Tissue microarray, tissue sample and immunohistochemistry
The tissue microarray enrolled 299 ESCC patients who underwent radical esophagectomy in the Department of Thoracic Surgery, Zhongshan Hospital, Fudan University, China, in 2007. Each patients’ information of follow-up was recorded in details. The follow-up was conducted as previous description [19]. Patients’ clinicopathological data were collected in details, including sex, age, tumor pathology, tumor size, tumor depth, lymph node metastasis, tumor-node-metastasis (TNM) stage (according to American Joint Committee on Cancer, eighth edition) [20], histological grade, and overall survival. The overall survival was defined as the interval from surgery to death. In addition, 10 pairs of fresh ESCC tissues and para-cancerous tissues were obtained from 10 ESCC patients at Zhongshan Hospital in 2017. The written informed consent was obtained from each patient. This study was approved by the ethics committee on human research of Zhongshan Hospital, Fudan University, and conducted according to the principles of the Declaration of Helsinki.
Immunohistochemistry (IHC) staining and evaluation were consistent to the previous research [21]. The antibody was listed in Supplementary Table S1. For the statistical purposes, the total immunoreaction scores of moderate and strong groups were classified as TET3high group, while the negative and weak groups were classified as TET3low group.
Cell lines culture, transfection and Western blot assay
The human ESCC cell line (ECa109, KYSE510, KYSE150, TE-1) and human normal esophageal mucosa epithelium cell line (HEEC) were purchased from the Chinese Academy of Sciences. All the cell lines were cultured in DMEM, supplemented with 10% fetal bovine serum and 100 IU/mL penicillin/streptomycin in a humidified incubator, under 95% air and 5% CO2 at 37℃.
The lentiviral vectors with TET3 was constructed by Hanyin Biotechnology Limited Company (Shanghai, China). The ov-TET3 lentiviral vectors was constructed according to NC_000002.12. One gRNA (AGTCTGATCGGACTCCGTAG) was designed for human TET3 against 0 to 1000bp upstream of the first exon of TET3. We introduced dCas9-VP64 and pMS2-p65-HSF1 into cells by infection and established stable transductants. The sequence of normal control group lentiviral vectors was a non-targeted scrambled sequence. The constructed virus and Polybrene were added into the culture medium according to the instruction and co-cultured for 24 hours. 48 hours later, Puromycin was added to the culture medium to exclude the cells without successful transfection. Lipofectamine 2000 (Invitrogen, Carlsbad, USA) was applied for transient transfection. Predesigned siRNA duplexes were purchased from Genomeditech Company. The sequences of siRNA were listed in Supplementary Table S2. The normal control group was cells transfected with a non-targeted scrambled sequence. The transient transfection procedure was as follows: 4×104 cells were seeded into a 24-well Corning culture plate at the previous day so that the adherent cell intensity could reach 60-80% at transfection. For each well, 2.5μl siRNA + 50μl Opti-mem and 1μl lipo2000 + 50μl Opti-mem were pre-mixed for 5 minutes at regular temperature. Next, siRNA and Lipo2000 were gently mixed, at a ratio of 1: 1, for 20 minutes at regular temperature. Subsequently, Pipetting 100μl of the pre-mixed siRNA + lipo2000 into a well, and adding 400μl culture medium free of penicillin/streptomycin to each well. After 6 hours of transfection, culture medium was replaced with regular culture medium containing penicillin/streptomycin. Total RNA and total protein were extracted at 24 and 48 hours after transfection, respectively.
The total protein was extracted with RIPA lysis buffer and protease inhibitor from the cell or tissues. For LPS or PBS pretreated group, cells were treated with LPS with corresponding concentration (1 or 8 μg/mL) or PBS for 48 hours before total protein extraction. The Western blot assay was performed in the same way as previously described [21]. The related antibodies which were used to detect the expression of the related protein were listed in the Supplementary Table S1.
RNA extraction and RT-qPCR analysis
The methods of total RNA extraction and reverse transcribe were performed in the same way with the previous work [21]. For LPS or PBS pretreated group, cells were treated with LPS with corresponding concentration (1 or 8 μg/mL) or PBS for 48 hours before total RNA extraction. RT-qPCR was performed on an ABI 7500 thermocycler (Applied Biosystems, Foster City, CA). The fold change for each target gene relative to the control group was calculated using the ΔΔCt method. The primers for the genes detected with RT-qPCR were listed in the Supplementary Table S3.
Fluorescence-Activated Cell Sorter (FACS) Analysis
FACS was performed on a CyAn ADP Analyzer (Beckman Coulter), and the data were analyzed with FlowJo software (Tree Star, Ashland, OR). The staining was performed following the manufacturer’s instructions. The related antibodies were listed in Supplementary Table S1 and applied according to the manufacturer’s instructions.
ELISA assay
The ELISA assay was performed with the LPS-ELISA Kit (LanpaiBIO). The fresh ESCC tissues and para-cancerous tissues were treated, according to the protocol, to obtain the corresponding supernatants. The supernatants were subjected into the wells provided with the LPS-ELISA Kit, following the standard protocol as described in the Kit instruction.
CCK-8, chemoresistance, colony-formation, transwell and sphere assay
For CCK-8 assay, cells proliferation and chemoresistance were detected according to the manufacturer’s instructions (CK04, Donjindo, Japan). The OD values were detected at 24 hours, 48 hours and 72 hours after transfection, respectively. For the analysis of PTX function, at 24 hours after transfection, the adherent cells were treated with PTX with 10nM, 20nM, 40nM or 80nM for another 24 hours. And then the OD values were detected.
For colony-forming assay, 1.5×102 cells were placed into a 60 mm cell culture dish. Cells of each group were cultured for 2 weeks. Then, cells were fixed with paraformaldehyde and stained with crystal violet. Colonies with more than 10 cells were counted from 5 random visual fields under microscope.
Transwell filters (Costar, Coring, NY) (6.5-mm insert, 8-mm polycar-bonate membrane) were chosen for Transwell assay. 5×103 cells were harvested in 200 mL of serum-free medium and pipetted in the upper chambers of the plates. Then, 600 mL medium supplemented with 20 % fetal bovine serum was added to the lower chambers of the plates. After incubation for 36 hours, cells that had migrated from the upper chamber into the pores of the membrane insert were fixed with paraformaldehyde and stained with crystal violet. The cells were counted from 5 random visual fields under microscope.
For sphere assay, 2×103 cells per well were cultured in ultralow attachment 24-well plates (Corning, USA) with the specific medium. The medium was DMEM/F12 supplemented with 10ng/mL bFGF, 20 ng/mL EGF, B27 supplement (1×), 100U/mL penicillin, and 100lg/mL streptomycin. The cells were cultured for 1 week to form primary spheres. Then, primary spheres were harvested and digested into single-cell suspensions, which were re-cultured in the specific medium for stem cells. After one-week culture, secondary spheres were recorded with microscopy and analyzed statistically. For LPS pretreated group, cells were treated with LPS for 1 week before the assay and kept on LPS stimulation till the end of assay.
Tumor xenograft assay
4-week-old male nude mice were obtained from Slaccas Company. 2×106 cells of each group were injected subcutaneously into either side of mice posterior flank. The normal Control group was implanted into the left posterior flank and the treatment group was implanted into the right of the same mouse. 3 weeks later, the mice were executed with cervical dislocation and the size of tumors were measured by caliper for statistical analysis. For LPS pretreated group, cells were treated with LPS for 1 week before the assay and the resuspension was also supplementary with LPS of corresponding concentration. The experiments complied with the ARRIVE guidelines and the care for the animals was in accordance with National Institutes of Health guide for the care and use of Laboratory animals.
Genomic DNA isolation and Dot blot assay
Cells genomic DNA was extracted with TIANamp Genomic DNA Kit (TIANGEN) and purified with Universal DNA purification Kit (TIANGEN). Then, the DNA was sonicated into fragments between 200-500 bp, and DNA concentration was measured with NanoDrop (Thermo Scientific). For Dot blot, nylon membrane was pre-wetted with methanol and following with TBST. Membrane was dried in regular temperature. DNA was diluted into specific concentration and spotted on membrane and placed under an ultraviolet lamp for 20 minutes to cross-link the DNA. Subsequently, the membrane was blocked with 5% milk in TBS-Tween for 1 hour and then incubated with the specific antibody at 4℃ overnight. After incubation with the species appropriate HRP-conjugated secondary antibody for 1 hour at regular temperature, the membrane was scanned by a Tanon 6100 scanner (Tanon, China). The 5hmC intensity was quantified by Image-J software (USA). For LPS or PBS pretreated group, cells were treated with LPS or PBS for 48 hours before the assay.
ChIP-qPCR and Nano-hmC-Seal-seq
Chromatin immunoprecipitation quantitative real-time PCR (ChIP-qPCR) assay was performed according to a standard protocol. Briefly, cross-linked cell lysate was sonicated and subjected to immunoprecipitation reaction. The immunoprecipitated DNA was purified with Universal DNA purification Kit (TIANGEN). 1% of each sample volume was separated as the “Input”. HOXB2 antibody (1:400) was used as the primary antibody (Supplementary Table S1). The purified DNA was analyzed with qPCR. For LPS or PBS pretreated group, cells were treated with LPS or PBS for 48 hours before the assay. The primer sequences are listed in Supplementary Table S3.
For Nano-hmC-Seal-seq, ov-control and ov-TET3 (lentiviral particles) cell lines with 3 independent samples were employed. Detailed information about extraction and quality inspection of gDNA samples, 5hmC-Seal library construction, sequencing and data processing was described elsewhere [22, 23]. Briefly, genomic DNA was extracted from cells using Quick-gDNA MicroPrep according to the manufacture’s instruction. After DNA purification, the captured DNA fragments were amplified by the Nextera kit. The sequencing steps were performed on the NextSeq 500 platform which contained a key pull-down step based on covalent chemistry. The raw 5hmC-Seal data was converted into clean reads through removing the reads with adapter, the reads whose proportion of “N” was over 10%, or the sequences less than 30 bp after quality trimming. The clean reads were aligned to the human genome reference [23]. The heatmap of signal distribution, K-means clustering, and genome-wide correlations were performed with deepTools software. The detection of 5hmC-enriched regions was performed with MACS. Peaks that were detected by all replicates were considered as high confident peaks. High quality alignments were summarized by counting overlaps with gene bodies. The raw data are available at the NCBI: PRJNA596911.
Luciferase reporter assay
The cMYC and NANOG promoters (-2000 to -1 bp) were amplified by PCR and then cloned into the PHY-803 luciferase reporter vector through digestion, ligation of vector and target fragment, and sequencing (Hanyin, China). The primer sequences were listed in Supplementary Table S3. Cells were transfected with luciferase reporter genes, including internal control plasmid. 24 hours after transfection, Firefly luciferase activities in cell lysates were measured with Luciferase Assay System (Promega), and normalized for transfection efficiency to the internal activity.
Statistical analysis
SPSS was applied for statistical analysis. The data was noted as “mean ± standard deviation”. Chi-square test or Student t test was applied for two-sample comparisons. The overall survival (OS) curve was drawn with the Kaplan-Meier product limit estimator and analyzed with the log-rank test. The correlations between risk factors and protein expression level were analyzed with two-tailed Pearson test. All the tests at p-value <0.05 were considered as significant difference.