Epigenetic silencing of DACH1 induces loss of transforming growth factor-β1 antiproliferative response in human hepatocellular carcinoma


  • Potential conflict of interest: Nothing to report.

  • Supported by grants from the National Basic Research Program of China (973 Program Nos. 2012CB934002, 2010CB912802); National High-tech R&D Program of China (863 Program Nos. SS2012AA020314, SS2012AA020821, SS2012AA020303); National Key Scientific Instrument Special Programme of China (Grant No. 2011YQ03013405); National Science Foundation of China (Grant Nos. 81121004, 81071953, 81161120432, 81072169, 81172422, 81261120395).


Human dachshund homolog 1 (DACH1) is a major component of the Retinal Determination Gene Network (RDGN) and functions as a tumor suppressor. However, the regulation of DACH1 expression and its function in hepatocellular carcinoma (HCC) remain unclear. In this study, epigenetic changes of DACH1 were analyzed in HCC cell lines and primary cancers. We found that promoter region hypermethylation was correlated with loss or reduction of DACH1 expression, and restoration of DACH1 expression was induced by 5-aza-2′-deoxycytidine (5-AZA) in HCC cell lines. Promoter region methylation was found in 42% of primary HCC. Reduced expression of DACH1 was associated with poor differentiation of HCC nodules and higher serum aspartate aminotransferase/alanine aminotransferase ratio. DACH1 suppressed cellular growth by reactivating transforming growth factor beta (TGF-β) signaling. Ectopic expression of DACH1 enhanced chemosensitivity to 5-fluorouracil (5-FU) by inducing p21 expression in HCC cells. Conclusion: DACH1 is frequently methylated in HCC and DACH1 expression is regulated by promoter hypermethylation. Down-regulation of DACH1 is a novel mechanism for gaining resistance to the antiproliferative signaling of TGF-β1 and 5-FU resistance. (Hepatology 2013; 58:2012–2022)




American Joint Committee on Cancer


alanine aminotransferase


aspartate aminotransferase




complementary DNA


dachshund homolog 1


DNA methylation transferases




glyceraldehyde-3-phosphate dehydrogenase


hepatocellular carcinoma


in vitro methylated DNA




messenger RNA


methylation specific polymerase chain reaction


normal blood lymphocyte DNA


nontumor tissue


Retinal Determination Gene Network


reverse-transcription polymerase chain reaction


tumor tissue


transforming growth factor


transcription start site



In China, hepatocellular carcinoma (HCC) is the fourth most frequently diagnosed cancer and the second leading cause of cancer-related death.[1] Molecular mechanisms leading to the development of HCC are extremely complicated. After initial insult by the carcinogens, it takes several years for the accumulated genetic and epigenetic changes to result in HCC through loss of tumor suppressor function and amplification/mutation of cancer genes. Aberrant methylation and dysfunction of the signaling pathways regulating cytostasis and apoptosis in hepatocytes are also crucial in the molecular pathogenesis of HCC.[2]

The pleiotropic effects of transforming growth factor (TGF)-β1 allow it to function as a “Janus-faced” cytokine during carcinogenesis. HCC can occur when the tumor cells lose the tumor-suppressive responses to TGF-β due to somatic mutations in the TGF-β/Smad signaling region or by acquiring resistance to the antiproliferative response to TGF-β. The latter may also induce p53 and p16Ink4a independent inhibition of cellular proliferation. The down-regulation of c-Myc expression by TGF-β is lost in the epithelium and cancer cells. It is concomitant with the loss of the growth-inhibitory response to TGF-β.[3] The activated Smad3 is believed to mediate activation of various transcriptional factors in the cytoplasm by the TGF-β receptor. The Smad3-transcription factor complex moves into the nucleus to combine with Smad4 and recognize a composite Smad-E2F site on c-Myc for repression.[4] The genetic dysfunction due to aberrant methylation possibly acts by causing resistance to the antiproliferative effects of TGF-β/Smad signaling.[5-7]

Dachshund homolog 1 (DACH1) is widely expressed in the epithelial cells as a major component of the Retinal Determination Gene Network (RDGN). It regulates gene expression either as a cointegrator or through direct binding with DNA.[8] Reduced DACH1 expression predicts poor outcome in cancers of the breast and endometrium.[9] Reintroduction of DACH1 into breast cancer cells inhibits tumor initiation and metastasis in vivo.[10, 11] DACH1 is thus a tumor suppressor gene. However, the regulation of DACH1 expression and its function in human HCC have not been elucidated. In this study we first analyzed epigenetic changes of DACH1, and then studied its expression and function in HCC in vitro.

Materials and Methods

The study protocol was approved by the Ethics Committees at the Chinese PLA General Hospital, and informed consent was obtained from each patient. The study was carried out in accordance with the guidelines of the 1975 Declaration of Helsinki and was consistent with good clinical practice guidelines and local regulatory requirements.

Cell Culture, Treatment, Reporter Genes, Expression Vectors, Plasmid Construction, and DNA Transfection

Human liver cancer cell lines (PLC/PRF/5, SK-Hep1, HepG2, SMMC-7721, and Bel-7402) were maintained in high-glucose Dulbecco's modified Eagle's medium (DMEM; Gibco, Invitrogen, Melbourne, Australia) supplemented with 10% fetal bovine serum (FBS; Hyclone, Logan, UT). HCC cell lines were split to low density (30% confluence) 24 hours before treatment. Conditioned with 5-aza-2′-deoxycytidine (5-AZA) (Sigma-Aldrich, St. Louis, MO) at 2 μM, the growth medium was changed every 24 hours for a total 96-hour treatment. The cells were serum-starved for 36 hours for detection of TGF-β and then stimulated by TGF-β1 (Peprotech, Rocky Hill, NJ) at a concentration of 5 ng/mL for 1 hour to determine the activity of phosphorylated Smad2/3. The cells were further starved for 48 hours to investigate the expression of c-Myc, and for another 96 hours to perform the cell viability assay with or without pretreatment with 5-AZA. For chemosensitivity assay, the cells were treated with 5-fluorouracil (5-FU) (Sigma-Aldrich) of eight concentrations from a highest concentration of 500 μg/mL diluted serially 2-fold for 72 hours. The vector expressing DACH1 was a gift from Dr. Cvekl and the reporter plasmids SBE4-Luc were described previously.[12] The cells (2.5 × 103) were seeded in 96-well plates, incubated for 24 hours, and transfected with an appropriate combination of the reporter, expression plasmid, and control vector using Fugene HD (Roche, Indianapolis, IN). They were serum-starved for 36 hours and then stimulated with or without TGF-β1 for 12 hours before luciferase assay. The transfection efficiency was normalized by cotransfection with 2 ng of pRL-TK plasmid (Promega, Madison, WI) and subsequently measured with the Promega dual-GLO reporter assay system. DACH1 was subcloned into lentivirus expression vector and expression of DACH1 was determined by immunofluorescence staining with anti-DACH1 antibody.

Patients and Collection of Tissue Samples

Postoperative cases of HCC were randomly recruited between January 2010 and November 2011. None of these patients had received preoperative anticancer treatment. The tumor sample and adjacent tissue (i.e., 1 cm zone of peritumoral parenchyma) were collected in pairs[13] and tumor staging was determined according to the American Joint Committee on Cancer (AJCC) Cancer Staging Manual, 2010 (7th Ed.)[14] (Table 1). Normal hepatic tissue was obtained from the edge of resected hemangioma of the liver. Prior to molecular genetic analysis, the presence of neoplastic tissue was confirmed by histopathological examination of hematoxylin and eosin (H&E)-stained sections.

Table 1. Clinicopathologic Features of the Study Patients
FeaturesNo. of Patients%
Age, years  
Preoperative ALT, U/L  
Preoperative AST, U/L 
α-Fetoprotein, μg/L  
Liver cirrhosis  
Tumor diameter, cm  
Standard deviation4.5
Tumor differentiation  
Multicentric occurrence  
Intrahepatic metastasis  
UICC TNM stage  

Methylation-Specific Polymerase Chain Reaction (PCR) (MSP)

Genomic DNA from HCC cell lines and primary HCC tissues were prepared as described.[15] The following MSP primers were designed according to genomic sequences flanking the transcription start sites: DACH1-M-sense, 5′-GGA AAA AAT TAT TAG TTT TCG CGG AC-3′; DACH1-M-antisense, 5′-AAA CCG AAA ACA CAA AAA TAA CGA TCG-3′; DACH1-U-sense, 5′-TTT GGA AAA AAT TAT TAG TTT TTG TGG AT-3′ and DACH1-U-antisense, 5′-AAA AAA CCA AAA ACA CAA AAA TAA CAA TCA-3′. The primer sequences were oligo-synthesized (Invitrogen, Beijing, China) for MSP to detect bisulfate-induced changes affecting unmethylated (U) and methylated (M) alleles.

Methylation-Sensitive Restriction Enzyme Digestion and Real-Time Quantitative PCR (MSRED-qPCR)

The quantitative method MSRED-qPCR was employed to evaluate the degree of DACH1 promoter region methylation in HCC cell lines and patients' specimens as described[16, 17] with modification. In brief, each sample was subsequently divided into two equal aliquots. The first aliquot was digested with 10 U Hha I (Takara, Dalian, China) at 37°C overnight in a final reaction volume of 20 μL. To maintain the homogeneity of the digestion between the test and control samples, Hha I was replaced with 50% glycerol in the sham-treated control reaction. After incubation, digested samples were incubated at 65°C for 15 minutes to inactivate the enzyme. To ensure complete enzyme digestion, it was run in parallel with a positive and a negative control digestion in which 200 ng of completely methylated or unmethylated control DNA (EpiTect Control DNA, Qiagen, Valencia, CA) was digested. After digestion, the same amount of digested or undigested plasma DNA along with control digestion was subjected to quantitative PCR with the Lightcycler 480II Real-Time PCR Detection System (Roche). Each reaction was performed in a final volume of 20 μL containing digested (1 μL) or undigested (1 μL) DNA, 5 μM each primer and 1 × QuantiFast SYBR Green RT-PCR Kit (Qiagen). The PCR reaction was performed at 95°C for 30 seconds, followed by 40 cycles of 95°C for 5 seconds, 68°C for 60 seconds, and 74°C for 60 seconds. At the end of the PCR cycles, melting curve analyses were performed to validate the specific PCR product. Primers used for this qPCR were: Forward, 5′-CGTGAAGTGGGGGCTGTGTTTTCGT and Reverse, 5′-CAGAGACTCCGAGAGCGCGAGACAC. The degree of DACH1 promoter methylation was expressed as 2-(Ctdigestion- Ctsham). Each sample was run in duplicate for analysis. For 100% digestion efficiency, the relative expression level of completely unmethylated control DNA must be close to zero, whereas the level of completely methylated control must be 1.

RNA Isolation and Semiquantitative Reverse Transcription-PCR

Total RNA was isolated using Trizol reagent (Life Technologies, Gaithersburg, MD) and chloroform/isopropanol precipitation. Reverse transcription of RNA (5 μg) was carried out using random 6-mer primers included in the Superscript III-reverse transcriptase kit (Invitrogen, Carlsbad, CA). The primer sets for DACH1 gene were: 5′-CGT GAA CAA GCA GAA CAG ACG-3′; 5′-CCC ATG ACG AAT GTC TGA CTG-3′.

Immunohistochemistry (IHC)

The sections of neoplastic tissue were subjected to IHC for detecting DACH1. The staining was assessed by two independent pathologists blinded to the origin of samples. The staining intensity and extent of the stained area were graded according to the German semiquantitative scoring system: staining intensity of the nucleus, cytoplasm and/or membrane (no staining = 0; weak staining = 1; moderate staining = 2; strong staining = 3); the extent of stained cells (0% = 0, 1%-24% = 1, 25%-49% = 2, 50%-74% = 3, 75%-100% = 4). The final immunoreactive score (0 to 12) was determined by multiplying the intensity score with the extent of stained cells.[12]

Western Blotting

Soluble proteins were resolved on sodium dodecyl sulfate (SDS)-polyacrylamide gel and electrophoretically transferred to a polyvinylidene fluoride membrane (Millipore, Bedford, MA). The primary antibodies were: DACH1 (Abcam, Cambridge, UK); c-Myc, Smad2 (P459), and Smad3 (P209) (BioWorld, Minneapolis, MN); rabbit monoclonal antibodies against phospho-Smad3 (Ser423/425), phospho-Smad2 (Ser465/467), and p53 (Cell Signaling Technology, Danvers, MA); p21 (Santa Cruz Biotechnology, Santa Cruz, CA), and β-actin (Beyotime, Nanjing, China). The data were normalized to β-actin.

Cell Growth Assay

The cells (5 × 103) were seeded in 96-well plates, incubated for 24 hours, and treated as above. After treatment, the cell proliferation was determined by the Cell Counting Kit 8 (CCK-8) (Beyotime). Absorbance at 450 nm and 630 nm were measured with the EXL800 microimmunoanalyzer (BioTek, Burlington, VT). The results were plotted as means ± SD. The data are expressed as the percentage of viable cells: relative viability (%) = [A450-630 (treated) − A450-630 (blank)] / [A450-630 (control) − A450-630 (blank)] × 100%. IC50 was defined as the concentration required for 50% inhibition of cell growth. The values were calculated by nonlinear regression analysis using SPSS 11.5 (Chicago, IL). For cell proliferation assay, growth was measured daily by CCK-8.

Cell Cycle and Apoptosis Analysis

Cells were processed by standard methods using propidium iodide (Sigma-Aldrich) staining of cellular DNA. Each sample was analyzed by flow cytometry with a FACScan Flow Cytometer (Becton-Dickinson Biosciences, Mansfield, MA) using a 488 nm laser. Histograms were analyzed for cell cycle compartments using ModFit v. 2.0 (Verity Software House, Topsham, ME). Annexin-V (Keygen Biotech, Nanjing, China) staining was conducted as per the manufacturer's protocol.

Colony-Forming Assays

2.0 × 103 cells were plated in triplicate in 2 mL complete growth medium. After 2 weeks incubation, colonies more than 50 μm in diameter were counted using an Omnicon 3600 image analysis system. The colonies were visualized after staining with 0.04% crystal violet in methanol for 1-2 hours.

Statistical Methods

Analysis was performed with SPSS v. 11.5. Data collected from multiple independent experiments are presented as mean ± standard error using Student t distribution with a 95% confidence interval. Parameters from multiple groups such as different histopathological grades of carcinoma were compared using one-way analysis of variance (ANOVA) with Student-Newman-Keuls (SNK)-q test. P < 0.05 was considered statistically significant.


DACH1 Expression Is Down-regulated by Promoter Region Hypermethylation in HCC Cell Lines

As shown in Fig. 1A, the expression of DACH1 mRNA was lost in SK-Hep1, very weak in Bel-7402, SMMC-7721, and reduced in PLC/PRF/5 and HepG2. MSP primers were designed around the transcription start site in the CpG island of the DACH1 gene promoter region (Fig. 1B). MSP results are shown in Fig. 1C. Complete methylation of the promoter region was found in SK-Hep1 cells and partial methylation was observed in Bel-7402, SMMC-7721, PLC/PRF/5, and HepG2 cell lines. The methylation results were further evaluated by MSRED-qPCR (Fig. 1D). Reexpression or increased expression was found in these HCC cell lines after 5-AZA treatment (Fig. 1E). The above results indicate that DACH1 expression was regulated by promoter region hypermethylation.

Figure 1.

Down-regulation of DACH1 by promoter region hypermethylation in HCC cell lines. (A) The endogenous expression of DACH1 mRNA was detected by semiquantitative RT-PCR in HCC cell lines. (B) CpG island of the DACH1 gene locus was predicted by “CpG island plot” (http://www.ebi.ac.uk/Tools/emboss/cpgplot/). CpG sites between −1,000 and +1,000 bp from transcription start site are presented (vertical bars). Sequence of MSP products (−394 to −264 bp) is shown. (C) MSP results of DACH1 in human HCC cell lines. U and M lanes indicate unmethylated and methylated alleles, respectively; in vitro methylated DNA (IVD) and normal human peripheral lymphocytes (NL) served as the methylation and unmethylation controls, respectively. (D) The methylation status of HCC cell lines was checked by way of MSRED-qPCR. (E) HCC cell lines were treated with 5-AZA for 96 hours and DACH1 was detected by semiquantitative RT-PCR. GAPDH was used as internal control. (−), 5-AZA untreated; (+), 5-AZA treated.

Reduced DACH1 Expression Is Associated With DNA Methylation in Human Primary HCC

DNA methylation status was further evaluated by MSP and MSRED-qPCR in human primary HCC. Of 55 archival genomic DNA samples isolated from HCC patients, the MSP results demonstrated that in 23 cases (42%) methylation was detected in cancer tissues and in nine cases (16%) methylation was found in adjacent tissues (Fig. 2A). No methylation was found in two (0%) samples of normal liver tissue. The methylation status was further evaluated by MSRED-qPCR. As shown in Fig. 2B, similar to the MSP results, methylation was more frequent in cancer tissues than in adjacent tissues. On IHC, DACH1 staining was mainly seen in the cytoplasm of cancer cells (Fig. 2C). The DACH1 expression was significantly decreased in neoplastic cells as compared with the adjacent normal tissue (P < 0.001, Fig. 2D,E).

Figure 2.

Reduced DACH1 expression was associated with DNA methylation in human primary HCC. (A) Representative MSP results of DACH1 methylation status in adjacent tissues (N) and HCC tissues (T). (B) The methylation level of HCC adjacent (left) and cancer (right) tissues was checked with MSRED-qPCR. The result is shown as scatterplots with horizontal lines representing the median. (C) Representative images of DACH1 expression in adjacent tissues (left) and paired cancer tissues (right) by IHC. Bars represent 100 μm. (D) The expression of DACH1 in each adjacent and carcinoma tissue. (E) DACH1 expression scores are shown as boxplots, with horizontal lines representing the median; the bottom and top of the boxes represent the 25th and 75th percentiles, respectively; the vertical bars represent the range of data.

Restoration of DACH1 Expression Activated TGF-β1 Signaling in an SK-Hep1 Cell Line

As reported before, TGF-β transactivates an artificial promoter containing the Smad-binding elements (SBE-4 Luc) through Smad3 and Smad4.[18] As shown in Fig. 3A, the promoter activity was increased in a dose-dependent manner by restoration of DACH1 expression combined with TGF-β1 treatment (P = 0.0005) and a 3-fold increase without TGF-β1 treatment (P = 0.0379). The level of phospho-Smad3 was associated with reexpression of DACH1, but the level of phospho-Smad2 remained unchanged (Fig. 3B). The expression level of c-Myc was similar in TGF-β1-treated and -untreated SK-Hep1 cells, and it was reduced when DACH1 was reexpressed. The reduced level of c-Myc was more significant when combined with reexpression of DACH1 and TGF-β1 treatment. The cell growth was inhibited by DACH1 overexpression regardless of stimulation with TGF-β1 (P = 0.0013 and P = 0.0412, respectively) (Fig. 3B,C). The expression of DACH1 was increased with 5-AZA treatment, and the expression was further augmented by combination of 5-AZA and TGF-β1 treatment. C-Myc expression was reduced after reexpression of DACH1 by 5-AZA treatment. Cell growth was inhibited by 5-AZA treatment (P = 0.0381) and reinforced in combination with TGF-β1 treatment (P = 0.003) (Fig. 3D,E).

Figure 3.

Reactivation of TGF-β signaling by restoration of DACH1 in SK-Hep1 cells. (A) Smad-binding elements (SBE)-4 Luc reporter was activated by DACH1 reexpression. Cotransfection of pSBE4-Luc and DACH1 resulted in slightly increased luciferase activity without TGF-β1 treatment. Luciferase activity was further increased after addition of TGF-β1. (B) Western blotting results showed gene expression level in DACH1 expressed or unexpressed SK-Hep1 cells treated or untreated with TGF-β1. Antibodies used are indicated. (C) Cell viability of DACH1 expressed or unexpressed SK-Hep1 cells treated or untreated with TGF-β1. C-Myc expression (D) and cell viability (E) in SK-Hep1 cells treated with or without 5-AZA and/or TGF-β1. *P < 0.05, **P < 0.01.

Association of DACH1 Expression and Clinicopathologic Features

The association between DACH1 methylation and clinicopathologic features was analyzed. As shown in Supporting Table 1, methylation of DACH1 was associated with poor differentiation (P = 0.0211) and aspartate aminotransferase (AST) / alanine aminotransferase (ALT) ratio (P = 0.0293). The expression of DACH1 was not associated with any clinicopathologic features in the adjacent tissues (Table 2). However, reduced expression of DACH1 was related to poor differentiation (P = 0.0322) (Fig. 4). Decreased DACH1 expression was associated with an AST/ALT ratio >2 (P = 0.0120). No significant association was found between DACH1 expression and liver cirrhosis, AFP level, tumor size, TNM stage, hepatitis B surface antigen (HBsAg) positivity, age, or gender.

Table 2. Relationship Between Adjacent Tissue and Carcinoma DACH1 and Clinicopathologic Features
VariableAdjacent Tissue DACH1 ScoresP valueCarcinoma Tissue DACH1 ScoresP value
Gender 0.7843 0.9913
Male (n = 48)5.982±1.706 4.294±1.917 
Female (n = 9)5.444±1.77 4.389±2.162 
Age (y) 0.7459 0.6368
<50 (n = 21)5.833±1.552 4.294±2.52 
≥50 (n = 36)6.032±1.763 4.153±1.619 
HBsAg 0.9616 1
Negative (n = 10)6.014±2.518 4.208±1.559 
Positive (n = 47)5.88±1.448 4.25±2.115 
α-Fetoprotein (μg/L) 0.9114 0.5637
<0 (n = 25)5.917±2.127 4.317±1.965 
≥20 (n = 32)5.763±1.66 3.897±2.096 
ALT (U/L) 0.5001 0.9473
≤75 (n = 32)5.88±1.829 4.124±1.793 
>75 (n = 25)5.37±1.785 4.167±2.586 
AST/ALT 0.9851 0.0120
≤1 (n = 27)5.764±2.098 4.104±2.085 
<2 (n = 18)5.843±1.644 4.49±2.071 
≥2 (n = 12)5.714±1.286 3.357±0.339 
Liver cirrhosis 0.1523 0.7739
Absent (n = 25)6.304±1.839 4.303±1.959 
Present (n = 32)5.505±1.613 4.505±1.976 
Tumor size 0.6776 0.4624
≤5 (n = 22)5.825±1.48 4.758±2.357 
>5 (n = 35)6.083±1.812 4.156±1.717 
Differentiation 0.6746 0.0322
Poor (n = 21)5.798±1.458 3.304±1.783 
Moderate (n = 29)5.907±1.54 4.8±1.662 
High (n = 7)6.667±3.859 4.875±1.931 
TNM stage 0.6763 0.3358
I (n = 37)6.056±1.768 4.281±1.96 
II (n = 7)5.133±2.142 5.60±1.342 
III (n = 13)5.788±1.483 4.24±1.499 
Figure 4.

Representative results of H&E and IHC for DACH1 staining in HCC tissues. Representative images of H&E and IHC for DACH1 in HCC with poor differentiation (case 1) and good differentiation (case 2) and corresponding adjacent tissues. Bars represent 25 μm.

Restoration of DACH1 Expression Inhibited Proliferation and Induced Apoptosis in HCC Cells

To observe the effect of DACH1 on proliferation, SK-Hep1 and SMMC-7721 cells stably expressing DACH1 or empty vector were established by lentivirus transduction (Fig. 5A). Cell viability was detected by daily measure of the CCK-8 value (Fig. 5B). Reexpression of DACH1 repressed proliferation and induced apoptosis in both cell lines (P < 0.01). The apoptotic cells were increased by DACH1 by 2 to 2.5-fold (Fig. 5C). Cell cycle progression analyzed by FACS showed that DACH1 increased cells in G0/G1 phase and decreased cells in S phase (Fig. 5D). Colony formation ability was inhibited by DACH1 (Fig. 5E). Together, our results demonstrated that DACH1 blocked cellular proliferation by inducing apoptosis and decreasing DNA synthesis.

Figure 5.

Ectopic expression of DACH1 inhibited proliferation and induced aopotosis. (A) Stable expression of DACH1 in SK-Hep1 and SMMC-7721 cells determined by fluorescent staining using antibody to DACH1. (B) Growth curves evaluated by Cell Counting Kit 8 (CCK-8) activity. (C) Apoptosis determined by Annexin-V positive cells. (D) Cell cycle distribution by flow cytometry. (E) Colony formation.

Overexpressing DACH1 Increased the Chemosensitivity to 5-FU in HCC Cell Lines

To determine whether differentially expressed DACH1 do have a functional impact, we raised its level by DACH1 transfection in SMMC-7721 cells and SK-Hep1 cells, and found that the IC50 to 5-FU was decreased in SMMC-7721 (19.27 ± 0.61 versus 9.58 ± 1.62 μg/mL) and SK-Hep1 (0.63 ± 0.17 versus 0.14 ± 0.06 μg/mL (Fig. 6A,B). Apoptotic cells increased from 0.78% to 1.82% at base level, from 8.3% to 12.72% with 5-FU treatment (Fig. 6C). Recently, DACH1 was found as a novel p53 binding partner that participates in p53-mediated induction of p21CIP1 and cell cycle arrest.[19] We investigated p53 and p21CIP1 abundance by western blot analyses. The results demonstrated that DACH1 did not change the p53 level, but enhanced p21 expression (Fig. 6D). Our data suggested that restoration of DACH1 induced the expression of p21CIP1 in SMMC-7721 cells, contributing to the inhibition of HCC cells proliferation.

Figure 6.

Overexpressing DACH1 increased the chemosensitivity to 5-FU in HCC cell lines. Dose-response curves of 5-FU-treated for 72 hours in SMMC-7721 (A) and SK-Hep1 (B) cells stably expressing DACH1 or vector control determined by CCK-8 activity. SMMC-7721 cells expressing DACH1 or vector control and treated with or without 5-FU at 10 μg/mL for 36 hours were analyzed for apoptosis (C) and protein expression using antibodies as indicated (D).


There are several reports about reduced DACH1 expression in human cancers.[10, 20] DNA methylation in the regions spanning the DACH1 gene is believed to be the underlying mechanism for carcinogenesis.[21] We have identified the presence of methylated CpG in the promoter region of DACH1. According to our observation, the expression of DACH1 can be reactivated/elevated by 5-AZA treatment. Our results imply that silenced or reduced expression of DACH1 by DNA methylation is related to HCC. This concept is supported by the absence of DACH1 expression in SK-Hep1 cells, which is poorly differentiated.[22] It is generally considered that poorly differentiated HCC cell lines are resistant to TGF-β-induced cell growth arrest.[7, 23] The HCC cells possibly retain intact machinery of TGF-β signaling and establish downstream alterations, which only disable the tumor-suppressive functions of this signaling pathway.[23-25] This conforms with our hypothesis that methylation silence of DACH1 is associated with resistance to a TGF-β-mediated antiproliferative effect in HCC cells. Our data further support that the resistance may be due to altered gene expression of DACH1 since restoration of DACH1 in SK-Hep1 cells resulted in elevated TGF-β signaling activity even without TGF-β stimulation. The c-Myc levels were decreased, leading to inhibited cell growth in SK-Hep1 cells, reinforced by TGF-β treatment. The DNA demethylation produced similar results in our study. Reexpression of DACH1 increased phosphorylated Smad3 under TGF-β treatment. However, TGF-β did not affect the total volume of Smad2 or Smad3. This activity may be the key factor about DACH1 impacting the TGF-β/Smad pathway. The phosphorylated Smad2 and Smad3 bind to Smad4 and move into the nucleus to form complexes that regulate transcription.[26] We have also inferred that elevated pSmad3C can promote tumor-suppressive activity of the TGF-β/Smad pathway. In fact, TGF-β activates both its type I receptor (TβRI) and c-Jun NH2-terminal kinase (JNK), which then phosphorylate Smad2 and Smad3 at the COOH-terminal (pSmad2/3C) and linker regions (pSmad2/3L), respectively.[27] When COOH-tail phosphorylation of Smad3 was prevented, pSmad3C mediated p21WAF1 transcription, and consequently the cytostatic effect of TGF-β/activin upon normal hepatocytes were inhibited.[28] Our study showed that pSmad3C (Ser423/425) was induced by DACH1 reexpression, and further caused TGF-β signaling inhibiting cell proliferation in SK-Hep1 cells.

Our data revealed that reduced DACH1 expression was associated with poor differentiation in primary human HCC, and loss of DACH1 expression was found in Sk-Hep1, which is poorly differentiated. It was reported that c-Myc inhibits differentiation in liver tumor of mouse.[29] Our study revealed that c-Myc was up-regulated by methylation silencing DACH1 expression, and restoration of DACH1 expression suppressed c-Myc expression in Sk-Hep1 cells. This indicates that up-regulation of c-Myc by DACH1 methylation may result in poor differentiation of HCC cells. Our results demonstrated that reduction of DACH1 expression is correlated with an elevated AST/ALT ratio. It was reported that this ratio tends to predict overall mortality in liver disease[30, 31] through grade of histological differentiation, biological behavior of the tumor, and prognosis.[32]

DACH1 was methylated in both HCC cell lines and HCC tissues. Ectopic expression of DACH1 repressed cellular growth through decreased proliferation and increased apoptosis, which are consistent with previous publications and indicate that DACH1 may be a novel tumor suppressor of HCC. A recent study found that DACH1 bound p53 and enhanced p53-dependent DNA damage response.[19] Our results showed that ectopic expression of DACH1 sensitizes HCC cells to 5-FU through enhanced p21 expression and apoptosis induction. The effects of DACH1 on other types of anticancer drugs needs further investigation.

We therefore conclude that epigenetic silencing of DACH1 is an important factor in proliferation, differentiation, and progression of HCC, possibly through deregulated TGF-β signaling. It has potential value for the stratification of patients, development of favorable therapeutic plans, as well as in providing new therapeutic targets. However, further work is required to explore the clinical significance for DACH1 in predicting prognosis and personalized treatment selection.