TR4 nuclear receptor enhances the cisplatin chemo-sensitivity via altering the ATF3 expression to better suppress HCC cell growth

Early studies indicated that TR4 nuclear receptor (TR4) may play a key role to modulate the prostate cancer progression, its potential linkage to liver cancer progression, however, remains unclear. Here we found that higher TR4 expression in hepatocellular carcinoma (HCC) cells might enhance the efficacy of cisplatin chemotherapy to better suppress the HCC progression. Knocking down TR4 with TR4-siRNA in HCC Huh7 and Hep3B cells increased cisplatin chemotherapy resistance and overexpression of TR4 with TR4-cDNA in HCC LM3 and SNU387 cells increased cisplatin chemotherapy sensitivity. Mechanism dissection found that TR4 might function through altering the ATF3 expression at the transcriptional level to enhance the cisplatin chemotherapy sensitivity, and interrupting ATF3 expression via ATF3-siRNA reversed TR4-enhanced cisplatin chemotherapy sensitivity in HCC cells. The in vivo HCC mouse model using xenografted HCC LM3 cells also confirmed in vitro cell lines data showing TR4 enhanced the cisplatin chemotherapy sensitivity. Together, these results provided a new potential therapeutic approach via altering the TR4-ATF3 signals to increase the efficacy of cisplatin to better suppress the HCC progression.


INTRODUCTION
Hepatocellular carcinoma (HCC) is the fifth most common cancer in the world [1] and ranks third among the most lethal cancers worldwide [2]. HCC is an aggressive cancer with a high mortality rate and many cases were diagnosed at a late stage when curative therapies are not easily to obtain [3]. Chemotherapy with several cytotoxic agents, such as cisplatin, doxorubicin and 5-florouracil, are widely used to treat advanced HCC, but chemo-resistance is often observed in the majority of the patients [4,5]. Thus, development of better chemotherapy strategies is urgently needed.
The TR4 nuclear receptor 4 (TR4) belongs to the nuclear receptor superfamily and was first cloned from human prostate and testis cDNA libraries [6]. Early studies suggested that TR4 might play important roles to alter several key signaling pathways via interacting with selective nuclear receptors including thyroid receptor, androgen receptor, retinoic acid receptor/retinoid X receptor, and estrogen receptor [7][8][9]. TR4 could influence oxidative stress-and ionizing radiation-induced damage [10], and altered TR4 led to change the chemo-resistance of PCa stem/progenitor cells [11].
Here we investigated the potential TR4 potential roles in the HCC progression, especially its impacts on the cisplatin chemotherapy sensitivity. The results suggest that TR4 may function through altering the ATF3 expression to enhance the cisplatin chemotherapy sensitivity to suppress the growth of HCC cells. www.impactjournals.com/oncotarget

TR4 expression is higher in HCC than surrounding normal liver cells in HCC patients
To study roles of TR4 in HCC progression, we first examined the TR4 expression in HCC patients liver samples, and results revealed higher TR4 mRNA expression in HCC as compared to surrounding normal liver cells in 14 HCC patients ( Figure 1A). We also found higher TR4 protein expression in HCC in 10 of 12 patients analyzed using western blot analysis ( Figure 1B) or IHC staining ( Figure 1C).
Together, results from Figure 1A-1C reveal that TR4 expression at both mRNA and protein levels is higher in HCC than surrounding normal liver cells, suggesting TR4 expression may be linked to the HCC development.

Higher expression of TR4 mRNA and protein in HCC cell lines correlate with greater cell chemosensitivity
We first examined the TR4 expression in various HCC cell lines and found TR4 expression was higher in Hep3B and Huh7 cells and lower in LM3 and SNU387 cells (Figure 2A, 2B). We then studied differential expression of TR4 impacts on altering the cell viability upon chemotherapy. We found adding cisplatin, the current used chemotherapy drug to treat HCC [12], suppressed HCC cells using MTS assays ( Figure 2B). Importantly, we found the cell viability was much higher in LM3 and SNU387 cells than in Huh7 and Hep3B cells ( Figure 2C), suggesting higher TR4 expression in HCC cells may be able to increase cisplatin chemotherapy sensitivity to better suppress HCC cells.

TR4 knockdown led to reduced chemosensitivity in Huh7 and Hep3B cells
To further confirm the above conclusion, we first knocked-down TR4 expression via TR4-shRNA in Huh7 cells ( Figure 3A, mRNA level and protein level), and then treated these cells with cisplatin and applied MTS assay to analyze the cytotoxicity of these cells. We found that Huh7 cells have less sensitivity to cisplatin treatment in the TR4 knocked-down (Huh7-shTR4) cells compared with the scrambled control (Huh7-scr) cells ( Figure 3B, 3C). Similar results were obtained when we replaced Huh7-shTR4 cells with Hep3B-shTR4 cells ( Figure 3D-3F). Similar results were obtained when we used another knocked-down TR4 plasimid (Supplementary Figure S1).

TR4 over-expression led to enhanced chemosensitivity in LM3 and SNU387 cell
We then applied an opposite approach to overexpress TR4 via adding TR4-cDNA in LM3 and SNU387 cells ( Figure 4A, mRNA level and protein level), and treated these cells with cisplatin and used MTS assay to analyze the cytotoxicity of these cells. The results revealed that adding TR4 increased LM3 cell sensitivity to cisplatin treatment ( Figure 4B, 4C). Similar results were obtained when we replaced LM3-wpi TR4 cells with SNU387-wpi TR4 cells ( Figure 4D-4F).

TR4 expression alters cell apoptosis of HCC cells treated with cisplatin
We also performed apoptosis assays using Annexin V-FITC/PI double staining in both TR4 knockeddown cells and TR4 overexpressed cells. The results revealed that in TR4 knocked-down cells (Huh7-shTR4), cisplatin-induced apoptotic death is significantly less than scramble control (Huh7-scramble) ( Figure 5A). In contrast, the apoptotic death was significantly more in TR4-overexpressed cells (LM3-wpiTR4) than vector control (LM3-wpi) ( Figure 5B). These results demonstrate that TR4 can enhance the chemo-sensitivity of cisplatin possibly by promoting cell apoptosis.

Mechanism dissection how TR4 alters the cisplatin chemo-sensitivity to HCC cells
Overwhelming evidence indicates that the ATF3 gene can be induced rapidly by a variety of stress stimuli in different cell types possibly via altering cell apoptosis [13][14][15][16][17][18][19][20][21]. Germain et al [22] also reported that cisplatin might suppress tumor growth via altering the ATF3 expression and knocked-down ATF3 might lead to attenuate the cisplatin-induced cytotoxicity in murine embryonic fibroblasts.
We first found that ATF3 expression was induced by cisplatin treatment and knocked-down TR4 significantly reduced its expression in the Huh7-shTR4 cells as compared with Huh7-scr cells ( Figure 6A). Cell apoptosis using cleaved-PARP also found that cisplatin-induced ATF3 is higher in Huh7-scr ( Figure 6A). As expected, overexpressing TR4 in LM3-wpiTR4 increased ATF3 expression as compared with LM3-wpi cells ( Figure 6B).
We then applied neutralization/interruption approaches to confirm if TR4 is required to alter the ATF3 signaling to modulate the cisplatin chemo-sensitivity in HCC cells. As shown in ( Figure 6C), we found interrupting ATF3 expression in the LM3-wpiTR4 cells reversed the overexpressed TR4-enhanced cisplatin chemo-sensitivity in HCC LM3 cells, suggesting that TR4 might function through altering the ATF3 expression to enhance cisplatin chemo-sensitivity in HCC LM3 cells. Similar results were obtained when we replaced LM3 cells with SNU387 cells ( Figure 6D).
Since the major biological function comes from transcriptionally regulating downstream genes, we assumed TR4 could directly bind to the promoter region of ATF3 to influence its transcription. We first demonstrated that TR4 altered the ATF3 expression at the transcriptional level in Huh7 cells and LM3 cells ( Figure 6E, 6F). We analyzed the sequence of ATF3 promoter region seeking for potential TR4-response-element (TR4RE) based on a previous publication [23] and identified a putative TR4RE with the sequence of 5′-AGGTCAGGATGATGGAA-3′. We then applied the ChIP assay and found TR4 could bind to this TR4RE in vivo ( Figure 6G). Finally, we constructed the reporter gene with ATF3 promoter linked to luciferase reporter and assayed its transcriptional activity after altering the TR4 expression in Huh7 cells and LM3 cells ( Figure 6H, 6I). The results proved the transcription activation of ATF3 was induced by TR4 ( Figure 6E-6I).

TR4 alters the cisplatin chemo-sensitivity in HCC mouse model
To confirm all above in vitro cell lines data in vivo, we generated the subcutaneous xenograft HCC mouse models by transplanting either LM3-wpi or LM3-wpiTR4 cells into nude mice. The nude mice were divided into two groups: LM3-wpi tumor group and LM3-wpiTR4 group. Both groups were further divided into two subgroups: control subgroup and treatment subgroup with 0.5 mg/kg body weight cisplatin (as known as CDDP) i.p injection. As shown in Figure 7A, significant inhibition of HCC growth was found in the two cisplatin treatment subgroups, and TR4 overexpression in the cisplatin treatment subgroup had better suppression of tumor growth.
IHC staining confirmed the different TR4 expression in the two groups ( Figure 7B), and the LM3-wpiTR4 subgroup with cisplatin injection clearly had better suppression of the LM3 tumors ( Figure 7C-7D).

DISCUSSION
The efficacy of chemotherapy drugs varies greatly among individuals due to the heterogeneity of cancer biology, and many attempts have been developed to improve their chemo-sensitivity to better suppress the  Quantitation is shown at right. All assays were performed in triplicate (*P < 0.05, **P < 0.01, ***P < 0.001 ***P < 0.001, ns = not significant). tumor growth [24][25][26][27]. For the HCC, it was reported that RYBP might be able to alter the HCC conventional chemotherapy via inducing HCC cells apoptosis [28]. Targeting NF-kappaB/miRNA-21/PTEN signaling pathway has also been suggested to alter the chemoresistance [29].
Early studies indicated that TR4 might play cytoprotective roles [8,10,[30][31][32][33] that could function through altering the cell survival signals including the TR4-Oct4-IL1Ra axis [11]. Here we found a reverse cytotoxic phenotype in HCC cells that might function through promoting cisplatin-mediated apoptosis, suggesting that the roles of TR4 in mediation of the tumor progression and chemotherapy efficacy is complex, and may depend on cell context or cytotoxicity inducers.
cisplatin has been used in the chemotherapy to suppress the HCC progression [34,35]. Early studies indicated Cisplatin's anticancer activity was associated with its DNA damaging effects and induction of cancer cell apoptosis [36,37], that might involve altering various signal transduction pathways, including activating ATF3 signaling pathways [22,38]. Here we found that TR4 may alter the cisplatin chemotherapy efficacy through activating ATF3. By activating ATF3, TR4 promoted HCC cell apoptosis under chemo-treatment. Using luciferase assay and ChIP assay, we demonstrated that TR4 might function as a transcription factor via binding to the ATF3-TR4RE sequence that is located on its 5' promoter region to regulate ATF3 expression in liver cells.
Interestingly, like other nuclear receptors that could also be trans-activated via phosphorylation, acetylation, or sumoylation [40], TR4 could also be modulated by the metformin that functions through activation of AMPK to phosphorylate TR4 [41]. Furthermore, androgen receptor co-activator ARA55 could also suppress TR4 transactivation by increasing acetylation in the DNA-binding domain of TR4 [42].
Here we found that TR4 may alter the cisplatin chemotherapy efficacy through activating ATF3. Future research using these potential TR4 ligands/activators or any potential upstream signals to increase TR4 function to increase cisplatin chemotherapy efficacy may help us to better suppress the HCC progression.

Western blotting
Cells were harvested and washed twice with cold PBS, then resuspended and lysed in RIPA buffer (1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS, 10 ng/ml PMSF, 0.03% aprotinin, 1 μM sodium orthovanadate) at 4°C for 30 min. Lysates were centrifuged for 10 min at 14,000 × g and supernatants were stored at −80°C as whole cell extracts. Total protein concentrations were determined by Bradford assay. Proteins were separated on 12% SDS-PAGE gels and transferred to polyvinylidene difluoride membranes. Membranes were blocked with 5% BSA and incubated with the indicated primary antibodies. Corresponding horseradish peroxidaseconjugated secondary antibodies were used against each primary antibody. Proteins were detected using the chemiluminescent detection reagents.

Immunohistochemical staining (IHC)
We collected 90 tumor and normal samples from HCC patients at Sir Run-Run Shaw Hospital. IHC was then performed to evaluate TR4 expression in these samples. IHC was also performed in subcutaneous tumors of node mice to evaluate TR4 expression.

In vitro cytotoxicity assay
Stable transfected cells (5 × 10 3 ) were seeded on a 96-well plate with 3 replicate wells and allowed to incubate for 48 hr with the treatment of various concentrations of cisplatin or incubated for 72 hr with the indicated concentration of cisplatin and assessed every 24 hr. After incubation, cell viability was assessed utilizing the tetrazolium-based MTS colorimetric assay (CellTiter 96 cell proliferation assay kit; Promega, Madison, WI, USA) according to the manufacturer's instructions. All experiments were performed at least in triplicate on three separate occasions. Dose-response curves were plotted.

Apoptosis assay
Cell apoptosis was evaluated by flow cytometry (FCM) assay. Briefly, Huh7-scr and Huh7-shTR4 cells were harvested and washed twice with PBS, stained with Annexin VFITC and propidium iodide (PI) in the binding buffer, and detected by FACS Calibur FCM (BD, CA, USA) after 15 min incubation at room temperature in the dark. LM3-wpi and LM3-wpiTR4 cells were harvested and washed twice with PBS, stained with Annexin V and 7-AAD in the binding buffer. The total apoptotic cells were quantified.

Subcutaneous xenograft model
Animal studies were conducted using Female 5 week-old nude mice. Subcutaneous implantation was performed as previously described (23) where mice were injected subcutaneously with LM3-wpi or LM3-wpiTR4 cells. After a week, mice were randomized into two groups: cisplatin solution at the dose of 0.5 mg/kg in treatment group every four days nine times and DMSO control group in the same manner. The two perpendicular diameters (W and L) of tumors were recorded. The tumor volume (V) was calculated according to the formula: V = (W 2 × L)/2. The mice were euthanized after experiments, and tumor tissues excised and weighed. All animal experiments were performed humanely in compliance with guidelines reviewed by the Animal Ethics Committee of the Biological Resource Centre of the Agency for Science, Technology and Research at the Sir Run-Run Shaw Hospital.

Statistical analysis
All results are expressed as mean ± standard deviation (SD). Statistical analysis of the difference between treated and untreated groups was performed with Student's t-test. Values of P < 0.05 were considered as significant differences.