes	21	7	14	0.017a
No	13	8	5
Lymphatic metastasis
Positive	22	8	15	0.008a
Negative	12	7	4
TNM stage
I-II	20	9	12	0.013a
III	14	6	7

a P<0.05. TNM, tumor, node, metastasis; FOXD2-AS1, FOXD2 adjacent opposite strand RNA 1; ER, estrogen receptor; PR, progesterone receptor; HER2, human epidermal growth factor receptor 2; EGFR, epidermal growth factor receptor.

FOXD2-AS1 knockdown reduces the breast cancer stem cell (BCSC) properties

It was investigated whether FOXD2-AS1 served a role in the maintenance of BCSC properties in BC. The FOXD2-AS1 expression level was significantly increased in BC cell lines, MCF-7, MDA-MB-231, MDA-MB-453 and MDA-MB-468, compared with normal human breast epithelial cells (MCF-10A; Fig. 2A; P<0.01). Furthermore, the FOXD2-AS1 expression level was significantly increased in MCF-7 and MDA-MB-468 sphere cells compared with parental cells (Fig. 2B; P<0.01). BCSC cells [MCF-7 cancer stem cells (CSCs) and MDA-MB-468 CSCs] were trans-fected with lentivirus-mediated interfering oligonucleotides targeting FOXD2-AS1 to knockdown the expression of FOXD2-AS1 (Fig. 2C and D). Knockdown of FOXD2-AS1 significantly decreased the percentage of CD44+/CD24− in the BCSC cell (MCF-7 CSC and MDA-MB-468 CSC) subpopulation (Fig. 2E; P<0.01). The results additionally demonstrated that downregulation of FOXD2-AS1 resulted in a significant decrease of stem factor expression, including Nanog, octamer-binding transcription factor 4 (Oct4) and sex determining region Y-box 2 (SOX2; Fig. 2F and G; P<0.01). Therefore, the data suggested that knockdown of FOXD2-AS1 decreased the properties of BCSC.

Figure 2 FOXD2-AS1 knockdown reduces BCSC properties in BC cells. (A) RT-qPCR determination of FOXD2-AS1 in BC cell lines, MCF-7, MDA-MB-231, MDA-MB-453 and MDA-MB-468, and human normal breast epithelial cells (MCF-10A). ##P<0.01 vs. MCF-10A. (B) Expression of FOXD2-AS1 in MCF-7 and MDA-MB-468 spheres and parental cells was determined using RT-qPCR. ##P<0.01 vs. respective parental. Knockdown efficiency of FOXD2-AS1 in (C) MDA-MB-468 and (D) MCF-7 cells. (E) Percentage of CD44+/CD24− cells in BCSC cells (MCF-7 and MDA-MB-468) was analyzed by flow cytometry. Expression of stem factors (Nanog, Oct4, SOX2) in (F) MCF-7 and (G) MDA-MB-468 cells with downregulation of FOXD2-AS1 was determined using RT-qPCR. ##P<0.05 vs. respective LV-NC. FOXD2-AS1, FOXD2 adjacent opposite strand RNA 1; BCSC, breast cancer stem cell; BC, breast cancer; RT-qPCR, reverse transcription-quantitative polymerase chain reaction; CD44, CD44 antigen; CD24, signal transducer CD24; Oct4, octamer-binding transcription factor 4; SOX2, sex determining region Y-box 2; LV, lentivirus; NC, negative control; sh, small hairpin.

Knockdown of FOXD2-AS1 inhibits the proliferation, migration and invasion of BC cells in vitro

The nuclear and cytoplasmic expression of FOXD2-AS1 was measured and it was identified that FOXD2-AS1 was primarily located in the cytoplasm in MCF-7 and MDA-MB-468 cells (Fig. 3A and B). Knockdown of FOXD2-AS1 inhibited cell proliferation in MCF-7 and MDA-MB-468 cells, as demonstrated by the CCK-8 results (Fig. 3C and D). As demonstrated in the wound-healing assay, FOXD2-AS1 knockdown significantly decreased the relative distance of migration (Fig. 3E–G; P<0.01). As demonstrated in the Matrigel assay, knockdown of FOXD2-AS1 decreased the number of invaded cells (Fig. 3H–J). Taken together, the data demonstrated that knockdown of FOXD2-AS1 suppressed the proliferation, invasion and migration of BC cells in vitro.

Figure 3 FOXD2-AS1 downregulation inhibits the proliferation, invasion and migration in breast cancer stem cells. Subcellular localization of FOXD2-AS1 in (A) MDA-MB-468 and (B) MCF-7 cells was detected. Effect of FOXD2-AS1 downregulation on proliferation in (C) MDA-MB-468 and (D) MCF-7 cells was examined by a Cell Counting Kit-8 assay. (E) Representative images of the wound-healing assay. Magnification, ×400. Effect of FOXD2-AS1 downregulation on migration in (F) MDA-MB-468 and (G) MCF-7 cells was determined by a wound-healing assay. (H) Representative images of the Matrigel assay. Effect of FOXD2-AS1 downregulation on invasion in (I) MDA-MB-468 and (J) MCF-7 cells was determined by a Matrigel assay. Data are presented as the mean ± standard deviation. ##P<0.05 vs. LV-NC. FOXD2-AS1, FOXD2 adjacent opposite strand RNA 1; LV, lentivirus; NC, negative control; sh, small hairpin.

Knockdown of FOXD2-AS1 suppresses BC tumor growth in vivo

To investigate the role of FOXD2-AS1 downregulation on BC growth, a xenograft mice in vivo assay was conducted. The knockdown of FOXD2-AS1 significantly decreased the tumor volume in xenograft mice (Fig. 4A; P<0.01). Additionally, knockdown of FOXD2-AS1 significantly decreased the weight of the tumor (Fig. 4B; P<0.01). The expression levels of Nanog, Oct4 and SOX2 were significantly decreased by FOXD2-AS1 downregulation (Fig. 4C; P<0.01). Furthermore, it was demonstrated that FOXD2-AS1 knockdown significantly decreased N-cadherin expression, and increased E-cadherin and vimentin expression (Fig. 4D; P<0.01). Therefore, the data suggested that FOXD2-AS1 knockdown suppressed the tumor growth of BC in vivo.

Figure 4 FOXD2-AS1 downregulation suppresses the tumor growth of breast cancer in vivo. (A) Tumor volume was measured in transplanted mice. (B) Tumor weight was determined. (C) Expression of Nanog, Oct4 and SOX2 was determined using RT-qPCR. (D) Expression of N-Cad, E-Cad and vimentin was determined using RT-qPCR. Data are presented as the mean ± standard deviation. ##P<0.05 vs. respective LV-NC. FOXD2-AS1, FOXD2 adjacent opposite strand RNA 1; RT-qPCR, reverse transcription-quantitative polymerase chain reaction; Oct4, octamer-binding transcription factor 4; SOX2, sex determining region Y-box 2; Cad, cadherin; LV, lentivirus; NC, negative control; sh, small hairpin.

FOXD2-AS1 promotes BC through modulation of PFN2 by sponging miR-150-5p

To examine the molecular mechanism underlying FOXD2-AS1-exhibited regulation of proliferation, invasion, migration and BCSC properties of BC, the possible targets of FOXD2-AS1 were investigated. Using bioinformatics analysis and validation assays, it was identified that there were complementary binding sites between miR-150-5p and the 3′-UTR of FOXD2-AS1 (Fig. 5A). The results of the reporter gene assay provided evidence that miR-150-5p was able to bind with the FOXD2-AS1 3′-UTR (Fig. 5B). The expression of miR-150-5p was lower in MDA-MB-468 and MCF-7 sphere cells compared with respective parental cells (Fig. 5C). Transfection of miR-150-5p mimics (Fig. 5D) significantly decreased mRNA expression levels of Nanog, Oct4 and SOX2 in MDA-MB-468 and MCF-7 cells (Fig. 5E and F; P<0.01). As observed in the wound-healing assay, miR-150-5p mimics significantly decreased the relative migration distance, demonstrating a decrease in migratory ability (Fig. 5G-I; P<0.01). As demonstrated in the Matrigel assay, miR-150-5p mimics decreased the number of invaded cells, suggesting a reduction of invasive ability (Fig. 5J–L). Taken together, these data demonstrated that miR-150-5p inhibited the proliferation, invasion, migration and BCSC properties of BC in vitro.

Figure 5 FOXD2-AS1 regulates BC malignancy through interaction with miR-150-5p. (A) Bioinformatics analysis of the binding between miR-150-5p and 3′UTR of wt FOXD2-AS1 and the mutation of putative binding sites. (B) Direct binding of miR-150-5p with FOXD2-AS1 3′-UTR was detected by a lucif-erase reporter assay. (C) RT-qPCR determination of miR-150-5p in MDA-MB-468 and MCF-7 spheres and parental cells. ##P<0.01 vs. respective parental. (D) MDA-MB-468 and MCF-7 cells were transfected with miR-150-5p mimics and transfection efficiency was evaluated. MDA-MB-468 and MCF-7 cells were transfected with miR-150-5p mimics. RT-qPCR determination of Nanog, Oct4 and SOX2 in (E) MDA-MB-468 and (F) MCF-7 cells. (G) Representative images of the wound-healing assay. Magnification, ×400. Migration in (H) MDA-MB-468 and (I) MCF-7 cells was determined by a wound-healing assay. (J) Representative images of the Matrigel assay. Invasion of (K) MDA-MB-468 and (L) MCF-7 cells was determined by a Matrigel assay. ##P<0.05 vs. respective NC. FOXD2-AS1, FOXD2 adjacent opposite strand RNA 1; BC, breast cancer; miR, microRNA; UTR, untranslated region; wt, wild-type; RT-qPCR, reverse transcription-quantitative polymerase chain reaction; Oct4, octamer-binding transcription factor 4; SOX2, sex determining region Y-box 2; NC, negative control; mut, mutant.

Using bioinformatics analysis, it was identified that there were complementary binding sites between miR-150-5p and the PFN2 3′-UTR (Fig. 6A). Furthermore, the direct interaction between miR-150-5p and PFN2 3′-UTR was confirmed by the results of the luciferase reporter assay (Fig. 6B). Transfection of miR-150-5p mimics decreased FOXD2-AS1 and PFN2 expression (Fig. 6C), whereas, miR-150-5p inhibitors (Fig. 6D) increased FOXD2-AS1 and PFN2 expression (Fig. 6E). FOXD2-AS1 knockdown significantly inhibited miR-150-5p inhibitor-induced increase of Nanog, Oct4 and SOX2 expression (Fig. 6F; P<0.01). Additionally, the miR-150-5p inhibitor-induced increase of N-cadherin, and decrease of E-cadherin and vimentin was inhibited by FOXD2-AS1 knockdown (Fig. 6G).

Figure 6 FOXD2-AS1 modulated the expression of PFN2 by sponging miR-150-5p. (A) Bioinformatics analysis of the binding between miR-150-5p and 3′UTR of wt PFN2 and the mutation of putative binding sites. (B) Direct binding of miR-150-5p with PFN2 3′-UTR was detected by a luciferase reporter assay. (C) RT-qPCR determination of FOXD2-AS1 and PFN2 expression in MCF-7 cells following transfection of miR-150-5p mimics. (D) MCF-7 cells were transfected with miR-150-5p inhibitors and the transfection efficiency was examined. (E) RT-qPCR determination of FOXD2-AS1 and PFN2 expression in MCF-7 cells following transfection of miR-150-5p inhibitors. MCF-7 cells were transfected with miR-150-5p inhibitors in the presence or absence of lentivirus-shFOXD2-AS1. (F) RT-qPCR determination of Nanog, Oct4 and SOX2. (G) RT-qPCR determination of N-Cad, E-Cad and vimentin. (H) RT-qPCR determination of PFN2 in 34 pairs of BC tissues and paired adjacent normal tissues. (I) Respective association of PFN2 expression between each BC tissue specimen and paired adjacent normal tissue. (J) A total of 34 tissues samples were divided into two groups: Higher PFN2 expression group and lower PFN2 expression group, according to the median expression of PFN2. (K) Overall survival in patients with BC with high/low PFN2 expression levels. (L) Correlation between FOXD2-AS1 and PFN2 expression in BC tissues. ##P<0.05 vs. respective NC; **P<0.05 vs. respective miR-150-5p inhibitor. FOXD2-AS1, FOXD2 adjacent opposite strand RNA 1; PFN2, profilin 2; miR, microRNA; UTR, untranslated region; wt, wild-type; RT-qPCR, reverse transcription-quantitative polymerase chain reaction; sh, small hairpin; Oct4, octamer-binding transcription factor 4; SOX2, sex determining region Y-box 2; BC, breast cancer; NC, negative control; mut, mutant; Cad, cadherin.

The expression of PFN2 was significantly increased in BC tissues specimens compared with adjacent normal tissues (Fig. 6H and I; P<0.01). According to the median expression of PFN2 (4.35), the BC tissues specimens were divided into two groups: A higher PFN2 expression group and the other group with lower expression of PFN2 (Fig. 6J). The differences of overall survival between groups with either higher or lower expression of PFN2 were compared using Kaplan-Meier curves and a log-rank test. The results demonstrated that overall survivals were poor in patients with BC with high expression levels of PFN2, compared with patients with low expression levels of PFN2 (Fig. 6K). Furthermore, the expression of FOXD2-AS1 was positively correlated with the PFN2 expression level in tumor tissues of patients with BC (Fig. 6L). The present results concluded that FOXD2-AS1 promoted BC through modulation of PFN2 by sponging miR-150-5p (Fig. 7).

Figure 7 Schematic figure of FOXD2-AS1-miR-150-5p/PFN2-induced promotion of BC. FOXD2-AS1 promotes BC proliferation, invasion, migration, stemness and tumor growth. FOXD2-AS1 regulates BC malignancy through modulation of PFN2 by sponging miR-150-5p. The FOXD2-AS1/miR-150-5p/PFN2 axis is critical in the development of BC. FOXD2-AS1, FOXD2 adjacent opposite strand RNA 1; miR, microRNA; PFN2, profilin 2; BC, breast cancer.

Discussion

BC is one of the most prevalent malignant tumors in women worldwide and one of the leading causes of cancer mortality (27). The high rate of metastasis and recurrence of BC typically contribute to accelerated progression (28). lncRNAs are emerging as important regulators in the process of cancer stemness and tumorigenesis (12,29). In the present study, it was demonstrated that FOXD2-AS1 expression was significantly increased in BC tissue, cells and sphere subpopulation. Additionally, the upregulation of FOXD2-AS1was closely associated with poor prognosis of patients with BC. Furthermore, downregulation of FOXD2-AS1 decreased cell proliferation, migration and invasion in BC cells, and inhibited tumor growth in the transplanted tumor in vivo. Knockdown of FOXD2-AS1 decreased the percentage of CD44+/CD24− cells and the ability to form spheres in the BCSC cell (MCF-7 CSC and MDA-MB-468 CSC) subpopulation, suggesting the inhibitory role of FOXD2-AS1 knockdown in BC stemness. Furthermore, knockdown of FOXD2-AS1 decreased the expression of stem factors, including Nanog, Oct4 and SOX2. It was identified that CSCs limit the efficiency of surgical resection or post-chemoradiotherapy as these cells may endow the growth or proliferation potential (30,31). The results suggested that FOXD2-AS1 serves an oncogenic role in BC and high FOXD2-AS1 expression was associated with poor prognosis of patients with BC.

FOXD2-AS1 was primarily located in the cytoplasm, suggesting that the primary reason for FOXD2-AS1-exhibited regulation of BC malignancy was due to post-transcriptional regulation. At present, serving as an miRNA 'sponge' is believed to be the most prevalent mechanism of lncRNA-mediated biological regulation (32). In the present study, it was demonstrated that the expression of FOXD2-AS1 was increased in the sphere subpopulation of BCSC, and FOXD2-AS1 and PFN2 expression was positively correlated. Furthermore, it was identified that miR-150-5p targeted the 3′-UTR of FOXD2-AS1 and PFN2 mRNA, and miR-150-5p expression was negatively associated with FOXD2-AS1 and PFN2 expression. miR-150-5p mimics decreased cell proliferation, migration and invasion in BC cells. Additionally, PFN2 expression was significantly upregulated in BC tissues. Furthermore, the upregulation of PFN2 indicated poor prognosis of patients with BC. FOXD2-AS1 and PFN2 expression was positively correlated. Previous studies demonstrated that miR-150-5p and PFN2 were able to regulate the development of certain cancer types (33-35). miR-150-5p inhibits cancer cell aggressiveness by targeting SPARC (osteonectin), cwcv and kazal like domain proteoglycan 1 in head and neck squamous cell carcinoma (33). PFN2 was correlated with poor prognosis of esophageal squamous cell carcinoma, which is proposed to be a therapeutic target (34). It was additionally demonstrated that PFN2 promoted migration, invasion and stemness of HT29 human colorectal cancer stem cells (35). However, at present, to the best of the authors' knowledge, there is no evidence for the direct interaction between FOXD2-AS1, miR-150-5p and PFN2. In summary, all the data suggested that FOXD2-AS1 regulated the expression of PFN2 and BCSC by serving as a sponge of miR-150-5p.

In conclusion, the findings demonstrated that FOXD2-AS1 was crucial for BC proliferation, invasion, migration and stemness, and tumor growth. FOXD2-AS1 regulates BC malignancy through modulation of PFN2 by sponging miR-150-5p. The present results suggested that the FOXD2-AS1/miR-150-5p/PFN2 axis is involved in the development of BC, and provides novel targets for the treatment of BC, and potential biomarkers for the diagnosis and prognosis of BC.

Funding

The present study was supported by General Project of Western Medicine of Guangzhou Health and Family Planning Commission (grant no. 20181A011099), Natural Science Foundation of Guangdong Province (grant no. 22017A030313551) and Guangzhou Key Medical Discipline Construction Project Fund.

Availability of data and materials

The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.

Authors' contributions

XA, MJ and NQ designed the study and wrote the paper. MJ, NQ, HX, HLia and HLi performed the experiments and analyzed the data. All authors read and approved the final manuscript.

Ethics approval and consent to participate

The present patient and animal studies were approved by the Ethics Committee of the Affiliated Cancer Hospital and Institute of Guangzhou Medical University (approval no. ACHIGMU-2012-1-3-05; Guangzhou, China). Written informed consent was obtained from all the enrolled patients.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Acknowledgments

Not applicable.

References