These authors contributed equally to this work.
Article first published online: 13 JAN 2012
Copyright © 2011 American Association for the Study of Liver Diseases
Volume 55, Issue 3, pages 910–920, March 2012
How to Cite
Liao, C.-H., Yeh, C.-T., Huang, Y.-H., Wu, S.-M., Chi, H.-C., Tsai, M.-M., Tsai, C.-Y., Liao, C.-J., Tseng, Y.-H., Lin, Y.-H., Chen, C.-Y., Chung, I.-H., Cheng, W.-L., Chen, W.-J. and Lin, K.-H. (2012), Dickkopf 4 positively regulated by the thyroid hormone receptor suppresses cell invasion in human hepatoma cells. Hepatology, 55: 910–920. doi: 10.1002/hep.24740
Potential conflict of interest: Nothing to report.
Supported by grants from Chang-Gung University, Taoyuan, Taiwan (CMRPD 34013, NMRP 140511) and from the National Science Council of the Republic of China (NSC 94-2320-B-182-052).
- Issue published online: 23 FEB 2012
- Article first published online: 13 JAN 2012
- Accepted manuscript online: 12 OCT 2011 12:50PM EST
- Manuscript Accepted: 27 SEP 2011
- Manuscript Received: 27 JAN 2011
Thyroid hormone (T3) mediates cellular growth, development, and differentiation by binding to the nuclear thyroid hormone receptor (TR). Recent studies suggest that long-term hypothyroidism is associated with human hepatocellular carcinoma (HCC) independent from other major HCC risk factors. Dickkopf (DKK) 4, a secreted protein, antagonizes the Wnt signal pathway. In this study, we demonstrate that T3 may play a suppressor role by inducing DKK4 expression in HCC cells at both the messenger RNA (mRNA) and protein levels. DKK4 was down-regulated in 67.5% of HCC cancerous tissues. The decrease in DKK4 levels was accompanied by a concomitant decrease in TR protein levels in the matched cancerous tissues in 31% of tissues compared by immunoblotting with the adjacent noncancerous tissues. Further, TR and DKK4 expression levels were positively correlated in both normal and cancerous specimens by tissue array analysis. In function assays, stable DKK4 transfected into J7 or HepG2 cells decreased cell invasion in vitro. Conversely, knocking down DKK4 restores cell invasiveness. DKK4-expressing J7 clones showed increased degradation of β-catenin, but down-regulation of CD44, cyclin D1, and c-Jun. To investigate the effect of DKK4 and TR on tumor growth in vivo, we established a xenograft of J7 cells in nude mice. J7-DKK4 and J7-TRα1 overexpressing mice, which displayed growth arrest, lower lung colony formation index, and smaller tumor size than in control mice, supporting an inhibitory role of DKK4 in tumor progression. Conclusion: Taken together, these data suggest that the TR/DKK4/Wnt/β-catenin cascade influences the proliferation and migration of hepatoma cells during the metastasis process and support a tumor suppressor role of the TR. (Hepatology 2012)
Thyroid hormone, 3,3′-5-triiodo-l-thyronine (T3), is a potent mediator of many physiological processes including embryonic development, cell differentiation, metabolism, and the regulation of cell proliferation.1, 2 The actions of T3 are mediated by nuclear thyroid hormone receptors (TRs). TRs are ligand-dependent transcription factors that comprise modular functional domains that mediate hormone binding (ligands), DNA binding, receptor homo- and heterodimerization, and interaction with other transcription factors and cofactors.3 TRs are derived from two genes, TRα and TRβ, located on human chromosomes 17 and 3, respectively. Transcripts of each of these genes undergo alternative promoter choice to generate TRα1 and TRα2 as well as TRβ1 and TRβ2 receptor isoforms.2–4
Using a complementary DNA (cDNA) microarray technique, we previously identified 148 genes that are positively regulated by T3 in a TRα1-overexpressing hepatoma cell line (HepG2-TRα1).5 Increasing evidence suggests that aberrant TR regulation or mutant TR genes may be associated with human neoplasia.6 Lin et al.7 reported truncated TRα1 and TRβ1 cDNA in 53% of human hepatocellular carcinomas (HCCs). Other groups8 have reported mutated TRs in HCC and cultured cells. However, an increasing number of studies have indicated that TR is a potent suppressor of tumorigenesis, invasiveness, and metastasis formation.9 This study focused on a set of genes (i.e., tumor suppressor genes) that are normally activated by the TR but are aberrantly repressed because of reduced TR expression or mutation during carcinogenesis.
The Dickkopf (DKK) family comprises secreted antagonists of Wnt signaling. Wnt/β-catenin signaling plays an important role in embryogenesis, tissue homeostasis, and tumor development.10 Wnt proteins participate in various types of cancer development and progression by binding to frizzled receptor and low density lipoprotein-receptor-related protein 5 and 6 (LRP5/6) and by signaling through the canonical and noncanonical Wnt pathways.11 The DKK family comprises four members, each of which possesses an N-terminal signal peptide and contains two conserved cysteine-rich domains separated by a linker region.12 The DKKs are frequently hypermethylated in gastrointestinal cancer, whereas knockdown of DKK4 enhances cell growth and invasiveness of esophageal cancer cells and colorectal cancer cells.13, 14 DKK1 and 3 are widely studied members of the DKK family. DKK3 is down-regulated in various cancers and its expression can inhibit cell proliferation.15 The expression pattern of DKK4 and its function are poorly understood in human HCC. Our study demonstrates that T3 up-regulates DKK4 expression in HepG2-TR cells. Our data show that DKK4 is expressed abundantly in noncancerous liver tissues and is down-regulated in cancerous tissues, and its role is elucidated.
Materials and Methods
Human HCC Specimens.
Clinical data for 117 HCC patients treated by total removal of liver tumors from January 1998 to December 2001 in Chang Gung Medical Center, Taiwan, were reviewed with the approval of the Institutional Review Board (IRB, No: 97-0234B) of Chang Gung Medical Center. All samples were frozen at −70°C immediately after surgical resection. The following clinicopathological data were retrospectively reviewed: gender, age, presence of liver cirrhosis, alcohol usage, Edmondson's histologic grade, microvascular invasion, macrovascular invasion, presence of tumor capsule, number of tumors, largest tumor size, presence of ascites upon surgery, α-fetoprotein (AFP), albumin, bilirubin, prothrombin time, creatinine, aspartate aminotransferase (AST), alanine aminotransferase, hepatitis B surface antigen (HBsAg), antibody against hepatitis C virus (anti-HCV), date of surgical resection, date of tumor recurrence, and date of last follow-up or HCC-related death. Of these patients, 90 were male and 27 were female, and their mean age was 55.7 years (range, 21-89 years); 75, 19, and 9 were positive for HBsAg, anti-HCV, and both viral markers, respectively, whereas 14 patients were negative for both markers.
Immunohistochemistry and Tissue Microarray.
Formalin-fixed and paraffin-embedded tissues from the lungs of SCID mice were examined by hematoxylin and eosin (H&E) staining. Tumor tissue microarrays were constructed with 40 formalin-fixed liver tissues and 40 hepatocellular carcinoma samples (US Biomax, Rockville, MD). The intensity of DKK4 staining was evaluated according to the following criteria: strongly positive (scored as 3+), dark brown staining in >30% of liver tissue completely obscuring the cytoplasm; weakly positive (scored as 1+), less brown staining in the HCC cytoplasm; and absent (scored as 0), no appreciable staining in tumor cells.
Statistical analyses were carried out using means standard deviations, one-way analysis of variance (ANOVA), and Tukey's honestly significant difference post-hoc test.
T3 Affects DKK4 Expression at the Messenger RNA (mRNA) and Protein Levels in HCC Cells.
The HCC cell lines used in this study were HepG2-TRα1, -TRβ1, -Neo, and J7-TRα1. The TR protein was overexpressed by 19- and 10-fold, respectively, compared with the control cell line (Fig. 1A). The DKK4 mRNA levels increased by 15.2- to 35-fold following incubation of HepG2-TR cells with 10 nM T3 for 48 hours (Fig. 1B,C). However, DKK4 was marginally induced by T3 in non-TR-expressing cell lines (-Neo, Fig. 1D).
The effect of TRs on DKK4 protein expression was assessed in HepG2 and J7 isogenic cell lines incubated for 12 and 24 hours in medium containing various levels of T3 (Fig. 2). DKK4 protein level increased by 5- to 9-fold following incubation of HepG2-TR cells with 1-10 nM T3 for 24 hours. These results demonstrate that the effect of T3 on the DKK4 protein level in TRα1- and TRβ1-overexpressing HepG2 cells is time- and TR-dependent. T3 significantly increased the level of DKK4 in the HepG2-TR stable cell lines in comparison with that in the HepG2-Neo control cell line (Fig. 2A-C). Similarly, T3 also induces DKK4 protein in J7-TRα1 cells (Fig. 2D).
T3 Induces DKK4 Protein In Vivo.
To determine the in vivo response of the DKK4 gene to T3 treatment, we thyroidectomized male Sprague-Dawley rats and divided them into four groups (n = 5/group) as described16 (Supporting Materials and Methods). Hypo- (Tx), hyper- (Tx+ T3; sham+T3), and euthyroid (sham) rats were established successfully (data not shown). Immunoblot analyses demonstrated that liver DKK4 protein levels were elevated in the T3-treated groups 7-fold in the sham + T3 group and 7.3-fold in the Tx + T3 group compared with the Tx group (Supporting Fig. 1).
DKK4 Is Down-Regulated in Human HCC.
To examine the expression levels of DKK4 in HCC and assess the correlation between TRs and DKK, we analyzed tissues and tissue microarrays by immunoblotting or immunostaining. A total of 117 consecutive patients with HCC were submitted for this study. An equal amount of protein (100 μg) from each specimen was resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and analyzed by immunoblotting. DKK4 protein was detected in most of the noncancerous tissues. However, DKK4 was down-regulated in 67.5% (79 of 117) of HCC cancerous tissues relative to the matched adjacent noncancerous tissues. Further, the decrease in DKK4 levels was accompanied by a concomitant decrease in TRα1/TRβ1 levels in the matched cancerous tissues in 31% (35 of 113) of tissues compared with the adjacent noncancerous tissues. The correlation between TR and DKK4 expression was analyzed. By using DKK4 T/N ratio as a dependent variable, linear regression analysis showed a positive correlation with either the TRα1 T/N ratio (regression coefficient = 0.437; 95% confidence interval [CI], 0.159-0.714; P = 0.002) or TRβ1 T/N ratio (regression coefficient = 0.343; 95% CI, 0.087-0.600; P = 0.009). The results from 12 representative paired-HCC specimens are shown in Fig. 3A.
In some HCC specimens the amount of protein obtained was insufficient to allow simultaneous immunoblotting analyses of correlations between DKK4 and TR proteins in normal liver and HCC. Therefore, we also performed immunohistochemical staining of tissue arrays containing tissue sections from 40 HCC and 40 liver tissues (normal) using a polyclonal antibody specific to DKK4. DKK4 staining intensity was classified on a 0-to-3 scale: 0, negative; 1+, weak; 1+ to 2+, moderate; 3+, strong. Of the 40 HCC samples, 12 (30%) were negative, 10 (25%) stained weakly, 14 (35%) stained moderately, and 4 (10%) stained strongly. In the 40 normal liver tissues, 13 (32.5%) stained moderately and 27 (67.5%) stained strongly. DKK4 staining was observed mainly in the cytoplasm of normal cells but was barely detected in tumor tissues (Fig. 3B). Three representative paired specimens are shown in Fig. 3C. The expression of TR was also analyzed. Of the 40 HCC samples, 15 (37.5%) were negative, 10 (25%) stained weakly, 12 (30%) stained moderately, and 3 (7.5%) stained strongly. In the 40 normal liver tissues, 24 (60%) stained moderately and 16 (40%) stained strongly.
The correlation between TR and DKK4 expression was analyzed. Because of the nonparametric nature of these data, Spearman's rank correlations were also calculated. TR and DKK4 expression levels were positively correlated in both normal tissues (Pearson correlation coefficient = 0.517, P = 0.001; Spearman's rank correlation coefficient = 0.464, P = 0.003) and cancerous tissues (Pearson correlation coefficient = 0.530, P < 0.001; Spearman's rank correlation coefficient = 0.553, P < 0.001). Regression analyses were performed to identify the clinical factors associated with the tumor/normal (T/N) ratio of DKK4 in HCC after surgical resection. The T/N ratio of DKK4 expression in HCC samples correlated with tumor size (beta = −0.050; 95% CI = −0.095 to −0.006; P = 0.027), histological grade (beta = −0.703; 95% CI = −1.024 to −0.382; P < 0.001), and liver cirrhosis (beta = 0.546; 95% CI = 0.119 to 0.972; P = 0.013). Patients were dichotomized into two groups based on higher and lower T/N ratios of DKK4 expression. Kaplan-Meier analyses showed that patients with a T/N ratio of DKK4 expression >0.75 had a longer disease-free survival than did those with a lower T/N ratio. The mean disease-free survival periods for higher and lower T/N groups were 52.3 months (95% CI = 37.4-67.3 months) and 28.7 months (95% CI = 18.8-38.6 months), respectively (P = 0.010) (Fig. 3D). Additionally, the mean overall survival period was longer in the higher-expression group (110.1 months; 95% CI = 97.5-112.7 months) than in the lower-expression group (78.5 months; 95% CI = 64.2-92.9 months; P = 0.008) (Fig. 3E). Figure 3D,E illustrates the cumulative survival curves of patients subgrouped according to lower and higher expression of DKK4 in HCC tissues.
DKK4 Overexpression Suppresses Cell Invasion.
DKK4 is a Wnt antagonist protein involved in the Wnt signaling pathway.17 To explore the biological role of DKK4 in hepatoma cells, we established a DKK4-overexpressing cell line and control cell line to examine whether DKK4 also participates in cell migration and invasion. Immunoblot analysis showed J7-DKK4 cells secreted more DKK4 into the culture medium than did J7-control cells (Fig. 4A). The overexpression of DKK4 was significantly decreased cellular proliferation compared with J7-control cells (Fig. 4B). To test the effect of the DKK4 gene on cell invasiveness was measured in vitro using Matrigel Transwell invasion assays. Further, the invasiveness of J7-DKK4#1 and #2 cells was inhibited by 75%-80% in J7 cells (Fig. 4C). Images of cell density were shown for two control and two overexpressing cell lines (Fig. 4C). The quantified results are shown in Supporting Fig. 2A.
Overexpression of DKK4 Affects the Wnt/β-Catenin Signaling Pathway.
To determine the effect of DKK4 on the Wnt-canonical signaling pathway, the expression of several proteins involved in this pathway was measured in J7 cells. β-Catenin was significantly down-regulated by 35%-40% in the two DKK4-overexpressing cell lines compared with control cell lines. In contrast, phosphorylated β-catenin protein was up-regulated by almost 1.6-fold in the two DKK4-overexpressing cell lines. The expression levels of CD44, cyclin D1, and c-Jun significantly decreased in DKK4-overexpressing cell lines in immunoblot analysis (Fig. 4D; Supporting Fig. 2B). This is consistent with a previous study indicating that c-Jun is a target gene of the Wnt/β-catenin pathway in human colorectal carcinomas.18
Overexpression of DKK4 decreased the activity of pro-matrix metallopeptidase (pro-MMP)-9 (92 kDa) and pro-MMP-2 (72 kDa) by 65%-70% and 18%-30% in J7 cells, respectively (Fig. 4E; Supporting Fig. 2C). To confirm the effect of DKK4 on β-catenin-ubiquitin complex formation, we performed immunoprecipitation assays. The data indicate the DKK4-mediated ubiquitination of β-catenin is involved in the degradation of β-catenin (Fig. 4F).
Effect of TR and DKK4 Reduces Tumor Growth and Lung Colony Formation In Vivo.
To investigate the effect of DKK4 and TR on tumor growth in vivo, we established a xenograft of J7 cells in BALB/c nude mice. Three J7 isogenic cell lines, J7-control, J7-DKK4, and J7-TRα1 were established. To verify the levels of expression of the DKK4 and TR proteins in the three J7 cell lines, cells were incubated with 1 nM T3 for 24 hours. The levels of the DKK4 protein were increased in J7-DKK4 and J7-TRα1 cells compared with J7-control cells (Fig. 5A). The three J7 cell lines were injected subcutaneously and monitored continuously for 21 days. Figure 4B shows the average tumor volume observed in each of the three groups (n = 4). J7-DKK4 and J7-TRα1-induced tumors grew significantly slower than did control tumors. On average, after 21 days, tumors detected in mice injected with J7-DKK4 and J7-TRα1 cells were 45%-90% smaller compared with the tumors observed in control mice (Fig. 5B). To determine whether the in vitro results (Fig. 4C) could be reproduced in vivo, we investigated the effect of DKK4 on lung colony-forming ability in SCID mice (Fig. 5C). Mice were intravenously injected with an equal number(1 × 106) of three stable J7 cell lines. A reduction in lung colony formation index was observed in animals injected with J7-DKK4 and J7-TRα1 cells compared with those injected with J7-control cells. The arrowheads indicate the lung colonies. All lines of SCID mice developed multiple macroscopic tumor nodules in the lung, as shown by H&E staining (Fig. 5C). However, the average lung colony formation index and tumor size were reduced 75%-95% in animals injected with J7-DKK4 and J7-TRα1 cells compared with those injected with J7-control cells (Fig. 5D).
Expression of TRβ1 and DKK4 Reduces Invasiveness In Vitro.
Similar results were observed using HepG2-TRβ1 stable cell lines using an in vitro assay (Fig. 6). Briefly, three HepG2 isogenic cell lines (HepG2-control, HepG2-DKK4, and HepG2-TRβ1) were established. Expression levels of TR and DKK4 proteins in those three cell lines were determined. The DKK4 protein was increased both in the HepG2-DKK4 and HepG2-TRβ1 cell lines (Fig. 6A). We verified the effect of DKK4-overexpression in HepG2-DKK4 and HepG2-TRβ1 cells, showing that invasiveness was inhibited by 60%-70% in both cell lines (Fig. 6B,C). Images of cell density for three cell lines are shown (Fig. 6B). An examination of the expression of downstream Wnt signaling molecules in HepG2-DKK4 and HepG2-TRβ1 cell lines showed that β-catenin, cyclin D1, and c-Jun were significantly decreased in HepG2-DKK4 and HepG2-TRβ1 compared with control cell lines (Fig. 6D; Supporting Fig. 3).
Knocking Down DKK4 Enhances Cell Invasion.
To examine whether invasion ability can be restored by knocking down DKK4, we established HepG2-TRβ1 knockdown (KD) cell lines expressing short hairpin RNA (shRNA) against DKK4 (TRβ1-DKK4-KD1 and 2) and a control cell line expressing scrambled shRNA (TRβ1-Scr). To verify the expression levels of both DKK4 and TR protein in the three HepG2-TRβ1 cells, we incubated the cell lines with 1 nM T3 for 24 hours to induce DKK4 (Fig. 7A). DKK4 protein level in TRβ1-DKK4-KD1 and TRβ1-DKK4-KD2 cells were decreased to 0.28- and 0.2-fold those in HepG2-TRβ1-Scr cells, respectively (Fig. 7A). The invasivity of HepG2-TRβ1-DKK4-KD cells was increased by 3- to 3.4-fold compared with HepG2-TRβ1-Scr cells (Fig. 7B; Supporting Fig. 4A). Images of cell density are shown (Fig. 7B). To determine the effect of DKK4 on the Wnt-canonical signaling pathway, we measured the expression of several proteins involved in this pathway in HepG2-TRβ1 cells. β-Catenin was significantly up-regulated (by 1.2- to 2.1-fold) in the two DKK4-KD cell lines compared with control cells. Similarly, cyclin D1 and c-Jun proteins were significantly up-regulated by 1.35- to 1.9-fold in the two DKK4-KD cell lines (Fig. 7C; Supporting Fig. 4B). Zymography assays revealed that these increases were associated with a 1.4- to 1.55-fold increase in MMP2 activity in TRβ1-DKK4-KD cells (Fig. 7D; Supporting Fig. 5C). These results indicate that TRs may act as tumor suppressors in hepatoma cells by inducing the expression of DKK4 and reducing signaling through the Wnt/β-catenin pathway.
In this study we attempted to elucidate the molecular mechanism responsible for DKK4 regulation by T3 in isogenic HepG2 cell lines and its physiological significance. DKK4 was down-regulated in 67.5% of HCC cancerous tissues. Furthermore, DKK4 levels were decreased concomitantly with TRα1/TRβ1 levels in 29.3% of the matched cancerous tissues. To investigate further the role of the β-catenin pathway in cell growth and metastasis of hepatoma cells, we overexpressed DKK4 in J7 cells to antagonize Wnt signaling. Overexpression of DKK4 led to increased β-catenin degradation, which decreased CD44, cyclin D1, and c-Jun expression and inhibited the cell growth and migration of J7-DKK4 cells. Previous reports demonstrated that β-catenin activation can control both hepatocyte growth and survival.19, 20 Activation of the β-catenin pathway appears to provide a potent proliferative and invasive advantage in a mouse model of accelerated liver carcinogenesis.21 The proto-oncogene c-Jun involved cellular progression, proliferation, and survival in cancer development.22 CD44 is overexpressed in many cancers, including colorectal carcinomas, and it promotes cell adhesion, migration, and invasion in breast cancer.23 Increasing DKK4 expression may influence the growth and migration of hepatoma cells. Ectopic expression of DKK4 leads to cell growth arrest and inhibition of cell migration both in vitro and in vivo.
In contrast, Baehs et al.24 demonstrated that DKK4 is a potent inhibitor of TCF-dependent signaling and growth in colorectal cancer cells. Moreover, DKK4 expression can be restored in colorectal cancer cell lines by treatment with trichostatin A.25 Our study showed that the endogenous DKK4 protein was not detectable in hepatoma cells (Figs. 4A or 6A), but was restored by TSA treatment (data not shown), which is consistent with the report of Baehs et al. Consequently, up-regulation of DKK4 may provide a native feedback loop for inhibition of the Wnt/β-catenin pathway in colon cancer.
Matsui et al. and Hirata et al.26, 27 reported that DKK4 was up-regulated in human colorectal cancer and renal cell carcinoma, respectively. Addition of recombinant human DKK4 protein decreased Wnt-canonical pathway activity in the human embryonic kidney HEK-293 cells, but not in colon cancer cell lines.26 These authors concluded that DKK4 acts as an inhibitor of the Wnt-canonical signaling pathway in nontumor cells. However, either loss of the adenomatosis polyposis coli (APC) gene or a mutation in β-catenin is frequently found in human colorectal cancer, an observation that explains why DKK4 is not an inhibitor in tumor cells. Hirata et al.27 also reported that DKK4 mRNA was up-regulated in renal cancer tissues compared with matched adjacent noncancerous tissues. In addition, DKK4 can activate the noncanonical c-Jun N-terminal kinase (JNK) signaling pathway while inhibiting the Wnt-canonical pathway in human renal cell carcinoma.27 To determine the expression of proteins involved in the noncanonical JNK signaling pathway, we used three HCC cell lines (J7, SK-Hep-1, HepG2) in which DKK4 overexpression had been established (data not shown). Our data showed that JNK and phospho-JNK levels were not significantly different in clones stably expressing DKK4 compared to control cells (data not shown), indicating that DKK4 did not activate the JNK-dependent noncanonical pathway in hepatoma cells. Thus, our study demonstrated a dual role for DKK4 in different tissues or cell types.
Recent studies have shown that TRs may function as tumor suppressors.9, 28, 29 These studies suggest that a partial loss of normal TR function caused by a decrease in the expression or complete loss of normal TR activity (due to mutation and/or aberrant expression) provides an opportunity for tumors to proliferate, metastasize, and invade other tissues. Consistent with our results, Martinez-Iglesias et al.9 defined a novel role for TR as a metastasis-suppressor gene, showing that it acts by suppressing activation of extracellular signal-regulated kinase and phosphatidylinositol 3-kinase signaling pathways. Further, that TRs/T3 exhibited anticarcinogenic or antitumoral effects have been reported on the carcinogen-induced rat HCC model through induction of a differentiation program of preneoplastic hepatocytes.30–32 TR can strongly inhibit the transcriptional response signaling pathway in HCC or breast cancer cells.33, 34 The same group further reported that an important functional role of endogenous corepressors in TRβ1-mediated suppression of ras-induced transformation and tumorigenesis.34 Consistent with previous reports16, 35 PTTG1 and Nm23-H1 are involved in liver regeneration and are overexpressed in HCC. However, PTTG1 and Nm23-H1 are also negatively regulated by T3. We found that DKK4 is up-regulated by the TR and suppresses cell invasion in human hepatoma cells. Alternatively, there is evidence suggesting that a mutated TR (mTR) can cause the development of HCC in transgenic mice, and that TR is involved in tumor development and progression.28, 36–38 Lin et al.39 and Chan and Privalsky38 reported that mTRs isolated from HCC tumors or cell lines aberrantly interacted with either corepressors or coactivators. Chan and Privalsky further reported that mTRs could alter their DNA recognition activity8 or target gene repertoire,40 thereby serving a regulatory role as a transcriptional sensor.8 Recent studies suggest that long-term hypothyroidism is associated with HCC, independent of other major HCC risk factors.41–43 TR may play a suppressor role by reducing PTTG1 and Nm23-H1 expression in the normal liver, which may increase the expression of DKK4. However, reducing TR expression in HCC causes TRs to lose their tumor-suppressor role during hepatocarcinogenesis.
In conclusion, we have shown for the first time that DKK4 is up-regulated by T3. We found that DKK4 expression is TR-dependent in some HCCs and might play a crucial role in the development of HCC. Our data suggest that the TR/DKK4/Wnt/β-catenin cascade inhibits the metastasis of hepatoma cells. TR may play a suppressor role by increasing DKK4 expression and reducing the Wnt/β-catenin signaling pathway in hepatoma cells.
- 2Thyroid hormone dependent regulation of target genes and their physiological significance. Chang Gung Med J 2008; 31: 325-334., , .
Additional Supporting Information may be found in the online version of this article.
|HEP_24740_sm_SuppFig1.tif||5202K||Supporting Information Figure 1|
|HEP_24740_sm_SuppFig2.tif||2295K||Supporting Information Figure 2|
|HEP_24740_sm_SuppFig3.tif||2624K||Supporting Information Figure 3|
|HEP_24740_sm_SuppFig4.tif||5078K||Supporting Information Figure 4|
Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.