Forty-nine clinical breast tumor tissues and the adjacent tissues (at least 5 cm away from the primary tumor) were collected at Xiangya Hospital, Changsha, China. All patients signed an informed consent form and the study was approved by the Independent Ethical Committee of Central South University, Changsha, China. Samples were stored at −80°C until use. Four aggressive breast cancer cell lines (MCF-7, MDA-MB-231, BT474 and BT549) and a normal breast cell line (MCF10A) were used. Cells were grown routinely in HEPES-buffered Dulbecco's modified Eagle's medium containing 10% fetal bovine serum (Gibco Life Technologies, Carlsbad, CA, USA) and cultured under 5% CO₂ humidified air.

Total RNAs were prepared using TRIzol reagent (Invitrogen Life Technologies, Carlsbad, CA, USA) according to the manufacturer's instructions. The extracted RNA was reverse-transcribed to cDNA using a PrimeScript reagent kit (Promega Corporation, Madison, WI, USA). The relative expression of CDCP1 mRNA was detected by SYBR-Green qPCR assay (Bio-Rad Laboratories, Inc., Hercules, CA, USA) performed on an ABI Prism 7700 (Applied Biosystems; Thermo Fisher Scientific, Inc., Waltham, MA, USA). β-actin was used as a control to normalize the starting quantity of RNA. The specific primers were as follows: CDCP1, F: 5′-TCTGCAAGGCTGTGACCAAG-3′, R: 5′-GCTCATTACTCAAGTCAACCAC-3′; β-actin, F: 5′-AGGGGCCGGACTCGTCATACT-3′, R: 5′-GGCGGCACCACCATGTACCCT-3′. The specific primers sets for miR-198, U6 and the PCR mix were purchased from GeneCopoeia, Inc. (Rockville, MD, USA). The expression of U6 was used as an endogenous control. Reactions for each sample were performed in triplicate. Relative expression levels were calculated using the 2^−∆∆Ct method.

Total cellular extracts were prepared from each group of cells with 200 ml lysis buffer and subjected to western blot analysis. Approximately 50 µg total protein was separated by sodium dodecyl sulphate-polyacrylamide gel electrophoresis, transferred to a polyvinylidene fluoride membrane and incubated with the indicated antibodies, followed by horseradish peroxidase-conjugated secondary antibody. Signals were visualized using enhanced chemiluminescence (ECL) substrates (Millipore, Billerica, MA, USA). The protein bands were visualized using an ECL detection kit (GE Healthcare Life Sciences, Chalfont, UK) as recommended by the manufacturer. β-actin was used for normalization. Antibodies of CDCP1 and β-actin were obtained from Abzoom (Abzoom Biolabs, Dallas, TX, USA).

A fragment of the 3′UTR of CDCP1 containing the predicted miR-198 target site was amplified and inserted into the psiCHECK-2 vector (Promega Corporation) downstream of the luciferase gene sequence. A psiCHECK-2 construct containing the 3′UTR of CDCP1 with a mutant sequence of miR-198 was synthesized. The wild-type 3′UTR of CDCP1 (Wt-3′UTR of CDCP1) and mutant 3′UTR of CDCP1 (Mut-3′UTR of CDCP1) primers were as follows: Wt-3′UTR of CDCP1, F: 5′-CTCGAGGCAAGCCCTGGATTCAGAGT-3′, R: 5′-GCGGCCGCGGATAACCACGAACCGACCTA-3′; Mut-3′UTR of CDCP1, F: 5′-GCGGCCGCGCAAGCCCTGGATTCAGAGT-3′, R: 5′-CTCGAGGGATAACCACGAACCGACCTA-3′. MCF-7 and MDA-MB-231 cells were plated in 96-well plates, then the Wt-3′UTR of CDCP1-psi-CHECK2 or the Mut-3′UTR of CDCP1-psi-CHECK2 was co-transfected with pre-miR-198 and pre-scramble mimics, respectively. The untreated group was used as a control. Luciferase activity was detected using a dual-luciferase reporter gene assay kit (Promega Corporation) and normalized to Renilla activity.

Loss of function is a powerful approach in the study of gene function. In this study, we used TALEN technology to knock out the CDCP1 gene in human breast cancer MCF-7 and MDA-MB-231 cells. TALENs designed to target CDCP1 gene were purchased from Sidansai Biotechnology (Shanghai, China). Cells in 24-well plates were transfected with 400 ng TALEN expression plasmids using Lipofectamine 2000 (Invitrogen Life Technologies) according to the manufacturer's instructions. Western blot analysis was used to examine the CDCP1 protein expression to validate the efficiency of TALEN plasmids.

Lentiviruses containing miR-198 (Lv-miR-198) or scramble (Lv-scramble) were purchased from GeneChem (Shanghai, China). The MCF-7 and MDA-MB-231 cells were cultured to 60–70% of the plates, and then a concentration of 3×10⁴ TU/well Lv-miR-198 or Lv-scramble lentivirus was added. RT-qPCR and western blot analysis were performed to determinate the mRNA and protein levels of CDCP1 in the MCF-7 and MDA-MB-231 cells after being infected for 7 days. The cells stably infected with lentivirus were expanded and harvested for further analysis.

3-[4,5-dimethylthiazol-2-yl]-2,5 diphenyl tetrazolium bromide (MTT) assay was performed to evaluate cell proliferation. Briefly, cells were allowed to grow in 96-well plates with 5,000 cells per well, and incubated for 24, 48 and 72 h, then 10 mg/ml MTT was added to the cells and incubated for 1 h. The reaction was then terminated by removal of the supernatant, and 200 µl dimethyl sulfoxide was added. After 1 h of incubation, the optical density at 570 nm of each well was measured with a microplate reader (Bio-Rad Laboratories, Inc.).

Cell migration was determined by Transwell assay. Cells suspended in serum-free medium were added into the upper chamber of the insert with Matrigel. Following 24 h incubation at 37°C, cells remaining on the upper side of the membrane were carefully removed, while cells that had migrated through the membrane were fixed with 75% alcohol and stained with crystal violet for 25 min, then washed with water and dried in air. The imaging and counting of cell numbers were performed using an inverted microscope (Nikon Corporation, Tokyo, Japan).

For adhesion assay, cells were seeded in a Matrigel-coated 96-well-plate. Following incubation for 1 h, the wells were washed twice with phosphate-buffered saline, fixed in 4% paraformaldehyde and stained with crystal violet. The imaging and counting of adherent cells were performed using an inverted microscope (Nikon Corporation).

All data are presented as the mean values ± standard deviation. Student's t-test was used to analyze the differences in the experiments. The Chi-squared test was used to demonstrate the differences in miR-198 or CDCP1 expression with clinicopathological features. P<0.05 was considered to indicate a statistically significant difference.

To examine the expression signature of miR-198 in human breast cancer progression, we first performed miRNA-based RT-qPCR analysis in 49 clinical tumor samples and matched adjacent tissues. As shown in Fig. 1A, we observed that the expression of miR-198 was significantly decreased in selected tumor tissues compared with the matched adjacent tissues. Correlation analysis of miR-198 expression with clinicopathological features revealed that downregulated miR-198 expression was significantly correlated with lymph node metastasis (P=0.036, Table I). We then moved to breast cell lines, and observed that the expression of miR-198 was significantly lower in the four invasive breast cancer cell lines (BT474, MDA-MB-231, MCF-7 and BT549 cells) than in the normal breast cell line MCF10A (Fig. 1B). Thus, our data suggest a strong link between downregulation of miR-198 and the pathogenesis of breast cancer. Contrary to miR-198, we observed that CDCP1 was significantly upregulated in selected clinical tumor samples (Fig. 1C) and invasive breast cancer cell lines (Fig. 1D). We also analyzed the association of CDCP1 expression with clinicopathological parameters, and observed that high CDCP1 expression levels were correlated with lymph node metastasis (P=0.028, Table II). These data suggest that upregulation of CDCP1 may be involved in breast cancer progression.

The findings above, as well as the computational prediction (Fig. 2A) prompted us to further investigate whether miR-198 directly targets CDCP1. To do so, we cloned the wild-type 3′UTR (Wt-3′UTR) of CDCP1 containing the predicted binding site of miR-198 downstream of a luciferase reporter gene (Fig. 2A). We also constructed its mutant version (Mut-3′UTR of CDCP1) by binding site mutagenesis. The vectors were co-transfected with miR-198 mimics (pre-miR-198) or corresponding scrambled mimics (pre-scramble) as controls into MCF-7 and MDA-MB-231 cells, respectively. The luciferase activity of cells transfected with miR-198 mimic was significantly decreased compared with that of control cells (Fig. 2A). Additionally, the miR-198-mediated repression of luciferase activity was abolished by the mutant putative binding site (Fig. 2A). Furthermore, we tested the inhibitory effect of miR-198 on CDCP1 expression in MCF-7 and MDA-MB-231 cells. RT-qPCR and western blot analysis revealed that enhanced miR-198 significantly decreased CDCP1 mRNA and protein levels compared with cells transfected with control in MCF-7 and MDA-MB-231, respectively (Fig. 2B-D). Taken together, our results suggest that CDCP1 is a direct functional target of miR-198 in breast cancer cells.

With the understanding that miR-198 is significantly downregulated in breast cancer tissues, we investigated whether miR-198 might serve as a tumor suppressor in breast cancer. We restored miR-198 expression in MCF-7 and MDA-MB-231 cells, which demonstrated a lower expression of miR-198 in the four selected breast cancer cell lines, by lentiviral infection with Lv-miR-198 or Lv-scramble lentivirus. RT-qPCR was performed to confirm that miR-198 was upregulated in MCF-7 and MDA-MB-231 cells following Lv-miR-198 infection (Fig. 3A). We then investigated the effect of miR-198 on cell proliferation, migration and adhesion, respectively. The MTT assay revealed that overexpression of miR-198 inhibited the proliferation of MCF-7 and MDA-MB-231 cells (Fig. 4A). Transwell assay indicated that the enhanced expression of miR-198 could significantly inhibit cell migration ability compared with the control group in MCF-7 and MDA-MB-231 cells (Fig. 4B). Moreover, cell adhesion assays revealed that miR-198 notably promoted MCF-7 and MDA-MB-231 cell adhesion (Fig. 4C). Taken together, our findings suggest that miR-198 may play a suppressive role in breast cancer cell growth and migration.

To investigate the precise function of CDCP1 in breast cancer cells, we next silenced CDCP1 in breast cancer cells. We knocked out CDCP1 using TALEN technology, which represents a promising approach for targeted knockout of genes in cultured human cells (22). Western blot analysis was performed to confirm TALEN-mediated knockout efficiency in MCF-7 and MDA-MB-231 cells. As shown in Fig. 3B, the CDCP1 gene was silenced effectively in CDCP1-TALEN vector-transfected cells. Similar to miR-198 restoration, silencing CDCP1 by TALEN inhibited cell proliferation and migration ability (Fig. 4A and B), and promoted cell adhesion (Fig. 4C). We further knocked out CDCP1 in MCF-7 and MDA-MB-231 cells stably infected with Lv-miR-198. As expected, a combination of silencing CDCP1 by TALEN and restored miR-198 had a more enhanced inhibitory effect on proliferation and migration ability than either silencing CDCP1 by TALEN or restoring miR-198 alone (Fig. 4A and B). Cell adhesion assays also revealed that cell adhesion was further promoted by the combinational treatment (Fig. 4C). Collectively, our results suggest that CDCP1 is involved in cell growth and migration of breast cancer cells.