The human normal thyroid Nthy-ori 3-1 cell line and two ATC cell lines (SW1736 and KAT-18) were obtained from the American Type Culture Collection (Manassas, VA, USA). The cell lines were authenticated using short-tandem repeat profiling performing by BMR Genomics (Padova, Italia). The cells were maintained in Dulbecco's modified Eagle's medium (HyClone; GE Healthcare Life Sciences, Beijing, China) with 10% fetal bovine serum (HyClone; GE Healthcare Life Sciences) in a humidified atmosphere with 5% CO₂ and 95% air at 37°C. For the Cell Counting Kit-8 (CCK-8) assay, SW1736 and KAT-18 cells were cultured with increasing doses of GA (Key Laboratory of Carcinogenesis and Intervention, China Pharmaceutical University, Nanjing, China; 0, 2, 4 and 6 µg/ml) for 48 h or treated with 2 µg/ml GA for the indicated time points (0, 24, 48 and 72 h) at 37°C. For the colony formation assay and flow cytometry, SW1736 and KAT-18 cells were cultured with increasing doses of GA (0, 2, 4 and 6 µg/ml) for 48 h at 37°C. To discover the association between BRD4 and GA, ATC cells were then divided into four groups, treated with BRD4 overexpression plasmid with dimethyl sulfoxide (DMSO), BRD4 overexpression plasmid with 2 µg/ml GA, control vector with DMSO, and control vector with 2 µg/ml GA.

Furthermore, human ATC specimens and their adjacent normal thyroid tissues (10 pairs) were collected from 10 patients (3 males and 7 females; age range, 51–70 years; median age, 62) who underwent surgery at the Weifang People's Hospital (Weifang, China) between January 2017 and October 2017. All ATC specimens were confirmed by pathological diagnosis according to the World Health Organization criteria (22). The present study was approved by the Ethics Committee of Weifang People's Hospital (Weifang, China). Written informed consent was obtained from each patient who participated in this study.

shRNA plasmids for BRD4, which were designed against the BRD4 gene and constructed in Phblv-u6-puro vectors, were purchased from Hanbio Biotechnology Co., Ltd. (Shanghai, China). A non-target scrambled oligonucleotide functioned as the negative control. To generate stable BRD4 knockdown cells, SW1736 and KAT-18 cells were grown in six-well plates until they reached 40% confluence. The medium was replaced with 1 ml Dulbecco's modified Eagle's medium (HyClone; GE Healthcare Life Sciences, Beijing, China) supplemented with 100 µl viral supernatant (1×10⁸ UT/ml) and 8 µg/ml polybrene (Hanbio Biotechnology Co., Ltd.) for 24 h. The SW1736 and KAT-18 cells were further cultured in Dulbecco's modified Eagle's medium (HyClone; GE Healthcare Life Sciences, Beijing, China) with 10% fetal bovine serum (HyClone; GE Healthcare Life Sciences) containing puromycin (Hanbio Biotechnology Co., Ltd.) at 5 and 3 µg/ml at 37°C for three passages, respectively. Individual puromycin-resistant colonies were isolated during drug screening. The shRNA sequences used in the present study were as follows: shBRD4, 5′-GatccGCCTGGAGATGACATAGTCTTATTCAAGAGATAAGACTATGTCATCTCCAGGTTTTTTc-3′; and shControl, 5′-GatccTTCTCCGAACGTGTCACGTAATTCAAGAGATTACGTGACACGTTCGGAGAATTTTTTg-3′. A non-target scrambled oligonucleotide functioned as the negative control and presented no homology to any human transcripts. Furthermore, BRD4 overexpression was examined using BRD4 overexpression plasmid and control vectors that were obtained from Santa Cruz Biotechnology, Inc. (Dallas, TX, USA). Cell transfection was conducted according to the manufacturer's protocol.

Cell viability was examined using a CCK-8 assay. Briefly, SW1736 and KAT-18 cells were plated into a 96-well plate (3,000 cells/well). After 0, 24, 48 and 72 h of incubation, 10 µl CCK-8 solution (Beyotime Institute of Biotechnology, Haimen, China) was added into each well at 37°C. Following 3 h of culturing, the absorbance of each well was determined using a Multiskan MK3 device (Thermo Fisher Scientific, Inc., Waltham, MA, USA) at a wavelength of 450 nm.

SW1736 and KAT-18 cells were cultured in a 6-well plate at a density of 300 cells/well. After 7 days, the cells were washed with phosphate-buffered saline (PBS), fixed with paraformaldehyde and stained with crystal violet. Subsequently, the number of the colonies that had migrated through the pores was quantified by randomly counting 10 independent visual fields using the images.

Cell apoptosis was assessed using flow cytometry with staining of the cells using an Annexin V/propidium iodide (PI) kit (Nanjing KeyGen Biotech Co, Ltd., Nanjing, China) according to the manufacturer's protocol. Briefly, cells seeded in 24-well plates were exposed to 0, 2, 4 and 6 µg/ml of GA for 48 h, then cells were collected and washed twice in ice cold PBS. The washed cells (2×10⁵/well) were resuspended in 100 µl binding buffer (included in the kit) and stained with 5 µl Annexin V and 5 µl PI for 15 min at room temperature. Following incubation for 15 min in the dark at room temperature, flow cytometry was performed. A flow cytometer (Cytomics FC 500 MPL; Beckman Coulter, Inc., Brea, CA, USA) was utilized to evaluate the apoptotic levels in each sample according to the manufacturer's protocol. Data were analyzed using ModFit LT 3.0 (Verity Software House, Inc., Topsham, ME, USA).

Total RNA was extracted from tissues and cells by using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. The concentration of total RNA was determined by spectrophotometry (TECAN Infinite^® F200PRO microplate reader; Tecan Group, Ltd., Mannedorf, Switzerland). For mRNA expression analysis, RNA was reverse transcribed into cDNA using a PrimeScript RT reagent kit with gDNA Eraser (Takara Biotechnology Co., Ltd., Dalian, China). qPCR was then conducted using SYBR Premix Ex Taq (Takara Biotechnology Co., Ltd.). The qPCR conditions were applied for detecting mRNAs: 95°C for 30 sec, followed by 40 cycles of 95°C for 30 sec, 60°C for 30 sec and 72°C for 30 sec using the following primers: BRD4 forward, 5′-CATGGACATGAGCACAATCA-3′, and reverse, 5′-TCATGGTCAGGAGGGTTGTA-3′; β-actin forward, 5′-GATCATTGCTCCTCCTGAGC-3′, and reverse, 5′-ACTCCTGCTTGCTGATCCAC-3′. β-actin served as the internal control. The specificity of amplification was examined using melting curve analysis and electrophoresis in 1% agarose gel. The relative expression level of the target gene mRNA was calculated as the inverse log of ΔΔCq and normalized to the internal control (23).

SW1736 and KAT-18 cells were lysed to extract the total protein by using RIPA lysis buffer (Promega Corporation, Madison, WI, USA) according the manufacturer's instructions. Protein concentration was determined using a BCA Protein Assay kit (Thermo Fisher Scientific, Inc.) according the manufacturer's protocol. Next, the same amount of protein (40 µg) from each cell line was subjected to 12% SDS-PAGE and transferred onto polyvinylidene difluoride membranes. The membranes were then respectively incubated with various diluted primary antibodies against BRD4 (1:200; cat. no. ab75898; Abcam, Cambridge, MA, USA), B-cell lymphoma 2 (Bcl-2; 1:500; cat. no. sc7382; Santa Cruz Biotechnology, Inc.), Bcl-2-associated X protein (Bax; 1:500; cat. no. sc7480; Santa Cruz Biotechnology, Inc.), cleaved caspase-3 (1:500; cat. no. #9661; Cell Signaling Technology, Inc., Beverly, MA, USA), cleaved poly(ADP-ribose) polymerase (PARP; 1:500; cat. no. #5625; Cell Signaling Technology, Inc.) or β-actin (1:800; cat. no. AA128; Beyotime Institute of Biotechnology, Nantong, China). Following incubation overnight at 4°C, horseradish peroxidase-conjugated goat anti-rabbit secondary antibody (1:2,000; cat. no. A0216; Beyotime Institute of Biotechnology) was added and incubated at room temperature for 2 h. Specific bands were visualized with an enhanced chemiluminescence reagent (KeyGen Biotech Co., Ltd.) on an autoradiographic film. For quantitative assay, images were analyzed using ImageJ software (version 1.48u; National Institutes of Health, Bethesda, MD, USA).

All animal studies were approved by the Animal Care and Welfare Committee of the Weifang People's Hospital and conducted in strict accordance with the guidelines of the National Animal Welfare Law of China. Female BALB/c nude mice (4-week-old; n=36; female; weight range, 20–25 g) were purchased from the Laboratory Animal Center of Yangzhou University (Yangzhou, China) and maintained in a specific-pathogen-free environment with a constant humidity (45–50%) and a constant temperature (25–27°C) under a 12 h light/dark cycle with ad libitum access to food and water. All mice were acclimatized for 1 week prior to the experiments. In order to identify the function of BRD4 in vivo, the mice (n=6/group) were subcutaneously injected into the right flanks with 2×10⁶ shControl or shBRD4 SW1736 cells. A total of 3 weeks later, the mice were euthanized with the dislocation of cervical vertebra. In order to identify the association between BRD4 and GA in vivo, the mice were subcutaneously injected with 2×10⁶ SW1736 cells transfected with BRD4 plasmid (n=12/group) and control plasmid (n=12/group). Following injection, the mice were divided into 4 groups (n=6/group) (BRD4 plasmid group injected with BRD4 plasmid + 4 mg/kg GA; control plasmid group injected with control plasmid + 4 mg/kg GA) and immediately received GA treatment or 0.9% saline solution. The intravenous treatments were performed once every 3 days for a total of 21 days. The tumor volume (mm³) was calculated using the following formula: Volume = 0.5 × length × width². After 3 weeks, the mice were euthanized, and the tumors were isolated, weighed, photographed and analyzed by immunohistochemistry.

Paraffin-embedded archival specimens were cut into sections 4-µm thick and baked at 65°C for 30 min. Sections were then dewaxed in xylene and rehydrated via graded alcohol solutions (100% ethanol, 5 min; 95% ethanol, 5 min; 70% ethanol, 5 min;), blocked in methanol containing 3% hydrogen peroxide for 10 min at room temperature, and then incubated with an BRD4 antibody (1:200; cat. no. ab75898; Abcam) anti-Ki67 antibody (1:200; cat. no. sc-23900; Santa Cruz Biotechnology, Inc.) at 4°C overnight. Following rinsing with PBS solution, biotinylated goat anti-rabbit serum IgG (1:2,000; cat. no. ab64256; Abcam) was used as a secondary antibody for 1 h at room temperature and the streptavidin peroxidase complex reagent was applied for 1 h at room temperature. Finally, the sections were incubated in a 3,3′-diaminobenzidine solution (DAB kit; Beyotime Institute of Biotechnology) at room temperature for 10 min and then counterstained with hematoxylin for 3 min at room temperature. Ten randomly selected visual fields per section were examined under a light microscope (magnification, ×200) to evaluate the BRD4 and Ki67 expression.

Data are expressed as the mean ± standard error. Statistical analysis was performed using SPSS version 13.0 software (SPSS, Inc., Chicago, IL, USA). Statistical significance was determined through unpaired Student's t-test or one-way analysis of variance. P<0.05 was considered as an indicator of a statistically significant difference.

To illustrate the function of GA on ATC cell viability, a CCK-8 assay was conducted. SW1736 and KAT-18 cells were cultured with increasing doses of GA (0, 2, 4 and 6 µg/ml) for 48 h. As shown in Fig. 1A, GA therapy for 48 h significantly inhibited the ATC cell viability in a dose-dependent manner. A CCK-8 assay was also performed on SW1736 and KAT-18 cells treated with 2 µg/ml GA for the indicated time points (0, 24, 48 and 72 h). The results demonstrated that GA treatment decreased the viability of the two ATC cell lines in a time-dependent manner (Fig. 1B).

In order to investigate the function of GA on ATC cell proliferation, a colony formation assay was then performed. As shown in Fig. 1C, treatment of ATC cells with GA at different concentrations for 7 days significantly decreased the number of colonies in the cultures. The function of GA on ATC cell apoptosis was also examined by flow cytometry, and the results revealed that treatment of SW1736 and KAT-18 cells with GA for 48 h significantly enhanced the apoptosis rate (Fig. 1D) in a dose-dependent manner.

Previous studies have demonstrated that BRD4 served as an oncogene in in papillary thyroid cancer (24,25). In the present study, we attempted to illustrate the association of GA and BRD4. The western blot assay demonstrated that GA treatment (2, 4 and 6 µg/ml) evidently decreased the protein expression of BRD4 in SW1736 (Fig. 2A) and KAT-18 (Fig. 2B) cells. Furthermore, the RT-qPCR assay illustrated that GA treatment significantly decreased the mRNA expression of BRD4 in the two cell lines in a dose-dependent manner (Fig. 2C and D). Therefore, BRD4 may serve a critical role in the antitumor effect of GA in ATC.

To determine the expression profile of BRD4 in ATC tissues, RT-qPCR was performed in 10 ATC specimens and their corresponding non-neoplastic thyroid tissues. Compared with the non-neoplastic thyroid tissues, the expression level of BRD4 was significantly enhanced in the ATC specimens (Fig 3A). Furthermore, the mRNA and protein expression levels in normal thyroid and ATC cells were examined by RT-qPCR and western blot analysis, respectively. The results revealed that the mRNA and protein expression levels of BRD4 were significantly lower in the human thyroid Nthy-ori 3-1 cell line in comparison with those in the ATC cell lines (Fig. 3B and C). In addition, immunohistochemical analysis was conducted in the patient specimens, and the results indicated that BRD4 was evidently overexpressed in ATC tissues as compared with that in the adjacent normal tissues (Fig. 3D).

BRD4 was silenced in the SW1736 and KAT-18 cells in order to analyze the influence of this protein on the biological behavior of ATC cells. Western blot analysis was used to determine the expression level of BRD4, demonstrating that the shRNA transfection successfully downregulated BRD4 in the cells (Fig. 4A and B). Subsequently, the viability of the SW1736 and KAT-18 cells was analyzed via a CCK-8 assay, and BRD4 silencing was observed to significantly decrease the viability of these cell lines (Fig. 4C and D). Furthermore, the colony formation assay demonstrated that the BRD4 knockdown significantly inhibited the proliferation of ATC cells (Fig. 4E and F), while the rate of apoptosis in these cells was markedly enhanced (Fig. 4G and H). All these data revealed that BRD4 is involved in the growth of ATC cells.

Western blotting was also performed to analyze the expression of apoptosis-associated proteins. In accordance with the aforementioned functional experiments, BRD4 silencing significantly decreased the expression of the anti-apoptotic factor Bcl-2, while it enhanced the expression levels of the pro-apoptotic factor Bax, cleaved caspase-3 and cleaved PARP in the SW1736 and KAT-18 cell lines (Fig. 4A and B).

To investigate the correlation of BRD4 expression with the ATC progression in vivo, SW1736 cells transfected with shBRD4 and shControl were subcutaneously injected into nude mice (n=6/group). At 3 weeks after injection, the tumors were removed, and images were captured (Fig. 5A). The tumor volume and weight of the excised tumors indicated that BRD4 knockdown significantly delayed the tumor growth (Fig. 5B and C). Histological analysis of tumor proliferation also demonstrated that shBRD4 tumors had significantly fewer Ki-67 positive cells compared with those in the shControl tumors (Fig. 5D and E). These data illustrated that BRD4 expression enhanced ATC growth in vivo.

To further illustrated the correlation of GA and BRD4, BRD4 expression was overexpressed by transfection with BRD4 overexpression plasmids. As shown in Fig. 6A and B, the BRD4 overexpression plasmid notably increased the expression of this protein in ATC cells compared with the control vector group.

ATC cells were then divided into four groups, treated with BRD4 overexpression plasmid + DMSO, BRD4 overexpression plasmid + 2 µg/ml GA, control vector + DMSO, and control vector + 2 µg/ml GA. The CCK-8 (Fig. 6C and D) and colony formation (Fig. 6E and F) assays revealed that overexpression of BRD4 was able to partially abrogate the inhibitory effects of GA on ATC cells. Furthermore, overexpression of BRD4 partially counteracted the suppression of cell apoptosis (Fig. 6G and H). Furthermore, the western blot analysis results demonstrated that BRD4 overexpression rescued the expression of the antiapoptotic protein Bcl-2, and counteracted the expression of pro-apoptosis gene Bax, which was induced by GA (Fig. 6A and B). Overexpression of BRD4 was demonstrated to counteract the effect of GA, so BRD4 down-regulation is critical for the antitumor growth effects of GA. These data manifested that GA inhibited the PTC progression through the inhibition of BRD4.

Tumor xenografts of SW1736 cells transfected with BRD4 overexpression and control vector were used to evaluate the antitumor effect of GA (4 mg/kg) in nude mice in vivo. At 21 days after injection, the tumors were removed and images are shown in Fig. 7A. The periodic measurements of the tumor volume (Fig. 7B) and the final weight (Fig. 7C) of the excised tumors demonstrated that BRD4 overexpression was able to significantly abrogate the antitumor effects of GA on ATC cells.