Patients’ cohort, clinical specimens, and data mining of online databases
A total of 110 HCC patients who underwent curative resection in the Department of Hepatobiliary Surgery, the First Affiliated Hospital of Xi’an Jiaotong University (Xi’an, China), were enrolled in this study. None of the patients received any chemotherapy or radiotherapy before the operations. The median follow-up of the patients was 34.3 months (1–86 months). The diagnosis of all HCC cases was validated by pathological examinations. Clinical specimens, including HCC tissues and adjacent non-tumor tissues, were collected. These samples were stored at −80 °C for RNA and protein extraction or formalin-fixed and embedded in paraffin. Written informed consent was obtained from each patient. The study protocol was approved by the research ethics committee of the First Affiliated Hospital of Xi’an Jiaotong University. The clinical characteristics of the HCC cohort are presented in Supplementary Table 1. The expression status of BRD9 in different types of human cancers and the corresponding normal tissues was obtained from the Gene Expression Profiling Interactive Analysis (GEPIA) website14 (http://gepia.cancer-pku.cn/index.html). The expression data of BRD9 in the TCGA database and GEO databases (GSE 14323, GSE 14520, GSE1898, GSE32649, and GSE 6764) were obtained from the Oncomine website15 (https://www.oncomine.org/). The statistical threshold used for determining the overexpression of BRD9 in the tumor samples vs the normal tissue in these public datasets was P < 0.05.
The antibodies used in this study included: anti-BRD9 (Active Motif, Carlsbad, CA, USA, #61537), anti-TUFT1 (Abcam, Cambridge, MA, USA, ab184949), anti-E-cadherin (Cell Signaling Technology, Beverly, MA, USA, #3195), anti-N-cadherin (Cell Signaling Technology, #14215), anti-vimentin (Cell Signaling Technology, #5741), anti-H3K27Ac (Abcam, ab4729), anti-H3K14Ac (Abcam, ab52946), anti-H3K9Ac (Abcam, ab4441), anti-P300 (Abcam, ab10485), anti-AKT (Cell Signaling Technology, #4691), anti-phospho-AKT (Ser-473, Cell Signaling Technology, #4060), anti-MYC (Cell Signaling Technology, #9402), and anti-GAPDH (Santa Cruz Biotechnology, Dallas, TX, USA, sc-47724).
The immortalized hepatocyte cell line LO2 and five HCC cells lines (HepG2, Hep3B, Huh7, MHCC97H, and HCCLM3) were obtained from the Cell Bank, Type Culture Collection, Chinese Academy of Sciences (Shanghai, China) and passed the test of DNA profiling (STR) in the year 2016. The cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM; Gibco, USA) supplemented with 10% fetal bovine serum (FBS; Gibco, USA), 100 units/mL penicillin and 100 μg/mL streptomycin (Sigma, St-Louis, MO, USA). All cell cultures were kept in a humidified cell incubator with 5% CO2 at 37 °C. To inhibit the activation of the AKT pathway, the cells were starved overnight and treated with MK2206 (10 µM)16. For BRD9 inhibition, the cells were starved overnight and treated with I-BRD9 (5 µM)13. All cells were routinely tested for mycoplasma contamination with a MycoAlert Detection Kit (Lonza).
Viral transduction and cell transfection
Brd9 shRNA (Addgene #75138) underwent enzymatic digestion with MluI and XhoI and was then inserted into the pGipZ lentiviral vector (NEB Quick Ligation Kit). Nontargeting hairpin shRNA in pGipZ was used as a control. The retroviral vector pMMP-BRD9 was generated by inserting the BRD9 cDNA into the pMMP vector. For retroviral packaging, the pMMP-BRD9 plasmid was transfected into 293T cells using Effectene (Qiagen, Valencia, CA, USA) along with the pMD.MLV plasmid and the pVSV.G plasmid. Viral supernatants were collected 72 h after transfection. Viral transduction was performed by incubating cells with the viral supernatant supplemented with polybrene (8 μg/mL) overnight at 37 °C. Further experiments were performed 48 h after viral transduction. The efficacy of viral transduction was confirmed by qRT-PCR and western blot. TUFT1 shRNA, P300 shRNA, TUFT1 vector, the corresponding NC shRNA, or the corresponding control vector was transfected into HCC cells as previously described16,17. HCC cells were subjected to western blotting and qRT-PCR to confirm the efficacy of TUFT1 or P300 knockdown.
Quantitative real-time PCR (qRT-PCR)
RNA extraction and qRT-PCR were performed as previously described18. 18S RNA was used as an internal control. Primer sequences for BRD9, TUFT1, E-cadherin, N-cadherin, vimentin, and 18S RNA are listed in Supplementary Table 2.
Immunohistochemistry (IHC) staining
Immunohistochemistry (IHC) staining was performed as described previously18. The following antibodies were used in the IHC staining: BRD9 antibody (1:50) and TUFT1 antibody (1:50). The intensity of IHC staining was graded as per the following criteria: 0, negative staining; 1, weak staining; 2, medium staining; and 3, strong staining. The percentage of positive staining was graded as per the following criteria: 0, <5%; 1, 5–25%; 2, 25–50%; 3, 50–75%; and 4, 75–100%. The IHC score of each slide was calculated by multiplying the scores of the staining intensity and the percentage of positive staining.
Protein extraction and western blotting were performed as described previously16. The following primary antibodies were used in this study: anti-BRD9 (1:1000), anti-TUFT1 (1:1000), anti-P300 (1:500), anti-E-cadherin (1:1000), anti-N-cadherin (1:500), anti-vimentin (1:1000), anti-p-AKT (1:1000), anti-AKT (1:1500), and anti-GAPDH (1:1500). A densitometric measurement was performed by using ImageJ software and normalized to GAPDH19.
The cell viability of HCC cells was evaluated by the MTT assay. The 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT, Roche, USA) was used for the MTT assay as described previously16. EdU and colony formation assays were performed for the evaluation of cell proliferation. For the EdU assay, a cell proliferation ELISA, the Cell-Light™ EdU Apollo®488 In Vitro Imaging Kit (#C10310-3, RiboBio Co., LTD, Guangzhou, China), was used as previously described20. For the colony formation assay, HCC cells (1 × 103) were seeded in 6-well plates and cultured for 2 weeks in DMEM medium with 10% FBS. The cell colonies containing more than 30 cells were counted as single colonies20.
The migration and invasion ability of the HCC cells was evaluated by a Transwell assay. In general, serum-starved HCC cells were resuspended in DMEM and seeded into the upper chamber. For the invasion assay, 150 µL Matrigel (1:8 dilution in serum-free DMEM) was added onto the membrane of the upper chamber before cell seeding. DMEM containing 20% FBS (600 µL) was added into the bottom chamber. Twenty-four hours after cell seeding, the HCC cells that migrated or invaded through the membrane of the upper chamber were stained with crystal violet. The cell number of the migrated or invaded cells was counted.
Chromatin immunoprecipitation (ChIP)
ChIP was performed to investigate whether BRD9 affected the histone acetylation status in the promoter of the TUFT1 gene and the engagement of P300 to the TUFT1 promoter. The HCC cells were fixed with 1% formaldehyde. ChIP assays were performed following the protocol of the EZ-Magna-ChIP HiSens kit (Millipore, 17-10461). Nuclear extracts were subjected to sonication on wet ice. The samples were sheared during three rounds of ten cycles of 30 s ON/30 s OFF with the Bioruptor® PLUS at the HIGH power setting (position H). Magnetic protein A/G beads were used for the immunoprecipitation (IP) of cross-linked protein/DNA. Antibodies against H3K9Ac (1:25), H3K14Ac (1:25), H3K27Ac (1:25), or P300 (1:50) or control IgG (1:25) were used to immunoprecipitate the sheared DNA. The precipitated DNAs containing fragments of the TUFT1 promoter were detected by qRT-PCR using seven pairs of walking primers along the TUFT1 promoter region16. The sequences of the seven pairs of walking primers are listed in Supplementary Table 2. The fold enrichment was relative to the input DNA.
In vivo growth and metastasis experiments
BALB/c nude mice (male, 4–6 weeks old) were randomly grouped for animal experiments (n = 6 mice per group). HCCLM3 cells with a BRD9 knockdown (HCCLM3-BRD9 shRNA) or negative control HCCLM3 cells (HCCLM3-NC shRNA) were employed for the tumor growth and metastasis experiments in nude mice. The subcutaneous injection experiment and tail-vein injection experiment were performed to evaluate the growth and metastasis of HCC cells as previously described16. Subcutaneous tumor tissues were dissected out for IHC staining of Ki67 (1:100), E-cadherin (1:50), and vimentin (1:100). The lungs of the nude mice were embedded in paraffin for H&E staining to identify the metastatic nodules. Animal protocols were approved by the Institutional Animal Care and Use Committee of Xi’an Jiaotong University.
The data in this study are shown as the mean ± S.E.M. At least three independent replicates were used to calculate S.E.M. Statistical analysis was performed by GraphPad Prism 5 software (San Diego, CA, USA) and the SPSS statistical package (Chicago, IL, USA). A two-tailed Student’s t-test, Pearson χ2 test, Kaplan–Meier plot, log-rank test, Spearman’s rank correlation coefficient, or ANOVA was used to determine the significant difference between groups. A P < 0.05 was statistically significant.