TY - JOUR
AU - Yu, Long
AB - Abstract Members of the metallothionein (MT) family are short, cysteine-rich proteins involved in metal metabolism and detoxification, suggesting that MT proteins protect cells from damage caused by electrophilic carcinogens and thereby constitute a critical surveillance system against carcinogenesis. However, the roles of MT proteins in human hepatocellular carcinoma (HCC) are not fully understood. We identified a member of the MT family, termed MT1M. MT1M is expressed in various normal tissues with the highest level in the liver. MT1M expression can be induced by heavy metals and protect Escherichia coli from heavy metal toxicity. However, MT1M expression markedly decreased in human HCC specimens. A methylation profiling analysis indicated that the MT1M promoter is methylated in the majority of HCC tumors examined. Moreover, restored expression of MT1M in the HCC cell line Hep3B, which lacks endogenous MT1M expression, suppressed cell growth in vitro and in vivo and augmented apoptosis induced by tumor necrosis factor α. Furthermore, stable expression of MT1M in Hep3B cells blocked tumor necrosis factor α-induced degradation of IκBα and transactivation of NF-κB. We conclude that MT1M is a novel member of the MT family. Frequent downregulation of MT1M in human HCC may contribute to liver tumorigenesis by increasing cellular NF-κB activity. Introduction Members of the metallothionein (MT) family are small cysteine-rich proteins with a molecular weight of approximately 6000Da that possess specific binding activity for metal ions. They are involved in metal detoxification and in the protection of cells against certain electrophilic carcinogens ( 1 , 2) . In humans, MTs are encoded by a family of genes located at chromosome 16q13 that contains 17 known members, of which 10 are functional and 7 are non-functional ( 3 , 4) . The functional human MTs, which include MT1 (A, B, E–H, X), MT2A, MT3 and MT4, have been reported previously ( 3 , 4) , and only one MT1 exists in the mouse genome. Others have reported dramatic loss of MT expression in primary human HCC by immunohistochemistry methods ( 5–9 ). Mice model results showed that MT1 and MT2 double knockout accelerates hepatocarcinogenesis in mice exposed to the carcinogen diethylnitrosamine ( 10 ). But it is unclear whether any specific isoform(s) of human MTs are preferentially deregulated in HCC. Thus, it is important to systematically assess the expression pattern of different MTs in normal and tumor tissues in order to clarify the cancer relevance of the members in the MT family. In the present study, we identified MT1M as a new functional member of the MT1 family. We demonstrated for the first time that MT1M is one of the most frequently downregulated MTs in HCC. We provided evidence that restored expression of MT1M in HCC cells impedes HCC cell growth in vitro and in vivo . We further showed that forced expression of MT1M enhances TNF-α-induced apoptosis but inhibits TNF-α-induced-activation of NF-κB. Materials and methods Differential display screening The quality of total RNA from 10 paired HCC specimens with their corresponding non-cancerous specimens was tested by northern blot analysis before the reverse transcription. Total cellular RNA and first-strand complementary DNA were prepared according to manufacturers instructions, and DD-PCR was performed as described Molecular Cloning (Third Edition). The differential fragments were subcloned, sequenced and confirmed by dot blot. Sequence homology was searched in GeneBank database. Human tissue samples and cell lines Surgical resection specimens were obtained from Chinese patients with HCC and one male patient with mixed hepatocellular-cholangiocarcinoma diagnosed at Qidong Liver Cancer Institute, Jiangsu province in China. Two fetal liver tissues were obtained by autopsy from the Medical College of Suzhou University, Jiangsu province in China. The resected specimens were immediately snap-frozen in liquid nitrogen and stored at −80°C for RNA analysis. Informed consent was obtained from each subject or subject’s guardian after approval by the hospital Ethics Committee. Twelve HCC cell lines (Hep3B, HepG2, SK-Hep1, YY-8103, Focus, Huh7, Bel-7402, Bel-7404, Bel-7405, QGY-7701, QGY-7703, SMMC-7721, PLC), hepatoblastoma cell line (HepG2), fetal liver–derived cell line (L02) and HEK 293T were employed in this study, maintained in RPMI1640 or Dulbecco’s modified Eagle’s medium with 10% fetal bovine serum at 37 ° C and 5% CO 2 . BrdU incorporation assay Proliferating cells were measured by colorimetric 5-bromo-2-deoxyuridine (BrdU) Cell Proliferation ELISA Kit (Roche, USA). Cells were incubated for additional 6h at 37°C by adding BrdU labeling solution, then fixed and denatured by FixDenat solution. After incubation with anti-BrdU-peroxidase working solution, substrate solution was added until the color development is sufficient for photometric detection. 1mM H 2 SO 4 was applied to stop the reaction. Absorbance was measured using an automatic enzyme-linked immunosorbent assay reader (450nm). Apoptosis assay Hep3B/Mp and Hep3B/Vp cells (3×10 5 cells/well) were plated in six-well plates. At 24h after plating, cells were treated with TNF-α (30ng/ml). The cells were collected at the indicated time points and fixed with 2ml of 70% ethanol at 4°C for 2h. Cells were washed with 1× phosphate-buffered solution and incubated for 15min at room temperature in 1× phosphate-buffered solution containing 100 µg/ml RNase A and 50 µg/ml propidium iodide. DNA content and cell-cycle analysis were assessed by FACScan (Becton Dickinson, USA). Based on propidium iodide staining, cells in the sub-G1 marker window were considered to be apoptotic. Subcutaneous tumor xenograft assay Four- to six-week-old female athymic BALB/c nude mice were obtained from Shanghai SLAC Laboratory Animal CO. LTD (Shanghai, China) and acclimated to laboratory conditions 1 week before tumor implantation. Tumor xenografts were established by s.c. injection of 4×10 6 cells suspended in 200 µl of 1× phosphate-buffered solution into both flanks of nude mice, with Hep3B cells stably transfected with pcDNA3.1A on the right flank and pcDNA3.1A-MT1M on the left. Tumors were measured every 3 days in two dimensions using calipers, and volume was estimated by the formula (length (mm) × width (mm) 2 )/2. The mice were killed 4 weeks after injection and the tumors were harvested for further examination. Experimental groups consisted of five mice in each group. The difference in mean tumor volume and weight was analyzed using the paired, two-tailed t -test with P < 0.05 considered statistically significant. Promoter methylation analysis The methylation status of the CpG islands in the MT1M promoter was analyzed by methylation-specific PCR (MSP) with sodium bisulfite–converted DNA from tissue samples. Briefly, 10 µg of DNA from each sample in 50 µl TE Tris, a common pH buffer, and EDTA was incubated with 5.5 µl of 3M NaOH at 37°C for 10min, followed by a 16-h treatment at 50°C with 30 µl of freshly prepared 10mM hydroquinone and 520 µl of freshly prepared 3.6M sodium bisulfite at pH 5.0. The DNA was desalted using a dialysis system with 1% agarose. The desalted DNA sample (approximately 100 µl in volume) was then denatured at 37°C for 15min with 5.5 µl of 3M NaOH followed by ethanol precipitation with 33 µl 10M NaOAc and 300 µl ethanol. After washing with 70% ethanol, the DNA pellet was dissolved with 30 µl TE at 65°C for 10min. The DNA sample was stored at −20°C until further use. About 50ng of the DNA sample was used for each PCR. MTS assay Stable transfectant Hep3B/Mp and control Hep3B/Vp cells were cultured at equal densities in 96-well plates. Cell proliferation was assayed every 24h using an MTS MTS, (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium) assay kit (Promega, USA). The resulting formazan product was quantitated with a multi-well spectrophotometer by measuring absorbance at 490nm. The experiments were performed in triplicate. Colony formation assay Hep3B, QGY-7703 and SK-Hep-1 cells were seeded at the same densities in 35-mm dishes (1×10 3 cells/dish). After 20h, cells were transiently transfected with pcDNA3.1A-MT1M or pcDNA3.1A vector. Transfectants were selected using G418 (750 µg/ml) for 2–3 weeks and stained with Giemsa. The total number of colonies in each plate from three independent transfections was counted. 5-Aza-dC treatment Cells (6×10 4 cells/dish) were plated in 35-mm dishes. After treatment with 5-aza-2-deoxycytidine (Sigma Aldrich, USA) at a final concentration of 5 µM or dimethyl sulfoxide for 4 days (with daily replacement of the culture medium), cells were harvested for semiquantitative reverse transcription (RT)–PCR analysis. Statistical analysis Microcal Origin software (version 6.0) was used for statistical analysis. The data was expressed as mean + standard deviations. The statistical significance of differences between the groups was evaluated using the Student’s t -test. P values < 0.05 was considered to be statistically significant. Other materials and methods are available at Supplementary Materials and Methods at Carcinogenesis Online. Results Identification of MT1M as a functional member of the MT family We identified a 300-bp cDNA fragment as one of the genes that are potentially downregulated in human HCCs in comparison with adjacent non-cancerous tissues during our initial screening for HCC relevant genes by differential display PCR ( Supplementary Table 1 , available at Carcinogenesis Online). Using a combination of approaches including 5′-RACE, EST database searching and RT–PCR analysis, we obtained a 447-bp cDNA clone containing an ORF that encodes a 61 amino acid protein. Both protein and DNA analysis indicated that this gene shares great homology to the MT1 family at both DNA and protein levels ( Figure 1A and Supplementary Figure 1 , available at Carcinogenesis Online). This gene is annotated as MT1M in NCBI database. The human MT1M gene is located within the MT gene cluster at chromosome 16q13 and flanked by the two functional MT genes MT1E and MT1A. MT1M was also named MT1K in a previous genomic sequencing report ( 4 ). The author treated it as a pseudogene based on the presence of several atypical amino acids in protein sequence that not seen in other known MTs, but further experimental evidences to support this conclusion is still lacking until now. In this study, we sought to determine whether MT1M is functional and plays roles in HCC. Fig. 1. Open in new tabDownload slide Identification of MT1M as a novel functional member of MT. (A) The alignment of the amino acid sequences of MT1M with other known functional members of the human MTs. (B) Northern hybridization ananlysis of MT1M in 16 human adult tissues. (C) The influence of Cd 2+ on bacterial growth. MT1M expressing E. coli was cultivated in medium containing 0.5mM Cd 2+ . The bacterial densities were measured at OD 600 . One representative experiment from three independent assays is shown. (D) Regulation of MT1M, MT1X and MT1G promoter activities by heavy metals in HEK 293T cells. At 24h after transfection with MT1M-Luc, MT1X-Luc and MT1G-Luc constructs, cells were exposed to 2 µM CdCl 2 (Cd 2+ ), 100 µM CuCl 2 (Cu 2+ ), 150 µM ZnCl 2 (Zn 2+ ), 10 µM dexamethasone (dex) and saline (sal) for 7h. Cells were then harvested and assayed for luciferase activities. * indicates statistical significance (* P < 0.05, * * P < 0.01) and three experiments were performed. (E) Semiquantitative RT–PCR was performed to analyze the mRNA levels of MT1M, MT1X and MT1G in HEK 293T, Hep3B and HepG2 cells. Cells were treated with Cd 2+ , Cu 2+ , Zn 2+ and dex for 7h before harvest. β2-microglobulin (β2-MG) was used as a loading control. First, we examined the expression status of the MT1M gene in 16 different human tissues by northern blot. Considering the high homology between MT1M and other members of the MT family, MT1M-specific primers were designed ( Supplementary Figure 1 , available at Carcinogenesis Online) and the PCR product amplified from the MT1M gene was confirmed by sequencing before it was used for northern blot. As demonstrated in Figure 1B , one transcript of MT1M with a size of approximately 0.5kb was detected in the majority of tissues examined. MT1M messenger RNA (mRNA) was found most abundant in liver, kidney, pancreas and skeletal musclebut expressed moderately or low in other tissues. Intriguingly, normal liver shows the highest expression of MT1M. Because MT1M contains all the conserved cysteine residues that are essential for metal chelation, we sought to determine whether MT1M has metal-binding activity. It is difficult to examine whether ectopic expression of MT1M confers metal resistance under physiological conditions in eukaryotic cells because endogenous MTs could be greatly induced by heavy metals. Thus, we adopted a useful Escherichia coli system that has been described previously ( 11 ). The results showed that more MT1M-expressed bacteria survived in culture media with a high concentration of Cd 2+ (0.5mM) than mock-transformed bacteria ( Figure 1C ), suggesting that MT1M expression can protect E. coli from heavy metal toxicity. We also found that MT1M proteins purified from bacteria were red in color, indicating that MT1M protein may bind to heavy metal ions. Next, we examined whether the gene promoter of the MT1M is responsive to stimulation of heavy metals under physiological conditions. The promoter region of the MT1M was cloned, and a luciferase reporter (MT1M-luc) was constructed. The reporter construct was transfected into HEK 293T cells and luciferase activities were measured in cells treated with saline (Mock), CdCl 2 (Cd 2+ ), CuCl 2 (Cu 2+ ), ZnCl 2 (Zn 2+ ) or dexamethasone (Dex). As controls, we also constructed two reporters from the promoters of other MT genes, MT1X-luc and MT1G-luc. Previous studies demonstrate that the promoters of the MT1X and MT1G are heavy metal responsive ( 6 ). As expected, the promoter activities of MT1X and MT1G were dramatically induced by heavy metal in HEK 293T cells ( Figure 1D ). Cd 2+ or Zn 2+ treatment also markedly augmented the activity of the MT1M promoter. In contrast, little or no effect on MT1M promoter activity was observed in cells treated with Cu 2+ and dexamethasone ( Figure 1D ). We also examined the effect of heavy metals on the expression of endogenous MT1s in HEK 293T, HepG2 and Hep3B cells. The expression of MT1M was at low to undectable levels in HEK 293T and Hep3B cells but can be induced by heavy metal at various degrees ( Figure 1E ). MT1M is highly expressed in HepG2 cells and a slight increase in the expression of MT1M was observed after treatment with heavy metals ( Figure 1E ). In addition, there was no induction in MT1M, MT1X and MT1G mRNAs by dexamethasone. This is not unexpected because these three genes do not contain the glucocorticoid receptor–binding sequences that are present in the promoter of the MT2A gene ( 4 ). Expression of the MT1M in HCC tumors and cell lines Because of the high protein homology among the members of the MT1 family ( Figure 1A ), it is almost impossible to examine the expression status of the MT1M protein in human tissues by western blot or IHC. To overcome this difficulty, we employed northern blot to assess MT1M expression in human HCC specimens. Total RNAs were isolated from cancer and adjacent non-cancerous tissues and transferred to nylon membranes for northern blot with a MT1M-specific probe. Significant downregulation of MT1M mRNA was observed in 7 out of 8 HCC tissues in comparison with the adjacent non-cancerous tissues. No significant difference in MT1M expression was detected between the normal and cancer tissues of a mixed hepatocellular-cholangiocarcinoma (CH01) ( Figure 2A ). We also examined MT1M expression in several HCC and hepatoblastoma cell lines by northern blot. Little or no expression of MT1M was observed in a number of HCC cell lines including Hep3B, YY-8103, L02, SK-hep-1 and BEL-7402 ( Figure 2B ). Fig. 2. Open in new tabDownload slide Analysis of expression of MT1M expression in HCC and adjacent non-cancerous liver tissues. (A) Expression of MT1M mRNA was examined by northern blot in two human fetal liver tissues, eight human HCCs (HCC01-08) and one cholangiohepatoma tissue (CH01). N denotes adjacent non-cancerous tissues and T denotes tumor tissues. Ethidium bromide staining of total RNA was used to ensure the equal loading. (B) Detection of the MT1M expression levels in nine HCC cell lines by northern blot. (C) Semiquantitative RT–PCR was performed to analyze the expression profile of the functional MT1s in 18 human tissues. (D) Semiquantitative RT–PCR was performed to analyze the mRNA levels of functional MT1s in a large cohort of HCC and adjacent non-cancerous tissues (case 09-38). β2-MG was used as an internal control. (D) MT1M mRNA expression levels were analyzed in 92 paired HCC and adjacent non-cancerous tissues (Case.39–131) by Quantitative RT–PCR methods. Detailed clinical data are available for this cohort of HCC patients (see Table I ). Table I. Correlation of the clinicopathological findings with tumor MT1M expression Variable . MT1M expression in HCC (T/N) . . Low(<0.125) . Moderate and high(>0.125) . P . Sex Female 14 2 0.173277 Male 54 22 Age-year <50 27 4 0.04887 >50 43 20 History of hepatitis type B No 25 10 0.658569 Yes 26 13 Family history of liver cancer No 66 22 0.265433 Yes 2 2 Hepatitis B virus Negative 16 6 0.752798 Positive 38 17 Tumor size <5 cm 28 14 0.146859 >5 cm 40 10 Tumor number Single 42 20 0.290761 Multiple 13 3 Hepatic cirrhotic nodule No 21 4 0.19742 Yes 46 19 Alpha-fetoprotein Negative(<20ng/ml) 31 13 0.364389 Positive(>20ng/ml) 37 10 Tumor encapsulation No 28 10 0.966554 Yes 40 14 Pathological differentiation I-II 45 14 0.522924 III-IV 21 9 TNM clinical stage I 21 13 0.042182 II–III 47 11 Survival <960days 30 14 0.534925 >960days 21 7 Variable . MT1M expression in HCC (T/N) . . Low(<0.125) . Moderate and high(>0.125) . P . Sex Female 14 2 0.173277 Male 54 22 Age-year <50 27 4 0.04887 >50 43 20 History of hepatitis type B No 25 10 0.658569 Yes 26 13 Family history of liver cancer No 66 22 0.265433 Yes 2 2 Hepatitis B virus Negative 16 6 0.752798 Positive 38 17 Tumor size <5 cm 28 14 0.146859 >5 cm 40 10 Tumor number Single 42 20 0.290761 Multiple 13 3 Hepatic cirrhotic nodule No 21 4 0.19742 Yes 46 19 Alpha-fetoprotein Negative(<20ng/ml) 31 13 0.364389 Positive(>20ng/ml) 37 10 Tumor encapsulation No 28 10 0.966554 Yes 40 14 Pathological differentiation I-II 45 14 0.522924 III-IV 21 9 TNM clinical stage I 21 13 0.042182 II–III 47 11 Survival <960days 30 14 0.534925 >960days 21 7 Fisher’s exact test was used. The bolding stands for the P -values with significant difference. Open in new tab Table I. Correlation of the clinicopathological findings with tumor MT1M expression Variable . MT1M expression in HCC (T/N) . . Low(<0.125) . Moderate and high(>0.125) . P . Sex Female 14 2 0.173277 Male 54 22 Age-year <50 27 4 0.04887 >50 43 20 History of hepatitis type B No 25 10 0.658569 Yes 26 13 Family history of liver cancer No 66 22 0.265433 Yes 2 2 Hepatitis B virus Negative 16 6 0.752798 Positive 38 17 Tumor size <5 cm 28 14 0.146859 >5 cm 40 10 Tumor number Single 42 20 0.290761 Multiple 13 3 Hepatic cirrhotic nodule No 21 4 0.19742 Yes 46 19 Alpha-fetoprotein Negative(<20ng/ml) 31 13 0.364389 Positive(>20ng/ml) 37 10 Tumor encapsulation No 28 10 0.966554 Yes 40 14 Pathological differentiation I-II 45 14 0.522924 III-IV 21 9 TNM clinical stage I 21 13 0.042182 II–III 47 11 Survival <960days 30 14 0.534925 >960days 21 7 Variable . MT1M expression in HCC (T/N) . . Low(<0.125) . Moderate and high(>0.125) . P . Sex Female 14 2 0.173277 Male 54 22 Age-year <50 27 4 0.04887 >50 43 20 History of hepatitis type B No 25 10 0.658569 Yes 26 13 Family history of liver cancer No 66 22 0.265433 Yes 2 2 Hepatitis B virus Negative 16 6 0.752798 Positive 38 17 Tumor size <5 cm 28 14 0.146859 >5 cm 40 10 Tumor number Single 42 20 0.290761 Multiple 13 3 Hepatic cirrhotic nodule No 21 4 0.19742 Yes 46 19 Alpha-fetoprotein Negative(<20ng/ml) 31 13 0.364389 Positive(>20ng/ml) 37 10 Tumor encapsulation No 28 10 0.966554 Yes 40 14 Pathological differentiation I-II 45 14 0.522924 III-IV 21 9 TNM clinical stage I 21 13 0.042182 II–III 47 11 Survival <960days 30 14 0.534925 >960days 21 7 Fisher’s exact test was used. The bolding stands for the P -values with significant difference. Open in new tab In normal human tissues, MT3 and MT4 are restricted expressed in brain and stratified squamous epithelia, respectively ( 5 ). Multiple MT1s are expressed in liver tissues. However, it is unclear which specific isoform(s) of human MT1s are expressed in normal liver tissues and preferentially deregulated in HCC. The gene-specific primers for seven known functional MT1s plus MT1M were designed ( Supplementary Figure 1 and Supplementary Table 2 , available at Carcinogenesis Online). The expression pattern of MT1s in normal human tissues was further confirmed by semiquantitative RT–PCR using specific primers and direct sequencing of PCR products (data not shown). As demonstrated in Figure 2C , each MT1 isoform exhibits a unique expression pattern in the normal human tissues. MT1B was not expressed in any tissues we examined, but other MT1s were abundantly expressed in liver tissues. Next, we compared the relatively expression of eight MT1s in a cohort of HCC samples. Consistent with the results obtained from northern blot, the levels of MT1M mRNA were significantly downregulated or lost in 14 out of 30 HCC tumors ( Figure 2D ). No difference in expression of MT1B, MT1E and MT1X were detected between HCC and normal tissues. However, decreased expression of MT1A (3/30, 10%), MT1F (3/30, 10%), MT1G (9/30, 30%) and MT1H (20/30, 66.7%) at various degrees was also detected between HCC and adjacent non-cancerous tissues. This result showed that not all of MT1s expressions are decreased in HCC. MT1M is one of the isoforms among the MT1 family that is mostly downregulated in human HCC. To further explore the clinico-pathological correlation of MT1M downregulation in HCC, we analyzed the MT1M expression with respect to various clinical parameters in 92 HCC patients using more accurate quantitative RT–PCR methods ( Figure 2E and Table I ). As shown in Table I , low levels of MT1M correlated with history of TNM clinical grade ( P = 0.042) and age of patients by the time of diagnosis of the disease ( P = 0.049). No correlation with other parameters was found. These findings indicate that downregulation of MT1M is frequent in human HCCs and might serve as a prognosis indicator of HCC. Methylation analysis of the MT1M promoter in HCC specimens and cell lines Methylation of CpG islands is important for gene silencing and imprinting ( 12 ). Because there is a CpG-rich region in the MT1M promoter ( Figure 3A ) and the expression of MT1M is markedly downregulated in HCC samples and cell lines, we hypothesized that the MT1M promoter is hypermethylated in HCC. To investigate this hypothesis, MSP assays were performed to examine the methylation status of the MT1M promoter. Methylated PCR products were detected in all of the cell lines except PLC and hepatoblastoma cell line HepG2 ( Figure 3B ). The methylation status of the MT1M promoter was also measured in eight paired HCC and adjacent non-cancerous tissues ( Figure 3C ). Methylated PCR products were detected in both tumor and non-cancerous tissues, whereas unmethylated PCR products were usually low or undetectable in tumors ( Figure 3C ), indicating that the methylation levels are greatly elevated in most of the HCC samples examined. The MSP results were confirmed by sequencing of the PCR products and a representative of this analysis is shown in Figure 3D . Consistent with these results, MT1M expression was restored by treatment of different HCC cell lines with the demethylating agent 5-Aza-dC ( Figure 3E ), but not the histone deacetylase inhibitor trichostatin A (data not shown). These findings suggest that decreased expression of MT1M in HCC tumors and cell lines is mediated, at least in part, by promoter methylation. In line with the finding that no change was observed in expression of MT1E and MT1X in human HCC tumors ( Figure 2D ), no methylation of their promoters were detected in HCC cell lines ( Figure 3B ). Methylation of the MT1F promoter was observed in a small portion of HCC cell lines examined ( Figure 3B ), which is consistent with the previous reports ( 9 ). Fig. 3. Open in new tabDownload slide Methylation profile of the putative MT1M promoter in HCC and adjacent non-cancerous liver tissues. (A) The CpG-rich region of the putative MT1M promoter. The CpG dinucleotides are shown in red. Unmethylation-specific primers are underlined. Methylation-specific primers are indicated by shading. (B, C) Methylation status of the MT1M promoter in HCC and hepatoblastoma cell lines and eight pairs of HCC and adjacent non-cancerous n tissues were examined by MSP assay. The electrophoretic patterns of PCR products are presented. U, amplification with the unmethylation-specific primers; M, amplification with the methylation-specific primers. (D) Representative DNA sequences of Methylation-Specific PCR (MS-PCR) products. The CpG dinucleotides are indicated by underlining. (E) Semiquantitative RT–PCR performed to determine the mRNA levels of MT1M in QGY-7703, SK-Hep1 and Hep3B cells before and after treatment with 5-Aza-dC. Expression of MT1M inhibits HCC cell growth in vitro and in vivo Since MT1M is significantly downregulated in human HCC, we investigated the potential role of MT1M in growth of HCC cells using colony formation assays. HCC cell lines QGY-7703, SK-Hep1 and Hep3B were transfected with an MT1M expression construct and a control vector, and stable clones were selected. A significant reduction (> 50%) in colonies was observed in MT1M transfected cells in comparison with vector-transfected cells in all three cell lines examined ( Figure 4A ). Thus, MT1M inhibits the growth of HCC cells in vitro . Fig. 4. Open in new tabDownload slide Effects of MT1M overexpression on HCC cell growth and proliferation. (A) Three cell lines (Hep3B, SK-Hep1 and QGY-7703) were transfected with pcDNA3.1A-MT1M or control vector and selected with G418 for 3 weeks and outgrowth colonies were stained. Representative photographs of transfected cells were shown. Total numbers of the colonies from three independent experiments were counted. * indicates statistical significance (* P < 0.01) (B) Analysis of MT1M protein expression in MT1M stable and control clones. V1, V2, V3: vector-transfected clones; M1, M2, M3: MT1M-transfected clones. Vp and Mp were mixed cell population from clones consisting of V1, V2, V3, and M1, M2, M3, respectively. (C) An equal number of cells from the indicated clones were plated and the MTS assay was performed to determine the effects of MT1M overexpression on Hep3B cell growth. (D) The DNA synthesis activity of Hep3B/Mp and Hep3B/Vp cells was assayed using BrdU ELISA Kit. (E) The cell-cycle distributions of Hep3B/Mp and Hep3B/Vp cells was analyzed by flow cytometry. (F) Xenografts were established by injecting s.c. 4×10 6 Hep3B/Vp and Hep3B/Mp cells to the left and right flanks of nude mice, respectively. The tumor volume of xenografts (left) at the indicated days after injection ( n = 5). Mice and dissected tumors were photographed at 28 days after injection (right). (G) Weight of tumors dissected as shown in (F, right). (H) MT1M protein expression in five paired tumors was examined by western blot. M: He3B/Mp V: Hep3B/Vp. To verify that the observed growth inhibitory effect was mediated by the exogenous expression of MT1M, individual Hep3B cell colonies were clonally expanded. Three colonies from vector-transfected cells and three colonies from MT1M - transfected cells were established. MT1M expression was determined by western blot ( Figure 4B ). The Myc-tagged MT1M protein was constitutively expressed, but at variable levels, in the three MT1M-transfected clones (Hep3B/M1, M2 and M3). However, no ectopically expressed MT1M protein was detected in control vector-transfected cells (Hep3B/V1, V2 and V3). To assess the effect of MT1M on HCC cell growth, we combined three individual MT1M clones and control clones, respectively, and the growth rates of the mixed cell populations (Hep3B/Mp versus Hep3B/Vp) was measured by MTS assays. The results showed Hep3B/Mp cells grew much slower than the control Hep3B/Vp cells ( Figure 4C ). BrdU incorporation assay results showed that the DNA synthesis activity in Hep3B/Mp cells was lower than in Hep3B/Vp cells ( Figure 4D ). We also found Hep3B/Mp cells have a lower percentage of cells in S phase and a higher percentage of cells in G1 phase than Hep3B/Vp cells by flow cytometry analysis ( Figure 4E ). Furthermore, there was no difference in spontaneous apoptosis between Hep3B/Mp cells and Hep3B/Vp cells as demonstrated by flow cytometry analysis. These results suggested that MT1M overexpression suppresses HCC cell growth by blocking DNA synthesis and cell-cycle progression but not by inducing spontaneous apoptosis. To further assess the growth inhibitory effect of MT1M, we examined the tumorigenicity of MT1M-transfected Hep3B cells in nude mice. Equal numbers of Hep3B/Mp and Hep3B/Vp cells were injected into the left and right flanks of nude mice, respectively, and followed by measurement of tumor growth at different time points. The in vivo growth of the Hep3B/Mp cells was significantly reduced throughout the observation period relative to Hep3B/Vp cells ( Figure 4F , left). Four weeks after injection, the tumors were removed and weighed. The tumor size ( Figure 4F , right) and weight ( Figure 4G ) was much smaller in the Hep3B/Mp cells in comparison with that in the control Hep3B/Vp cells. Expression of MT1M in tumors grown in mice was confirmed by western blot ( Figure 4H ). Our data indicate that expression of MT1M inhibits growth of HCC cells in vivo . MT1M overexpression represses TNF-α-induced activation of NF-κB NF-κB plays an essential role in tumor growth by protecting cells from apoptosis induced by inflammatory cytokines such as TNF-α ( 13 ). The functional association between MT and NF-κB has been investigated. However, the reported results are contradictory, which is very probably due to the fact that different members of the MT family may function differently ( 14–18 ). To further understand the molecular mechanism by which MT1M promotes HCC cell growth, we examined the effect of MT1M on TNF-α-induced cell death. Hep3B/Mp cells were more sensitive than Hep3B/Vp cells to apoptosis induced by TNF-α ( Figure 5A ). We next examined whether MT1M could inhibit TNF-α-mediated activation of NF-κB using an NF-κB luciferase reporter assay. An approximately 2.5-fold increase in luciferase activity was detected in Hep3B/Vp cells following TNF-α treatment ( Figure 5B ). However, little or no increase in luciferase reporter activity was observed in Hep3B/Mp cells ( Figure 5B ). This result indicates that MT1M overexpression represses TNF-α-induced activation of NF-κB in Hep3B cells. Fig. 5. Open in new tabDownload slide MT1M overexpression represses TNF-α-induced apoptosis and NF-κB activation. (A) Hep3B/Mp and Hep3B/Vp cells were treated with TNF-α (30ng/ml). At 24h after treatment, apoptosis was measured using propidium iodide staining assays. (B) Effect of MT1M overexpression on the activity of the NF-κB dependent luciferase reporter. Hep3B/Vp and Hep3B/Mp cells were transiently transfected with the pNF-κB-Luc luciferase reporter. At 36h after transfection, cells were treated with TNF-α (30ng/ml) for 6h and then assayed for luciferase activities. * indicates statistical significance (* P < 0.01) (C) Effect of MT1M overexpression on TNF-α-induced phosphorylation and degradation of IκBα and phosphorylation of p65. Hep3B/Mp and Hep3B/Vp cells were treated with TNF-α (10ng/ml) for different time periods and then harvested. Protein levels of total and p-IκBα and p-p65 and β-actin were determined by western blot. The blots correspond to one representative experiment and the graph (D) shows the quantification of IκBα, p-IκBα and p-p65 levels using actin for standardization. The error bars represent mean ± standard deviation from three independent experiments. (E) Hep3B/Mp and Hep3B/Vp cells were treated with TNF-α (10ng/ml) for 15min, harvested and lysed, and the nuclei were separated from the cell lysates. Nuclear protein from each sample was incubated with biotin-labeled probes containing consensus NF-κB binding sites in the presence and absence of unlabeled competitive wild-type oligo or mutated oligo. Products of binding reactions were resolved by 6% polyacrylamide gel electrophoresis under non-denaturing conditions. (F) Quantitative RT–PCR analysis of the expression of NF-κB target gene IL-8, ICAM-1 and TRAF2 in Hep3B/Mp and Hep3B/Vp cells treated with TNF-α (30ng/ml) for 8h. A critical step in the regulation of NF-κB activity is phosphorylation and degradation of the IκBα ( 19 ). To determine whether inhibition of NF-κB activation by MT1M was mediated by the inhibition of IκBα degradation, Hep3B/Vp and Hep3B/Mp cells were treated with TNF-α for different periods of time, and the levels of IκBα protein were examined by western blot. We demonstrated that TNF-α decreased the levels of IκBα in Hep3B/Vp cells in a time-dependent manner ( Figure 5C , left panel and Figure 5D ). However, the TNF-α-induced decrease in the levels of the IκBα protein was attenuated in Hep3B/Mp cells ( Figure 5C , right panel and Figure 5D ). Because TNF-α-induced phosphorylation of IκBα leads to its rapid degradation, we examined the TNF-α-induced IκBα Ser32 phosphorylation. TNFα-induced IκBα phosphorylation was also reduced in Hep3B/Mp cells ( Figure 5C and 5D ). Since phosphorylation of p65 at Ser536 residue was suggested as an additional step in the cascade of events leading to NF-κB activation ( 20 ), we also examined the effect of MT1M on TNF-αmediated p65 Ser536 phosphorylation. Similar to IκBα Ser32 phosphorylation, TNF-α-induced Ser536 phosphorylation of p65 was also reduced in Hep3B/Mp cells ( Figure 5C and 5D ). NF-κB activation was also assayed by electrophoretic mobility shift assay. As shown in Figure 5E , compared with Hep3B/Vp cells, the relative density of NF-κB-containing band was greatly reduced in Hep3B/Mp cells when exposed to TNF-α for 15min. To further assess the NF-κB inhibition activity of MT1M, the mRNA levels of NF-κB target genes were analyzed by quantitative RT–PCR. As shown in Figure 5F , TNF-α treatment significantly induced the expression of interleukin-8, ICALM and TRAF2, but this effect was greatly abolished in Hep3B/Mp cells. The above results further suggest that MT1M can inhibit the activation of the NF-κB pathway induced by TNF-α treatment. Discussion In the present study, MT1M was identified and regulation and function of this gene in HCC was characterized. MT1M expression can be induced by heavy metal under physiological conditions and protect cells from heavy metal toxicity. By using both MT1M gene-specific PCR and northern blot with MT1M-specific probe, we demonstrated that MT1M is frequently downregulated in human HCC tumors and cell lines. It has been shown previously with immunohistochemistry that overall expression of MT proteins are downregulated in HCC ( 5–9 ). Due to the high homology among MT proteins and low recognition specificity of MT antibody, the key question as to which members in the MT family account for this downregulation has not been fully addressed. For example, it has been shown by IHC that MT proteins are lost in human HCC samples ( 8 ). By systematical analysis using gene-specific primers for each member of the MT1 family, no change in expression of MT1E and MT1X was detected in the cohort of HCC samples examined in this study. Moreover, we demonstrated for the first time that expression of other MT1 proteins, such as MT1G, and MT1H is also markedly downregulated in HCC samples. Thus, our study confirm previous findings with IHC studies that MT genes are downregulated in HCC, but we provided further evidence that MT1M, MT1G and MT1H are the members in the MT family that are most frequently downregulated in a large cohort of HCC specimens. We demonstrated with methylation-specific PCR that the MT1M promoter is highly methylated in human HCC tumors and cell lines examined. Consistently, we further showed that treatment of HCC cell lines, in which the MT1M is shown to be methylated, with the demethylating agent 5-Aza-dC restored the expression of MT1M. Interestingly, aberrant DNA methylation of MT1M was also identified in a large methyaltion analysis of non-cancerous esophageal mucosae in association with smoking history ( 19 ). By using the same approach, we also demonstrated that MT1E and ME1X promoters are not methylated in HCC cell lines. Our findings confirm that promoter methylation is one of the important mechanisms that lead to downregulation of MT proteins such as MT1M and MT1F in human HCC. In agreement with low expression of MT1M and methylation of MT1M promoter in human HCCs, we further showed that restored expression of MT1M in MT1M-deficient HCC cells blocks HCC cell growth in culture and HCC tumors in mice. These findings support the notion that MT1M is a novel tumor suppressor in human HCCs. NF-κB is a transcription factor that plays important roles in cell proliferation, survival and immunity against viral infections. The role of NF-κB in HCC is not understood fully. Tai et al. ( 21 ) and Chiao et al. ( 22 ) reported that NF-κB was constitutively activated in HCC. Increasing evidence suggests that upregulation of NF-κB and downregulation of MTs are correlated with the growth of HCC cells. In the present study, we demonstrated that restored expression of MT1M in HCC cells lacking endogenous MT1M not only diminished phosphorylation and degradation of IκBα but also markedly inhibited NF-κB activity in HCC cells treated with TNFα. Taken together, the results from this study and previous reports suggest that the loss of expression of MTs in HCC may play a critical role in HCC carcinogenesis via abnormal activation of the NF-κB pathway. We further show that restored expression of MT1M in HCC cells that lack endogenous MT1M decrease TNF-α-induced activation of NF-κB and inhibits HCC cell growth in vitro and in vivo . Therefore, MT1M may play an important role in liver tumorigenesis due to its frequent downregulation in HCC, which may lead to aberrant activation of the NF-κB pathway. Supplementary material Supplementary Materials and Methods, Figure1 and Tables 1–3 can be found at Supplementary Data Funding National Key Sci-Tech Special Project of China (2008ZX10002-020, 2008ZX10002-021); National Natural Science Foundation of China (30872947). Conflict of Interest Statement: None declared. Abbreviations: References 1. Suzuki Y.J. et al. 1997 Oxidants as stimulators of signal transduction. Free Radic. Biol. Med. 22 269 – 285 Google Scholar Crossref Search ADS PubMed WorldCat 2. Theocharis S.E. et al. 2004 Metallothionein expression in human neoplasia. Histopathology 45 103 – 118 Google Scholar Crossref Search ADS PubMed WorldCat 3. West A.K. et al. 1990 Human metallothionein genes: structure of the functional locus at 16q13. Genomics 8 513 – 518 Google Scholar Crossref Search ADS PubMed WorldCat 4. Stennard F.A. et al. 1994 Characterisation of six additional human metallothionein genes. Biochim. Biophys. Acta 1218 357 – 365 Google Scholar Crossref Search ADS PubMed WorldCat 5. Cai L. et al. 1998 Metallothionein and apoptosis in primary human hepatocellular carcinoma (HCC) from northern China. Anticancer Res. 18 ( 6B ) 4667 – 4672 Google Scholar PubMed OpenURL Placeholder Text WorldCat 6. Datta J. et al. 2007 Metallothionein expression is suppressed in primary human hepatocellular carcinomas and is mediated through inactivation of CCAAT/enhancer binding protein alpha by phosphatidylinositol 3-kinase signaling cascade. Cancer Res. 67 2736 – 2746 Google Scholar Crossref Search ADS PubMed WorldCat 7. Huang G.W. et al. 2002 Metallothionein expression in hepatocellular carcinoma. World J. Gastroenterol. 8 650 – 653 Google Scholar Crossref Search ADS PubMed WorldCat 8. Tao X. et al. 2007 Downregulated expression of metallothionein and its clinicopathological significance in hepatocellular carcinoma. Hepatol. Res. 37 820 – 827 Google Scholar Crossref Search ADS PubMed WorldCat 9. Lu D.D. et al. 2003 The relationship between metallothionein-1F (MT1F) gene and hepatocellular carcinoma. Yale J. Biol. Med. 76 55 – 62 Google Scholar PubMed OpenURL Placeholder Text WorldCat 10. Majumder S. et al. 2010 Loss of metallothionein predisposes mice to diethylnitrosamine-induced hepatocarcinogenesis by activating NF-kappaB target genes. Cancer Res. 70 10265 – 10276 Google Scholar Crossref Search ADS PubMed WorldCat 11. Zheng W.J. et al. 2004 Expression of human metallothionein III and its metalloabsorption capability in Escherichia coli. Prep. Biochem. Biotechnol. 34 265 – 278 Google Scholar Crossref Search ADS PubMed WorldCat 12. Bird A. 1999 Molecular biology—DNA methylation de novo Science 286 2287 – 2288 Google Scholar Crossref Search ADS PubMed WorldCat 13. Karin M. 2006 Nuclear factor-kappaB in cancer development and progression. Nature 441 431 – 436 Google Scholar Crossref Search ADS PubMed WorldCat 14. Abdel-Mageed A.B. et al. 1998 Activation of nuclear factor kappaB: potential role in metallothionein-mediated mitogenic response. Cancer Res. 58 2335 – 2338 Google Scholar PubMed OpenURL Placeholder Text WorldCat 15. Butcher H.L. et al. 2004 Metallothionein mediates the level and activity of nuclear factor kappa B in murine fibroblasts. J. Pharmacol. Exp. Ther. 310 589 – 598 Google Scholar Crossref Search ADS PubMed WorldCat 16. Kanekiyo M. et al. 2001 Zinc-induced activation of the human cytomegalovirus major immediate-early promoter is mediated by metallothionein and nuclear factor-kappaB. Toxicol. Appl. Pharmacol. 173 146 – 153 Google Scholar Crossref Search ADS PubMed WorldCat 17. Sakurai A. et al. 1999 Regulatory role of metallothionein in NF-kappaB activation. FEBS Lett. 455 55 – 58 Google Scholar Crossref Search ADS PubMed WorldCat 18. Crowthers K.C. et al. 2000 Augmented humoral immune function in metallothionein-null mice. Toxicol. Appl. Pharmacol. 166 161 – 172 Google Scholar Crossref Search ADS PubMed WorldCat 19. Oka D. et al. 2009 The presence of aberrant DNA methylation in noncancerous esophageal mucosae in association with smoking history: a target for risk diagnosis and prevention of esophageal cancers. Cancer 115 3412 – 3426 Google Scholar Crossref Search ADS PubMed WorldCat 20. Ghosh S. et al. 2002 Missing pieces in the NF-kappaB puzzle. Cell 109 Suppl S81 – S96 Google Scholar PubMed OpenURL Placeholder Text WorldCat 21. Tai D.I. et al. 2000 Constitutive activation of nuclear factor kappaB in hepatocellular carcinoma. Cancer 89 2274 – 2281 Google Scholar Crossref Search ADS PubMed WorldCat 22. Chiao P.J. et al. 2002 Role of Rel/NF-kappaB transcription factors in apoptosis of human hepatocellular carcinoma cells. Cancer 95 1696 – 1705 Google Scholar Crossref Search ADS PubMed WorldCat Author notes The sequence of MT1M reported in this paper has been deposited in GenBank Database with the accession number AF136177, AAL83902. © The Author 2012. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com © The Author 2012. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com
TI - Metallothionein MT1M is a tumor suppressor of human hepatocellular carcinomas
JF - Carcinogenesis
DO - 10.1093/carcin/bgs287
DA - 2012-12-01
UR - https://www.deepdyve.com/lp/oxford-university-press/metallothionein-mt1m-is-a-tumor-suppressor-of-human-hepatocellular-XKEarp24fa
SP - 2568
EP - 2577
VL - 33
IS - 12
DP - DeepDyve
ER -