Bioinformatics data analysis

GRB7 expression in different cancers and BC subtypes was analyzed by UALCAN (https://ualcan.path.uab.edu/) and TIMER (https://cistrome.shinyapps.io/timer/). We downloaded GRB7 expression data for human HER2+ BC and non-HER2+ BC or normal tissue samples from the Gene Expression Omnibus database (GEO, https://www.ncbi.nlm.nih.gov/geo/) and The Cancer Genome Atlas (TCGA, https://cancergeno.me.nih.gov/). BC patient prognosis was analyzed by Kaplan‒Meier Plotter (https://kmplot.com/analysis/). CancerSEA was used to analyze the biological function of GRB7. The cistrome was used to predict potential transcription factors. TIMER was used to analyze the correlation between genes.

Cell lines and culture

All human BC cell lines including ER+ BC cell lines MCF7 and T47D, HER2+ BC cell lines SK-BR-3, MDA-MB-453, and TNBC cell lines MDA-MB-231 and MDA-MB-468 and the immortalized mammary epithelial cell line MCF10A were purchased from the American Type Culture Collection (ATCC) (Manassas, VA, USA). These cell lines were authenticated by short tandem repeat (STR) profiling and tested for mycoplasma contamination. Cells were cultured in DMEM, RPMI-1640, L15 or Ham’s F-12 medium (Gibco, USA) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin (Procell, Wuhan, China), and streptomycin at 37 °C in a 5% CO2 humidified atmosphere.

RNA and plasmid transfection

For transient transfection, the indicated small interfering RNAs (siRNAs) (100 pM) or plasmid DNA (5 μg) were introduced into cells in the presence of Lipo8000 (Beyotime, Shanghai, China) and Opti-MEM (Gibco, USA) for 48 h according to the manufacturer’s instructions. Plasmids for GRB7 and TCF12 were from Sino Biological Co., Ltd. (Beijing, China). The siRNAs for GRB7 and TCF12 were obtained from Tsingke Co., Ltd. (Beijing, China). The following GRB7 and TCF12 siRNAs were used: siGRB7#1, sense, 5′- CAGAUGUGAACGAGUCCAATT-3′, antisense, 5′- UUGGACUCGUUCACAUCUGTT-3′; siGRB7#2, sense, 5′- GGGUGCAGCUGUACAAGAATT-3′, antisense, 5′- UUCUUGUACAGCUGCACCCTT-3′; siGRB7#3, sense, 5′- GGUUCAGGACGGAAGCUUUTT-3′, antisense, 5′- AAAGCUUCCGUCCUGAACCTT-3′; siTCF12#1, sense, 5′- CAUUCAGUCCUGUCUAGUATT-3′, antisense, 5′- UACUAGGACUGAAUGTT-3′; siTCF12#2, sense, 5′- CCGUGAAUCUCCUAGUUAUTT-3′, antisense, 5′- AUAACUAGGAGAUUCACGGTT-3′. Unless otherwise stated, a cocktail of GRB7 siRNA and TCF12 siRNA was used separately in each of the following experiments.

RT‒qPCR

Total RNA was extracted from cells using TRIzol reagent (Invitrogen, Waltham, United States) and subjected to reverse transcription in the presence of PrimeScript RT Master Mix (Takara, Japan). Quantitative PCR was performed using SYBR Green reagent, and β-Actin was used as the internal control. The sequences of primers used are shown in Table 1.

Table 1 Sequences of the primers used in this studyWestern blot analysis and immunoprecipitation

The cells were harvested and lysed in RIPA lysis buffer containing protease inhibitor cocktail (Beyotime, Shanghai China). Protein samples (30 μg) were separated by 8%–12% SDS‒PAGE and transferred to PVDF membranes for immunoblotting overnight at 4 °C with primary anti-GRB7 (1:1000, Proteintech, 10045-1-Ig), N-Cadherin (1:3000, Proteintech, 22018-1-AP), E-Cadherin (1:10,000, Proteintech, 20874-1-AP), Vimentin (1:5000, Proteintech, 10366-1-AP), Slug (1:1000, Proteintech, 12129-1-AP), Snail (1:500, Proteintech, 13099-1-AP), β-Actin (1:5000, Santa, sc-56459), and anti-TCF12 (#78,), Notch1 (#3608), Jagged1 (#70109), β-catenin (#8480), MMP2 (#40994), p-AKT (#4060), AKT (#4691), ERK1/2 (#9102) antibodies (all from Cell Signaling Technology) with 1:1000 dilution, followed by incubation with secondary antibody (1:5000, ZSBio, ZB-2301 and ZB-5305) for 2 h at room temperature. Protein bands were visualized with a chemiluminescence imaging system (Tanon, Shanghai, China) using ECL substrate. For immunoprecipitation, cell lysates were incubated with S-protein agarose or primary antibodies plus protein A/G agarose (Beyotime, Shanghai, China) and were then analyzed by western blot.

MTT and EdU assays

3-(4,5)-Dimethylthiahiazo (-z-y1)-3,5-di-phenytetrazoliumromide (MTT) and 5‑ethynyl‑2′‑deoxyuridine (EdU) assays (Beyotime, Shanghai, China) were used to detect cell proliferation. For MTT assay, 2000–3000 cells were inoculated in 96-well plates, and 10 μL MTT solution was added at 24, 48, 72, 96, and 120 h after treatments as indicated. The absorbance was measured at 570 nm by a multiscan spectrophotometer. For EdU assay, 48 h after transfection, SK-BR-3 and MDA-MB-453 cells were incubated in the manufacturer's recommended final concentration of 10 μM EdU solution for 2 h. Next, the cells were reacted with a fluorescent dye mixture, and images were acquired by fluorescence microscopy. Cell colony formation ability was assessed as described previously [22].

Wound healing assay

For the wound healing assay, after cells were inoculated in 6-well plates and grown to 90% confluence, the cell layer was scraped with a 20 µL pipette tip, and a wound was formed. Images were obtained by electron microscopy at the indicated time intervals, and wound closure rates were assessed using ImageJ software.

Migration and invasion assays

Cell migration and invasive activity were assessed by transwell plates with or without extracellular interstitial gel coating, respectively. Briefly, the cells were designated treated and placed in FBS-free medium in the upper chamber, and fresh medium containing 10% FBS as a chemoattractant was placed in the lower chamber. The cells were incubated under the corresponding culture conditions for 48 h, fixed with 4% paraformaldehyde, and stained with 0.1% crystal violet. The membranes were further washed and scored for cell number after drying.

ChIP assay

ChIP analysis was performed using a chromatin immunoprecipitation kit (Cell Signaling Technology, USA) according to the manufacturer's instructions as previously reported [23]. Briefly, SK-BR-3 cells were soaked in 1% formaldehyde for 15 min and then incubated in glycine for 5 min; cells that had been washed in PBS and incubated with protease inhibitors were collected and sonicated to produce genomic DNA fragments of approximately 200 to 500 bp in length. Chromatin extracts were immunoprecipitated using an anti-TCF12 antibody (1:50, Cell Signaling Technology, #78245) at 4 °C. IgG was used as a negative control.

Luciferase reporter assay

Wild-type and mutant promoter binding region sequences were inserted into the luciferase system. In 96-well plates, the indicated cells were transfected using Lipo8000 with reporter vectors with knockdown or overexpression of GRB7 and TCF12 for 48 h. The luciferase reporter gene assay was performed using the Dual Luciferase Reporter Assay System (Promega Corporation, Madison, WI, USA).

Animal experiments and treatments

Female NOD/scid-IL-2Rgamma(c)(null) (NSG) mice aged 4–5 weeks were obtained from Shanghai Model Organisms Center, Lnc (Shanghai, China) and maintained in a well-ventilated animal transit room with a 12 h light/dark cycle, relative humidity of 60 ± 10%, and a controlled temperature of 22 °C. An orthotopic xenograft model was established in NSG mice by implanting MDA-MB-453 cells (5 × 107 cells/60 μL) mixed with 20 μL Matrigel into the mammary fat pad. Ten days later, the size of the xenograft tumor was measured every 2 days with a caliper. Tumor volume was calculated using the following formula: length × width2/2. After the xenograft tumors grew to an appropriate volume, the nude mice were randomly divided into groups. As previously described [24], in vivo processing with siRNA in a mouse model was used to explore effects of GRB7 knockdown in vivo. Specifically, cholesterol-modified siRNA targeting GRB7 or negative control (Tsingke, 5 nmol/kg) was dissolved in 25 μL sterile physiological saline, mixed with 5 μL in vivo transfection reagent (Engreen, Beijing, China) and then injected into the tumor every 2 days for 2–3 weeks. For in vivo rescue assays, a cholesterol-modified siRNA targeting TCF12 combined with lentivirus containing GRB7 or pCDH was injected into tumor-bearing NSG mice. The lentivirus containing GRB7 with a virus titer of 108/UI was injected into the tumor mass once every 5 days for 3 weeks. Animals were sacrificed with 3% sodium pentobarbital at an appropriate time, and then the orthotopic grafts were exfoliated for follow-up examination. Protocols for animal use were reviewed and approved by the Animal Care and Use Committee of Chongqing Medical University in accordance with Institutional Animal Care and Use Committee guidelines.

Immunofluorescence (IF) and immunohistochemical (IHC) staining

A tissue microarray containing 159 BC pathological samples and normal breast tissue samples was provided by Shanghai Outdo Biotech Co., LTD. Immunofluorescence analysis was performed as described previously [25]. For IHC staining, all xenograft tumor specimens were immersed in formalin. Before staining, tissues were cut to 5 μm thickness and placed on slides. Sections were deparaffinized in xylene and graded alcohols and boiled in sodium citrate buffer (pH 6.0) for antigen recovery, followed by blocking endogenous peroxidase activity with 3% hydrogen peroxide. Treated sections were incubated with anti-GRB7 (1:200, Proteintech, 10045-1-Ig), anti-N-cadherin (1:5000, Proteintech, 22018-1-AP), anti-E-cadherin (1:5000, Proteintech, 20874-1-AP), anti-vimentin (1:5000, Proteintech, 10366-1-AP), anti-slug (1:500, Proteintech, 12129-1-AP), and anti-Ki67 (1:5000, Proteintech, 27309-1-AP) antibodies at 4 °C overnight. Next, the sections were incubated with biotinylated goat anti-rabbit IgG secondary antibody for 30 min at room temperature and visualized with a 3,5-diaminobenzidine (DAB) substrate kit, and microscopy images were produced by light microscopy.

Statistical analysis

All experiments were conducted in triplicate. All of the data were analyzed by GraphPad Prism v.9.0 and are presented as the mean ± standard deviation (SD). Differences between the indicated groups were calculated using Student’s t test or ANOVA. p < 0.05 was considered statistically significant.