Tumor specimens and adjacent tissues were collected from patients who were diagnosed with breast cancer (stages I–IV) and underwent surgery at Ningbo No. 2 Hospital (Ningbo, China) between March 2015 and August 2017. None of these patients underwent local or systemic therapy prior to surgery. The clinical characteristics of these 160 patients are described in Table I. The present study was approved by the Research Ethics Committee of Ningbo No. 2 Hospital, and all patients provided written informed consent.

Human breast cancer cell lines MCF-7, T47D and MDA-MB-231, and normal MCF-10A cells were purchased from the American Type Culture Collection (Manassas, VA, USA). All cells were cultured in Dulbecco's modified Eagle's medium (DMEM; HyClone; GE Healthcare Life Sciences, Logan, UT, USA) supplemented with 10% fetal bovine serum (FBS; ExCell Bio, Shanghai, China) at 37°C in an incubator containing 5% CO₂.

Cells (MCF-7 and T47D) were transfected with small interfering RNA (si)SKA2 (sense, 5′-GGCUGGAAUAUGAAAUCAATT-3′ and antisense, 5′-UUGAUUUCAUAUUCCAGCCTT-3′), si-negative control (NC; sense, 5′-UCCUCCGAACGUGUCACGUTT-3′ and antisense, 5′-ACGUGACACGUUCGGAGAATT-3′) (both Shanghai GenePharma Co., Ltd., Shanghai, China), SKA2cDNA plasmid (Ref Seq: M_001100595.1, Atgg cctcggaggt ggggcacaat ttggagtcgc cggaaactcc gcggcgga ggctggacca gagtcgagtt ctcctcct gcaccaaagg gagccgccac tctggtgt ctaaaccgcc tcggttccag gctgagt ctgatctgga ttacattcaa caggctgg aatatgaaat caagactaat tcctgatt cagcaagtga gctgtcacca ctga) or NC plasmid (both Shanghai Genechem Co., Ltd., Shanghai, China) using Lipofectamine^® 2000 (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA), according to the manufacturer's protocol. The cells were seeded in 6-well plates (siRNA transfection: 100 pmol siRNA and 5 µl Lipofectamine 2000, plasmid DNA transfection: 4 µg DNA and 10 µl Lipofectamine 2000, at room temperature) and were grown to 50% confluence prior to transfection. RNA and protein were extracted 24 h post-transfection. The MOCK was not transfected with any agent.

A total of 24 h post-transfection, MCF-7 cells were harvested and washed three times with cold PBS. Cytoplasmic and nuclear protein fractions were extracted using a Mammalian Nuclear and Cytoplasmic Protein Extraction kit (Beijing Transgen Biotech, Co., Ltd., Beijing, China), according to the manufacturer's protocol; these proteins were then used for western blotting.

Total protein extracted from the cells by 1XSDS (Sigma-Aldrich; Merck KGaA) was quantified by bicinchoninic acid analysis (Beyotime Institute of Biotechnology, Shanghai, China). Cellular proteins (40 µg) were separated by 12% SDS-PAGE and were transferred onto polyvinylidende difluoride membranes (EMD Millipore, Billerica, MA, USA) for immunoblotting. After blocking with 5% bovine serum albumin (BSA; Beijing Solarbio Science & Technology Co., Ltd., Beijing, China) for 2 h at room temperature, the membranes were incubated with specific primary antibodies diluted in 1X Tris-buffered saline-0.1% Tween-20 overnight at 4°C. The following antibodies were used: SKA2 (1:1,500; cat. no. ab91551), MMP2 (1:1,500; cat. no. ab7033), MMP9 (1:1,500; cat. no. ab137651), vimentin (1:2,000; cat. no. ab8978), fibronectin (1:2,000; cat. no. ab2413; all Abcam, Cambridge, UK), N-cadherin (1:2,000; cat. no. 14215), E-cadherin (1:2,000; cat. no. 3195), β-actin (1:1,500; cat. no. 8457), Histone H3 (1:1,500; cat. no. 9728) and GAPDH (1:1,500; cat. no. 5174; all Cell Signaling Technology, Inc., Danvers, MA, USA). Subsequently, membranes were incubated for 1 h at room temperature with goat anti-rabbit immunoglobulin (Ig)G-horseradish peroxidase (HRP) or goat anti-mouse IgG-HRP (1:5,000; cat. nos. BA1054/BA1050; Wuhan Boster Biological Technology, Ltd., Wuhan, China). The protein bands on the membrane were visualized using a chemiluminescence imaging system (LI-COR Biosciences, Lincoln, CA, USA) and detected by chemiluminescence. Densitometric analysis was performed by Tanon GIS version 4.1.2 software (Tanon Science and Technology Co., Ltd., Shanghai, China).

The mRNA expression levels of SKA2 were detected using RT-qPCR. Briefly, total RNA was isolated using TRIzol^® reagent (Invitrogen; Thermo Fisher Scientific, Inc.) and 1 µg total RNA was subjected to first-strand cDNA synthesis for 15 min at 37°C and 5 sec at 85°C using a reverse transcription kit (Thermo Fisher Scientific, Inc.). cDNA was amplified by RT-qPCR using SYBR-Green PCR Master Mix (Roche Applied Science, Madison, WI, USA) on a LightCycler^® 480 system (95°C/10 min/1 cycle, 95°C/10 sec/45 cycles, 60°C/60 sec/45 cycles; Roche Applied Science) and fold-changes were calculated by relative quantification (2^−∆∆Cq) (25). The primers used were as follows: SKA2 forward, 5′-CTGAAACTATGCTAAGTGGGGGAG-3′, reverse, 5′-TTCCAAACATCCTGACACTCAAAAG-3′; and GAPDH forward, 5′-AAGCCTGCCGGTGACTAAC-3′ and reverse, 5′-GCATCACCCGGAGGAGAAAT-3′.

Cells were seeded in 6-well plates at a density of 1×10⁵ cells/well. Once cellular density reached ~70%, the cells were scraped in a straight line with a 1-ml blue micro-pipette tip. After rinsing the dislodged cells with PBS, the cell culture medium was replaced with fresh serum-free medium. The widths at 0 and 24 h were compared to assess the distance of migration using fluorescence microscopy.

Cell invasion assays were conducted using a Matrigel-coated invasion chamber (24-well plates, 8-µm pore size; Corning Inc., Corning, NY, USA), in accordance with the manufacturer's protocol. A total of 5×10⁴ cells were seeded in the upper chambers of the wells in 100 µl FBS-free medium, and the lower chambers were filled with DMEM supplemented with 20% FBS, in order to stimulate cell invasion. After 24 h incubation at 37°C in an incubator containing 5% CO₂, the cells on the filter surface were stained with 0.1% crystal violet for 30 min and images were captured with a fluorescence microscope (Nikon Corporation, Tokyo, Japan). Absorbance was measured at 590 nm.

Immunohistochemical staining was performed on formalin-fixed paraffin-embedded tissue sections (4-µm; formalin fixation was performed at room temperature with no time limit set for fixation). All sections were blocked with 5% BSA for 30 min at room temperature and incubated with anti-SKA2 (1:500; cat. no. ab91551; Abcam) at 4°C overnight. The sections were then incubated with a horseradish peroxidase-goat anti-rabbit IgG secondary antibody for 40 min at room temperature (1:100; cat. no. TA140003; OriGene Technologies Inc., Beijing, China). Finally, Myer's hematoxylin was used for background staining for 3 min at room temperature and photographs were taken using fluorescence microscopy. The appropriate positive (tumor tissues at different stages) and negative controls (normal tissue) were included. The positive cells and staining intensity of tumor tissues were scored respectively. According to the percentage of positive cells, <5% was 0, 5% ~<25% was 1, 25% ~<50% was 2, 50% ~<75% was 3, and <75% was 4. According to the coloring strength score, it could be divided into 4 grades: No coloring was 0, pale brown was 1, brown was 2, dark brown was 3. The result of the two products: 0 was negative, 1 was positive, in which <6 was low expression and ≥6 is high expression (26).

MCF-7 cells were fixed in 4% paraformaldehyde for 30 min at room temperature and then permeabilized with 0.1% Triton X-100 in PBS at room temperature for 15 min. The cells were blocked with 1% BSA overnight and incubated overnight at 4°C with Alexa Fluor^® 488-conjugated β-tubulin rabbit monoclonal antibody to (1:500; cat. no. 2116) or E-cadherin antibody (1:1,000; cat. no. 3195; both Cell Signaling Technology, Inc.). Then all cells were counterstained with 4′,6-diamidino-2-phenylindole (5:100; cat. no. C0060; Beijing Solarbio Science & Technology Co., Ltd., Beijing, China) for 15 min at room temperature and E-cadherin stained cells were counterstained with CY3-labeled goat anti-rabbit IgG (1:50; cat. no. BA1032; Wuhan Boster Biological Technology, Ltd.) for 1 h at room temperature. Finally, the coverslips were washed twice with PBS and images were captured using confocal scanning microscopy.

All experiments were repeated three times and data are expressed as the means ± standard deviation. The values between the two groups in Fig. 1A was compared with independent t-test. One-way analysis of variance and followed by Fisher's least significant difference tests was used to calculate P-values between the cell groups in the invasion and wound healing assays, and to compare measurements of the protein and mRNA expression levels in breast cancer cell lines. Data in Table I were analyzed using the χ² test of four by four table. Statistical analyses were performed using SPSS 15.0 software (SPSS, Inc., Chicago, IL, USA). P<0.05 was considered to indicate a statistically significant difference.

To investigate the role of SKA2 in breast cancer, SKA2 expression levels were measured in 80 patient samples by RT-qPCR and the levels were normalized to GAPDH. The results demonstrated that SKA2 mRNA was significantly upregulated in breast cancer tissues compared with in adjacent normal tissues (P<0.001; Fig. 1A). In addition, the mRNA (Fig. 1B) and protein (Fig. 1C) expression levels of SKA2 in three human breast cancer cell lines (MCF-7, T47D and MDA-MB-231) were evaluated and compared with the MCF-10A normal human mammary epithelial cell line. SKA2 was significantly increased in all three cell lines compared with in MCF-10A cells (P<0.05).

To investigate the clinical significance of SKA2 in patients with breast cancer, SKA2 immunohistochemical staining in a cohort of 160 specimens was analyzed. As demonstrated in Table I, high intratumoral SKA2 levels were significantly associated with clinical stage (P=0.001) and all lymph node metastasis groups (Luminal A, B HER2 and basal; P<0.001) but not with patient age, tumor size, histological grade, pathological type or molecular typing (Table I). In addition, intratumoral SKA2 levels increased gradually with disease progression from TNM stage I to IV (Fig. 2).

To investigate the function of SKA2 in breast cancer cell migration, siSKA2 specifically targeting SKA2 was used. Transfection with siSKA2 significantly decreased the protein expression levels of SKA2 in breast cancer cells (MCF-7 and T47D which were chosen as a result of their differential expression levels of SKA2; P<0.01; Fig. 3A). Following knockdown of SKA2 expression, its effects on cell migration and invasion in vitro were investigated using Transwell and scratch wound healing assays. Compared with the control cells, breast cancer cell migration and invasion were significantly inhibited in siSKA2-transfected cells (P<0.05; Fig. 3B and C).

It has previously been demonstrated that the extracellular matrix (ECM) serves a critical role in tumor invasion and metastasis (27). Since MMPs, particularly MMP2 and MMP9, destroy ECM processing, they are associated with tumor invasion and metastasis (28). The results of the present study demonstrated that siSKA2 transfection significantly decreased the expression levels of MMP2 and MMP9 in breast cancer cells (P<0.05; Fig. 3D).

MCF-7 cells were immunostained with anti-β-tubulin antibodies to determine the presence of microtubules. NC-transfected cells exhibited clear and intact microtubules with well-formed dendrites. In siSKA2-transfected cells, the microtubules were polymerized and the structure was spindle-shaped, which may lead to inhibited cell migration (Fig. 4).

EMT is a process in which epithelial cells lose epithelial characteristics and gain mesenchymal features, including motility and invasiveness (29). To identify the mechanisms underlying the decreased migration and invasion of siSKA2-transfected breast cancer cells (MCF-7 and T47D), the expression levels of E-cadherin, which is an epithelial marker that is downregulated during EMT, were measured. In addition, vimentin, fibronectin and N-cadherin, which are mesenchymal markers that are upregulated during EMT, were also measured (30). Notably, knockdown of SKA2 significantly decreased the expression levels of vimentin, fibronectin and N-cadherin (P<0.05); however, no significant effects were detected on the expression levels of E-cadherin (Fig. 5A). Subsequently, SKA2 expression was upregulated with SKA2cDNA, and the expression levels of E-cadherin were significantly decreased (P<0.05), whereas N-cadherin, fibronectin and vimentin levels were significantly increased (P<0.05; Fig. 5B and C). Therefore, SKA2 may mediate invasion and metastasis in human breast cancer via EMT.

To investigate the effects of SKA2 on the translocation of E-cadherin, cytoplasmic and nuclear proteins were isolated and extracted from MCF-7 cells (as SKA2 is highly expressed in this cell line and MCF-7 is frequently used in breast cancer studies this cell line was selected), and western blotting was performed. The results demonstrated that there were no obvious alterations in E-cadherin in the cytoplasm compared with in the MOCK group; however, E-cadherin levels were significantly increased in the nucleus following transfection with siSKA2 (P<0.05; Fig. 6A). Similar results were obtained by immunofluorescence (Fig. 6B). These findings indicated that the suppression of SKA2 facilitated the translocation of E-cadherin from the cytoplasm to the nucleus.