Cell culture

MCF-7 was purchased from the American Type Culture Collection (Manassas, VA, USA). Tamoxifen-resistant MCF-7 cells (TAMR) and adriamycin-resistant MCF-7 cells (ADR) were generous gift from Prof. Keon Wook Kang (Seoul National University, Republic of Korea). MCF-7 cells were cultured in DMEM medium supplemented with 10% FBS and 1% penicillin/streptomycin whereas ADR and TAMR cells were cultured in DMEM medium supplemented with 10% FBS, 1% P/S and 3 µM doxorubicin and tamoxifen, respectively. All cells were incubated at 37 °C in a 5% CO2 atmosphere.

The passage (P) numbers of cells utilized for in vitro studies were as follows: P28 to P106 for MCF-7 cells, P42 to P84 for adriamycin-resistant (ADR) cells, P47 to P83 for tamoxifen-resistant (TAMR) cells, P5 to P10 for SUZ12-overexpressing (OV) cells, and P2 to P10 for EZH2-OV cells.

Stable cell line generation

MCF-7 cells stably overexpressing SUZ12 or EZH2 were generated by transfecting MCF-7 cells with 5 µg SUZ12 (Cone ID: TRCN0000475819) or EZH2 plasmid (Clone ID: TRCN0000467064) (Mission shRNA plasmid DNA, Sigma-Aldrich) in the presence of lipofectamine 3000 (Invitrogen, Carlsbad, CA, USA), following the protocol outlined in the technical bulletin of MISSION lentiviral transduction particles (Sigma-Aldrich). After 48 h of transfection, transfected cells were selected by culturing with DMEM medium supplemented with 10% FBS and 2 µg/mL puromycin. Confirmation of the overexpression of each target gene was achieved through immunoblotting analysis.

The generation of luciferase-expressing TAMR cells followed the protocol outlined in the Luciferase Plasmid Guide (Addgene). Briefly, 5 µg of pCAG-Luciferse plasmid DNA (55,764, Addgene) was transfected to TAMR cells along with Lipofectamine 3000 (Invitrogen). Luciferase-expressing cells were selected by culturing them at geneticin (10 mg/mL)-containing media.

Transcriptome and bioinformatics analysis

mRNAs were extracted and analyzed using the Nanostringn Ncounter Pancancer Pathway array kits and systems, provided by PhileKorea (Seoul, South Korea). Genes showing ≥ 2-fold change (p < 0.05) were considered differentially expressed genes (DEG) and were analyzed using Panther 16.0 software [37], followed by GO Enrichment Analysis tool (http://geneontology.org/) to validate the GO terms associated with Epithelial-Mesenchymal Transition (EMT), and StemChecker (http://stemchecker.sysbiolab.eu/) [38] to predict the stemness signature molecule.

To analyze the association between the expression of ITGA11 or EZH2, and the overall survival (OS) and relapse-free survival (RFS) in breast cancer patients, the Kaplan‑Meier‑plotter (KM plotter, http://kmplot.com/analysis/) was utilized. The cohort was classified based on high and low expression levels and with an autoselect best cutoff. The gene expression profiles of GSE20711, GSE20685, and GSE3494 were obtained from the NCBI Gene Expression Omnibus (GEO) database.

siRNA transfection

siRNA transfection followed the siTran siRNA transfection application guide. Briefly, cells were seeded in antibiotic-free media and transfected with 100 nM siRNA targeting ITGA11, ITGB1, EZH2, SUZ12, HIF-1α, or GLI-1 (ORIGENE, Rockville, MD, USA) for 72 h using DharmFECT reagent 1 (Thermo Scientific, Waltham, MA, USA).

Cell viability assay

Cell viability was measured by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay [39]. Cells were seeded in 96-well plates in culture media containing 1% FBS for 24 h. Then, the cells were treated with vehicle, tamoxifen, or doxorubicin at the indicated concentrations. After 48 h, MTT solution (Merck, Burlington, MA, USA) was added. After 4 h of incubation, the media with MTT solution was removed and dimethyl sulfoxide (DMSO) was added to dissolve the formazan crystal. Absorbance was measured at 540 nm using a microplate reader (BMG LABTECH GmbH, Ortenberg, Germany).

Measurement of apoptosis and caspase 3 activity

Apoptosis was measured using the FITC Annexin V apoptosis detection kit (BD Biosciences, San Jose, CA, USA) following to the manufacturer’s instructions. Briefly, cells treated with 100 nM non-target siRNA (siNT) or ITGA11 siRNA (siITGA11) for 72 h were trypsinized and washed with ice-cold PBS. Thereafter, 1 × 105 cells were suspended in 1× binding buffer (100 µL) and were stained with propidium iodide (5 µL) and Annexin V-FITC (5 µL) and incubated in the dark for 15 min at 25 °C. After that, 400 µL of 1× binding buffer was added and analyzed by flow cytometry (FACSVerse Cytometer, BD Biosciences, San Jose, USA).

Caspase-3 activity assay was performed using a caspase-3 assay kit (Abcam, Cambridge, MA, USA) following the manufacturer’s instructions. Cells transfected with siNT or siITGA11 were treated either with tamoxifen or doxorubicin (10 µM for 48 h). Then, cells were lysed and quantified for total protein using BCA protein assay reagent (Pierce; Rockford, IL, USA). Samples containing 200 µg of total protein were assayed for caspase-3 activity with DEVD-pNA (200 µM) as a caspase-3-specific substrate. Samples were incubated at 37 °C for 120 min and absorbance was measured at 405 nm using a microplate reader (BMG LABTECH).

Western blot analysis

Cells were lysed with radioimmunoprecipitation assay (RIPA) buffer (Thermo Scientific, Waltham, MA, USA) containing 1X protease and phosphatase inhibitor cocktail (Thermo Scientific) for total protein extraction. Nuclear and cytoplasmic proteins were extracted using NE-PER nuclear and cytoplasmic extract reagents (Thermo Scientific), respectively. The proteins separated by SDS-PAGE were transferred to a nitrocellulose membrane (Whatman GmbH, Dassel, Germany) and subjected to immunoblotting with primary and secondary antibodies. Immunoblots were visualized using an ECL kit (Thermo Scientific). Images were captured using the LAS-4000 mini system (Fuji, Tokyo, Japan).

Co-immunoprecipitation assay (co-IP)

Immunoprecipitation was conducted using a 100 µg of total proteins, employing an IP-grade Integrin α11 antibody or IgG (1 mg/mL) (Sigma-Aldrich), incubated for 16 h at 4 °C. Subsequently, Protein A agarose beads (50 µL) (Thermo Scientific) were added in the immunoprecipitated solution and allowed to incubate for 4 h at 4 °C. The resultant immune complexes were collected through centrifugation at 3000× g for 2 min at 4 °C. The collected pellet was washed with PBS and then re-suspended in 25 µL of 1× sample buffer (62.5 mM Tris-HCl pH 6.8, 2.5% SDS, 0.002% Bromophenol Blue, 0.7135 M (5%) β-mercaptoethanol, 10% glycerol), followed by heating at 95 °C for 5 min. After centrifugation at 12,000× g for 30 s at 4 °C, supernatant containing the IP samples was collected.

Sphere formation assay

Cells (1 × 103) were seeded onto 24-well low attachment plates (Corning Costar, Corning, NY, USA) in prEGM media (Lonza, Basel, Switzerland) and grown in spheroids. After 15 days, an inverted microscope (TE2000-U; Nikon, Tokyo, Japan) was used to capture images of the spheres. The number of spheroids larger than 50 μm in diameter was counted using Image J 1.48v software (National Institute of Health, Bethesda, MD, USA).

Flow cytometry analysis for stem cell population

Single cell suspension (1 × 107 cells/mL) in cold PBS containing 3% FBS were stained with APC-anti-human CD24, FITC-anti-human CD44, and their respective isotype control antibodies (APC-anti-mouse IgG 2a, and FITC-anti-mouse IgG1) for 30 min in the dark at 4 °C. Stained cells were washed twice and analyzed by flow cytometry (FACSVerse Cytometer, BD Biosciences).

Immunocytochemistry

Breast cancer cells (1 × 105 cells) were seeded on confocal dishes (SPL Life Science, Pocheon, Korea). After 24 h incubation at 37 °C, cells were fixed with 4% paraformaldehyde in PBS (pH 7.4) for 10 min at 25 °C. After fixation, cells were washed three times for 5 min with PBS and permeabilized with 0.1% Triton X-100 for 10 min at 25 °C. Cells were then incubated with 1% BSA (bovine serum albumin) in 1× PBST for 1 h to prevent non-specific binding. After blocking, cells were incubated with E-cadherin and Vimentin primary antibodies in 1% BSA in PBST overnight at 4 °C, stained with an Alexa fluor 488 anti-mouse and Alexa fluor 647 anti-rabbit secondary antibody in the dark for 1 h at 25 °C, washed, counterstained with 1 µg/ml DAPI, and rinsed with PBS. Images were captured at 400X magnification using an inverted fluorescence microscope (TE2000-U; Nikon).

Transcription factor (TF) activation profiling array

A Cancer Stem Cell TF Activation Profiling Plate Array (Signosis, Santa Clara, CA, USA) was used to compare the activity of stemness-associated TFs. In brief, nuclear lysates were treated with biotin-labeled probes containing consensus sequences of TF DNA binding sites for 30 min at 25 °C. The spin column purification method was used to separate the TF/probe complex mixtures. Using an elution buffer, the bound probes were released from the complex and centrifuged at 9,800 g for 2 min. After the eluents were denatured at 98 °C for 5 min, the denatured sample was added to TF hybridization buffer. The resulting mixture was then put to each well of the hybridization plate, the plate was sealed with aluminum adhesive and the hybridization plate was then incubated at 42 °C for 16 h. In order to find the bound DNA probe, a streptavidin-horseradish peroxidase conjugate was used. Endpoint luminescence readings of the samples were observed using Fluostar omega (BMG LABTECH, Ortenberg, Germany).

Chromatin immunoprecipitation (ChIP) assay

Chromatin extraction and subsequent ChIP assay were performed using chromatin extraction kit (Abcam) and ChIP Kit-One Step (Abcam) according to manufacturer’s instructions. After chromatin cross-linking with 1% formaldehyde, chromatin was extracted and sheared by sonication. The lysates were incubated with antibodies against HIF-1α, GLI-1, EZH2, RNA Pol II or IgG for 2 hours at 25°C. Then, samples were digested using proteinase K, and DNA was subjected to qPCR using SYBR Green (Qiagen; Germantown, MD, USA) and primers (Bioneer; Daejeon, Korea). The primers used were ITGA11 (Forward 5′-CACGACATCAGTGGCAATAAG-3′ and Reverse 5′-GACCCTTCCCAGGTTGAGTT-3′), ITGB1 (5′-GCAAGCTCAGGCATAACAGC-3′ and 5′-CCCTGGCTCAGAGAGAATGC-3′), ITGB8 (5′-CTGTTTGCAGTGGTCGAGGAGT-3′ and 5′-TGCCTGCTTCACACTCTCCATG-3′), EZH2 (5’- CCCTGACCTCTGTCTTACTTGTGGA − 3’ and 5’-ACGTCAGATGGTGCCAGCAATA-3’ or GAPDH (5′-ACCACAGTCCATGCCATCAC-3′ and 5′-TCCACCACCCTGTTGCTGTA-3′). Input DNA (1%) was used for percentage binding analysis.

Migration assays (wound healing and transwell migration)

For wound healing assay, TAMR or ADR cells transfected with siNT, siITGA11, siSUZ12 or siEZH2 were seeded in 24 well plates. Cells were scratched with sterile 10 µL disposable plastic pipette tips and washed with PBS. The cells were incubated with serum-free medium containing 10 µg/mL Mitomycin C (to inhibit cell proliferation). After 2 h, the cells were treated with 5% serum as an inducer. The wound healing procedure was observed by microscopy, and images were captured at 0 and 48 h after treatment with an inducer. The distance between the wound edges was measured using Image J 1.48v software.

For transwell migration assay, 1 × 106 cells per well were seeded in the upper chamber (8.0-µm pore membrane) in 100 µL of 1% FBS containing medium, and 600 µL of 5% FBS containing medium was added to the bottom chamber. After 18 h incubation at 37 °C in a 5% CO2, the cells on the upper surface of the membrane were removed with cotton swab, and the migrated cells on the lower surface of the membrane were fixed using methanol, and stained with hematoxylin and eosin. The numbers of cells migrated per field were captured at 200× magnification using a digital camera fitted to the inverted microscope (Nikon, Tokyo, Japan). The captured images were used to count the number of migrated cells.

Animal experiment ethics

The animal experiments (chick embryo and mouse experiments) were approved beforehand by the Institutional Animal Care and Use Committee (IACUC) of Yeungnam University and were performed following the institutional guidelines of the Institute of Laboratory Animal Resources (1996) and Institutional Animal Care and Use Committee of Yeungnam University (2009).

Measurement of anti-tumor and anti-metastatic activity using chick chorioallantoic membrane (CAM)

Fertilized eggs were procured from Byeolbichon Farm (Gyeongbuk, South Korea), and eggs were incubated at 37 °C in 55% relative humidity. A small hole was made using a hypodermic needle on the wider side of 9-day-old fertilized eggs after selection of bifurcated vessels. Another hole was made on the broad side by applying negative pressure to the first hole and creating a false air sac that was later sealed. A window (1 cm2) was made above the false air sac. MCF-7, siRNA (NT, SUZ12, EZH2, or ITGA11) transfected TAMR or ADR cells (1.5 × 106 cells/CAM) were inoculated on the CAM. Eggs were returned to the incubator after sealing the window. CAM tissue beneath the tumor region was resected from the embryo and then harvested. Blood vessels in the tumor region were viewed using an optical microscope (Olympus Corporation, Tokyo, JAPAN) and counted. Tumor tissues detached from the CAM were weighed.

For metastatic study, 1.5 × 106 cells/CAM labelled with cell-tracking red-florescent dye were mixed in 50% Matrigel and implanted on the exposed CAM. On the 5th day of implantation, the lower CAM and liver of developing chicken embryo were collected to evaluate metastatic cells using fluorescence-aided Leica L2 microscope (Leica, Tokyo, Japan). The lower CAM and liver tissues were further analyzed to detect human DNA hypoxanthine phosphoribosyl transferase (HPRT) using PCR.

P34 and P48 for MCF-7, P54 and P62 for ADR, and P58 and P64 for TAMR were used for inoculation onto the CAM.

Bioluminescence imaging and survival rate measurement in a mouse tumor model.

Seven weeks-old female BALB/c nude mice (OrientBio, Gyeonggi, South Korea) were injected subcutaneously with 1.5 × 106 TAMR-Luc cells (P5) in 200 µL of DMEM/Matrigel (1:1) into the right flank. Tumors were allowed to grow untreated until they reached approximately 200 mm3. A total of 20 tumor-bearing nude mice were randomly divided into 4 groups and were used to evaluate the efficacy of treatment. The mice were then administered twice a week for 2 weeks with siRNA complex targeting none (siNT), integrin α11 (siITGA11), or EZH2 (siEZH2) by intratumoral injection. Each siRNA complex contained 5 µg of siRNA and 3.5 µL Oligofectamine (Invitrogen). Tumor size was calculated using the equation (l × b2)/2, where l and b were the larger and smaller dimensions of each tumor, respectively. In addition, the body weights of the mice were also measured to evaluate tumor progression degree. For in vivo imaging of the tumor at the 14th day of the first drug administration, mice were administered with D-luciferin (150 mg/kg) (Promega, Madison, WI) by intraperitoneal injection and placed in a light-tight mouse imaging chamber (Xenogen IVIS Imaging System, Xenogen, US) under anesthesia of 2.5–3% isoflurane. At 10 min after D-luciferin injection, the converted D-luciferin was measured in the value of emitted photons using Living Image software 4.0 (Caliper, Alameda, CA.). Afterward, the mice were returned to their cages, and the survival rate was measured.

Statistical analysis

Data from more than three independent experiments were averaged. Data are expressed as mean ± SEM. Statistical significance was assessed by one-way analysis of variance (ANOVA) followed by Newman-Keuls comparison using Graph Pad Prism 8.0 (San Diego, CA, USA). p values lower than 0.05 were considered significant.