UNC0638 was purchased from MedchemExpress LLC (Princeton, NJ, USA) and dissolved in dimethyl sulfoxide (DMSO). Dulbecco's modified Eagle's medium (DMEM) and fetal bovine serum (FBS) were purchased from Thermo Fisher Scientific, Inc. (Waltham, MA, USA). 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) and low melt point soft agar were purchased from Sigma-Aldrich (Merck KGaA, Darmstadt, Germany). Invasion chambers and matrigel were purchased from BD Biosciences (Franklin Lakes, NJ, USA). Gentian violet was purchased from Beijing Solarbio Science & Technology Co., Ltd. (Beijing, China). The Annexin V-fluorescein isothiocyanate/propidium iodide apoptosis detection kits (cat. no. 556570) were purchased from BD Biosciences. Cell lysis buffer (cat. No. P0013), a BCA protein assay kit (cat. no. P0011) and Primary Antibody Dilution Buffer (cat. no. P0023A) were purchased from Beyotime Institute of Biotechnology (Shanghai, China). Polyvinylidene fluoride (PVDF) membrane and chemiluminescent reagents were obtained from EMD Millipore (Billerica, MA, USA). Antibodies to E-cadherin (20874–1-AP), vimentin (10366–1-AP), Slug (12129–1-AP), Snail (13099–1-AP) and β-actin (60008–1-Ig) were purchased from Proteintech Group, Inc. (Chicago, IL, USA) and diluted at 1:1,000 with Primary Antibody Dilution Buffer (Beyotime, Beijing, China).

Human breast cancer cell lines, MDA-MB-231 and BT-549 were obtained from the American Type Culture Collection (Manassas, VA, USA) and cultured in DMEM medium supplemented with 10% FBS and 1% penicillin/streptomycin in a 5% CO₂ atmosphere at 37°C. The expression of E-cadherin was knocked down by stable transfection with lentivirus and selected with puromycin. The lentiviral vector, pLKO.1 puro shRNA E-cadherin was provided by Bob Weinberg (Addgene plasmid no. 18801) (17).

Cell viability was measured via MTT assay (cat. no. M2128 Sigma-Aldrich; Merck KGaA). A total of 5×10³ cells per well were seeded into 96-well plates. After 24 h incubation at 37°C, cells were treated with various concentrations of UNC0638 for 24 h. At the end of drug treatment, 20 µl per well MTT reagent (5 mg/ml in phosphate-buffered saline; PBS) was added to each well and incubated at 37°C for 4 h. Then the MTT-containing medium was replaced with 150 µl DMSO per well and the plates were shaken for 10 min in the dark. Absorbance of each well was determined using a microplate reader at a wavelength of 490 nm. All experiments were repeated three times.

The cell migration and invasion assay were performed using the BD transwell system (cat. no. 353097; BD Biosciences) according to the manufacturer's protocols. Briefly, cells were treated with 0.5 and 1 µM UNC0638 or diluent for 24 h. Subsequently, 2×10⁴ treated cells in 200 µl serum-free medium were seeded in the upper chambers of the transwell inserts and culture medium with 10% FBS was added to the lower chambers. For the invasion assay, inserts were pre-coated with Matrigel. After 24 h incubation at 37°C, cells that had migrated to the reverse side of the inserts were fixed with methanol and stained with Gentian violet at room temperature for 30 min. Cells in randomly selected fields were counted under an inverted microscope (Olympus CKX31; Olympus Corporation, Tokyo, Japan).

Cells were seeded in 6-well plates and grown into a confluent monolayer in complete medium. An artificial wound across the cell culture monolayer was formed by scratching with a 10-µl pipette tip. The wounded monolayers were then washed twice with PBS to remove non-adherent cells, and serum-free medium with or without UNC0638 was added to the 6-well plate. Images of each wounded monolayer were obtained at 0 and 24 h using a phase-contrast microscope.

Following treatment with UNC0638 or diluent, cells were harvested and lysed with cell lysis buffer. The concentration of total protein was determined using the BCA protein assay kit. Equal quantities (30 µg per load) of protein were size-fractionated by 10% SDS-polyacrylamide gel electrophoresis and transferred onto PVDF membranes (90 V for 2 h). After blocking with 10% non-fat milk, the membrane was reacted with primary antibodies at 4°C overnight, followed by incubation with horseradish peroxidase-conjugated secondary antibodies at room temperature for 2 h. The protein bands were detected using an enhanced chemiluminescence detection system (cat. no. WBKLS0500; Merck KGaA) according to the manufacturer's protocols.

Cells (10⁵ per well) were plated in ultra-low attachment 6-well plate (Corning Incorporated, Corning, NY, USA) in serum-free DMEM-F12 medium supplemented with 20 ng/ml basic fibroblast growth factor, 20 ng/ml epidermal growth factor, 5 µg/ml insulin and 0.4% bovine serum albumin. Following incubation at 37°C for 48 h, UNC0638 or diluent was added to the medium and the suspension cultures continued for 1–2 weeks until tumorspheres formed. Tumorspheres with diameter >50 µm were counted. The colony numbers were obtained from five random views using a microscope. Experiments were repeated three times with duplication in each experiment.

The 6-well plate was pre-coated with 1 ml per well 2% low-melting agarose. Following treatment with 0.5 and 1 µM UNC0638 or diluent for 24 h, cells were detached, counted and resuspended in 0.5 ml 1% low-melting-point agarose mixed with 0.5 ml 2X complete culture medium. The cell suspensions were then layered on top of 2% low-melting agarose and incubated at 37°C in a humidified atmosphere of 5% CO₂ for 2 weeks. Colonies containing at least 50 cells were counted. All experiments were repeated three times.

The E-cadherin promoter sequences (from-108 to 125 of the human E-cadherin gene) were synthesized by Genewiz, Inc. (Suzhou, China) and cloned into pGL3-basic vector (18). For the luciferase assay, 2×10⁵ cells per well were seeded in a 6-well plate. Following incubation at 37°C for 24 h, cells were co-transfected with E-cadherin promoter-luciferase plasmid together with pRL-CMV (Renilla luciferase) using Lipofectamine^® 2000 transfection agents (cat. no. 11668-019; Invitrogen; Thermo Fisher Scientific, Inc.). After 6 h, the medium was replaced with fresh medium containing UNC0638 or diluent and cultured for another 24 h. luciferase activity was measured using the Dual-Luciferase Assay kit (cat. no. E1910; Promega Corporation, Madison, WI, USA) according to the manufacturer's protocol. The firefly luciferase activity value was normalized to the Renilla activity value. All procedures were performed in triplicate.

All experiments were repeated three times and statistical analysis was performed using SPSS software (version 19; IBM Corp., Armonk, NY, USA). Data were analyzed using Student's t-tests and P<0.05 was considered to indicate a statistically significant difference.

UNC0638 was identified as a synthetic small molecular compound with potent inhibitory activities against G9a (Fig. 1A). In the present study, the cytotoxicity of UNC0638 was initially determined in two TNBC cell lines, MDA-MB-231 and BT-549. The two cell lines were treated with UNC0638 at various concentrations (ranging from 0.5–8 µM) for 24 h followed by MTT assay. In comparison with that of the solvent control (DMSO), the cellular viabilities of the two TNBC cell lines were not significantly affected by treating with UNC0638 at a concentration between 0.5 and 4 µM (Fig. 1B and C). Therefore, non-cytotoxic concentrations of UNC0638 (<4 µM) were applied in all subsequent experiments.

Metastasis, which is the critical reason for treatment failure and poor prognosis in patients with TNBC, involves migrations and invasive abilities of cancer cells to pass through the basement membrane and to invade surrounding tissues (19,20). Therefore, a transwell assay was performed to examine the effect of UNC0638 on the migratory and invasive ability of TNBC cells. For detection of cell invasive ability, a chamber membrane was pre-coated with Matrigel to stimulate the basement membrane. As demonstrated in Fig. 2A-D, the number of cells migrating to the lower chamber and invading the Matrigel membrane were significantly reduced by UNC0638 in a dose-dependent manner.

The inhibitory effect of UNC0638 on migration of MDA-MB-231 and BT-549 cells was evaluated by wound-healing assay. After a single layer of cells was scratched to generate the wounds, the two TNBC cell lines were treated with 0.5 and 1 µM UNC0638 for 24 h. As demonstrated in Fig. 2E and F, UNC0638 treatment significantly decreased the ability of TNBC cells to migrate and recover the wounded area. The quantitative data presented in Fig. 2H indicated that TNBC inhibited the migration of TNBC cells in a dose-dependent manner. These results indicate that UNC0638 is effective in preventing TNBC cell migration and invasion.

G9a promotes EMT activation of TNBC cells by directly repressing the expression of E-cadherin (14,21,22). Therefore, in the following experiment, the effect of UNC0638 on the expression of various EMT markers was determined in the two TNBC cell lines. As shown in Fig. 2I, the expression levels of E-cadherin were dose-dependently upregulated by UNC0638 treatment in the two cell lines, whereas the expression levels of vimentin were downregulated. In addition, UNC0638 treatment decreased the expression level of Slug and Snail1 in the two cell lines, which are key transcription factors that strongly promote the EMT process.

The invasive and metastatic features of TNBC cells are closely linked to the EMT activation and CSC properties. Acquisition of CSC activities block cellular differentiation and promote cancer cells to convert from their differentiated epithelial state to a mesenchymal phenotype, which provides a distinct advantage in tumor progression and metastasis (23,24). As undifferentiated CSCs were hypothesized to generate tumorspheres in non-adhesive and serum-free culture conditions, tumorsphere assays are useful for enriched CSCs and investigating their behavior (25,26). In the present study, MDA-MB-231 and BT-549 cells were cultured in serum-free suspension conditions in the presence or absence of UNC0638 for a week, and the number and size of tumorspheres were monitored under phase contrast microscopy. As presented in Fig. 3A and B, the size and numbers of tumorspheres were significantly decreased by UNC0638 indicating its inhibitory effect on tumorsphere generation of TNBC cells.

A critical feature of CSCs is the self-renewal capacity, which enables a single cell to form a colony under anchorage independent conditions (27, 28). In the current study, a soft agar colony assay was used to determine the effect of UNC0638 on the CSC self-renewal capacity. As summarized in Fig. 3C, the numbers of colonies formed by UNC0638-treated cells are dose-dependently decreased, when compared with the cells treated with diluent (DMSO). In addition, the size of the colonies formed in UNC0638-treated groups were markedly smaller than that in the untreated groups (data not shown). Together, these results demonstrate that UNC0638 exhibited potent inhibitory activities on tumorigenicity and self-renewal of breast CSCs.

In addition, E-cadherin was silenced in the two TNBC cell lines to determine whether restoring E-cadherin is essential to UNC0638 inhibitory activities against EMT-mediated invasion and migration of TNBC cells. As presented in Fig. 4A, shRNA stable transfection efficiently attenuates the UNC0638 induced upregulation of E-cadherin. Furthermore, E-cadherin knockdown restores the repressive effect of UNC0638 on cellular migration and invasion (Fig. 4B and C). In addition, as studies have revealed that G9a, combined with other transcription factors, directly binds to the promoter of E-cadherin and represses its mRNA transcription, the effect of the G9a inhibitor on mRNA expression levels and promoter activity of E-cadherin was examined in the MDA-MB-231 cell line. As presented in Fig. 4D and E, UNC0638 dose-dependently increased the mRNA expression levels and promoter activity of E-cadherin. Thus, these results indicate that UNC0638 exhibited its inhibitory activities by restoring E-cadherin transcription.