Activation of the RhoB Signaling Pathway by Thyroid Hormone Receptor β in Thyroid Cancer Cells

Thyroid hormone receptor (TR) mediates the crucial effects of the thyroid hormone (T3) on cellular growth, development, and differentiation. Decreased expression or inactivating somatic mutations of TRs have been found in human cancers of the liver, breast, lung, and thyroid. The mechanisms of TR-associated carcinogenesis are still not clear. To establish the function of TRβ in thyroid cancer cell proliferation, we constructed a recombinant adenovirus vector, AdTRβ, which expresses human TRβ1 cDNA. Thyroid cancer cell lines in which TRβ protein levels were significantly decreased as compared to intact thyroid tissues were infected with AdTRβ and the function of TRβ on cell proliferation and migration was analyzed. Ligand-bound TRβ induced HDAC1 and HDAC3 dissociation from, and histone acetylation associated with the RhoB promoter and enhanced the expression of RhoB mRNA and protein. In AdTRβ-infected cells, T3 and farnesyl transferase inhibitor (FTI)-treatment induced the distribution of RhoB on the cell membrane and enhanced the abundance of active GTP-bound RhoB. This RhoB protein led to p21-associated cell-cycle arrest in the G0/G1 phase, following inhibition of cell proliferation and invasion. Conversely, lowering cellular RhoB by small interfering RNA knockdown in AdTRβ-infected cells led to downregulation of p21 and inhibited cell-cycle arrest. The growth of BHP18-21v tumor xenografts in vivo was significantly inhibited by AdTRβ injection with FTIs-treatment, as compared to control virus-injected tumors. This novel signaling pathway triggered by ligand-bound TRβ provides insight into possible mechanisms of proliferation and invasion of thyroid cancer and may provide new therapeutic targets for thyroid cancers.


Introduction
Thyroid hormone receptors (TRs) are ligand-dependent transcription factors that mediate the actions of the thyroid hormone (T3) in cellular development, growth and differentiation. Two human TR genes, THRA and THRB that are located on different chromosomes, encode T3-binding isoforms (TRa1, b1, b2, and b3) that are expressed in a tissue-and development-dependent manner [1]. Over the past decades, significant advances have been made in the understanding of TR actions in maintaining normal cellular function. However, the roles of TRs in human cancer are not well understood. The reduced expression of TRs because of hypermethylation or deletion of TR genes in human cancers suggests that TRs could function as tumor suppressors [2]. A close association of somatic mutations of TRs with thyroid cancers further supports the notion that the loss of normal functions of TR could lead to uncontrolled growth and loss of cell differentiation [3].
To understand the functional consequences of ligand-bound TRb effects on downstream signaling pathways in thyroid cancer cells, we focused on RhoB that is a member of the Ras superfamily of isoprenylated small GTPases, which regulate actin stress fibers and vesicle transport [4]. Other RhoGTPases, which include RhoA and RhoC, promote oncogenesis, invasion, and metastasis [5]. In contrast, RhoB has antiproliferative and proapoptotic effects in cancer cells, and overexpression of RhoB can inhibit cell migration, invasion, and metastasis [6]. Membrane association of RhoB protein occurs through either geranylgeranylated (RhoB-GG) or farnesylated modifications. RhoB responds to farnesyl transferase inhibitor (FTI)-treatment by a gain-of-function mechanism that is characterized by elevation of the RhoB-GG form that inhibits proliferation or apoptosis of cancer cells [7]. Thus, altered expression and activity of RhoB may be crucial for cancer progression and therapeutic responses.
In the present study, we explored the function of ligand-bound TRb in thyroid cancer cells. Ligand-bound TRb induced RhoB protein expression, leading to increased expression of p21 followed by decreased cell proliferation and motility. FTI-treatment enhanced these antiproliferative functions of ligand-bound TRb. Our results identify RhoB upregulation as a key step for targeting thyroid cancer cell proliferation and tumor progression. This novel signaling pathway triggered by ligand-bound TRb provides insight into possible proliferation and invasion mechanisms of thyroid cancer.

Materials and Methods
Cell culture BHP18-21 cells, which were reported by Ohta et al. [8], were provided as a gift by Dr. Jerome Hershman, UCLA. BHP18-21v cells, that express Pax-8, but do not express either the thyroglobulin or the thyroid transcription factor-1 gene, were isolated from BHP18-21v cells [9]. FRO and WRO cells, which were reported in citation [10,11] were kindly provided by Dr. Shunichi Yamashita, University of Nagasaki. These cell lines are not de novo cell lines but have already been reported [8,10,11] and were kindly provided as gifts by our collaborators. All cells were grown in RPMI 1640 medium with 10% (v/v) fetal bovine serum (FBS) in a humidified incubator under a 5% CO 2 atmosphere. DNA profiling of cancer cell lines was analyzed by Promega Japan (Tokyo, Japan) and is shown in Table 1. Thyroid hormone, triiodothyronine (T3) and the farnesyl transferase inhibitor (FTI), FTI-277, were purchased from Sigma Aldrich (St. Louis, MO). The cells were incubated with resin-stripped (T 3 -depleted) FBS [12] and were then infected with 30 MOI of AdTRb or control AdLacZ. The cells were incubated in medium with or without 30 nM of T3 or 5 mM of FTI.

Construction of the recombinant adenoviral vectors
AdTRb is a DE1-DE3 recombinant adenovirus that expresses the human TRb gene [13] under the control of the immediate early promoter of cytomegalovirus (CMV). Construction of the AdTRb virus has been previously described [13]. AdLacZ, which contains the CMV promoter-controlled lacZ gene, was purchased from Quantum Biotechnologies (Montréal, Canada) and was used as a control. AdTRbPV is a recombinant adenoviral vector that expresses human TRbPV, which is a dominant negative mutant of TRb [14], under the control of the cytomegalovirus immediate early promoter. The FLAG-TRbPV plasmid [15] was used as the template for cloning TRbPV into pShuttle2 (Clontech, Mountain View, CA) using the polymerase chain reaction. The PCR primers were: FLAG-ApaI-59 (AAGGGCCCGCCGCCATGGACTACAAAGACGATGACGAC), and TRbPV-39 KpnI (AATGGTACCTTCAGTCTAATCCTCGAACACTTCC) in which the underlines indicate restriction sites. An adenovirus vector was then constructed using the Adeno X Expressing System (Clontech) according to the manufacturer's protocol. Recombinant adenoviruses were plaque-purified, harvested 48 h after infection of 293 cells, and purified by double cesium chloride gradient ultracentrifugation [16]. Viral titers were determined by plaque assays with cultured 293 cells [16] and an adenovirus titration kit (Clontech).

Plasmid construction and luciferase assays
The RhoB reporter plasmid 2726/+86 RhoB-Luc plasmid was kindly provided by Dr. Dimitris Kardassis (University of Crete Medical School) [17]. Transient cotransfections were carried out in triplicate. The Renilla luciferase plasmid was cotransfected for normalization. Plasmid transfections were performed in AdTRb or control virus-infected thyroid cancer cells using Lipofectamine 2000 (life technologies, Grand Island, NY). Twenty-four hours after transfection, the cells were treated with or without T3 for an additional 24 h. Dual luciferase assays were carried out according to the manufacturer's instructions (Promega, Madison, WI).

ChIP assay
ChIP assays were performed using a simple ChIP enzymatic chromatin immunoprecipitation kit (Cell Signaling Technology, Danvers, MA). AdTRbinfected cells (5610 7 ) that were treated with or without T3 were fixed with 1% formaldehyde and digested chromatin was immunoprecipitated with anti-histone 3 (positive control), normal rabbit IgG (negative control), anti-acetyl histone 3 (Lys 9), anti-histone deacetylase (HDAC)1 or anti-HDAC3 antibodies. Five hundred ml of diluted chromatin were immunoprecipitated with 5 mg of each antibody and protein G magnetic beads according to the manufacturer's instructions. Quantitative real-time (RT)-PCR was performed using a SYBR green real-time PCR kit (Life Technologies). The primer sequences used for RT-PCR were: forward 59-GCAGCAGCAGCGCAGACT-39 and reverse 59-ACTCGGCCTAGCTCTCTC-39.

Quantitative real-time reverse transcriptase PCR
Total RNA was extracted from cells by using an RNeasy mini kit (Qiagen, Valencia, CA) according to the manufacturer's instructions. After quantification by spectrophotometry, 5 mg of total RNA was reverse transcribed to obtain cDNA by using 160 mM deoxynucleotide triphosphate, 50 ng of random hexamer primers and 200 units of SuperScript II according to the manufacturer's recommendations (Life Technologies). TaqMan probes for human RhoB, TRb, and 18S rRNA were purchased from Life Technologies. mRNA levels were determined by real-time RT-PCR as previously described by Furuya et al. [18].

Western blot analysis
This study was approved by the ethics committee of the University of Yamanashi. Normal thyroid tissues from the opposite lobe of carcinomas were obtained by surgery after the patients gave written informed consent. The documents regarding informed consent, which were approved by the ethics committee, were kept in the university hospital storage cabinet. Normal tissue from 4 of the patients was pooled for the Western blotting experiment. Tissues were frozen in liquid nitrogen immediately after resection. Protein lysates of normal thyroid tissue and of cancer cell lines were prepared by using cell lysis buffer (Cell Signaling Technology) according to the manufacturer's instructions. Membrane protein was purified from AdTRb-infected BHP18-21v cells by using a membrane protein purification kit (GE healthcare Life Science, Pittsburgh, PA) according to

Immunoprecipitation analysis
To analyze the protein expression of TRb in cells, 500 mg of protein lysate from thyroid tissue, HepG2 or virus-transfected thyroid cancer cell lines was immunoprecipitated with 5 mg of the C3 mouse monoclonal antibody against the C terminus of TRb (Santa Cruz Biotechnology) or with 5 mg of mouse IgG as indicated in the text by using the Dynabeads protein G immunoprecipitation kit (Life Technologies) according to the manufacturer's protocol. Western blot analysis was performed using an anti-TRb antibody (Santa Cruz Biotechnology) against an N-terminal TRb peptide or an anti-TRa antibody (Santa Cruz Biotechnology) against an N-terminal TRa peptide. The cell lysates of AdTRbinfected cells were immunoprecipitated with Rhotekin RBD tagged agarose beads and the abundance of GTP-bound RhoB was analyzed by pull-down assay, by using RhoB activation assay kit (Cell Biolabs, Inc. San Diego, CA).

Cell-cycle analysis
The cells were stained with 20 mg/ml of propidium iodide (Life Technologies) and were analyzed using a BD Biosciences FACS Calibur flow cytometer (Franklin Lakes, NJ) and Cell Quest software version 3.0. The percentage of cells in the G 1 , S, or G 2 /M phase was calculated by using ModFitLT version 3.1.

Cell proliferation and invasion assays
The number of cells was measured by using a nonradioactive cell proliferation assay (Cell Counting Kit-8; Dojindo, Kumamoto, Japan) as previously described by Furuya et al [19]. Cellular invasion was measured using the Cytoselect Cell Invasion assay kit (Cell Biolabs, Inc.), according to the manufacturer's instructions. Cell culture medium with 10% FBS was used as a chemoattractant in the lower well of the chamber. After rehydration of the basement membrane, thyroid cancer cell lines were seeded in the upper compartment of the chamber in serum-free medium (1610 4 cells per well) with T3 or FTI-treatment. After incubation at 37˚C in 5% CO 2 for 24 h, the non-invading cells were removed from the upper surface, and the cells that had invaded the membrane (8 mm pore size) to reach the lower surface were stained with CyQuant GR Fluorescent Dye. Their fluorescence was analyzed at 480 nm using a fluorescence plate reader.

Xenografts of BHP18-21v cells and vector injection
BHP18-21v cells (1610 7 ) were transplanted into the abdominal subcutaneous tissue of 8-week-old male nude mice (BALB/C nu/nu). Three weeks later, when the tumors had reached approximately 1 cm in diameter, mice were randomly assigned to the experimental or control group (6 mice per group). Five610 8 plaque-forming units of AdTRb or AdLacZ in 100 ml PBS were injected into the tumors. Adenovirus administration was repeated every 7 days. At each time point, all mice were also intraperitoneally injected with 100 mg/kg/body weight of FTI-277. On day 30, mice were euthanized by CO 2 inhalation. Tumors which were removed from the mice were fixed in 10% buffered formalin and embedded in paraffin. Sections were permeabilized with 0.2% Triton X-100 in PBS for 10 min at room temperature. Serum free T3 and TSH levels were determined by using the free T3 ELISA kit (Phoenix Pharmaceuticals, Inc, CA) and a TSH assay kit (Uscn Life Science Inc, Wuhan, China), respectively, according to the manufacturers' instructions. The institutional biosafety committee of the University of Yamanashi approved the animal procedures and the use of recombinant DNA.

Statistics
Data are expressed as means ¡ S.D. Statistical analysis was performed using oneway analysis of variance or the unpaired 2-tailed Student's t-test, and probability values of less than 0.05 were considered to be statistically significant.

Results
To explore TRb function in thyroid cancer cell proliferation, we focused on analysis of RhoB, which is a target of TRb and is expressed at very low levels in 130 human cancer cell lines [20]. For the present study, we used the 3 thyroid cancer cell lines, BHP18-21v, FRO and WRO, which were infected with 30 MOI of AdTRb or AdLacZ. We first analyzed TRb protein expression in these cells. Five hundred mg of whole cell lysate protein was immunoprecipitated with a monoclonal antibody that recognizes the TR C-terminal region, followed by Western blot analysis with an antibody against an N-terminal TRb peptide. As shown in Fig. 1A, TRb expression was clearly observed in the positive controls; intact thyroid tissue (lane 1) and HepG2 cells that express functional TR [21] (lane 2), and also in AdTRb-infected BHP18-21v, FRO and WRO cells (lanes [6][7][8]. No TRb protein expression was observed in AdLacz-infected BHP18-21v, FRO or WRO cells (lanes [3][4][5] or in HepG2 cells that were immunoprecipitated with a mouse monoclonal IgG antibody as a negative control (lane 9). Tubulin expression in the protein lysate (10 mg) was also analyzed as a loading control (Fig. 1B). We also analyzed TRa expression in these cell lysates by Western blot analysis of the immunoprecipitated proteins with an antibody against an Nterminal TRa peptide. TRa expression was observed in the positive controls; thyroid tissue and HepG2 (lanes 1 and 2, respectively). No TRa protein expression was observed in BHP18-21v, FRO or WRO cells that were infected with AdLacZ (lanes [3][4][5] or AdTRb (lanes 6-8). Down-regulation of PPARc expression and its activity is considered to be one mechanism of thyroid carcinogenesis [22]. We therefore also analyzed the protein expression of PPARc in BHP18-21v, FRO or WRO cells (Fig. 1C). PPARc expression was decreased in these cancer cells compared with that of intact thyroid tissue but infection with AdTRb had no effect on the expression of PPARc in thyroid cancer cells. These data indicate that the three AdTRb or AdLacZ infected cell lines tested are a good model for analysis of TRb function. We next analyzed the effect of T3 treatment of AdTRb-or AdLacZ-infected thyroid cancer cell lines on RhoB expression ( Fig. 2A-C). T3-treatment (30 nM) of AdTRb-infected (lanes 1-4) but not of AdLacZ-infected (lanes 5-7) BHP18-21v, FRO and WRO cells for 6, 12 or 24 h significantly enhanced the expression of RhoB mRNA in a time-dependent manner compared to non-treated non-virally infected cells. Furthermore, Western blotting analysis indicated that T3-treatment for 12 or 24 h induced RhoB protein expression in AdTRb-infected BHP18-21v, FRO and WRO cells but not in AdLacZ-infected cells (Fig. 2D-F, lanes 2-4 and lanes 5-7 respectively). Thus, induction of RhoB mRNA and protein expression are T3/AdTRb-dependent in thyroid cancer cell lines.
To clarify the mechanisms through which ligand-bound TRb enhanced the expression of RhoB in AdTRb-infected BHP18-21v, FRO and WRO cells, we analyzed the effect of T3 treatment of these cell lines on the transcriptional activity of the RhoB promoter. AdTRb-or control AdLacZ-infected cells were therefore transiently transfected with the RhoB promoter-luciferase reporter construct, 2726/+86 RhoB-Luc. Luciferase activity driven by the RhoB promoter was significantly enhanced in AdTRb-infected cells, but not in AdLacZ-infected cells following T3-treatment ( Fig. 3A-C).
Transcription of the RhoB promoter is inhibited by transcriptional repressors that recruit histone deacetylases (HDACs) to remove the acetylation of lysine residues on histones3 (H3) tails [23]. It is well established that the acetylation status of the tail domains of histones3 (H3) change during transcriptional activation of the RhoB promoter [24]. To determine if induction of RhoB gene expression by ligand-bound TRb correlates with changes in the acetylation status of H3 in the promoter region of the RhoB gene, chromatin immunoprecipitation (ChIP) assays were performed. AdTRb-infected BHP18-21v, FRO or WRO cells were incubated with or without T3 for 24 h and ChIP assays were performed using antibodies to H3, acetylated H3, HDAC1 and HDAC3. As shown in Fig. 3D-F, H3 was found in fragments of the RhoB promoter (lane 3) in AdTRbinfected cells with or without T3-treatment. However, HDAC1 and HDAC3 were associated with fragments of the RhoB promoter only when the cells were incubated without T3, and not when the cells were incubated with T3 (HDAC1/3; lanes 5 and 6, respectively). Consistent with these results, no acetylated H3 was associated with the promoter in the absence of T3 treatment, whereas in the presence of T3, the H3 that was associated with the promoter was highly acetylated (lane 4). These results indicated that ligand-bound TRb induced RhoB transcription via alteration of the histone acetylation status of thyroid cancer cells.
We next analyzed the effect of ligand-bound TRb on the cellular localization and function of RhoB. For this purpose we employed the agent FTI. FTI treatment of cells induces the membrane association of RhoB by modification of RhoB through the addition of lipid moieties [4]. Membrane association of RhoB is important for its activity. We first determined the effect of 5 mM of FTI treatment on RhoB protein expression in AdTRb-infected BHP18-21v cells that were treated with or without T3 (30 nM). Western blotting of whole cell lysates of treated/ untreated AdTRb-infected BHP18-21v cells with RhoB antibody (Fig. 4A) indicated that FTI-treatment did not alter T3 induction of RhoB expression (lanes 2 and 4) and did not enhance RhoB protein expression in AdTRb-infected BHP18-21v cells in the absence of T3-treatment (lane 3).
We then analyzed the effect of T3 treatment with/without FTI co-treatment on RhoB activity. Like other small GTPases, RhoB regulates molecular events by cycling between an inactive GDP-bound form and an active GTP-bound form. In the active GTP-bound form, RhoB specifically binds to the Rho-binding domain (RBD) of Rhotekin to regulate downstream signaling cascades. We therefore analyzed the abundance of GTP-bound RhoB in cell lysates of AdTRb-infected BHP18-21v cells that were treated with or without T3 and/or FTI by immunoprecipitation of cell lysates with Rhotekin RBD-tagged agarose beads followed by immunoblotting for RhoB (Fig. 4B). These pull-down assays indicated that not only the abundance of RhoB but also the abundance of the active GTP-bound form of RhoB in AdTRb-infected BHP18-21v cells was

). No GTP-bound RhoB was detected in FTI-treated cells without T3-treatment (lane 3). These results indicated that the ligand-bound
AdTRb-induced RhoB protein in BHP18-21v cells was an active kinase whose activity was enhanced by co-treatment with FTI.
To determine if T3 and FTI could induce membrane association of RhoB in AdTRb-infected BHP18-21v cells, cell surface proteins were biotinylated. Biotinylated proteins in cell lysates were then pulled down with streptavidin beads and analyzed by immunoblotting with an anti-RhoB antibody (Fig. 4C). T3treatment induced the expression of RhoB at the cell membrane (lane 2), which was greatly enhanced by co-treatment with FTI (lane 4). FTI-treatment did not enhance RhoB protein expression at the cell membrane of AdTRb-infected BHP18-21v cells in the absence of T3-treatment (lane 3). These results indicated that T3-induced membrane association of RhoB that was enhanced by FTI cotreatment and were consistent with T3 induction of RhoB activity in AdTRbinfected BHP18-21v cells.
The cyclin-dependent kinase (CDK) inhibitor, p21, is a negative regulator of cell-cycle progression and is one of the downstream targets of RhoB. To confirm the activity of RhoB in AdTRb-infected BHP18-21v cells treated with T3 and/or FTI, we analyzed the protein expression of p21 by Western blot analysis (Fig. 5A-C). p21 expression was analyzed in cells transfected with RhoB-targeted or control siRNA to confirm the RhoB dependence of p21changes. Three thyroid cancer cells were pre-cultured in T3-free medium with stripped serum, as described in ''Materials and Methods''. FTI-treatment did not enhance p21 protein expression in siRNA-control AdTRb-infected cells without T3-treatment (lane 2). In AdTRbinfected cells, the expression level of p21 was increased in cells treated with T3 (lane 3) and was greatly enhanced by co-treatment with FTI (lane 4) compared to non-treated cells (lane 1). However, p21 was not induced by T3 plus FTI treatment of AdTRb-infected cells 24 h after transfection with RhoB-targeted siRNA (lane 5). p21 inhibits the activities of CDKs, leading to hypo- phosphorylation of retinoblastoma protein (Rb), which in turn causes cell-cycle arrest at the G0/G1 phase [25]. We analyzed the phosphorylation of Rb (p-Rb) in AdTRb-infected thyroid cancer cells treated with T3 and/or FTI [ (Fig. 5 (A-C)). Rb was phosphorylated in AdTRb-infected BHP18-21v, FRO or WRO cells with or without T3-treatment (lanes 1 and 3). FTI-treatment alone did not inhibit the phosphorylation of Rb in AdTRb-infected cells (lane 2). Rb phosphorylation was significantly decreased in AdTRb-infected cells by co-treatment with T3 and FTI (lane 4). On the other hand, inhibition of Rb phosphorylation was not observed by T3 plus FTI treatment of AdTRb-infected cells after transfection with RhoBtargeted siRNA (lane 5).
We further studied the cell cycle stages of these cells using flow cytometry (Fig. 5D-F). In si-control AdTRb-infected cells, treatment with T3 plus FTI induced a significant increase in the G0/G1 phase population compared to nontreated or T3-treated cells, whereas in RhoB-targeted siRNA-transfected cells, T3+ FTI treatment had no effect on the cell population in the G0/G1 phase. Furthermore, treatment with T3 plus FTI reduced the number of cells in the G2/ M phase in si-control-but not in RhoB-targeted siRNA-AdTRb-infected cells.
These results indicate that the enhanced RhoB expression in AdTRb-infected cancer cells that were treated with T3 and FTI, was accompanied by increased expression of p21 and inhibition of cell-cycle progression. To confirm that the observed changes in p21 expression reflected changes in cell proliferation, we next analyzed the effects of TRb on cancer cell proliferation using the MTT assay. For this assay, BHP18-21v, FRO or WRO cells were infected with 30 MOI of AdTRb and, after 12 h of incubation in adenovirus-containing medium, the cells were cultured with or without T3 (30 nM) and 5 mM of FTI for 24 or 72 h. After 24 and 72 h of incubation cell number had increased in the absence of T3-treatment but was decreased in the presence of T3 and was significantly decreased in the presence of T3 plus FTI (Fig. 6A-C).
As a further measure of RhoB activity, we next analyzed cell invasion as RhoB has been reported to negatively affect cell migration. We analyzed the effect of T3 and FTI treatment on the invasive properties of AdTRb transfected BHP18-21v, FRO or WRO cells using a cell invasion assay chamber. The cancer cells that invaded through the 8 mm pores of the dividing membrane to the lower chamber were quantified using fluorescence measurements. T3 plus FTI treatment significantly inhibited cell invasion of BHP18-21v, FRO and WRO cells compared the control group (80%, 71% and 85% reduction respectively; Fig. 6D-F). FTI treatment alone had no significant effect on migration. The combined results indicated that ligand-bound TRb induced the expression of RhoB which acts as a critical mediator of the cellular response to FTI and inhibits thyroid cancer cell proliferation and migration.
To evaluate the efficiency of AdTRb in inhibition of cancer cell growth in vivo, 1610 7 BHP18-21v cells were injected into the left flanks of nude mice. The BHP18-21v cell line was selected for these experiments because these cells formed large tumors in nude mice, compared with FRO and WRO cell lines (data not shown). One week after injection, the cells formed small tumors at the injection sites. The tumors had grown to 1 cm in a diameter by 3 weeks after injection. At this point (day 1), as well as on days 7, 14 and 21, AdTRb, AdLacZ, or AdTRbPV, which expresses a ligand-binding domain mutant of TRb, was injected into the tumors (5610 8 PFU in 100 ml PBS). At the same time points all mice were intraperitoneally injected with 100 mg/kg/body weight of FTI. To analyze the effects of a hyperthyroid state, AdTRb-or AdLacZ-infected mice were intraperitoneally injected with or without T3 (250 mg/kg/body weight) for 3 days. In the hyperthyroidism group, adenovirus injection into the tumors was repeated on days 7, 14 and 21 following T3-treatment. Tumor volumes were calculated at various time points from caliper measurements, as described in Materials and Methods. In AdTRb-infected mice, there was a period of continued growth (5 days) followed by a decrease in tumor size. As shown in Fig. 7A, after 13 days the mean tumor volume in the AdTRb-treated groups with or without T3-treatment was significantly less than that of the control group (p,0.01). The final tumor volumes of AdTRb-, AdLacZ-, and AdTRbPV-treated mice with or without T3treatment on day 22 were 136¡30.1 vs. 142¡41.7, 4898¡265 vs. 5087¡334, and 4786¡727 vs. 5081¡421 mm 3 , respectively. The final tumor volumes of AdTRbtreated mice were significantly smaller than those of mice that were infected with AdLacZ or AdTRbPV. In AdTRb-infected mice without T3 treatment, serum free T3 and TSH concentrations were 1.84¡0.39 and 2.16¡0.76, respectively ( Fig. 7B). There were no differences in these values between AdLacZ-and AdTRbPV-treated mice. The free T3 or TSH levels of these mice were almost within the euthyroid range as indicated by a previous report [26]. In contrast, in T3-treated mice the concentration of free T3 was significantly elevated and serum TSH levels were suppressed compared with euthyroid mice. These results indicated that ligand-bound TRb inhibits thyroid cancer cell growth and that the T3 level of euthyroid mice is sufficient for induction of TRb-associated tumor growth suppression.
Immunohistochemical analysis indicated that ligand-bound TRb induced the expression of RhoB in the cytoplasm (Fig. 7C) of AdTRb-treated tumors in the euthyroid state, whereas RhoB expression was not observed in AdLacZ-treated tumors. Ligand-bound TRb-induced p21 expression was clearly observed in the nucleus of AdTRb-infected tumors, but p21 expression was not observed in AdLacZ-treated tumors. Immunohistochemical analysis of the proliferation marker Ki-67 in xenograft tissue sections showed that the fraction of proliferating cells is lower in AdTRb-treated tumors versus AdLacZ-treated tumors ( Fig. 7D; b and a respectively). Quantification of proliferation indicated that 14.0% ¡4.2% and 92.0% ¡4.8% of the cells were proliferating in AdTRb-and AdLacZ-treated tumors, respectively (p,0.01) (Fig. 7E). Furthermore, cleaved caspase 3 expression was induced in AdTRb-treated tumors but not in AdLacZ-treated tumors (Fig. 7F). Thus, TRb overexpression reduced the proliferation and induced the apoptosis of human thyroid carcinoma cells xenografted in mice and strongly inhibited tumor growth.

Discussion
The present study focused on the functional characterization of ligand-bound TRb in human cancer cells. We provide evidence that ligand-bound TRb activates the RhoB signaling pathway and inhibits cancer cell proliferation both in vitro and in vivo. Previous reports of poorly differentiated thyroid cancers in knock-in mice with a dominant-negative TRb mutation strongly suggest the involvement of TRb in carcinogenesis [2,3,27]. Our findings support the hypothesis that TRb acts as a tumor suppressor in several types of cancer.
RhoB differs from the other GTPases RhoA and RhoC in that it is postulated to be a tumor suppressor because its expression is decreased in a number of tumor cell types [4]. Indeed, RhoB-null mice have increased susceptibility to skin tumor carcinogenesis [28]. Marlow et al. reported that reactivation of suppressed RhoB is a critical step for inhibition of the proliferation of anaplastic thyroid cancers [29]. In the current study, we analyzed 3 thyroid cancer cell lines, BHP18-21v; papillary thyroid carcinoma, FRO; anaplastic thyroid carcinoma, and WRO; follicular thyroid carcinoma [11] that had lost the expression of endogenous TR and in which RhoB protein expression was also not observed by Western blotting analysis. These findings suggested the possibility that endogenous TR, when present, might function as a tumor suppressor by inducing RhoB protein expression.
Indeed TR has been associated with transcription regulation of a number of genes via control of histone acetylation. Thus, unliganded TR binds with the nuclear receptor corepressor (NCoR) that recruits histone deacetylase (HDAC)containing complexes to the promoter regions of genes [30]. These HDACs deacetylate histones and repress promoter-activity [31,32]. On the other hand, an increasing number of coactivators have been implicated in T3-dependent gene activation including SRC1 [33], CBP and p300 [34]. These coactivators have been shown to possess intrinsic histone acetyltransferase (HAT) activity. Acetylation of core histones in chromatin has long been proposed to facilitate transcription. Our results showing T3/TRb dependent RhoB promoter transcription from a reporter plasmid, and T3/TRb dependent regulation of Histone H3 acetylation on the RhoB promoter, are consistent with these data. Our data are also consistent with other reports regarding a role for HDACs in the regulation of RhoB expression. Thus, HDAC1 represses RhoB expression in lung cancer cells and treatment with the HDAC inhibitor, trapoxin A enhanced the transcription and expression of RhoB mRNAs [23]. The T3-dependent increased levels of RhoB mRNA, protein, and promoter activity observed in our study, coupled with the ChiP assay data, suggested that the RhoB gene was transcriptionally regulated by the T3-bound TRb via HDAC inhibition. Our findings therefore establish RhoB as a direct transcriptional target of liganded-TRb in thyroid cancer cell. Although previous reports indicated that FTI-treatment enhances the expression of RhoB in gastric cancer [35] or breast cancer [36] cells, in the present study, FTI-treatment did not enhance the expression of RhoB in BHP18-21v, FRO, or WRO cells. These differences may be caused by the HDAC status of the RhoB promoter region of these cancer cells.
In our study, T3-treatment of AdTRb-infected BHP18-21v cells enhanced the expression of RhoB both in the cytoplasm and on the cell surface and enhanced the active GTP-bound form of RhoB. Furthermore, concomitant FTI-treatment greatly enhanced the expression of surface RhoB and of the active, GTP-bound form of RhoB. These data are consistent with the known localization of RhoB and the effect of FTI. Thus RhoB is known to be posttranslationally modified by the addition of lipid moieties (geranyl-geranyl or farnesyl groups), which enable RhoB to interact with components of the cell membrane [4]. RhoB localizes to both the plasma membrane and the membrane of endosomes [37] and has specific functions in endosomal trafficking. RhoB responds to FTI-treatment by a gain-offunction mechanism that is characterized by elevation of the geranyl-geranylated isoform of RhoB that inhibits the proliferation of cancer cells [4]. The combined data suggest that the liganded-TRb-induced repression of cell proliferation that we observed might be mediated by localization of the induced RhoB to cellular membranes, from where it exerts its anti-proliferative activity.
Our results indicated that the ligand-bound TRb signaling pathway led to tumor growth inhibition, not only in vitro but more importantly in vivo in a tumor xenograft model. Biochemical analysis in vitro suggested that this inhibition occurred via a RhoB-dependent mechanism that is upstream of p21. The p21 protein is a major player in cell-cycle control. Once activated, p21 exerts a negative effect on cell-cycle progression by preventing formation of the CDK2/ cyclin E complex. Although p53 plays a key role in p21 regulation, induction of p21 expression can also occur in a p53-independent manner [38,39]. The effect of liganded TRb mediated-inhibition of cell-cycle progression and upregulation of RhoB and p21, occurred in cancer cells in which p53 function is disrupted [10]. Therefore, in our experimental model, p53 is unlikely to promote TRb-induced p21 expression. Furthermore, when the activation of RhoB that was induced by T3/TRb was inhibited by siRNA, the induction of p21 was not observed in these thyroid cancer cell lines. In AdTRb-infected cancer cells treated with T3, ligandbound TRb enhanced the expression of p21, while Rb protein was still phosphorylated. On the other hand, FTI and T3 treatment of AdTRb-infected cancer cells significantly enhanced the expression of p21 leading to hypophosphorylation of Rb, which in turn caused cell cycle arrest at the G0/G1 phase. Consistent with these findings, treatment of AdTRb-infected cells with T3 alone did not enhance the number of cells in the G0/G1 phase whereas treatment with T3 and FTI did enhance the number of cells in the G0/G1 phase. Our results therefore identify RhoB upregulation as a key step for inhibition of thyroid cancer cell proliferation and therefore tumor progression via activation of p21.
The thyroid gland, which is a relatively common site for the development of malignant neoplasms, gives rise to 90% of all endocrine cancers. Thyroid cancer generally has a good prognosis. However, poorly differentiated or anaplastic thyroid carcinoma, which constitutes about 5%-14% of all thyroid carcinomas, is highly malignant and has a median survival time of only 2-6 months; anaplastic thyroid carcinoma rapidly invades adjacent structures and metastasizes throughout the body, especially to the lung [40,41]. Since no effective therapy is available for these aggressive types of thyroid carcinoma, novel therapeutic strategies including gene therapy are urgently needed. Our findings indicate that the TRbinduced pathways acted in concert to delay tumor progression and block metastatic spread. We therefore speculate that this TRb-induced suppression of cancer cell growth may confer therapeutic effects on poorly differentiated and anaplastic thyroid carcinomas and that the underlying biochemical pathway may provide novel therapeutic targets for these cancers.