TPC1 (iCell-h309), IHH-4 and BCPAP (iCell-h022) cells were purchased from iCell Bioscience Inc (Shanghai, China). All thyroid papillary carcinoma cell lines were cultured in Dulbecco’s modified Eagle’s Medium (DMEM, Hyclone, Logan, Utah, USA) supplemented with 10% fetal bovine serum (FBS, ZhejingTianhang Biotechnology Co. Ltd, Hangzhou, China). A human normal cell line Nthy-ori 3-1 was required from iCell Bioscience Inc and cultured in RPMI medium 1640 with 10% FBS. All cells were maintained in a humidified atmosphere at 5% CO${}_2$ and 37 °C.

TPC1, IHH-4 and BCPAP cells were seeded into 6-well plates, and transfected with two siRNA targeting KMT2D or negative control (NC) using lipofectamine 2000 (Thermo Fisher Scientific, Waltham, MA, USA). The efficacy of RNA interference in thyroid papillary carcinoma cells was assessed by qRT-PCR. Compared with sh-NC group, the mRNA level of KMT2D was significantly decreased in si-KMT2D (1) and si-KMT2D (2) group (Supplementary Fig. 2).

RNA was isolated from Nthy-ori3-1, TPC, IHH-4 and BCPAP cells with Trizol (Sangon Biotech, Shanghai, China). The total RNA was reverse transcribed into cDNA at the following condition: 95 °C, 10 min; 95 °C, 15 s; 60 °C, 60 s; 40 cycles. The mRNA expression was quantified by the SYBR Premix Ex TaqII (Takara, Shiga, Japan) and normalized by Glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The primers were listed in Table 1.

The cells were trypsinized, harvested and lysed in Radio-Immunoprecipitation Assay (RIPA) buffer containing phenylmethylsulfonyl fluoride (PMSF) (Beyotime, Shanghai, China) and protease inhibitor (CWBIO, Beijing, China). The concentration of total protein was measured by the BCA Protein Assay Kit (Beijing Solarbio Science & Technology Co., Ltd, Beijing, China). After loading samples, proteins were migrated by Sodium dodecyl-sulfate polyacrylamide gel electrophoresis (SDS PAGE) and transferred to polyvinylidene fluoride (PVDF) membranes, which were then blocked with non-fat milk for 2 h. The primary antibody was used to incubate membrane overnight at 4 °C. The antibodies were bought from Affinity Biosciences company (KMT2D Antibody: DF9646; H3K4me2 Antibody: AF0589; H3K9me2 Antibody: DF6937; NCOA6 Antibody: DF13176; THRB Antibody: AF0357; H3 Antibody: AF0863; GAPDH Antibody: AF7021). The second antibody was diluted with non-fat milk and incubated membrane for 1.5 h at room temperature. ECL detection kit and ChemiScope 6100 (Clinx Science Instruments Co., Ltd, Shanghai, China) were used to visualize protein bands.

High-sensitivity ChIP Kit (ab185913, abcam, Cambridge, UK) was used in this study. We assessed the association between KMT2D and NCOA6. The protocol was as follow: (1) Formaldehyde cross-linking and ultrasonic crushing of cells: the cells were removed, formaldehyde was added to make the final concentration of formaldehyde 1%, and the cells were incubated at 37 °C for 10 min. The cross-linking was terminated, the medium was sucked up, and the cells were washed twice with cold Polybutylene Succinate (PBS). The cells were collected in a 15 mL centrifuge tube with a cell scraper. After pre-cooling, the cells were collected by centrifugation, and the supernatant was poured out. (2) Impurity removal and antibody feeding: after ultrasonic crushing, centrifuge to remove insoluble substances; A 900μL dilution Buffer and 20 μL of 50$\times$polyisocyanopeptide (PIC) are added to 100 μL of the ultrasonic fragmentation product. Then 60 μL of Protein A arose/Salmon Sperm DNA was added. Toss and mix at 4 °C for 1 h; After 1 h, the samples were left at 4 °C for 10 min to precipitate and centrifuged at 700 rpm for 1 min. Take the supernatant. 20 μL of each will be reserved as input. Antibody was added to one tube and no antibody was added to the other overnight at 4 °C. (3) Test the effect of ultrasonic crushing: Take 100 μL of the product after ultrasonic crushing, add 4 μL 5M NaCl, and treat it at 65 °C for 2 h to uncross-link. Half was extracted with phenol/chloroform. Ultrasonic effect was detected by electrophoresis. (4) Precipitation and cleaning of immune complexes: After overnight incubation, 60 μL Protein A Agarose/Salmon Sperm DNA was added to each tube at 4 °C for 2 h; After 10min, the supernatant was removed by centrifugation. Wash the precipitated complex. Then, 20 μL NaCl was added to elution. (5) Recovery of DNA samples: After unlinking overnight, l μL RNaseA was added and incubated at 37 °C for 1 h. 10 μL EDTA, 20 μL Tris HCl (pH 6.5), 2 μL 10 mg/mL proteinase K was added for 2 h, subsequently. The final sample was dissolved in 100 μL H${}_2$O. (6) The bound target DNA fraction was amplified by PCR.

Cells were seeded in a 96-well plate for 24 h, and 10 $\mu$L CCK-8was added. The availability of cells was detected by CCK-8 kit (MedChemExpress, New Jersey, USA) at 0 h, 24 h, 48 h and 72 h after siRNA transfection. The results were determined by microplate reader (Molecular Devices, Silicon Valley, USA) at 450 nm.

Cells were resuspended in serum-free medium, added to a six-well plate (5 $\times$ 10${}^5$ cells/well) overnight. The wounds were made using a 10 $\mu$L pipette tip, cell debris was removed by washing with PBS for three times. Wound healing was observed under an inverted microscope at 0 h and 12 h. Cells migration was assessed by ImageJ v1.8.0 software (National Institutes of Health, New York, USA).

Cells were added into the upper chambers of 24-well transwell chambers (Corning, New York, USA) with or without BD Matrigel (invasion assay and migration assay). After 24 h of incubation, cells on the surface of upper chamber were removed with cotton swabs. The invaded/migrated cells on the lower surface were fixed with 4% paraformaldehyde and stained with 0.1% crystal violet. Cells were imaged and counted in five random fields.

Statistical analysis was performed with SPSS 18.0 software (IBM Corp., Chicago, IL, USA) and GraphPad Prism 8 software (GraphPad Software Inc., San Diego, CA, USA). Each experiment was repeated three times. Differences between two groups were made using Independent sample T test or Kruskal-Wallis H Test. All data were expressed as the means $\pm$ standard deviations, p 0.05 was considered statistically significant.

We compared the expression of KMT2D in thyroid papillary carcinoma cell lines and normal thyroid follicular cell line by qRT-PCR and Western blotting (Fig. 1). The results demonstrated that KMT2D mRNA levels were significantly higher in TPC, IHH-4 and BCPAP cells than in Nthy-ori3-1 cell. Western blotting analysis of KMT2D protein confirmed the RNA expression data, showing remarkably higher expression in thyroid papillary carcinoma cells than in normal thyroid cell.

Fig. 1.

The mRNA and protein expressions of KMT2D in Nthy-ori3-1, TPC, IHH-4 and BCPAP cells. (a) qRT-PCR analysis. (b) Western blotting analysis. A: Nthy-ori3-1 cell; B: TPC cell; C: IHH-4 cell; D: BCPAP cell. ${}^{\mathrm{▲}\mathrm{▲}}$p 0.01.

TPC, IHH-4 and BCPAP cells were transfected with KMT2D inhibitors or NC to evaluate the effect of KMT2D in thyroid papillary carcinoma cells. CCK-8 assay demonstrated the cell viability in si-KMT2D (1) and si-KMT2D (2) groups was lower than in si-NC group (Fig. 2a). It indicated that KMT2D knockdown suppressed the proliferation of TPC, IHH-4 and BCPAP cells. Then, Wound healing and Transwell assay were used to investigate the role of KMT2D in the cell invasion and migration. As shown in Fig. 2b, the migration distances in si-KMT2D (1) and si-KMT2D (2) groups were less than si-NC group at 12 h after scratch. Similarly, the Transwell assay showed a significantly decrease in the number of migrated cells in si-KMT2D group (p 0.01). The number of invaded cells in KMT2D gene silence group decreased remarkably compared with si-NC group (p 0.01, Fig. 2c). These results indicated that KMT2D promotes the proliferation, migration and invasion of thyroid papillary carcinoma cells.

Fig. 2.

KMT2D regulates migration and invasion in TPC, IHH-4 and BCPAP cells. (a) CCK8 assay. (b) Wound healing. (c) Transwell assay. ${}^{\mathrm{▲}}$p 0.05, ${}^{\mathrm{▲}\mathrm{▲}}$p 0.01.

ChIP assay showed that KMT2D is associated with NCOA6 in TPC, IHH-4 and BCPAP cells (Fig. 3a). Compared with anti-IgG group, the input of NCOA6 is significantly increased in anti-KMT2D group. To understand why KMT2D is required for the activities of NCOA6, we examined the levels of H3K4me2 and H3K9me2, which are closely correlated with active enhancers and promoters. Western blotting showed that silencing KMT2D (si-KMT2D) was significantly upregulated H3K4me2 and H3K9me2 at protein levels (Fig. 3c). In addition, western blotting and qRT-PCR analysis were used to explore the association of KMT2D and NCOA6. As shown in Fig. 3b,c, KMT2D knock out markedly decreased the expression of NCOA6 and THRB compared with si-NC group. These results demonstrated that KMT2D regulates the expression of NCOA6 and THRB through demethylation of H3K4me2 and H3K9me2 in thyroid papillary carcinoma cells.

Fig. 3.

The mechanism of KMT2D on papillary thyroid carcinoma cells. (a) ChIP assay. (b) qRT-PCR analysis. (c) Western blotting analysis. A, si-NC; B, si-KMT2D (1); C, si- KMT2D (1). ${}^{\mathrm{▲}}$p 0.05, ${}^{\mathrm{▲}\mathrm{▲}}$p 0.01.