The BRD4 inhibitor JQ1 suppresses tumor growth by reducing c-Myc expression in endometrial cancer

Dimethyl sulfoxide (DMSO) and 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) were purchased from Sigma-Aldrich (St. Louis, MO, USA). The BRD4 inhibitors, JQ1, I-BET151 and OTX015, were all purchased from MedChemExpress (Shanghai, China), dissolved in DMSO, and stored in small aliquots – 20 ℃.

Human EC cell lines (HEC-1A, Ishikawa, RL-95 and An3C cells) were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). HEC-1A cells were cultured in McCoy’s 5A medium (Thermo Fisher Scientific Inc., Waltham, MA, USA) supplemented with 10% fetal bovine serum (FBS) (Biological Industries, Kibbutz Beit-Haemek, Israel) and 1 mM sodium pyruvate (Thermo Fisher Scientific Inc.). Ishikawa and An3C cells were grown in RPMI 1640 medium (Thermo Fisher Scientific Inc.) with 10% FBS. RL-95 cells were maintained in DMEM/F12 medium (Thermo Fisher Scientific Inc.) with 10% FBS. All cell culture media were supplemented with 100 U/ml penicillin and 100 μg/ml streptomycin (Thermo Fisher Scientific Inc.). Cells were cultured at 37 ℃ in a 5% CO₂ atmosphere. For in vitro experiments, BET inhibitors were added to cultured cancer cells at indicated concentrations. Equivalent concentrations of DMSO were added to cells as controls.

Fresh endometrial tissue samples from 50 patients with EC (Type I, n = 33; Type II, n = 17) and 14 patients with leiomyoma who had undergone hysterectomy between January 2010 and June 2012 were obtained from the Department of Gynecology and Obstetrics of Qilu Hospital. Staging and histological subtyping were performed according to the 2009 guidelines of the International Federation of Gynecology and Obstetrics. Fully informed written consent was acquired from the patients before the collection of tissue samples. This study was approved by the Ethics Committee of Qilu Hospital of Shandong University (KYLL-2019(KS)-376). The clinical characteristics of the EC patients are listed in Table 1. All patients were followed up until June 2019. The patients in this study had no history of other previous cancers and did not undergo any preoperative chemotherapy, radiotherapy, or other hormonal therapies. Overall survival (OS) was defined as the time between the date of surgery and the date of patient death directly due to EC or last follow-up. Some tissues were preserved immediately in liquid nitrogen for subsequent western blot assays. All samples were fixed in 4% paraformaldehyde for 24 h, dehydrated, embedded in paraffin blocks, and sectioned (4 μm thickness) for immunohistochemistry (IHC) assays.

Endometrial tissue sections were deparaffinized with xylene and rehydrated in different concentrations of ethanol. Hydrogen peroxide was used to block endogenous peroxidase activity, and nonspecific antigens were blocked by incubation in goat serum for 30 min. The sections were then incubated with primary antibodies overnight at 4 °C. The primary antibody against BRD4 was purchased from Abcam (1:50 dilution, Cambridge, UK). After sequential incubation with biotin-labeled goat anti-rabbit IgG polymer and horseradish peroxidase-labeled streptavidin for 30 min each, positive signals were detected using 3,3′-diaminobenzidine (DAB) substrate (Zhongshan Jinqiao Biotechnology, Beijing, China) following the manufacturer’s recommendations. Finally, the stained tissues were evaluated and scored by two blinded investigators.

Strong positive staining was defined as a brown signal in the nucleus or cytoplasm; moderate staining as a yellow–brown reaction; and weak staining as a light-yellow reaction. The staining intensity was graded as follows: 0 = no staining; 1 = weak staining; 2 = moderate staining; and 3 = strong staining. The percentage of tumor cell staining was graded based on the following criteria: 0 = no staining; 1 =  ≤ 10% positive tumor cells; 2 = 11–50% positive tumor cells; 3 = 51–80% positive tumor cells; and 4 =  ≥ 81% positive tumor cells. The staining score was calculated by multiplying the intensity score by the quantity score and ranged from 0 to 12. Scores ≥ 8 indicated overexpression of BRD4 in endometrial tissues.

Cells (2000 cells in 100 μl/well) were propagated in 96-well culture plates overnight and then exposed to the indicated concentrations of JQ1, OTX015, I-BET151 or transfected with siRNA or BRD4 overexpression lentivirus. After treatment for indicated time, 10 μl of MTT solution (5 mg/ml) was added to each well, and the plates were incubated at 37 ℃ for another 4 h. Then, the medium was removed, and DMSO (100 μl/well) was added to dissolve the formazan product. The absorbance of each well was measured at 570 nm using a microplate reader (Tecan Group Ltd., Männedorf, Switzerland). Three replicate wells were included for each experiment, and the experiments were performed in triplicate.

Cells (800 cells in 2 ml/well) were seeded in 6-well plates and cultured for 48 h. Next, the cells were treated with JQ1 or OTX015 at the indicated concentrations for 14 days at 37 ℃ in 5% CO₂. The cells were fixed with paraformaldehyde for 20 min and stained with crystal violet (Beyotime, Beijing, China) for 30 min, and colonies (> 50 cells) were counted.

RIPA buffer with protease inhibitor cocktail, PMSF (0.1 mM) and NaF (10 mM) was used to lyse cells treated with JQ1 or transfected with siRNA. Protein concentrations were determined by the BCA protein assay kit (Beyotime, Beijing, China). Equal amounts of protein (30 μg) were separated by SDS-PAGE on a 10% gel and transferred to a nitrocellulose membrane. After blocking with 5% defatted milk for 1 h at room temperature, the membrane was incubated overnight at 4 ℃ with primary antibodies against BRD4 (Abcam), c-Myc, CDK4, Bax (Santa Cruz, CA, USA), β-Actin, cleaved-Caspase-3 (cl-Caspase3), PARP, cleaved-PARP (cl-PARP), CDC25A, CyclinD1, P21, Rb, phosphorylated Rb (p-Rb), PI3K, AKT, phosphorylated AKT (p-AKT), mTOR, phosphorylated mTOR (p-mTOR) (Cell Signaling Technology, MA, USA). Secondary antibodies were purchased from Cell Signaling. Protein bands were detected using an enhanced chemiluminescent substrate (Thermo Fisher Scientific Inc.). ImageJ software (National Institutes of Health, USA) was used to quantify the immunoblots. β-Actin served as a control.

Apoptosis of JQ1 or OTX015-treated EC cells or cells infected with lentivirus was detected by flow cytometry. Cells were washed with PBS, digested with trypsin without EDTA and resuspended in 100 μl of binding buffer. Next, the cells were stained with 5 μl of Annexin V-FITC and 5 μl of propidium iodide (PI) (BD, NJ, USA). After incubation at room temperature in the dark for 15 min, another 400 μl of binding buffer was added to each tube. Cell apoptosis was analyzed by flow cytometry (BD Biosciences, San Jose, CA, USA) according to the manufacturer’s instructions.

Cell cycle analysis was performed by flow cytometry. Synchronized cells treated with JQ1 or OTX015 for 24 h were washed with PBS, digested with trypsin with EDTA, and fixed in 75% cold ethanol overnight at 4 ℃. The cells were washed twice with PBS and stained with 400 μl of PI/RNase Staining Buffer (BD Biosciences) at room temperature in the dark for 30 min. The cell cycle was then analyzed by flow cytometry according to the manufacturer’s instructions.

Small interfering RNAs (siRNAs) targeting human BRD4 or c-Myc and silencer negative control siRNAs were constructed by GenePharma (Shanghai, China). Cells were seeded into 6-well plates at 2.5 × 10⁵ cells/well and incubated overnight before transfection with 150 nM (final concentration) BRD4 or c-Myc siRNA using Lipofectamine 3000 (Invitrogen) according to the manufacturer’s instructions. For western blot assays, the cells were collected 72 h following transfection. For MTT assays, cells were collected on days 1–5 following transfection.

To establish stable cell lines, lentivirus was purchased from GeneChem (Shanghai, China). Ishikawa cells were infected with the pUbi-BRD4 and pUbi-NC lentivirus. After 48 h, cells were screened with 2 μg/mL puromycin (Solarbio, Beijing, China) for 1–2 weeks to obtain stable cell lines.

Animal experiments were approved by Institutional Animal Care and Use Committees of Qilu Hospital of Shandong University. Ishikawa cells (1 × 10⁷ cells in 100 μl of PBS) were subcutaneously injected into the left armpits of 5-week-old nude mice. When the tumor volume reached approximately 50 mm³, the mice were randomly allocated to the treatment or control group. Depending on the group, JQ1 (50 mg/kg/d) or placebo was administered intraperitoneally daily for 3 weeks. Body weight and tumor volume were measured every other day. Tumor volume was calculated using the following formula: (length × width²)/2. Blood was sampled by extirpation of the eyeball before death and used for routine blood examination. The mice were sacrificed 21 days after initiating JQ1 treatment, and tumors were immediately removed and weighed.

Survival probabilities were calculated by the Kaplan–Meier method. Because the number of events was low for overall survival, we could not evaluate the independent prognostic value of BRD4 in a multivariable model adjusting for BMI, disease stage, histological type, and histological grade simultaneously. All experiments were repeated at least three times. Student’s t-test was used to compare differences between two groups. The results are presented as the means ± standard deviations (SD). All statistical tests were two-sided; *P < 0.05 was considered statistically significant, **P < 0.01 moderately significant, and ***P < 0.001 highly significant. GraphPad Prism Version 7.00 was used to perform the statistical analyses.

Tissue samples derived from 50 patients with EC and 14 patients with leiomyoma undergoing hysterectomy between January 2010 and June 2012 were used to validate and estimate the potential role of BRD4. To determine the expression levels of BRD4 in EC tissues and normal endometrial samples, western blot and IHC analyses were performed. As expected, BRD4 protein expression was upregulated in EC tissue (n = 8) compared with normal endometrial tissue (n = 6) (Fig. 1A). IHC staining demonstrated that BRD4 protein expression was primarily located in the nucleus (Fig. 1B). In addition, high BRD4 expression was significantly associated with serous carcinoma and clear cell carcinoma (Fig. 1C). The stained EC tissue samples were scored and divided into groups with high (n = 28) or low (n = 22) expression of BRD4. Strong nucleic BRD4 expression was significantly associated with shorter EC patient survival in Kaplan–Meier analysis (P = 0.0218; Fig. 1D).

TCGA and the GSE17025 dataset were used to further evaluate the expression of BRD4 and its prognostic value in EC. As shown in Fig. 2A, a paired-comparison test of data from TCGA showed that BRD4 expression was higher in EC cancer tissues than in corresponding adjacent tissues. In addition, BRD4 expression was significantly higher in serous carcinoma, a form of type II EC, than in type I EC (Fig. 2B). The cohort from TCGA was then divided into groups with high or low BRD4 expression. Compared with the low BRD4 expression group, OS was significantly lower in the high BRD4 expression group (P = 0.0021, Fig. 2C). Similar trends were observed in the analyses of the GSE 17025 dataset, as shown in Fig. 2D–F. However, in the Type I EC group, there were no significant differences in BRD4 expression according to grade or stage in TCGA and the GSE 17025 dataset. These results suggest that elevated BRD4 expression is a negative prognostic factor for EC patients.

MTT and colony formation assays were used to detect the inhibitory effect of JQ1 on the proliferation of EC cells. The MTT assays revealed that JQ1 inhibited cell growth in a dose-dependent and time-dependent manner in all four EC cell lines (Fig. 3A). JQ1 concentration of 1.0 µM was sufficient to completely inhibit the growth of the two most sensitive cell lines, HEC-1A and Ishikawa, which were therefore chosen for subsequent experiments. Colony formation assays showed that treatment with JQ1 for 14 days at a dose as low as 0.1 µM JQ1 inhibited colony formation by EC cells (Fig. 3B). Low doses of JQ1 (0.1 µM) reduced colony numbers, while 1.0 µM JQ1 led to almost complete inhibition of colony formation (data not shown). Detection of the expression of BRD4 in the four EC cell lines by western blot showed that BRD4 was highly expressed in HEC-1A and Ishikawa cell lines, which were relatively sensitive to JQ1 (Fig. 3C). To verify the role of BRD4 in EC cells, we performed experiments using two other common BET inhibitors, I-BET151 and OTX015. MTT assay demonstrated that OTX015 has shown a significant anti-proliferative activity against EC cells, superior to that of I-BET151 (Fig. 3D). Combined with literature reports, OTX015 has been used in several clinical trials. Therefore, we used OTX015 for the subsequent experiments. Similarly, OTX015 and I-BET151 suppressed cell colony formation in EC cells (Fig. 3E). Taken together, our results demonstrate that BET inhibitors reduce the proliferation of human EC cells and that the sensitivity to JQ1 depends on the expression level of BRD4.

To investigate the mechanism underlying the antiproliferative effects of BET inhibitors in EC cells, we analyzed cell apoptosis and apoptotic-signaling pathways using flow cytometry and western blot. Compared with the control group, JQ1 treatment for 48 h increased early and late apoptosis of EC cells (Fig. 4A). For HEC-1A and Ishikawa cells, the apoptosis rates were 43% and 11% in the groups treated with 5 µM JQ1, respectively, compared with 14% and 5% in the corresponding control groups. Consistent with the flow cytometry analysis, JQ1 activated caspase-3 and induced PARP cleavage (Fig. 4B). JQ1 also led to upregulation of the proapoptotic protein Bax and downregulation of the antiapoptotic protein Bcl-2 in EC cells. Like JQ1, OTX015 regulated expression of Bcl-2/Bax, and triggered cell apoptosis (Fig. 4C and D).

Given the important role of the PI3K/AKT/mTOR signaling pathway in cell growth and proliferation, we further investigated the impact of JQ1 on this pathway by western blot. As shown in Fig. 5A-D, in both HEC-1A and Ishikawa cells, JQ1 downregulated the total expression of PI3K but had no effect on the total expression of AKT or mTOR. However, JQ1 significantly downregulated the phosphorylation of AKT and TOR in a time- and dose-dependent manner.

The effect of JQ1 and OTX015 on the cell cycle in EC cells was evaluated by flow cytometry and western blot. Flow cytometry showed that both of JQ1 and OTX015 suppressed the cell cycle in G1 phase in Ishikawa cells (Fig. 6A and C) but not HEC-1A cells (Additional file 1: Fig. S1). Treatment with different concentrations of JQ1 increased the share of Ishikawa cells in G1 phase from 47.03 ± 2.06 to 62.96 ± 2.70%. As for OTX015, the rate of G1 phase in the control group was 35.32 ± 3.14%, whereas, the rate rose to 58.45 ± 6.83% in OTX015 group of high concentrations. To evaluate the underlying mechanism, we analyzed cell cycle signaling pathways by western blot. JQ1 suppressed the expression of CDC25A, CDK4, and CyclinD1 and phosphorylation of Rb. By contrast, JQ1 upregulated P21 expression (Fig. 6B and D).

JQ1 has been reported to exert its antitumor effects by suppressing c-Myc expression in other solid tumors. We therefore examined the influence of JQ1 on BRD4 and c-Myc expression in EC cell lines. Western blot showed that JQ1 significantly reduced BRD4 and c-Myc protein expression (Fig. 7A and B). siRNA knockdown of BRD4 also inhibited c-Myc transcription and cell proliferation, suggesting that JQ1 reduces c-Myc expression through inhibition of BRD4 (Fig. 7C). siRNA knockdown of c-Myc in EC cell lines inhibited cell growth (Fig. 7D and E), but the inhibitory effect was weaker than that elicited by treatment with 2.5 µM JQ1, suggesting that JQ1 may have effects other than repression of c-Myc expression. Western blot assay showed that knockdown of BRD4 inhibited the PI3K/AKT/mTOR pathway, which was basically the same as JQ1 (Fig. 7F). Analysis of TCGA data using the online tool GEPIA (Gene Expression Profiling Interactive Analysis) showed that BRD4 expression is positively correlated with c-Myc expression in EC tissues (Fig. 7G).

We established Ishikawa-NC cells as well as Ishikawa-BRD4 overexpression cells. BRD4 protein level was significantly increased in stable cells with the overexpression vector (Fig. 8A). Compared with control cells, the cell viability rate was higher in Ishikawa-BRD4 overexpression cells following the treatment with different concentrations of JQ1 (Fig. 8B). Cell apoptotic rates mediated by JQ1 were partially reversed by forced overexpression of BRD4 (Fig. 8C). Our data demonstrated that the antitumor effect of JQ1 to some extent was BRD4-dependent in endometrial cancer cells.

To further validate the suppressive effect of JQ1 in vivo, we used Ishikawa cells to construct mouse xenograft models. The xenograft models were then administered JQ1 (50 mg/kg/d) or placebo intraperitoneally daily for 3 weeks. There was no obvious difference in body weight between the two groups (Fig. 9A). JQ1 administration clearly reduced EC tumor growth in  vivo. Compared with the placebo group, tumor volume and tumor weight were smaller in the JQ1-treated group (Fig. 9B and D). The difference in tumor volume between the JQ1-treated group and the placebo group became significant on day 12. Representative EC tumors are shown in Fig. 9C. At the end of the experiment, the mean tumor weight was 598.8 ± 328.4 mg in the placebo group (n = 5) but 316.0 ± 156.3 mg in the JQ1-treated group (n = 5) (P < 0.05). Moreover, routine blood counts showed that JQ1 had no obvious toxicity with respect to hematopoietic function (Fig. 9E). Taken together, these data suggest that JQ1 suppresses EC progression in vivo.