The cell lines MCF7, T47D, BT549 and MDA-MB-231 breast cancer were obtained from the Type Culture Collection of Chinese Academy of Sciences (Shanghai, China). MCF7 and M231 cells were maintained in Dulbecco's modified Eagle's medium (Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA). T47D and BT549 cells were maintained in RPMI-1640 medium (Gibco; Thermo Fisher Scientific, Inc.). All media were supplemented with 10% fetal bovine serum (FBS; Biowest, SAS, Nuaille, France) and cell lines were cultured at 37°C with 5% CO₂.

Total protein was extracted from cells using a radioimmunoprecipitation lysis buffer (cat. no. sc-24948; Santa Cruz Biotechnology, Inc., Dallas, TX, USA). A bicinchoninic acid protein Assay (Thermo Fisher Scientific, Inc.) was used to quantify total protein concentration. Equal amounts of proteins (50 µg/well) were separated by SDS-PAGE on 10% gels and electrophoretically transferred onto polyvinylidene fluoride membranes (cat. no. IPFL00010; Merck KGaA, Darmstadt, Germany). The membranes were blocked with 5% milk in Tris-HCl buffered solution for 2 h at room temperature, and then incubated overnight at 4°C with primary antibodies and secondary antibody (1:2,000, cat. no. 7074; Cell Signaling Technology, Inc., Danvers, MA, USA) at 37°C for 1 h. The signals were visualized using an enhanced chemiluminescent substrate (Thermo Fisher Scientific, Inc.) and detected by a FluorChem Q imaging system (ProteinSimple, San Jose, CA, USA). Images from western blot analysis were quantified using Quantity One^® software 4.6.6 (Bio-Rad Laboratories, Inc., Hercules, CA, USA). The expression level was normalized with respect to an internal control. Primary antibody against SETDB1 was obtained from ProteinTech Group, Inc. (Chicago, IL, USA; cat. no. 11231-1-AP), and primary antibodies against E-cadherin (cat. no. 3195), β-catenin (cat. no.8480), vimentin (cat. no. 5741), Snail (cat. no. 3879) and Slug (cat. no. 9585) were obtained from Cell Signaling Technology, Inc. (EMT Antibody Sampler Kit; cat. no. 9782). β-actin rabbit monoclonal antibody (mAb) internal control was obtained from Cell Signaling Technology, Inc. (cat. no. 8457).

Total RNA was obtained using RNAiso plus (cat. no. 9108; Takara Biotechnology Co., Ltd., Dalian, China), and 2 µg total RNA was reversed transcribed to cDNA using PrimeScript RT Master Mix (cat. no. DRRO36A; Takara Biotechnology Co., Ltd.) according to the manufacturer's instructions. Primers used to detect SETDB1 (forward, 5′-GATGAGGAACTGGAGAAGATG-3′ and reverse, 5′-ATTAGTCACTGCCCTGGATG-3′); transforming growth factor 1 (TGF1 forward, 5′-ACCTGAACCCGTGTTGCTCT-3′ and reverse, 5′-GAACCCGTTGATGTCCACTT-3′); zinc finger E-box binding homeobox 1 (ZEB1 forward, 5′-GAAAATCCAGTCGCTACAA-3′ and reverse, 5′-CACACAGAAGACAAGTGCTA-3′); Snail (forward, 5′-CCCAATCGGAAGCCTAACT-3′ and reverse, 5′-GCTGCTGGAAGGTAAACTCT-3′); Slug (forward, 5′-CTCAGAAAGCCCCATTAGT-3′; reverse, 5′-GCCCAGAAAAAGTTGAATAG-3′); tight junction protein 1 (ZO-1 forward, 5′-GCAGCAAGAGATGGCAATA-3′ and reverse, 5′-CAGGGACATTCAATAGCGTA); and GAPDH (internal control forward, 5′-CTGACTTCAACAGCGACACC-3′ and reverse, 5′-TGCTGTAGCCAAATTCGTTGT-3′); transcripts were obtained from BioTNT (Shanghai, China). qPCR was performed using SYBR Premix Ex Taq (cat. no. RR420A; Takara Biotechnology Co., Ltd.) on a StepOnePlus Real-Time System (Applied Biosystems; Thermo Fisher Scientific, Inc.). The thermal cycle conditions consisted of one cycle at 95°C for 10 min, followed by 40 cycles of amplification at 95°C for 15 sec, and then 60°C for 1 min. The expression level of mRNA was obtained by calculating the 2^−∆∆Cq values (25).

pLVX-SETDB1-wt-IRES-Puro and pLKD-EF1a-EGFP-P2A-LUC-F2A-Puro-U6 shSETDB1 vectors were generated by Obio Technology (Shanghai) Co., Ltd. (Shanghai, China). The empty vector acts as control group. The three shRNA sequences were as follows: S1, 5′-GGGCAGTGACTAATTGTGA-3′; S2, 5′-GCATGCGAATTCTGGGCAAGA-3′; S3, 5′-GGGAGGACATAGAAGACATCT-3′. The negative control shRNA (shNC) sequences were as follows: Y009 forward, 5′-CCGGTTCTCCGAACGTGTCACGTTTCAAGAGAACGTGACACGTTCGGAGAATTTTTTG-3′; Y009 reverse, 5′-AATTCAAAAAATTCTCCGAACGTGTCACGTTCTCTTGAAACGTGACACGTTCGGAGAA-3′. Cells (5×10⁴) were seeded onto 24-well plates. Lentiviral vectors were introduced into breast cancer cells at different multiplicity of infection (MOI) values (MDA-MB-231 and BT549, MOI 10; MCF7 and T47D, MOI 40). Different final concentrations (2, 4, 6, 8 and 10 µg/ml) of puromycin were applied to the breast cancer cells for 24 h, and the concentration of 6 µg/ml was selected to kill non-transfected cells for 2 weeks to obtain stably expressed cells. Transfection efficiency was validated with RT-PCR and western blot analysis.

A total of 2×10³ cells/well were cultured in 96-multiwell plates for 96 h or 7 days. Cells were incubated with CCK-8 reagent (10 µl/well; cat. no. CK04-11; Dojindo Molecular Technologies, Inc., Kumamoto, Japan) for 1 h at 37°C. Absorbance at 450 nm was read using a Multiscan Go-1510 microplate reader (Thermo Fisher Scientific, Inc.).

Cells (~5×10⁴) were resuspended in 100 µl serum-free medium and plated in the upper chamber of a Transwell assay system (cat. no. 3422; Corning Incorporated, Corning, NY, USA). Then, 10% FBS-containing medium was added to the lower chamber. For the invasion assay, the Transwell membrane was precoated with Matrigel (cat. no. 354234; BD Biosciences, Franklin Lakes, NJ, USA) and incubated at 37°C for 4 h. After cells were cultured for 24 h, the upper chambers were fixed with 100% methanol for 10 min at room temperature, and stained with 0.1% crystal violet for 20 min at room temperature. Cells were counted under a light microscope in five predetermined fields and averaged.

Cells (~5×10⁵) were seeded onto 6-well plates and incubated overnight. A 1,000 µl sterile pipette tip was used to scratch the cell monolayer (point of zero migration). After cell incubation for 24 and 48 h, phase-contrast images of the scratched area were captured using computer-assisted light microscopy to assess gap closure.

Female nude mice aged 4–6 weeks (n=12) were obtained from Shanghai SLAC Laboratory Animal Co. Ltd. (Shanghai, China). Cells (~1×10⁵ cells in 100 µl) were resuspended in serum-medium and injected into the tail vein of mice. The animals were monitored twice weekly. Body condition score was used to assessment of overall health of the animal from 1 (emaciated/wasted) to 5 (obese) (25). After 2 months of observation, the maximum body weight loss was 5% of original weight. When the body condition score was 1/5, all mice were euthanized and organs were dissected. Lungs were fixed with 10% buffered formalin for 24 h at room temperature, and embedded in paraffin. Sections (4 µm-thick) were dewaxed in xylene, rehydrated through decreasing concentrations of ethanol (100, 95 and 80%) and washed in PBS. Then stained with hematoxylin for 10 min and eosin for 30 sec at room temperature. Following staining, sections were dehydrated through increasing concentrations of ethanol (80, 95 and 100%) and xylene. Animal experiments were approved by the Ethics Committee of Fudan University (Shanghai, China).

CHIP assay was performed with a SimpleCHIP Enzymatic Chromatin IP kit (cat. no. 9003; Cell Signaling Technology, Inc.) according the manufacturer's instructions. Briefly, ~4×10⁶ MCF7 cells overexpressing SETDB1 for each immunoprecipitation (IP) were crosslinked with 1% formaldehyde at room temperature for 10 min. Then, 0.5 µl micrococcal nuclease was added/IP preparation to digest DNA to 150–900 bp. IP antibodies histone H3 (D2B12) XP Rabbit mAb (10 µl; positive control; cat. no. 4620), normal rabbit IgG (2 µl; negative control; cat. no. 2729) and SETDB1 rabbit polyclonal antibody (2 µg; cat. no. 11231–1-AP; Proteintech Group, Inc.) were added to 500 µl 1X CHIP buffer containing 10 µg fragmented chromatin DNA at 4°C with rotation overnight. Then, 30 µl CHIP-Grade Protein G Magnetic Beads (cat. no. 9006; Cell Signaling Technology, Inc.) were used to pull-down the antibody/chromatin complex at 4°C with rotation for 2 h. Chromatin was eluted from the antibody/protein G magnetic beads at 65°C for 30 min and reversed cross-linked with 6 µl 5 M NaCl and 2 µl proteinase K (cat. no. 10012) at 65°C for 2 h. DNA was purified using spin columns (centrifuged at 18,500 × g in a microcentrifuge for 30 sec.) and quantified by PCR. A master reaction mix was prepared including 12.5 µl nuclease-free H₂O, 2.0 µl 10X PCR buffer, 1.0 µl 4 mM dNTP Mix, 2.0 µl 5 µM primers, 0.5 µl Taq DNA Polymerase (cat. no. EP0404; Thermo Fisher Scientific, Inc.). The following PCR reaction program was used: Initial denaturation, 95°C 5 min; denaturation, 95°C 30 sec; annealing, 62°C 30 sec; extension, 72°C 30 sec; denaturation, annealing and extension were repeated for a total of 34 cycles; final extension of 72°C 5 min. PCR products (10 µl) were resolved on 2% agarose gels and ethidium bromide was used to visualize DNA. Six pairs of Snail primers were used (P2 forward, 5′-GGAAGCCAGCGTGAAAGATC-3′ and reverse, 5′-TCCTCTCCTCAGCCAACTCG-3′; P3 forward, 5′-GAGAGGACTTTGGCTTTTAC-3′ and reverse, 5′-TCATTAAGCGGAATACTCCC-3′; P4 forward, 5′-GTATTCCGCTTAATGACTGC-3′ and reverse, 5′-AAAACCTATAAGCACCCCAC-3′; P5 forward, 5′-GTGGTGTGGGGTGCTTATAG-3′ and reverse, 5′-CTGTAACACGGCTCCATAGG-3′; P6 forward, 5′-AGCCGTGTTACAGCCTTTAG-3′ and reverse, 5′-CGTAGGAGTTTGGACTTTGC-3′; P7 forward, 5′-AAAGTCCAAACTCCTACGAG-3′ and reverse, 5′-ATTATCAAGGGAAAAGGCCC-3′).

Single guide RNA (sgRNA) was designed using an online Optimized Crispr Design Tool (genome-engineering.org). The Snail gene sequence was downloaded from Ensemble Genome Browser (ensembl.org). The 25 nucleotide sequences followed by protospacer adjacent motif sequence were designed as sgRNA (sense, 5′-CACCGTGTAGTTAGGCTTCCGATTG-3′ and antisense, 5′-AAACCAATCGGAAGCCTAACTACAC-3′). The sequence of NC sgRNA was as follows: 5′-GTATTACTGATATTGGTGGG-3′. Following phosphorylation and annealing (37°C for 30 min, 95°C for 5 min, and then ramping down to 25°C at 5°C/min), the sgRNA oligos were cloned into a BbsI-digested pX330 vector in a thermocycler (37°C for 5 min, 23°C for 5 min for a total of six cycles and then held at 4°C hold until analysis). Then, 1–2 µl of the final product was transformed into DH5α competent cells. Colonies were selected and verified by sequencing performed by Sangon Biotech Co., Ltd. (Shanghai, China). Cells (5×10⁴) were seeded into 24 well plates. DNA (1 µg) and 2 µl Lipofectamine^® 3000 (cat. no. L3000001; Thermo Fisher Scientific, Inc.) were added to cells for 24 h.

Statistical analyses were performed using GraphPad Prism 6 software (GraphPad Software, Inc., La Jolla, CA, USA). Three independent experiments were performed. The data of western blot analyses were quantified using Quantity One^® software 4.6.6 (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Student's t-test was used to compare data the difference between two groups. One-way analysis of variance multiple comparison with Tukey's post hoc test was used for multiple groups. P<0.05 was considered to indicate a statistically significant difference.

SETDB1 was overexpressed in the MCF7 and T47D cells, and knocked down in BT549 and MDA-MB-231 cells via lentiviral-vector infections. Transfection efficiency was validated by RT-qPCR and western blot analysis (Fig. 1A). Proliferation assays revealed that knocking down SETDB1 expression resulted in a decrease in the proliferation of BT549 cells at 84 h and at 24 h in MDA-MB-231 cells (Fig. 1B). By contrast, overexpression of SETDB1 increased proliferation of T47D cells after 84 h and MCF7 cells after 2 days (Fig. 1C).

Transwell assay results revealed that downregulation of SETDB1 in BT549 (Fig. 2A) and MDA-MB-231 (Fig. 2B) cells significantly decreased migration and invasiveness, particularly in breast cancer cells transfected with lentiviral containing the sh2 sequence compared with the negative control. Ectopic expression of SETDB1 facilitated migration and invasiveness in MCF7 cells (Fig. 2C). A wound-healing assay revealed that SETDB1 overexpression in MCF7 cells significantly increased cell migration compared with controls (Fig. 2D). In MDA-MB-231-treated immunodeficient mice, H&E staining revealed that multiple tumor nodules were observed in lungs of MDA-MB-231/NC animals (4/4 in M231 shNC group), and few or no tumors were found in mice injected with SETDB1-deficient cells (0/4 in M231 sh2 group; 1/4 in M231 sh3 group) (Fig. 2E). No tumor nodules were observed in other tissues in shNC and shSETDB1 groups.

MDA-MB-231 cell morphology changed from long spindle shapes to round shapes following knockdown of SETDB1 expression (Fig. 3A). Thus, SETDB1 may be involved EMT in breast cancer. In SETDB1-overexpressing MCF7 cells, there was no difference between MCF7/SETDB1 and MCF7/NC cells after puromycin treatment for 2 weeks. However, most MCF7/SETDB1 cells changed from epithelial to mesenchymal phenotypes on the 26th day (Fig. 3A). Immunoblotting data revealed that E-cadherin and β-catenin protein was significantly decreased in MCF7/SEDB1 cells on the 26th day after infection and selection, with no difference was observed on the 14th day. By contrast, expression of vimentin was significantly increased in MCF7/SETDB1 cells on the 26th and 14th days (Fig. 3B). Thus, SETDB1 overexpression induced EMT in breast cancer cells. SETDB1 RNA and protein levels decreased after EMT formation on the 26th day compared with the 14th day in the MCF/SETDB1 cells (Fig. 3C). To further address the mechanism of how SETDB1 induces EMT in breast cancer, RT-qPCR to detect the RNA level of EMT markers, including TGF1, ZEB1, Snail, Slug and ZO-1. Slug expression was significantly increased following SETDB1 overexpression in MCF7 cells (Fig. 3D). Western blot analysis revealed no significant difference in Slug protein expression on the 26 and 14th days, whereas Snail protein was significantly increased in MCF7/SETDB1 cells on the 26th day (Fig. 3E).

Subsequently, CHIP was performed to investigate whether SETDB1 regulated the Snail gene by directly binding to its promoter region. The nucleotide segments from 0–2,000 bp downstream from the initiation start site (ATG) were divided into 13 regions. An enrichment of fragments that hybridized with the CHIP-grade SETDB1 antibody was observed in the promoter regions of P2, P3, P4, P5, P6 and P7 (Fig. 3F). Thus, SETDB1 may be involved in EMT by upregulating the expression of Snail transcription factor.

Following Crispr-Cas9 knockout of Snail expression in SETDB1-overexpressing MCF7 cells, MCF7/SETDB1 cell morphology changed to an epithelial phenotype following Snail knockout, whereas this did not occur in the Crispr-NC group (Fig. 4A). Thus, depletion of Snail may reverse EMT in SETDB1-overexpressing MCF7 cells. Cell migration and invasiveness was significantly decreased in MCF7/SETDB1 + Crispr-Snail cells compared with MCF7/SETDB1 and MCF7/SETDB1 + Crispr-NC cells (Fig. 4B). Immunoblotting results revealed that E-cadherin and β-catenin protein were significantly increased following knockout of Snail expression, and vimentin was decreased in MCF7/SETDB1 + Crispr-Snail cells (Fig. 4C). These results highlight a critical role of Snail in SETDB1-induced EMT of breast cancer cells. However, compared to MCF7/NC group cells, the expression of E-cadherin and β-catenin was still significantly decreased and vimentin was increased in MCF7/SETDB1 + Crispr-Snail cells (Fig. 4C).