We obtained 172 paraffin-embedded surgical BrCa tissue specimens from the Department of Pathology at the First Affiliated Hospital of Anhui Medical University (Hefei, Anhui, China). We recruited 68 female patients with benign breast diseases from the same hospital. The 68 female patients were used as the control group. No patients had received preoperative adjuvant therapy. Two pathologists confirmed the diagnoses of breast tumors in all the samples used. All cases were followed-up for 60 months. Overall survival (OS) was calculated on the basis of data. Patient samples were categorized into two groups in accordance with median expression (high vs. low expression) (Table I) and assessed using a Kaplan-Meier survival plot. The protocol of the present study was approved by the Institutional Review Board of Anhui Medical University. Detailed information of patients including ER/PR expression, lymph node metastasis and age are listed in Table II. The agreement on the use of these tissue samples was approved by the Biomedical Ethics Committee of Anhui Medical University and the confidentiality of patient information was maintained. We obtained informed consent from all patients before enrolling participants in the present study.

The Oncomine microarray database (http://www.oncomine.org) was used for analysis. We screened for relevant entries on ARPC2 expression in BrCa tissues and normal breast tissues. With ‘Cancer vs. normal’ analysis as the primary filter. Data sets were used in the present study in accordance with Ma et al (23), Curtis et al (24), Perou et al (25) and Zhao et al (26). We screened for differentially expressed genes (Pearson's correlation >0.4) in several important node locations of ARPC2 in BrCa (Pearson's correlation >0.4) by using BioPortal for Cancer Genomics (http://www.cbioportal.org). All searches were performed according to the cBioPortal's online instructions.

Human BrCa cell lines (MCF-10A, MCF-7, T47D, BT474 and MDA-MB-231) were obtained from the American Type Culture Collection (ATCC; Manassas, VA, USA) and cultured under ATCC-recommended conditions. Cells were routinely incubated at 37°C and 5% CO₂ under a humidified atmosphere.

The full length of Homo sapiens ARPC2 was synthesized and subcloned into an entry vector by OriGene Technologies Technologies, Inc., (Beijing, China). siRNA-ARPC2 (siARPC2) was purchased from Shanghai GenePharma Co., Ltd. (Shanghai, China). MCF-7 and MDA-MB-231 cells were transfected using Lipofectamine 2000 (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA) in accordance to the manufacturer's instructions. Transfection efficiency was confirmed by performing reverse transcription-quantitative real-time polymerase chain reaction (qRT-PCR) and western blotting.

We used RT-PCR to evaluate the mRNA levels of ARPC2. Total RNA was isolated from cultured cells using TRIzol reagent. RNA was reverse transcribed into cDNA using the PrimeScript RT reagent kit (Takara Biotechnology Co., Ltd., Dalian, China). qRT-PCR was performed using the SYBR Prime Script kit (Takara Bio, Inc., Otsu, Japan). ARPC: Forward primer, 5′-TCCGACTCTACCAGCTGATGC-3′ and reverse primer, 5′-AAGCTGGACTCATCCCACAGC-3′. PCR was conducted at 95°C for 15 min, followed by 40 cycles of 94°C for 15 sec, 55°C for 30 sec, and 64°C for 30 sec. Experiment was repeated three times and the expression of ARPC2 was quantified using the 2^−ΔΔCq method (27). Cells were lysed via a lysis buffer which consisted of RIPA (Beyotime Institute of Biotechnology, Haimen, China). The protein concentration was determined using BCA kit. Equal amounts of total protein (10 µg) were separated by 10% SDS polyacrylamide gels and transferred onto polyvinylidene fluoride (PVDF) membranes, followed by blocking with 5% non-fat milk for 1 h. The membranes were incubated with primary antibodies overnight at 4°C. Protein expression was quantified by densitometry (Quantity One software; Bio-Rad Laboratories, Hercules, CA, USA). The antibodies used for western blotting were: N-cadherin (cat. no. 4061; Cell Signaling Technology, Inc., Danvers, MA, USA), E-cadherin (cat. no. 5296; Cell Signaling Technology, Inc.), vimentin (cat. no. 5296; Cell Signaling Technology, Inc.), ZEB1 (cat. no. 3396; Cell Signaling Technology, Inc.) and MMP-9 (cat no. 3852; Cell Signaling Technology, Inc.), MMP-3 (cat. no. ab52915; Abcam, Cambridge, UK), β-actin (cat. no. ab5694; Abcam), ARPC2 (cat. no. ab11798; Abcam) and GAPDH (cat. no. g9545; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany). The bands were displayed by ECL illumination and analyzed by the Tanon 5200 image acquisition system (Tanon Science and Technology Co., Ltd., Shanghai, China).

For the colony formation assay, transfected cells were grown in a fresh 6-well plate with RPMI-1640 medium (Gibco; Thermo Fisher Scientific, Inc.) supplemented with 10% fetal bovine serum. For the cell colony formation assay, the transfected cells were seeded in 6-well plates (1×10³/well) and cultured for 14 days at 37°C. After 14 days, all visible colonies per well were manually counted. Assays were performed in triplicate, and the results were calculated as the mean ± standard deviation (SD).

The siRNA-control (siControl) and siRNA-ARPC2 (siARPC2) were transfected into MCF-7 and MDA-MB-231 cells for 48 h. For the migration assay, 5×10⁴ cells were seeded with serum-free RPMI-1640 medium in the upper chamber of Transwell inserts (8-µm; Corning Incorporated, Corning, NY, USA). For the invasion assay, inserts were coated with Matrigel (Sigma-Aldrich; Merck KGaA). For the migration and invasion assays, the lower chamber of the Transwell unit was filled with culture medium supplemented with 10% serum. After 36–48 h, the cell filters were fixed using methanol and stained with 0.1% crystal violet. The number of cells was manually counted. Each experiment was assessed in triplicate.

Cells (5×10⁶) were seeded in 6-well plates. Once the cells achieved 100% confluence, the culture plate was scratched along its central axis of with 200-µl sterile pipette tips. An inverted microscope was used to observe wound healing at three time-points (0, 24 and 48 h) after wounding. The wound healing rate was calculated and wounds were observed and images were captured. The assays were performed three times.

Five-week-old, female, and BALB/c nude mice (7 in each group) were used for the xenograft tumor formation assay. All animal testing was performed at the Model Animal Research Center of The University of Science and Technology of China (Hefei, China). All of the experiments were approved by the Institute Research Ethics Committee of Anhui Medical University (Hefei, China). The suspended cells (5×10⁶) were injected into each mouse on the right side of the posterior flank. Seven days after injections, tumor nodules were checked every 5 days and tumor volumes were calculated by width × length × (width + length)/2. All tumors were removed from the nude mice after 40 days.

For IHC staining, 4-µm sections of the TMA blocks were incubated overnight with a pre-diluted goat antihuman ARPC2 monoclonal antibody (Fuzhou Maixin Biotech Co., Ltd., Fuzhou, China). Immunoreactivity was evaluated semi-quantitatively, and the result was calculated on the basis of the percentage of positive cells. Immunostaining was considered as low positive when the percentage of the stained tumor cells was <30%, and as high positive when the percentage of stained tumor cells exceeded 30%.

Cells (0.5×10⁶−1×10⁶) were washed twice with cold phosphate-buffered saline. Cells were stained with propidium iodide (PI; Cycletest™ Plus DNA reagent kit; BD Biosciences, Franklin Lakes, NJ, USA) in accordance with the manufacturer's protocol. Analyses and measurements were performed using FACSVerse flow cytometer (BD Biosciences). The percentage of cells in the S, G0/G1 and G2/M phases were counted and compared. Similarly, cell apoptosis was analyzed by flow cytometry using the Annexin V-FITC/PI Apoptosis Detection kit (BD Biosciences) 48 h after transfection.

Data were analyzed with SPSS 19.0 software (SPSS Inc., Chicago, IL, USA). The association between the ARPC2 expression level and the clinopathological characteristics of patients was analyzed by the Chi-square test. Student's t-test was used for comparing the differences between groups. Cumulative recurrence and survival rates were analyzed by the Kaplan-Meier method. P<0.05 was considered to indicate a statistically significant difference.

We quantified the expression of ARPC2 in different types of human BrCa cell lines (MCF-7, T47D, BT474 and MDA-MB-231) and human breast epithelial cell line (MCF-10A) by RT-PCR and western blotting. As revealed in Fig. 1A and B, we found that ARPC2 was highly expressed in BrCa cell lines. However, changes in ARPC2 expression in four types of BrCa cell lines did not exhibit a consistent pattern. Since MCF-7 and MDA-MB-231 cell lines had a good tumorigenic effect, we selected MCF-7 and MB-MDA-231 cell lines as experimental models.

We performed IHC to assess ARPC2 protein expression in different types of BrCa (n=172) and paired normal breast (n=68) tissues. As revealed in Fig. 1C, the stain for ARPC2 protein expression was predominantly located in the cytosol. Our results demonstrated that the expression level of ARPC2 protein in normal tissues was significantly lower than in the cancer tissues (Table I). The Kaplan-Meier curve and log-rank test analyses revealed that ARPC2 expression was associated with the OS of patients with BrCa. We evaluated the prognostic value of ARPC2 using the protein expression data and survival information of 172 patients with BrCa. We divided patient samples into two groups in accordance with the median expression (high vs. low expression). As revealed in Fig. 1D, high ARPC2 levels were associated with a low OS rate. Moreover, to illustrate the potential relationship between ARPC2 and BrCa progression, we used the Oncomine database (http://www.oncomine.org) for the analysis of ARPC2 mRNA expression. We found that ARPC2 mRNA expression was significantly upregulated in BrCa relative to that in normal breast tissues (Fig. 1E).

We analyzed the relationship between ARPC2 protein expression and the clinicopathological parameters of patients with BrCa. As presented in Table II, ARPC2 protein expression was significantly and positively associated with lymph node metastasis, tumor size and grade of BrCa (P<0.05). The protein expression of ARPC2, however, was not significantly associated with other clinical parameters, including age, HER-2 expression, and estrogen and progesterone receptor status of patients with BrCa.

To delineate the role of ARPC2 in BrCa tumorigenesis, we knocked out ARPC2 in MDA-MB-231 cells and overexpressed ARPC2 expression in MCF-7 cell lines. We confirmed ARPC2 expression by performing western blotting and qRT-PCR (Fig. 2A and B). To further confirm the role of ARPC2 in BrCa migration and invasion, we used a colony formation and scratch wound healing assays (Fig. 2C and D). The colony forming ability of cells was decreased and the percentage of wound closure was increased by ARPC2 silencing in MDA-MB-231 cells. Similarly, the migration and the invasion of MDA-MB-231 was also inhibited by ARPC2 silencing as determined by Transwell assays. Thus, the inducive effect of ARPC2 in BrCa cell migration and invasion was revealed.

We assessed the effects of pcDNA3.1-ARPC2 transfection on cellular proliferation and growth through colony formation assays. Our results demonstrated that pcDNA3.1-ARPC2 displayed a relatively rapid proliferation rate compared to control cells when MCF-7 cells were transfected with pcDNA3.1-ARPC2. The colony forming ability (Fig. 2C) as well as the migration and invasion abilities (Fig. 2E) were enhanced as determined by colony formation and Transwell assays.

We downregulated ARPC2 expression in the MDA-MB-231 cell line through transfection with an ARPC2-specific siRNA or a non-specific siRNA control. As revealed in Fig. 3A, ARPC2 silencing altered the cell cycle of the MDA-MB-231 cell line. We found that a higher number of siARPC2 MDA-MB-231 cells were in the G1 phase compared with the control cells. In addition, the transfection of siARPC2 increased apoptosis of MDA-MB-231 cells, as determined by performing Annexin V-FITC/PI staining (Fig. 3B).

ARPC2 overexpression significantly increased the level of phosphorylated Smad2 (p-Smad2) protein, a downstream effector of the TGF-β pathway (Fig. 4A). Moreover, ARPC2 overexpression efficiently increased phosphorylated Smad2 levels. To further validate that TGF-β signaling is responsible for ARPC2-induced EMT, we used TGF-β receptor kinase inhibitor SB43 to block TGF-β signaling. The results revealed that suppression of TGF-β signaling reduced Smad2 phosphorylation levels and downregulated vimentin expression (Fig. 4B). To investigate whether ARPC2 is required for TGF-β-induced EMT, we transfected MDA-MB-231 cells with siARPC2 or non-target control siRNA and examined their responses to TGF-β treatments after 48 h. It was determined that TGF-β stimulation induced EMT transfer in MDA-MB-231-siCtrl cells, but not in MB-MDA-231-siARPC2 cells. TGF-β treatment inhibited E-cadherin expression and increased vimentin expression. However, under the same conditions E-cadherin and vimentin expression levels were not altered in MB-MDA-231-siARPC2 cells (Fig. 4C). These experiments suggested that ARPC2 is involved in TGF-β-induced EMT.

Cellular adhesive ability is associated with EMT. To determine whether ARPC2 expression triggers EMT in BrCa cells, we examined MMP-3, MMP-9 (cell migration marker), E-cadherin (epithelial cell marker), N-cadherin, vimentin (mesenchymal marker) and ZEB-1 protein expression after transfection with siARPC2 and pcDNA3.1-ARPC2 (Fig. 4D and E). Compared to the control cell line, the expression levels of, vimentin, MMP-3, MMP-9 and ZEB-1 were upregulated by transfection with pcDNA3.1-ARPC2 and the level of E-cadherin was downregulated. These results demonstrated that ARPC2 stimulated the expression of EMT-related proteins and EMT-TFs. The transfection of pcDNA3.1-ARPC2 resulted in a markedly high invasiveness of BrCa. Knockdown of ARPC2 by siRNA resulted in the inhibition of invasion and metastasis of BrCa. To examine whether the TGF-β could stimulate the expression of ARPC2 in MDA-MB-231 cells, cells were incubated with the presence 10 ng/ml TGF-β for 12, 24 and 48 h. Subsequently, we used 2.5, 5, and 10 ng/ml TGF-β to culture cells for 48 h. qPCR analysis revealed that ARPC2 mRNA expression was time- and dose-dependent on TGF-β stimulation (Fig. 4F).

We demonstrated the biological role of ARPC2 in MDA-MB-231 cells through an in vivo experiment. We transfected the ARPC2 vector into MDA-MB-231 cells which were then subcutaneously injected into BALB/c nude mice. MDA-MB-231 and ARPC2-MDA-MB-231 groups were examined 40 days after implantation (Fig. 5). The ARPC2-MDA-MB-231 group of mice formed significantly larger tumor sizes than the MDA-MB-231 control group (Fig. 5C). As revealed in Fig 5, a high ARPC2 level increased tumor growth and weight (Fig. 5A and B), and generated lung or lymphonodus metastases in nude mice (Fig. 5D). We subsequently explored the effects of ARPC2 on tumor growth in vivo by IHC, the results of which revealed that tumour growth and tumor weight were significantly increased in the ARPC2-overexpression group compared with the negative control group and that the Ki-67 expression levels in the ARPC2-overexpression group were higher than those in the negative control group.