Dysregulation of the TGFBI gene is involved in the oncogenic activity of the nonsense mutation of hepatitis B virus surface gene sW182*
Shih Sheng Jiang1*, Shiu-Feng Huang2,3*, Min-Syuan Huang2, Yng-Tay Chen2,4, Hsiang-Ju Jhong2, Il-Chi Chang2, Ya-Ting Chen2, Jer-Wei Chang1, Wen-Ling Chen1, Wei-Chen Lee5, Miin-Fu Chen5, Chau-Ting Yeh6, Isao Matsuura2
Running title: HBV sW182* mutant and TGFBI *These two authors have equal contributions.
1National Institute of Cancer Research, National Health Research Institutes, Zhunan, Taiwan
2Institute of Molecular and Genomic Medicine, National Health Research Institutes, Zhunan, Taiwan
3Department of Pathology, Chang Gung Memorial Hospital at Linko, Taoyuan, Chang Gung University College of Medicine, Linko,Taiwan
4Present address: Human Genetic Center, China Medical University Hospital, Taichung, Taiwan
5Department of Surgery, Chang Gung Memorial Hospital at Linko, Taoyuan, Chang Gung University College of Medicine, Linko, Taiwan
6Liver Research Center, Chang Gung Memorial Hospital at Linko, Chang Gung University College of Medicine, Taoyuan, Taiwan
Corresponding Author for reprint: Isao Matsuura, Ph.D.
Institute of Molecular and Genomic Medicine, National Health Research Institutes,
35 Keyan Road, Zhunan, Miaoli, 350, Taiwan
TEL: 886-37-246166, ext.35369 FAX: 886-37-586459
Electronic word count: 4992 words
List of abbreviations in the order of appearance: Hepatitis B virus (HBV)
HBV surface (S) gene
Transforming growth factor--induced (TGFBI) gene Hepatocellular carcinoma (HCC)
Hepatitis B core antigen (HBcAg) Gene set enrichment analysis (GSEA)
Quantitative polymerase chain reaction (qPCR)
Cyclic AMP-responsive element binding protein (CREB), Combined bisulfite restriction analysis (COBRA)
5-azacytidine (5-Aza)
Green fluorescent protein (GFP) DNA methyltransferase (DNMT)
Conflict of interest:
The authors who have contribution to this study declared that they do not have anything to disclose regarding funding or conflict of interest in this report.
Financial support:
This work was supported by grants from National Health Research Institutes (NHRI -A1-093-PP-04, NHRI -A1-096-PP-04, NHRI-A1-98-PP-04, NHRI-A1- MG-100-PP-04, NHRI-A1-MG-100-PP-09, NHRI-A1-MG-101-PP-09, NHRI-A1-MG-102-PP-09) and National Science Council, Taiwan (NSC92-2320-B- 400-002, NSC97-2320-B-400-006-MY3, NSC102-2319-B-4-001)
Abstract
Background & Aims: The nonsense mutations of hepatitis B virus (HBV) surface (S) gene have been reported to induce transcriptional transactivation activity. We have previously identified several transforming nonsense mutations of HBV S gene and the sW182* mutant had the most potent oncogenicity. This study aimed to understand the molecular mechanisms leading to the oncogenic activity of the sW182* mutant. Methods: NIH3T3 cells were stably transfected with plasmids encoding the wild type PreS/S gene or the nonsense mutant (sW182*) of HBV PreS/S gene. Gene expression microarray analyses were performed to identify genes that are differentially regulated in the mutant clones. Additional functional study was performed on one of the differentially expressed genes, transforming growth factor--induced (TGFBI) gene, and its expression was analyzed in human hepatocellular carcinoma (HCC) tissues.
Results: Compared with the expression patterns of wild type clones, the sW182* mutant clones were significantly enriched by gene sets associated with cell cycle regulation, DNA repair, and genome instability. The TGFBI gene was downregulated in the sW182* clones, and irresponsive to TGF- treatment. The level of cyclin D1, a negatively regulated TGFBI target, was highly elevated in the sW182* mutant cells. Exogenous expression of TGFBI alleviated the oncogenic activity of sW182* in mouse xenograft model. In human HBV-related HCC cancerous tissue, expression of TGFBI was downregulated in 25 of the 55 (45%) patients.
Conclusions: Dysregulation of the TGFBI gene is involved in the oncogenic activity of the sW182* mutant of hepatitis B virus S gene. This has never been described before.
Electronic word count: 250 words
Key words: Nonsense mutation, HBV PreS/S gene, HCC, TGFBI, cyclin-D1, hepatocarcinogenesis
Introduction
Hepatocellular carcinoma (HCC) is one of the major cancers in the world and among the leading causes of cancer death [1]. Although the association between chronic hepatitis B virus (HBV) infection and HCC has been well-established [2-4], whether HBV could play a direct role in hepatocarcinogenesis remains uncertain. In most of the HBV-related HCC patients, the development of liver cancer is a long-term process, and liver cirrhosis precedes the cancer. In some patients, however, HCC development without cirrhosis has been reported suggesting that viral factor(s) alone could also play a direct role in
hepatocarcinogenesis without invoking chronic liver inflammation.
Nonsense mutations of HBV PreS/S gene have been reported to induce transcriptional transactivation activity previously, suggesting these mutations may contribute to HBV-associated oncogenesis [5-6]. The PreS/S gene has one long open reading frame with three in-frame “start” (ATG) codons. Accordingly, the gene is divided into three
sections: pre-S1, pre-S2, and S, and three proteins with different sizes are produced: large S, middle S, and small S by translation from pre-S1 + pre-S2 + S, pre-S2 + S, and S, respectively. Previously, we have identified multiple nonsense mutations of HBV S gene in HBV-related HCC patients, who developed HCC after lamivudine treatment [7, 8]. Our studies demonstrated that such nonsense S mutants could transactivate oncogene promoters and increase tumorigenicity in xenograft model using the construct of the whole PreS/S gene with the nonsense mutation of S gene.
Furthermore, we identified 25 HBV-related HCC patients, who were positive for hepatitis B core antigen (HBcAg) in their HCC tissue by immunohistochemical stain, suggesting
the presence of replicative HBV in the cancerous part. We performed whole HBV genome mutation analyses in these 25 paired HCC tissue samples, and identified a recurrent nonsense mutation at the codon for tryptophane182 (sW182*) in the cancerous part of the HCC tissues (Huang SF, unpublished data). Functional studies revealed that this mutant had potent oncogenic activity by mouse xenograft study and in vitro
migration assay (Huang SF, unpublished data). Recently, Lee et al. also reported the presence of the sW182* mutation in HBV patients in Korea. They also demonstrated the oncogenic potential of the sW182* mutant [9].
We have established NIH3T3 cell lines stably transfected with wild type PreS/S gene or PreS/S gene with the sW182* mutation, and performed gene expression microarray analysis to search for candidate genes involved in hepartocarcinogenesis. Marked downregulation of transforming growth factor--induced (TGFBI) gene in the sW182* mutant clones was identified. Thus, we have performed a series of functional analyses to clarify the significance of TGFBI downredulation in the oncogenic activity of the
sW182* mutant.
TGFBI, also known as ig-H3, was originally discovered as a TGF--inducible gene in human adenocarcinoma cell line [10]. Its association with corneal dystrophy has been widely accepted [11]. TGFBI has also been reported to play a role in a wide variety of pathological conditions including tumorigenesis. Loss or reduced expression of TGFBI has been associated with lung, breast and ovarian cancers [12-14]. Moreover, TGFBI- deficient mice develop various kinds of tumor including liver tumor, which was
accompanied by upregulation of cyclin D1 [15]. On the other hand, a significant number of reports have described that TGFBI may have a tumor promoter function as well
[14,16-18]. It has been suggested that this functional difference of TGFBI may depend on the tumor microenvironment[14]. High expression of TGFBI seems to relate to more aggressive tumor [17,18], whereas the low level of TGFBI appears to cause tumor initiation [15].
This study provides for the first time a link between the nonsense mutation of HBV S gene and TGFBI pathway in the hepatocarcinogenesis.
Materials and Methods
Cell culture and Constructs
NIH3T3 cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% calf serum, 2mM glutamine, 100 unit/ml penicillin and 100g/ml streptomycin. The vector pIRESbleo (Addgene, Cambridge, MA, USA) was used for the construct. Transfection with plasmids containing wild type of the whole PreS/S gene and PreS/S gene with the sW182* mutation, respectively, were performed using Lipofectamine 2000 reagent (Invitrogen, Carlsbad, CA, USA). After Zeocin (Invitrogen, Carlsbad, CA, USA) selection, several stable cell clones were isolated for “vector”, “wild type” and “mutant”, respectively. These cells were maintained with occasional treatments with 0.1mg/ml Zeocin.
Human TGFBI cDNA was cloned in pBABEpuro vector (Addgene, Cambridge, MA, USA) by reverse transcriptase polymerase chain reaction (RT-PCR) from HaCaT human keratinocytes. The sW182* mutant cell lines (clone 6) were retrovirally transduced with this construct or pBABE-GFP (control) as described previously [19].
The reporter plasmid for the TGFBI promoter was constructed as follows: A fragments containing 1000 bp upstream of the TGFBI transcription initiation site were generated by PCR from genomic DNA from HaCaT cells. The primer sequences are as follows: GAA TGG TAC CCT TCA TGG AAC ATC ATT GGC TTG GG-3’ (forward) and 5’-GTT TAA GCT TGG AGC GGG ACG ACG CGC ACC-3’ (reverse). KpnI (forward) and HindIII (reverse) sites (underlined) were introduced in the primers for cloning into pGL3-Basic vector (Promega, Madison, WI).
Gene expression microarray profiling and pathway/gene set analysis
Gene expression microarray profiling experiments were previously described [20]. The
microarray data has been deposited in GEO with accession number GSE 50180. Gene set enrichment analysis (GSEA) [21] was used to obtain molecular pathways or annotated gene sets enriched in genes differentially expressed between wild type and the sW182* mutant stable clones. Alternatively, this GSEA was applied to analyze
pathways/gene sets enriched in genes significantly correlated with expression levels of TGFBI mRNA among tested cell lines.
Quantitative polymerase chain reaction (qPCR) for TGFBI
Cells were serum-starved (0.2 % FBS) for 16 h before treated (or untreated) with 0.5 nM TGF- (PeproTech Asia, Rehovot, Israel) for 7 h. Total RNA was prepared by using TRIzol (Invitrogen, Carlsbad, CA, USA). The first strand of cDNA was synthesized from 1 μg total RNA using High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA). Q-PCR was run on CFX96 Real-Time System (Bio-Rad, Berkeley, CA, USA) with KAPA SYBR FAST qPCR kit (KAPA Biosystems, Cape Town, South Africa). GAPDH gene was used as an internal standard for normalization.
Control, wild type and mutant cells were transfected with the reporter construct along with a pRL-TK Renilla reporter. Cells were treated with 0.5 nM TGF- for 24 hours in low serum medium (0.2% FBS). Luciferase activity was measured using the Dual-Luciferase Reporter Assay System (Promega, Madison, WI).
Antibodies, Western blot, and immunofluorescence microscopy
Antibodies used in the study were purchased from following suppliers: cyclin D1 [EPR2241], CREB [E306], phospho-CREB (pSer133) [E113], phospho-Smad3
(pSer423/425) [EP8235], GeneTex (San Antonio, TX, USA); actin, Santa Cruz (Dallas, TX, USA); TGFBI (k1H12), Smad3 [MO5], Abnova (Walnut, CA, USA); -tubulin, Epitomics (Burlingame, CA, USA). Horseradish peroxidase-conjugated secondary antibodies were from Pierce (Rockford, IL, USA). For the protein lysate preparation, cells were cultured in low serum medium for 24 h. Then cells were lysed or further treated with 0.5 nM TGF- for the various time periods before lysed. Western blot procedure was previously described [19].
For immunofluorescence microscopy, cells with 50-60% confluence were fixed and incubated with an -tubulin antibody (1:100). -Tubulin was visualized by Alexa Flour 594 goat anti-rabbit IgG (Life Technologies, Carlsbad, CA, USA). Nuclei were stained with DAPI (Sigma-Aldrich, St. Louis, MO, USA).
Cells were treated or untreated with TGF- for 7 h in low serum medium. After treatment, genomic DNA was prepared by QIAamp mini kit (QIAGEN, Germantown, MD, USA). Bisulfite conversion of genomic DNA (500 pM) was performed using the EZ DNA Methylation-Gold Kit (Zymo Research, Irvine, CA, USA). Combined bisulfite restriction analysis (COBRA) primer-amplified PCR products were purified using DNA Clean & Concentrator-5 (Zymo Research, Irvine, CA, USA), digested by BstUI (New England BioLabs, Ipswich, MA, USA) and separated on a 3% agarose gel.
The primers used for COBRA were designed based on ‘‘The Li Lab’’ program (http://www.urogene.org/methprimer/). The primers used were as follows:
5’-TGAATATTTTAAGAAATTGAAATTTGAG-3’, 5’-AAACCCACAAACAATACCAAAAC-3’.
A 320 bp PCR product amplified from unmethylated template that was resistant to the restriction enzyme (BstUI) was indicated as “unmethylated” (Fig.3). For DNA
demethylation, cells were treated with10 M 5-azacytidine (5-Aza) Sigma-Aldrich, St. Louis, MO, USA) for 4 days before subjected to qPCR analysis. 5-Aza was present throughout the experiment.
Mouse xenograft assay
Nude mice were obtained from National Laboratory Animal Center, Taiwan. Cells were grown to ~60% confluence before harvested and mixed with matrigel (BD
the cell lines of vector, wild type and mutant, respectively (1 x 106 cells/injection). At the end of the experiment, the tumor was extracted and weighed. The p value was obtained from Students t-test. All animals in this experiment received human care. The study protocol complied with our institution's guideline and has been approved by the Animal Research Ethics Committee at National Health Research Institutes, Taiwan (IACUC Approval No. 100077).
Analysis of TGFBI expression in human HCC tissue by qPCR
Fresh frozen tumor tissues and paired benign liver tissue of 55 HBV-related HCC
patients, who received surgical operation in Chang-Gung Memorial Hospital during year 2000 to 2002 and with signed informed consent, were obtained from the tumor bank of CGMH for the HBV-related carcinogenesis studies. The patient’s clinical data were obtained from medical records. This protocol was approved by the Institutes of Reviewing Board of Chang- Gung Memorial Hospital. RNA was extracted from the fresh frozen tissues by Trizol and subjected to qPCR for TGFBI expression.
Nucleotide sequence of HBV PreS/S gene
For sequencing analysis, we used 3 pairs of primer to cover the preS/S gene of HBV. For amplification, we used the following 3 pairs of primers:
preSF 5’-GAGAGTCCACACGTAGCG-3’, preSR 5’-GACTCTGTGGTATTGTGAGG-3’,
SR 5’-GTTCCTGTGGCAATGTGCC-3’, preSF2 5’-CTCACAACTGTGCCAGCAGC-3’,
SR2 5’-CCTTGATAGTCCAGAAGAACC-3’.
The PCR amplicons were subjected to direct sequencing on an ABI3730 genetic analyzer (Applied Biosystems, Foster City, CA). Sequence variations were determined by using Seqscape software (Applied Biosystems, Foster City, CA, USA). The HBV reference sequence was: AY167089 for genotype B, and AY167095 for genotype C (National Center for Biotechnology Information).
Results
Gene expression microarray profiling
A total of 349 gene probes were found to have at least 2-fold (average) differential expression between cells of mutant and wild type groups. In order to know what
biological pathways or curated gene sets could be involved in the phenotypic differences between wild type and mutant clones, we performed gene set enrichment analysis
(GSEA). According to GSEA, the differentially expressed genes between two groups were significantly enriched by gene sets associated with a wide variety of cancer-related pathways. For examples, many genes involved in regulation of mitotic cell cycle or mismatch repair were significantly downregulated in sW182* mutant group (Fig. 1A-B). Likewise, gene set involved in genome instability [22] was also significantly
downregulated in the mutant cells (Fig 1C). One thousand permutations and statistics of “difference of classes” were used. In the analysis we noted that the TGFBI gene was significantly downregulated in the cells bearing the sW182* mutant (Fig. 1D).
Dysregulation of TGFBI in the sW182* mutant cells
To validate the result above, levels of the TGFBI gene as well as protein were examined in detail by means of qPCR, reporter assay and Western blot. Fig. 2A showed the level of Smad5, which was used as a control, was comparable among the three cell lines. Strikingly, the TGFBI gene was very low in the mutant cells. Furthermore, whereas the gene was responded to TGF- in vector control and wild type cells, the response was lost in the mutant cells. All independent clones tested showed the similar result (Fig. 2B),
indicating that the observed TGFBI dysregulation in the sW182* mutant cells was not due to clonal variation. Vector clone 4, wild type clone 3 and mutant clone 6,
respectively, were used as representative clones in this study unless otherwise indicated. In a reporter assay with a TGFBI promoter construct, luciferase activity was induced by TGF- in control vector and wild type cells, but not in the mutant cells (Fig. 2C). This result was consistent with that of qPCR (Fig. 2B). The TGFBI protein levels in vector and wild type cells was increased in response to TGF- treatment, but remained low in the mutant cells (Fig.2D). Smad3 was equally phosphorylated in each cell line in
response to TGF-, indicating that TGF- signaling per se was not affected by wild type PreS/S or the mutant (Fig.2E). These results together implied that the TGFBI gene and protein were rather specifically dysregulated in the sW182* mutant cells.
Hypermethylation of the TGFBI promoter in the sW182* mutant cells
Promoter hypermethylation is one of the important mechanisms for the TGFBI gene silencing in cancer cells [14,23]. To test this possibility, methylation status of the TGFBI promoter region was analyzed by COBRA (Fig. 3A). In the putative CpG island of the promoter, methylation was detected in each cell line. A significant amount of the unmethylated form also existed in the vector control and wild type cells. However, the unmethylated form was absent in the sW182* mutant cells, indicating hypermethylation of the promoter in mutant cells. When the cells were treated with a demethylation agent (5-Aza), it resulted in an increase of the TGFBI transcription level by 5-fold in the mutant, whereas the drug had no effect on control vector or wild type cells (Fig. 3B).
Increased expression of cyclin D1 in the sW182* mutant cells
The level of cyclin D1, The level of cyclin D1, a protein negatively regulated by TGFBI, was found to be highly elevated in mutant cells (Fig. 4A), in which the TGFBI level was dysregulated. cyclin D1 is regulated through various signaling and transcription factors. In TGFBI deficient mice, the high cyclin D1 level is accompanied by aberrant activation of cyclic AMP-responsive element binding protein (CREB), one of the transcription factors regulating cyclin D1[15]. Consistent with this observation, an elevated CREB activation, judged by its phosphorylation level, was also observed in the sW182* mutant cells (Fig. 4A).
Abnormal mitosis in the sW182* mutant cells
TGFBI has been described in microtubule stabilization, and loss of TGFBI can induce mitotic abnormalities [24]. Immunofluorescence staining with -tubulin was performed to reveal microtubule structures and nuclear morphology. A significantly higher
proportion of abnormal cells with multinuclei was observed in the mutant cells compared to wild type cells (mutant 11%, wild type 0.8%, n = 500) (Fig. 4B). In addition, abnormal mitotic figures with multipolar spindle were frequently found in cells harboring the sW182* mutant (16.7% of mitotic cells), but rare for wild type cells (Fig. 4B).
Tumorigenic activity of the sW182* mutants and its inhibition by exogenous TGFBI
To test the tumorigenic activity, vector, wild type and mutant cells, respectively, were xenografted into mice. All the 5 nude mice subcutaneously injected with the sW182*
cells had tumor formations 4 weeks after injection, whereas those with wild type or vector control cells did not (Fig. 4C). To show that these effects by the mutant were attributed to dysregulated TGFBI, the protein was exogenously expressed by retroviral transduction. Green fluorescent protein (GFP) was also transduced as a control. The exogenous TGFBI significantly reduced cyclin D1and pCREB demonstrated by Western blot (Supplementary Fig. 1). Four nude mice were subcutaneously injected with the sW182* cells with GFP (control) or TGFBI, respectively, and the resulted tumors were compared. Both GFP- and TGFBI-transduced mutant cells had tumor formation with 100% incidence, but the tumor size was significantly reduced with TGFBI-transduced cells (Fig. 4D).
TGFBI expression in human HCC tissue
Among the 55 HBV-related paired HCC and non-tumor liver tissue samples, twenty-five tumors (25/55, 45.45%) displayed downregulated TGFBI in the tumor compared to the paired benign liver tissue (Supplementary Fig. 2). Clinicopathological characteristics of the 55 patients were shown in Table 1. Downregulation of TGFBI was significantly associated with higher tumor stage (p=0.0323). Sequencing of HBV PreS/S gene revealed 6 patients with the sW182* mutation in the cancerous part of the liver tissue. Among these 6 patients, the level of TGFBI expression was downregulated in 3 (3/6, 50%).
Discussion
We have previously identified 4 different nonsense mutations of HBV S gene with transformation activity: sL21*, sW156*, and sW172* [7,8], and sW182* (Huang SF, unpublished data) in HBV-related HCCs, suggesting these mutations may play an important role in hepatocarcinogenesis. But the underlying mechanism remained uncertain. In the current study, we have identified TGFBI as a target gene by the gene expression microarray study. TGFBI was validated to be indeed dysregulated in the cells containing the sW182* mutant, but not in wild type PreS/S gene (Fig. 2). This study provides for the first time a link between HBV PreS/S gene and TGFBI pathway. Silencing of tumor suppressor genes is thought to be one of the major causes of cancer. As a tumor suppressor, the TGFBI gene is often silenced in many types of cancers, and hypermethylation is a major mechanism for inactivation of the promoter[14,23,25,26]. Our result clearly showed that the sW182* mutant could induce hypermethylation of the TGFBI promoter (Fig. 3). Certain factors have been reported to bring DNA
methyltransferases (DNMTs) to promoter region of target genes [27,28]. It is possible that the sW182* mutant could gain similar ability to that of those factors, although products of PreS/S gene or its mutants have never been reported to directly bind to DNA or a DNA methyltransferase. HBV X gene has been shown to promote hypermethylation of certain gene promoters such as p16INK4A [29,30]. In such a case, HBx transcriptionally upregulates DNMT1 through activation of cyclin D1-CDK4/6-Rb-E2F pathway [29]. In the sW182* cells, this pathway might also be activated due to the elevated cyclin D1 expression. If so, it is possible that, in the mutant cells, the TGFBI promoter
hypermethylation might be achieved by a mechanism similar to that seen with HBx. Further study is necessary for delineation of the mechanism.
Zhang et al. reported that TGFBI deficiency increased the incidence of spontaneous tumor formation, including liver tumor, through upregulation of cyclin D1 [15]. Their study implied that TGFBI deficiency per se could induce tumor and that it also increased sensitivity to carcinogen. This implication is reasonable, as cyclin D1 is a well-known proto-oncogene [30,31]. Our observation is reminiscent of their report (Fig. 4A and C). TGFBI could also induce microtubule stabilization, and loss of TGFBI would result in mitotic splindle abnormalities [24]. This was exactly what we found in the sW182* mutant cells (Fig. 4B). Taken together, our results suggested that the sW182* mutant was sufficient to induce mitotic abnormalities, which could be due to downregulation of
TGFBI.
We also have checked the TGFBI expression in HBV-related HCC tissue by qPCR and near half of the tumors (25/55, 45.45%) showed downregulation of TGFBI compared with paired benign liver tissue. In addition, the sW182* mutation was identified in 6 HCCs, and half of the above 6 tumors had downregulated TGFBI, which was similar to the conditions in our mutant cell lines.
In this study, forced expression of TGFBI alleviated, but did not abolish tumor formation by the sW182* mutant (Fig.4 E). This suggests that the oncogenic activity of the mutant involves multiple mechanisms. Our microarray data support this assumption (Fig. 1A-C). Three gene sets including cell cycle regulation, DNA repair, and genome instability were all differentially regulated in the sW182* mutant cells compared with wild type cells.
In summary, the present study has provided the evidence that the nonsense mutation (sW182*) of HBV S gene could directly induce tumor growth and dysregulation of
TGFBI is involved in this oncogenic process. This will provide new insights into the
Acknowledgements
References
[1] Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global cancer statistics. CA Cancer J Clin 2011; 61: 69-90.
[2] Breasley, R. P. Hepatitis B virus: The major etiology of hepatocellular carcinoma. Cancer 1988; 61: 1942-1956.
[3] Feitelson M. Hepatitis B virus infection and primary hepatocellular carcinoma. Clin Microbiol Rev 1992; 5: 275-301.
[4] Robinson WS. Molecular events in the pathogenesis of hepadnavirus- associated hepatocellular carcinoma. Annu Rev Med 1994; 45: 297-323.
[5] Kekulé AS, Lauer U, Meyer M, Caselmann WH, Hofschneider PH, Koshy R. The pre-S2/S region of integrated hepatitis B virus encodes a transcriptional
transactivator. Nature 1990; 343: 457-461.
[6] Lauer U, Weiss L, Hofschneider PH, Kekulé AS. The hepatitis B virus pre-S/S(t) transactivator is generated by 3' truncations within a defined region of the S gene. J Virol 1992; 66: 5284-5289.
[7] Lai MW, Huang SF, Hsu CW, Chang MH, Liaw YF, Yeh CT. Identification of nonsense mutations in hepatitis B virus S gene in patients with hepatocellular carcinoma developed after lamivudine therapy. Antivir Ther 2009; 14: 249-261. [8] Yeh CT, Chen T, Hsu CW, Chen YC, Lai MW, Liang KH, et al. Emergence
of the rtA181T/sW172* mutant increased the risk of hepatoma occurrence in patients with lamivudine-resistant chronic hepatitis B. BMC Cancer 2011; 11: 398 doi: 10.1186/1471-2407-11-398
[9] Lee SA, Kim K, Kim H, Kim BJ. Nucleotide change of codon 182 in the surface gene of hepatitis B virus genotype C leading to truncated surface protein is associated with progression of liver diseases. J Heaptology 2012; 56: 63-69. [10] Skonier J, Neubauer M, Madisen L, Bennett K, Plowman GD, Purchio AF. cDNA
cloning and sequence analysis of beta ig-h3, a novel gene induced in a human adenocarcinoma cell line after treatment with transforming growth factor-beta. DNA Cell Biol 1992; 11: 511-522.
[11] Klintworth GK. The molecular genetics of the corneal dystrophies--current status. Font Biosci 2003; 1: 687-713.
[12] Zhao Y, El-Cabry M, Hei TK. Loss of Betaig-h3 protein is frequent in primary lung carcinoma and related to tumorigenic phenotype in lung cancer cells. Mol Carcinog 2006; 45: 84-92.
[13] Calaf G, Echiburu-Chau C, Zhao YL, Hei TK. BigH3 protein expression as a marker for breast cancer. Int J Mol Med 2008; 21: 561-568.
[14] Ween MP, Oehler MK, Ricciardelli C. Transforming growth factor-beta-induced protein (TGFBI)/(ig-H3): A matrix protein with dual functions in ovarian cancer. Int J Cancer 2012; 13: 10461-10477.
[15] Zhang Y, Wen G, Shao G, Wang G, Lin C, Fang H, et al. TGFBI deficiency predisposes mice to spontaneous tumor development. Cancer Res 2007; 69: 37-44. [16] Lin B, Madan A, Yoon JG, Fang X, Yan X, Kim TK, et al. Massively parallel signature sequencing and bioinformatics analysis identifies upregulation of TGFBI and SOX4 in human glioblastoma. PLos One 2010; 5: e10210.
matrix protein betaig-h3/TGFBI promotes metastasis of colon cancer by enhancing cell extravasation. Genes Dev 2008; 22: 308-321.
[18] Zajchowski DA, Bartholdi MF, Gong Y, Webster L, Liu HL, Munishkin A,
et al. Identification of gene expression profiles that predict the aggressive behavior
of breast cancer cells. Cancer Res 2001; 61: 5168-5178.
[19] Matsuura I, Lai CY, Chiang KN. Functional interaction between Smad3 and S100A4 (metastatin-1) for TGF--mediated cancr cell invasiveness. Biochem J 2010; 426: 327-335.
[20] Jiang SS, Fang WT, Hou YH, Huang SF, Yen BL, Chang JL, et al. Upregulation of SOX9 in lung adenocarcinoma and its involvement in the regulation of cell growth and tumorigenicity. Clin Cancer Res 2010; 16: 4363-4373.
[21] Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A 2005; 102: 15550.
[22] Boyault S, Rickman DS, de Reyniès A, Balabaud C, Rebouissou S, Jeannot E, et al. Transcriptome classification of HCC is related to gene alterations and to new
therapeutic targets. Hepatology 2007; 45: 42-52
[23] Shao G, Berenguer J, Borczuk AC, Powell CA, Hei TK, Zhao Y. Epigenetic inactivation of betaig-h3 gene in human cancer cells. Cancer Res 2006; 66: 4573.
extracellular matrix protein TGFBI induces microtubule stabilization and sensitizes ovarian cancers to paclitaxel. Cancer Cell 2007; 12: 514-527.
[25] Shah JN, Shao G, Hei TK, Zhao Y. Methylation screening of the TGFBI promoter in human lung and prostate cancer by methylation-specific PCR. BMC Cancer 2008; 8: 284 doi:10.1186/1471-2407-8-284
[26] Kang S, Dong SM, Park NH. Frequent promoter hypermethylation of TGFBI in epithelial ovarian cancer. Gynecol Oncol 2010; 118: 58-63.
[27] Suzuki M, Yamada T, Kihara-Negishi F, Sakurai T, Hara E, Tenen DG, et al. Site specific DNA methylation by a complex of PU.1 and DNmt3a/b. Oncogene 2006; 25: 2477-2488.
[28] Tachibana M, Matsumura Y, Fukuda M, Kimura H, Shinkai Y. G9a/GLP complexes independently mediate H3K9 and DNA methylation to silence transcription. EMBO J 2008; 27: 2681-2690.
[29] Jung JK, Arora P, Pagano JS, Jang KL. Expression of DNA methyltransferase 1 is activated by hepatitis B virus X protein via a regulatory circuit involving the p16INK4a-cyclin D1-CDK 4/6-pRb-E2F1 pathway. Cancer Res 2007; 67: 5778.
[30] Zhu YZ, Zhu R, Shi LG, Mao Y, Zheng GJ, Chen Q, et al. Hepatitis B virus X protein promotes hypermethylation of p16 (INK4A) promoter through upregulation of DNA methyltransferases in hepatocarcinogenesis. Exp Mol Pathol 2010; 89: 268-275.
[31] Wang TC, Cardiff RD, Zukerberg L, Lees E, Arnold A, Schmidt EV. Mammary hyperplasia and carcinoma in MMTV-cyclin D1 transgenic mice. Nature 1994;
Figure legends
Fig. 1. Downregulated gene sets/pathways and TGFBI in the sW182* mutant cells. Three pathways/gene sets associated with (A) cell cycle regulation, (B) DNA repair, and (C) genome instability, were significantly enriched in the differential gene expression profile between groups of wild type cells and the mutant cells, as analyzed by GSEA. ES, enrichment score. A negative value of ES means that the gene set/pathway is negatively correlated with the referred group of cells (downregulation). (D) TGFBI level measured by Illumina Beadchips microarray analysis. vector, vector clone; WT, wild type clones; MT, sW182* mutant clone.
Fig. 2. The TGFBI gene and protein were dysregulated by the sW182*mutant. (A) Levels of the Smad5 in vector control, wild type and the mutant cells measured by qPCR. (B) TGFBI level measured by qPCR in multiple clones and their response to TGF-. The level in the vector clone 3 cells was used for normalization. (C) TGFBI reporter assay. Cells were treated with TGF- for 24 hours before luciferase activity was measured. (D) TGFBI protein level analyzed by Western blot after TGF- treatment. (E) Smad3 phosphorylation level analyzed by phospho-Smad3 Western blot after TGF- treatment.
Fig. 3. Hypermethylation of the TGFBI promoter in the sW182* mutant cells. (A) Putative CpG island region of the TGFBI promoter from vector control, wild type and the sW182* mutant cells was analyzed for methylation status. The unmethylated form (unmethylated) was absent in the mutant cells, indicating hypermethylation of the promoter in mutant cells. PCR products from methylated template are indicated
(methylated). (B) Effect of a DNA methyl transferase inhibitor (5-Aza) on the TGFBI transcription. In each panel, TGFBI expression was compared with that in untreated cells (DMSO). Relative value to control (=1) is shown.
Fig. 4. TGFBI is involved in the oncogenic activity of the sW182* mutant. (A) Increased expression of cyclin D1 in the sW182* cells compared with vector control and wild type were demonstrated by Western blot. (B) Frequent abnormal mitosis in the sW182* cells compared with wild type. Left panel, interphase cells. Arrowheads indicate cells with multinuclei). Right panel, cells in metaphase. (C) The nude mice xenograft study showed only sW182* mutant cells had tumor growth after 4 weeks, (D) The tumor size was significantly reduced with TGFBI-transduced cells in the xenograft study. Two representative mice and the quantitation are shown.
