MDA-MB-231, MCF-7, MCF-10, BT549 and BT474 breast cancer cell lines were obtained from the Shanghai Institute of Life Sciences Cell Bank and cultured according to the manufacturer's instructions.

DMEM and FBS were purchased from Gibco (Thermo Fisher Scientific, Inc.). The empty plasmid pEGFP-n1-APP (cat. no. 69924) and pENTR APP short hairpin (sh)RNA (cat. no. 30135) plasmids were supplied by Addgene Inc. The transfection reagent polyetherimide (PEI; cat. no. 03880) was supplied by Sigma-Aldrich (Merck KGaA). PrimeScript RT reagent kit (Takara Bio, Inc.) and One Step SYBR-Green PrimeScript RT-PCR kit II (Takara Bio, Inc.) kits were used for reverse transcription (RT) and quantitative-PCR (q-PCR), respectively. Transwell chambers and Matrigel were purchased from BD Biosciences. Rabbit anti-human APP (1:2,000 for western blot analysis; 1:300 for immunohistochemistry; cat. no. 2452S), mouse anti-human E-cadherin (1:2,000; cat. no. 14472), mouse anti-human N-cadherin (1:2,000; cat. no. 14215), mouse anti-human cytokeratin (1:2,000; cat. no. 4545), mouse anti-human vimentin (1:2,000; cat. no. 49636), mouse anti-human MMP-9 (1:2,000; cat. no. 3852), rabbit anti-human MMP-2 (1:2,000; cat. no. 4022), rabbit anti-human MMP-3 (1:2,000; cat. no. 14351) and rabbit anti-human mitogen-activated protein kinase kinase kinase 11 (MLK3) primary antibodies (1:2,000; cat. no. 2817) were purchased from Cell Signaling Technology, Inc. Rabbit anti-human MEK4 (1:2,000; cat. no. ab33912), rabbit anti-human phosphorylated (p)-MEK4 (1:2,000; cat. no. ab131353), rabbit anti-human p-MLK3 (1:2,000; cat. no. ab191530), rabbit anti-human JNK3 (1:2,000; cat. no. ab126591), rabbit anti-human p-JNK3 (1:2,000; cat. no. ab124956) and rabbit anti-human β-actin primary antibodies (1:4,000; cat. no. ab179467), as well as horseradish peroxidase (HRP)-conjugated goat anti-rabbit (1:5,000; cat. no. ab6721) and goat anti-mouse (1:3,500; cat. no. ab6789) secondary antibodies were purchased from Abcam. TRIzol^® reagent was obtained from Thermo Fisher Scientific, Inc. qPCR primers were synthesized by Shanghai Biotech.

MDA-MB-231, MCF-7 and BT474 cells were cultured in DMEM containing 10% FBS and 1% streptomycin mixture, and then placed in a humidified atmosphere with 5% CO₂ at 37°C. Cell passaging was conducted using 0.25% trypsin + EDTA.

A total of eight female patients with breast cancer (age, 37-62 years) underwent clinical and histopathological diagnosis at the First Affiliated Hospital of Xiamen University between January and December 2018. All patients included in the study had clinical TNM stage III or IV breast cancer, and had not been treated with radiotherapy or chemotherapy prior to surgery. Written informed consent was obtained from each patient. The study protocol was approved by the Ethics Committee of the First Affiliated Hospital of Xiamen University. All immunohistochemical samples were retrieved from the Department of Pathology of the First Affiliated Hospital of Xiamen University. Immunohistochemical staining was performed on tissues. Samples were soaked in 10% neutral buffered formalin at 4°C and embedded in paraffin. Subsequently, the paraffined slices (thickness, 5 µm) were dewaxed and rehydrated, then placed on glass slides. Antigen retrieval was performed in a pressure cooker for 3 min at 120°C. Slices were then incubated with primary antibody (1:500) at 4°C overnight, followed by incubation with corresponding secondary antibodies for 2 h at room temperature. After washing in PBS, the slices were added with diaminobenzidine, stained with 10% hematoxylin for 10 sec at room temperature and sealed with neutral gum. Sections were visualized and photographed using an optical microscope (Leica Microsystems GmbH; magnification, ×200). Results were quantified using Image Pro Plus software (version 6.0; Media Cybernetics).

MDA-MB-231, MCF-7 and BT474 cells in the logarithmic growth phase were trypsinized and counted, and then seeded into a 6-well plate at a density of 5×10⁵ cells/well. To determine the effects of APP overexpression, five groups were defined as follows: i) Untransfected cells (PEI but no plasmid); ii) control plasmid (PEI with 2 µg control pEGFP plasmid); iii) control shRNA (PEI with 2 µg control pENTR plasmid); iv) APP overexpression (PEI with 2 µg pEGFP-n1-APP plasmid); and v) APP shRNA (PEI with 2 µg pENTR APP shRNA plasmid). For the subsequent experiments, cells were transfected and categorized into three groups as follows: i) Control (PEI but no plasmid); ii) overexpression (PEI with 2 µg pEGFP-n1-APP plasmid); and iii) silencing (PEI with 2 µg pENTR APP shRNA plasmid) groups. The transfected cells were used in subsequent experiments after 48 h.

MDA-MB-231 cells were cultured in six-well plates at a density of 5×10⁵ cells/well, and then treated with the MEK inhibitor PD0325901 (Sigma-Aldrich; Merck KGaA; 1 nM) for 48 h at 37°C. When drug treatment and transfection were combined, MDA-MB-231 cells were cultured in six-well plates at a density of 5×10⁵ cells/well, and then transfected with APP overexpressing plasmid for 24 h, followed by treatment with the MEK inhibitor PD0325901 (1 nM) for 48 h.

MDA-MB-231, MCF-7 and BT474 cells in the logarithmic growth phase were trypsinized and counted, then seeded in a 6-well plate at a density of 5×10⁵ cells per well. After reaching complete adherence and ~60% confluence, the cells were transfected with the corresponding plasmids. After 48 h of incubation, the cells were digested and counted, resuspended in serum-free medium, and inoculated into the upper Transwell chamber at a density of 1×10⁴ cells/well. The lower chamber was filled with 500 µl complete medium containing 10% FBS in a 24-well plate. After the upper and lower chambers were assembled, they were placed in a cell culture incubator for 4 h, and the upper chamber was removed. The Transwell inserts were scraped with cotton swabs to remove residual cells. Cells on the lower surface of the well were stained with 0.1% crystal violet solution for 30 min at room temperature and visualized using an optical microscope (Leica Microsystems GmbH; magnification, ×100).

The matrix collagen solution (8 mg/ml) was diluted 8X with serum-free medium. The diluted Matrigel (50 µl) was then added to the upper chamber of the Transwell insert in advance, and placed in a cell culture incubator for a minimum of 1 h at 37°C. After treating the cells as described in the migration assay, cells resuspended in the serum-free medium were seeded at a density of 1×10⁴ cells/well into the upper chamber of the Transwell insert, where the Matrigel solution had been previously added at room temperature and then put into culture incubator at 37°C for 30 min. The lower chamber was filled with 500 µl complete medium containing 10% FBS. After the upper and lower chambers were assembled, they were placed in a cell culture incubator for 12 h at 37°C. Next, the cells were treated as described in the migration assay section. Stained cells were photographed using an optical microscope (Leica Microsystems GmbH; magnification, ×100) and counted in five randomly selected fields.

A total of 1×10⁶ MDA-MB-231 cells were seeded into a 6-well plate, on which the UV-sterilized polydimethylsiloxane blocks (1 mm ×2 cm) were placed before. The culture medium was supplemented with 10% FBS. When the cells reached a confluence of 90%, the blocks were removed and the width was photographed using an optical microscope (Leica Microsystems GmbH; magnification, ×40) and measured at 24, 48 and 72 h. The gap closure values were calculated as follows: Percentage of gap closure= 1 -(width_t/width₀) ×100% (36).

Total RNA was extracted from MDA-MB-231, MCF-7 and BT474 cells using TRIzol reagent (Thermo Fisher Scientific, Inc.) at 4°C, according to the manufacturer's instructions. A ratio of A260/A280 was considered to be acceptable within the range of 1.8-2.0. The RT reaction was performed as follows: Initial incubation at 37°C for 5 min, followed by sequential steps at 42°C for 15 min, at 85°C for 10 sec and at 4°C for 60 min. β-actin was selected as the reference gene. For the q-PCR, the thermocycling conditions were as follows: Pre-denaturation at 95°C for 5 min, followed by 40 cycles of denaturation at 95°C for 15 sec and of annealing at 60°C for 15 sec. The 2^−ΔΔCq method (37) was used to quantify gene expression. The primer sequences used were as follows: β-actin forward, 5′-CCT CGC CTT TGC CGA TCC-3′ and reverse, 5′-GGA TCT TCA TGA GGT AGT CAG TC-3′; MMP-9 forward, 5′-AAT CTC ACC GAC AGG CAG CT-3′ and reverse, 5′-CCA AAC TGG ATG ACG ATG TC-3′; APP forward, 5′-TCT CGT TCC TGA CAA GTG CAA-3′ and reverse, 5′-GCA AGT TGG TAC TCT TCT CAC TG-3′; cytokeratin forward, 5′-ACC AAG TTT GAG ACG GAA CAG-3′ and reverse, 5′-CCC TCA GCG TAC TGA TTT CCT-3′; vimentin forward, 5′-GCC CTA GAC GAA CTG GGT C-3′ and reverse, 5′-GGC TGC AAC TGC CTA ATG AG-3′; MMP-2 forward, 5′-TAC AGG ATC ATT GGC TAC ACA CC-3′ and reverse, 5′-GGT CAC ATC GCT CCA GAC T-3′; MMP-3 forward, 5′-CTG GAC TCC GAC ACT CT G GA-3′ and reverse, 5′-CAG GAA AGG TTC TGA AGT G AC C-3′; E-cadherin forward, 5′-CGA GAG CTA CAC GTT CAC GG-3′ and reverse, 5′-GGG TGT CGA GGG AAA AAT AGG-3′; N-cadherin forward, 5′-TCA GGC GTC TGT AGA GGC TT-3′ and reverse, 5′-ATG CAC ATC CTT CGA TAA GA C TG-3′.

The cells were lysed with RIPA buffer (Beyotime Institute of Biotechnology) to extract the total protein from each group. Protein lysates were centrifuged at 12,000 × g for 15 min at 4°C and the protein concentration was determined using a bicinchoninic acid assay. Samples (30 µg per lane) were isolated using 8-10% SDS-PAGE and transferred to PVDF membranes, followed by blocking with 5% skimmed milk at room temperature for 2 h. Target proteins were detected with specific antibodies overnight at 4°C and observed on Bio-Rad Gel Imager using HRP-conjugated secondary antibodies (1:5,000; incubated for 3 h at room temperature) and enhanced chemiluminescence reagents (Western Lightning, Inc.). The bands were quantified using Quantity One 4.6 software (Bio-Rad Laboratories, Inc.).

Data analysis were performed using SPSS 21.0 (IBM Corp.). Graphs were drawn using GraphPad Prism 6.0 (GraphPad Software, Inc.). Multivariate mean comparisons were performed using one-way ANOVA with Dunnett's test. Student's t-test was used to analyze the difference between two groups. All experimental data were obtained from ≥3 independent experiments. Data are the presented as mean ± SD. P<0.05 was considered to indicate a statistically significant difference.

The expression of APP was detected by immunohistochemistry in human breast cancer tissues. The results shown in Fig. 1A indicated a significant upregulation of APP in breast cancer tissues compared with adjacent tissues (P<0.001). Subsequently, the expression levels of APP in five breast cancer cell lines were examined (Fig. 1B), and three cell lines (MDA-MB-231, MCF-7 and BT474) showing a moderate expression of APP were selected for further experiments. APP overexpression and knockdown were then performed on these selected cell lines.

Following transfection with the corresponding plasmids, it was found that the empty plasmids had no effects on the expression of APP in MDA-MB-231 cell line, therefore cells treated with PEI alone were used as the control in the subsequent experiments (Fig. 1C). The expression levels of APP in MDA-MB-231, MCF-7 and BT474 cells were detected by RT-qPCR and western blot analysis. The mRNA and protein expression levels of APP in cells transfected with the pEGFP-n1-APP plasmid were significantly increased compared with control cells (Fig. 1D-F; P<0.05). APP mRNA and protein expression levels in pENTR APP shRNA cells were significantly decreased compared with control cells (P<0.05). The present results suggested that the transfection experiments were successful.

The results of the Transwell assay showed that the migratory (Fig. 2A-C) and invasive (Fig. 2E-G) abilities of pEGFP-n1-APP-transfected cells were significantly enhanced compared with the control group (P<0.05), whereas pENTR APP shRNA-transfected cells exhibited significantly impaired migratory and invasive abilities compared with the control group (P<0.05). The results of the gap closure assay revealed that APP overexpression significantly promoted gap closure in MDA-MB-231 cells, while APP knockdown inhibited gap closure (Fig. 2D; P<0.05).

APP overexpression significantly upregulated the expression levels of the mesenchymal-related genes MMP-9, MMP-2, MMP-3, N-cadherin and vimentin in MDA-MB-231 cells (P<0.05), whereas the expression levels of the epithelial-related genes E-cadherin and cytokeratin were significantly attenuated (Fig. 3; P<0.05). APP knockdown exhibited the opposite effects (P<0.05).

APP overexpression notably increased the phosphorylation of MLK3, MEK4 and JNK3 in breast cancer MDA-MB-231 cells compared with the control group (Fig. 4; P<0.05). However, no significant differences were observed in the expression levels of total MLK3, MEK4 and JNK3 in any of the groups compared with the control. However, the phosphorylation levels of MLK3, MEK4 and JNK3 were significantly decreased following APP silencing (P<0.05).

MDA-MB-231 cells were cultured and seeded into 6-well plates at a density of 5×10⁵ cells/well. They were then treated with the MEK inhibitor PD0325901 (1 nM) for 48 h. Cells not treated with the inhibitor were used as the control group. After 48 h, total protein was collected to detect APP protein expression. In addition, Transwell and Matrigel assays were performed. PD0325901 treatment alone did not affect the expression levels of APP (Fig. 5A), but significantly inhibited the migratory and invasive abilities of breast cancer cells compared with the control group (Fig. 5B and C; P<0.05). Moreover, following APP overexpression, MDA-MB-231 cells (Fig. 5D) were treated with MEK inhibitor PD0325901, and it was found that PD0325901 could significantly reduce the migratory and invasive ability increased following APP overexpression, as shown by migration, invasion and gap closure assays (Fig. 5E-H; P<0.05).

PD0325901 treatment alone significantly inhibited the expression levels of MMP-9, N-cadherin and vimentin in MDA-MB-231 breast cancer cells (P<0.05), but significantly increased the expression levels of E-cadherin and cytokeratin (Fig. 6A-C; P<0.05). A decreased expression of epithelial-associated markers (E-cadherin and Cytokeratin) and increased expression of mesenchymal markers (MMP-9 and N-cadherin) was identified in APP overexpressing breast cancer cells compared with the control group (Fig. 6D-F; P<0.05). Conversely, PD0325901 treatment significantly reversed the expression levels of EMT-related genes caused by APP overexpression (Fig. 6D-F; P<0.05).

The present results suggested that PD0325901 treatment significantly inhibited the activation of JNK3, which is downstream of MEK (Fig. 7A-C; P<0.05), but did not affect the expression of total JNK3 compared with untreated cells. In addition, no significant differences were observed in the expression and activation levels of MEK4 and its upstream factor, MLK3. The overexpression of APP increased the phosphorylation level of MEK, MLK3 and JNK3. In addition, in MDA-MB-231 cells transfected with APP overexpressing plasmid and treated with the MEK inhibitor PD0325901 (1 nM), the phosphorylation levels of MEK4 and MLK3 were not significantly changed compared with the overexpression group (Fig. 7F). By contrast, PD0325901 treatment markedly reversed the phosphorylation of JNK3 caused by APP overexpression (P<0.05). Moreover, cell transfection and drug treatment had no effect on the protein expression levels of the unphosphorylated forms of MEK4, JNK3 and MLK3 in MDA-MB-231 cells (Fig. 7D-F).