The human breast epithelial cell line MCF10A and breast cancer cell lines MCF7, T47D, BT549, MDA-MB-231 and SKBR3 were obtained from The Cell Bank of Type Culture Collection of The Chinese Academy of Sciences. MCF10A cells were cultured in DMEM/F12 (Hyclone; Cytiva) supplemented with 5% horse serum (Thermo Fisher Scientific, Inc.), 10 µg/ml insulin (Sigma-Aldrich; Merck KGaA), 20 ng/ml epidermal growth factor (R&D Systems, Inc.), 0.1 µg/ml cholera toxin (Sigma-Aldrich; Merck KGaA) and 0.5 µg/ml hydrocortisone (Sigma-Aldrich; Merck KGaA). BT549, MDA-MB-231 and SKBR3 cells were maintained in RPMI-1640 (Hyclone; Cytiva) supplemented with 10% FBS (Thermo Fisher Scientific, Inc.). MCF7 and T47D cells were cultured in DMEM (Hyclone; Cytiva) with 10% FBS. All cells were supplemented with 100 mg/ml streptomycin (Hyclone; Cytiva) and 100 IU/ml penicillin (Hyclone; Cytiva) and cultured in an atmosphere containing 5% CO₂ at 37°C.

miR-454-5p mimics (5′-GAAGUAAAGGGGCAAGAUAGGGC-3′) and negative control (5′-UUCUCCGAACGUGUCACGUUU-3′) oligonucleotides were purchased from Guangzhou RiboBio Co., Ltd. Transient transfection was conducted using Lipofectamine^® 3000 (Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions. Cell were seeded in a 6-well plate at a density of 2×10⁵ cells/well and then transfected with 200 pmol oligonucleotides at 37°C for 24 h. Experiments were performed 48 h after transfection.

The miR-454-5p antisense strand (5′-GCCCUAUCUUGCCCCUUUACUUC-3′) or inhibitor control (5′-UCUACUCUUUCUAGGAGGUUGUGA-3′; anti-Control) was cloned into the pEZX-AM04 lentiviral vector (GeneCopoeia, Inc.) and short hairpin RNA (sh)FoxJ2 (5′-GCAAGCCACGATACAGCTATT-3′) or shControl (5′-CCTAAGGTTAAGTCGCCCTCG-3′) was cloned into the pLKO.1 lentiviral vector (GeneCopoeia, Inc.). 293T cells were transfected with 10 µg lentiviral vectors and 10 µg packaging vectors when the density of cells reached 80–90%. After transfection for 48 h at 37°C, supernatants containing virus particles were harvested and purified. The lentiviral particles were used to infect targeted cells at 60% confluency with a multiplicity of infection of 30 and stable cells were selected with 800 mg/ml puromycin. The FoxJ2 knockdown efficiencies were confirmed by reverse transcription-quantitative PCR (RT-qPCR) and western blotting (Fig. S1).

Total RNA was extracted from breast cancer cells using TRIzol^® reagent (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. The isolated RNA was reverse transcribed into cDNA using a PrimeScript™ RT Master Mix kit (Takara Biotechnology Co., Ltd.) according to the manufacturer's protocol. qPCR analysis was performed using SYBR Green Mix (Takara Biotechnology Co., Ltd.) on a CFX96 Touch™ Real-time PCR System (Bio-Rad Laboratories, Inc.); the thermocycling conditions were as follows: Initial denaturation for 8 min at 95°C; followed by 40 cycles of two-step PCR at 95°C for 20 sec and 60°C for 1 min. The expression of mRNA or miRNA was normalized to GAPDH or U6, respectively. Data were analyzed using the 2^−ΔΔCq method (42). The qPCR primers are listed in Table I.

Total protein was isolated from cells using RIPA lysis buffer containing PMSF (both from Cell Signaling Technology, Inc.). Protein concentrations were determined using a BCA Protein Assay kit (Thermo Fisher Scientific, Inc.). All samples (40 µg per lane) were separated by SDS-PAGE on 10% gels and then transferred to polyvinylidene fluoride membranes (EMD Millipore). After blocking with 5% skimmed milk for 1 h at room temperature, the membrane was incubated with primary antibodies overnight at 4°C, followed by washing with PBS and incubation with horseradish peroxidase-conjugated goat anti-rabbit (cat. no. 7074) or horse anti-mouse (cat. no. 7076) (both at a dilution of 1:2,000; Cell Signaling Technology, Inc.) secondary antibodies for 1 h at room temperature. The bands were visualized with ECL reagent (EMD Millipore). Antibodies against N-cadherin (cat. no. sc-393933), E-cadherin (cat. no. sc-71008), Vimentin (cat. no. sc-80975), FoxJ2 (cat. no. sc-514265) (all from Santa Cruz Biotechnology, Inc.) and GAPDH (cat. no. 5174; Cell Signaling Technology, Inc.) were used at a dilution of 1:1,000.

For FoxJ2 3′-UTR activity analysis, the 3′-UTR of FoxJ2 mRNA and the corresponding sequence with mutations (FoxJ2-1: 5′-GCTGGCAAAGAAATGGATAGGGA-3′ to 5′-GCTGGCAAAGAAATGGATAAA A-3′; FoxJ2-2: 5′-GAAGTAAAGGGGCAAGATAGGGC-3′ to 5′-GAAGTAAAGGGGCAAGATAAAAC-3′) were cloned into a psiCHEK2 luciferase reporter plasmid (Promega Corporation). MDA-MB-231 cells were seeded into a 24-well plate at 1×10⁴ cells/well and co-transfected with psiCHEK2-FoxJ2 (500 ng) and miR-454-5p mimics (50 pmol) or negative control using Lipofectamine 3000 for 48 h at 37°C. For E-cadherin promoter activity analysis, the E-cadherin promoter was cloned into a pGL3-basic luciferase reporter plasmid (Promega Corporation) and was transfected into MDA-MB-231 cells stably expressing anti-miR-454-5p or shFoxJ2, as well as into the control cells using Lipofectamine 3000 for 48 h at 37°C. Luciferase activity was analyzed after transfection for 48 h and the cell lysates were measured for luciferase activity using Dual-Luciferase Reporter Assay System (Promega Corporation) according to the manufacturer's instructions. The firefly luciferase activity was normalized to Renilla luciferase activity.

Stably transfected MDA-MB-231 cells were cross-linked with 1% formaldehyde for 15 min at room temperature. After quenching with 125 mM glycine, the cells were collected, washed and sonicated in RIPA buffer. The cross-linked lysate was sonicated (power, 20 W; duration, 30 sec/cycle for 40 cycles; temperature, 5–6°C) to obtain 500–1,500 bp fragments, which were immunoprecipitated with IgG or anti-FoxJ2 antibody (cat. no. sc-514265; Santa Cruz Biotechnology, Inc.) at 4°C for 3 h. This was followed by incubation with 50 µl protein A/G beads overnight at 4°C. qPCR was used to identify and quantify the precipitated DNA. Primers for ChIP assays are presented in Table I.

Breast cancer cells were seeded into six-well plates at 500 cells/well. Transient transfection with miR-454-5p mimics was conducted using Lipofectamine^® 3000, aforementioned, or MDA-MB-231 cells stably expressing anti-miR-454-5p or shFoxJ2 were used, as well as their control cells. Cells were cultured for 20 days at 37°C, and colonies were washed with PBS, fixed with 4% paraformaldehyde for 15 min at room temperature, and stained with hematoxylin. The colonies with >50 cells were counted under a light microscope at ×100 magnification.

MDA-MB-231 cells stably expressing anti-miR-454-5p or shFoxJ2, as well as their control cells, were seeded in 6-well plates at a density of 5×10⁴ cells/well and cultured at 37°C for 24 h. Cell morphology was examined under a light microscope at ×100 magnification.

The Cell Counting Kit-8 (CCK-8; Dojindo Molecular Technologies, Inc.) was used to determine cell viability, according to the manufacturer's protocols. Cells were seeded into a 96-well plate at 5×10³ cells/well. After transfection for 24 h, CCK-8 reagent (10 µl) was added to the medium and incubated at 37°C for 2 h. The absorbance of each sample was measured using a microplate reader (Thermo Fisher Scientific, Inc.) at 450 nm.

Cells were plated at a density of 2×10⁴ on glass coverslips, washed with ice-cold PBS and fixed in 4% formaldehyde solution for 15 min at room temperature. The coverslips were blocked with 2% BSA (Cell Signaling Technology, Inc.) in PBS for 30 min at room temperature and incubated with a primary antibody against E-cadherin (1:500; cat. no. sc-71008; Santa Cruz Biotechnology, Inc.) overnight at 4°C. Subsequently, cells were incubated with Alexa Fluor^® 488-conjugated anti-mouse secondary antibody (1:300; cat. no. 4408; Cell Signaling Technology, Inc.) for 1 h and then stained with DAPI for 10 min, both at room temperature. The coverslips were washed with PBS and observed under a fluorescence microscope at ×400 magnification.

Cells were seeded into a 6-well plate at a density of 5×10⁴ cells/well. After culture at 37°C for 24 h, cells were collected and fixed with 70% ethanol for 24 h at −20°C, followed by incubation with 10 µg/ml PI and 50 µg/ml RNase (BD Biosciences) on ice for 15 min. Cells were then assessed by flow cytometry using a BD FACSCalibur flow cytometer (BD Biosciences).

Cell invasion and migration were assessed using BD Transwell chambers pre-coated with or without Matrigel, respectively (BD Biosciences). Cells were seeded (2×10⁴ cells/well) into the upper Transwell chamber containing RPMI-1640 medium without serum; RPMI-1640 medium with 10% FBS was added to the bottom chamber. After incubation for 16 h at 37°C, the invaded or migrated cells on the lower surface were fixed in 4% formaldehyde solution for 15 min at room temperature, and subsequently stained with crystal violet for 15 min at room temperature. Images of the invaded or migrated cells were captured under a light microscope at ×100 magnification.

TargetScan Human release 7.2. (http://www.targetscan.org) was used for prediction of miR-454-5p targets. The top three genes containing two potential miR-454-5p target sites were selected for further study. LASAGNA-Search 2.0 (https://www.bitnos.com/info/lasagna-search) was used to predict FoxJ2 binding sites on the E-cadherin promoter region.

The expression levels of miR-454-5p were analyzed in breast cancer tissues (n=749) and normal breast tissues (n=76) obtained from The Cancer Genome Atlas (TCGA) database by using UALCAN (http://ualcan.path.uab.edu/index.html). The relationship between miR-454-5p and FoxJ2 was determined by ENCORI (http://starbase.sysu.edu.cn). The prognostic value of miR-454-5p expression was examined by using the online database, KM-Plotter (www.kmplot.com/mirpower). Patients were divided into two groups by ‘auto select best cutoff’ feature and assessed using a Kaplan-Meier survival plot.

Female BALB/c-nude mice (n=12; age, 5 weeks; weight, 22 g) were purchased from Vital River Lab Animal Technology Company. The mice were maintained in a specific pathogen-free environment with a 10/14-h light/dark cycle in 40–60% humidity at 27°C and had ad libitum access to food and water. The experimental procedures were approved by the Animal Experimentation Ethics Committee of Qingdao Central Hospital (Qingdao, China). A total of 1×10⁶ stably transfected MDA-MB-231 cells were injected subcutaneously into the right mammary fat pad. Tumor volume was calculated using the following equation: (Length × width²)/2. The mice were sacrificed on day 35 after tumor implantation, and the tumors were excised and weighed. The tumor burden did not exceed the recommended dimensions (tumor diameter_max ≤16.5 mm; volume_max ≤1,200 mm³). The animals were anesthetized (60 mg/kg ketamine and 5.0 mg/kg xylazine) and then sacrificed by cervical dislocation.

Statistical analyses were performed using SPSS 20.0 (IBM Corp.). Data are presented as the mean ± SD from at least three independent experiments. The unpaired Student's t-test was used to compare differences between two groups, and one-way ANOVA followed by Tukey's multiple comparison post hoc test was used when comparing >2 groups. P<0.05 was considered to indicate a statistically significant difference.

To determine the role of miR-454-5p in breast cancer, the expression levels of miR-454-5p were analyzed in breast cancer tissues (n=749) and normal breast tissues (n=76) obtained from The Cancer Genome Atlas (TCGA) database by using UALCAN. Increased miR-454-5p expression levels were observed in breast cancer compared with normal tissues (Fig. 1A), and miR-454-5p expression was significantly upregulated in triple-negative and Her2⁺ breast cancer types (Fig. 1B). Furthermore, high miR-454-5p expression was associated with a poor prognosis in patients with breast cancer as determined using the KM-Plotter database (Fig. 1C). Subsequently, the expression levels of miR-454-5p were examined in five breast cancer cell lines (MCF7, T47D, BT549, MDA-MB-231 and SKBR3) and a normal breast epithelial cell line MCF10A. The expression levels of miR-454-5p were significantly higher in BT549, MDA-MB-231 and SKBR3 breast cancer cell lines compared with MCF10A cells (Fig. 1D). Taken together, these results indicated that miR-454-5p may serve an important role in breast cancer progression.

As miR-454-5p was upregulated in BT549, MDA-MB-231 and SKBR3 compared to MCF10A cells, aforementioned, which suggested that miR-454-5p may serve an important role in these cell lines, MDA-MB-231 and BT549 were choose to further study. To evaluate the potential role of miR-454-5p in breast cancer progression, miR-454-5p mimics or negative controls were transfected into MDA-MB-231 and BT549 cells. The expression of miR-454-5p was effectively elevated in MDA-MB-231 and BT549 cells after transfection with miR-454-5p mimics compared with miR-NC-transfected cells, as determined by RT-qPCR (Fig. 2A). CCK-8 and colony formation assays were conducted to evaluate the effects of miR-454-5p on breast cancer proliferation. It was found that the overexpression of miR-454-5p significantly enhanced cell viability and the number of colonies in MDA-MB-231 and BT549 cells compared with the miR-NC group (Fig. 2B and C, respectively). Furthermore, the proportion of cells at the S phase was significantly increased in miR-454-5p-overexpressing MDA-MB-231 and BT549 cells compared with that of control cells (Fig. 2D). Transwell analyses were performed to assess the effects of miR-454-5p on breast cancer cell migration and invasion (Fig. 2E and F, respectively). The number of migrated and invaded cells was significantly increased in miR-454-5p-overexpressing MDA-MB-231 and BT549 cells compared with those in control cells. These results suggested that miR-454-5p may function as an oncogene in breast cancer progression.

TargetScan was used to predict target genes of miR-454-5p, and the top three genes containing two potential target sites in their 3′-UTRs were ATXN7L3B, RIPPLY3 and FoxJ2 (Fig. 3A). Of the three genes, only FoxJ2 has been reported to be involved in cancer progression (43–46); therefore, it was selected for further study. The luciferase reporter assay was used to determine whether miR-454-5p could directly bind to the 3′-UTR of FoxJ2. The two putative miR-454-5p target sites in the 3′-UTR of FoxJ2 were cloned into the psiCHEK2 reporter plasmid and subsequently transfected into MDA-MB-231 cells along with miR-454-5p mimics or negative control. The luciferase activity of the FoxJ2 3′-UTR reporter construct was significantly decreased in the miR-454-5p-transfected MDA-MB-231 cells compared with luciferase activity in the miR-NC cells (Fig. 3B). This effect was abolished in the mutated FoxJ2 3′-UTR group in which both target sites for miR-454-5p were inactivated (FoxJ-1 + 2) by site-directed mutagenesis (Fig. 3C). The expression levels of FoxJ2 were notably decreased in MDA-MB-231 and BT549 cells transfected with miR-454-5p mimics compared with the expression levels in control cells, as demonstrated by RT-qPCR (Fig. 3D) and western blotting (Fig. 3E).

To further investigate the relationship between miR-454-5p and FoxJ2, the expression levels of miR-454-5p and FoxJ2 from TCGA database was analyzed by ENCORI. As presented in Fig. 3F, the expression of miR-454-5p exhibited a negative co-expression trend with FoxJ2. Thus, these results support the bioinformatics prediction that FoxJ2 was a direct target of miR-454-5p.

To further confirm the regulation of the miR-454-5p/FoxJ2 axis in breast cancer progression, stably transfected MDA-MB-231 cell lines overexpressing the anti-miR-454-5p inhibitor with or without shFoxJ2 co-transfection, as well as anti-Control cells, were generated. The expression of miR-454-5p was significantly downregulated in miR-454-5p-silenced MDA-MB-231 cells compared with that of control cells, as determined via RT-qPCR (Fig. 4A). The expression of FoxJ2 was upregulated in miR-454-5p-silenced MDA-MB-231 cells, whereas it was notably downregulated in miR-454-5p and FoxJ2-silenced MDA-MB-231 cells compared with that of control cells, as determined using western blotting (Fig. 4B). It was also found that knockdown of FoxJ2 significantly reversed the inhibitory effects of miR-454-5p silencing on the proliferation, migration and invasion of MDA-MB-231 cells as determined via CCK-8 (Fig. 4C), colony formation (Fig. 4D), cell cycle (Fig. 4E) and Transwell assays (Fig. 4F), respectively.

To investigate the biological effect of the miR-454-5p/FoxJ2 axis on breast cancer progression in vivo, MDA-MB-231-control, MDA-MB-231-anti-454 and MDA-MB-231-anti-454 + shFoxJ2 cells were injected into the mammary fat pads of female BALB/c-nude mice. As presented in Fig. 4G-I, tumor volumes and weights were significantly decreased in mice injected with MDA-MB-231-anti-miR-454 cells compared with those in mice injected with MDA-MB-231-control cells. Moreover, it was identified that knockdown of FoxJ2 could abolish the inhibitory effects of miR-454-5p silencing on tumor growth. Collectively, these results indicated that silencing of miR-454-5p may inhibit breast cancer progression through the upregulation of FoxJ2 expression.

Accumulating evidence has suggested that EMT could promote progression in breast cancer (41). Therefore, it was investigated whether miR-454-5p regulated EMT to affect breast cancer progression. It was identified that miR-454-5p-silenced MDA-MB-231 cells had a cobblestone-like morphology, whereas anti-miR-454-5p + shFoxJ2 MDA-MB-231 cells and anti-Control cells maintained their spindle-like fibroblast morphology (Fig. 5A). In addition, it was demonstrated that the expression of epithelial marker E-cadherin was increased in miR-454-5p-silenced MDA-MB-231 cells, whereas the expression levels of mesenchymal markers Vimentin and N-cadherin were decreased in miR-454-5p-silenced MDA-MB-231 cells compared with the expression levels in control cells, as determined by western blotting (Fig. 5B). Moreover, knockdown of FoxJ2 could reverse this effect caused by miR-454-5p silencing (Fig. 5B). In total, three putative FoxJ2 binding sites on the E-cadherin promoter region were identified using LASAGNA-Search (47). ChIP analysis results demonstrated that FoxJ2 bound to region 1 and region 2 from the E-cadherin promoter region (Fig. 5C). Luciferase assay results indicated that knockdown of FoxJ2 significantly abolished miR-454-5p depletion-induced E-cadherin promoter activity (Fig. 5D). E-cadherin mRNA expression levels were increased in miR-454-5p-silenced MDA-MB-231 cells, whereas this effect was abolished by FoxJ2 knockdown as shown by RT-qPCR (Fig. 5E) and immunofluorescence (Fig. 5F). E-cadherin mRNA was determined to be positive co-expression trend with FoxJ2 mRNA by ENCORI (Fig. 5G). Together, these results indicated that miR-454-5p may regulated EMT through the transcriptional upregulation of E-cadherin by FoxJ2.