Hepatitis B virus X protein (HBx)-related long noncoding RNA (lncRNA) down-regulated expression by HBx (Dreh) inhibits hepatocellular carcinoma metastasis by targeting the intermediate filament protein vimentin†‡§
The Department of Medical Genetics, Second Military Medical University, Shanghai, China
Potential conflict of interest: Nothing to report.
This work was supported in part by grants from the National Natural Science Foundation of China (No. 81071700, 81171936, 81071680 and 31171251).
The hepatitis B virus X protein (HBx) has been implicated as an oncogene in both epigenetic modifications and genetic regulation during hepatocarcinogenesis, but the underlying mechanisms are not entirely clear. Long noncoding RNAs (lncRNAs), which regulate gene expression with little or no protein-coding capacity, are involved in diverse biological processes and in carcinogenesis. We asked whether HBx could promote hepatocellular carcinoma (HCC) by regulating the expression of lncRNAs. In this study we investigated the alteration in expression of lncRNAs induced by HBx using microarrays and real-time quantitative polymerase chain reaction (PCR). Our results indicate that HBx transgenic mice have a specific profile of liver lncRNAs compared with wildtype mice. We identified an lncRNA, down-regulated expression by HBx (termed lncRNA-Dreh), which can inhibit HCC growth and metastasis in vitro and in vivo, act as a tumor suppressor in the development of hepatitis B virus (HBV)-HCC. LncRNA-Dreh could combine with the intermediate filament protein vimentin and repress its expression, and thus further change the normal cytoskeleton structure to inhibit tumor metastasis. We also identified a human ortholog RNA of Dreh (hDREH) and found that its expression level was frequently down-regulated in HBV-related HCC tissues in comparison with the adjacent noncancerous hepatic tissues, and its decrement significantly correlated with poor survival of HCC patients. Conclusion: These findings support a role of lncRNA-Dreh in tumor suppression and survival prediction in HCC patients. This discovery contributes to a better understanding of the importance of the deregulated lncRNAs by HBx in HCC and provides a rationale for the potential development of lncRNA-based targeted approaches for the treatment of HBV-related HCC. (HEPATOLOGY 2013)
As one of the most common malignancies in the world, hepatocellular carcinoma (HCC) has a very high morbidity and mortality. It is a major global health challenge that affects an estimated 500,000 people worldwide each year.1 The leading cause of HCC is attributable to persistent hepatitis B virus (HBV) infection, which can result in endstage liver disease, including liver cirrhosis and HCC. The smallest open reading frame of the HBV genome, HBX, encodes the hepatitis B virus X protein (HBx) and has been implicated in hepatocarcinogenesis and considered to be oncogenic.2 Furthermore, it has been observed that about 60% of HBx transgenic mice develop HCC after the age of 18 months and that some of these tumors eventually metastasize, which mimics the history of human HCC developed from a chronic HBV infection.3, 4 However, the molecular mechanisms underlying HBx-mediated tumorigenesis are not entirely clear. It is currently recognized that HBx is involved in epigenetic modifications, as much as genetic regulation, during hepatocarcinogenesis.5, 6
Long noncoding RNAs (lncRNAs) are a class of noncoding RNA transcripts longer than 200 nucleotides with little or no protein-coding capacity.7 Recent reports have demonstrated that lncRNAs have a very important role in epigenetic regulation. Through regulating gene expression by a variety of mechanisms, including transcription, posttranscriptional processing, chromatin modification, genomic imprinting, and the regulation of protein function, lncRNAs are involved in diverse biological functions and pathological processes.8-10 Faghihi et al.11 have found that the aberrant expression of lncRNA-BACE1AS could lead to Alzheimer's disease by regulating the expression of its antisense gene BACE1 that codes for β-secretase.
Increasing evidence relates changes in expression levels of lncRNAs to cancers. Therefore, lncRNAs can potentially be used as diagnostic markers or therapeutic targets for cancer in the clinic. For example, the abnormal expression of MALAT1 is associated with a wide range of cancers, including breast cancer, lung cancer, prostate cancer, and liver cancer, and is related to tumor metastasis. Its overexpression can be used as an early prognostic indicator that suggests a lower survival rate.12-14 Using lncRNA microarrays, it has also been found that the lncRNAs HULC and HEIH are aberrantly expressed in HCC tissues and are involved in hepatocarcinogenesis.15, 16 Given that the contribution of HBx to the expression of lncRNAs has never been discussed before, we wonder whether HBx could affect the expression of lncRNAs, thereby promoting HCC.
In this study we determined the lncRNA expression profiles in the livers of HBx transgenic mice and wildtype mice by lncRNA microarrays and real-time polymerase chain reaction (PCR). Our results indicate that some lncRNAs are dysregulated in HBx transgenic mice and that their expression is HBx-related. We further investigated the biological function of an HBx down-regulated lncRNA, Dreh, in vivo and in vitro, and found that it plays an important role in hepatocarcinogenesis.
Dreh, down-regulated expression by HBx EMT, epithelial-to-mesenchymal transition; HBx, hepatitis B virus X protein; HCC, hepatocellular carcinoma; H&E, hematoxylin and eosin; IF, intermediate filament; lncRNA, long noncoding RNA; OS, overall survival; qRT-PCR, quantitative reverse transcription-polymerase chain reaction; RACE, rapid amplification of cDNA ends; RIP, RNA immunoprecipitation; RFS, recurrence-free survival.
Materials and Methods
Animal and Patient Samples.
The HBx transgenic mice were constructed by the Model Animal Research Center of Nanjing University (Nanjing, China). The male BALB/C nude mice (4 weeks old) used in this study were purchased from the Shanghai Experimental Animal Center of the Chinese Academy of Sciences (Shanghai, China). All mice were bred and maintained in a specific pathogen-free facility and were used in accordance with the institutional guidelines for animal care. One hundred HBV-related HCC tissues and the corresponding adjacent noncancerous livers used were obtained with informed consent from patients who underwent hepatectomy in Eastern Hepatobiliary Surgery Hospital (Second Military Medical University, Shanghai, China). Fifty pairs of tissues used for quantitative real-time PCR analysis were randomly chosen from these 100 cases. Recurrence-free survival (RFS) was defined as the interval between the date of tumor resection and the date of diagnosis of any type of relapse (intrahepatic recurrence and extrahepatic metastasis), death, or the last observation point. Overall survival (OS) was defined as the interval between surgery and death or the last follow-up examination. For surviving patients, the data were censored at the last follow-up. The study was performed in accordance with the guidelines of the Institutional Review Board of the Liver Cancer Institute. The clinicopathological characteristics of the 100 patients are summarized in Supporting Table 1.
Microarray and Computational Analysis.
Six each of 20-month-old male HBx-transgenic mice and wildtype C57BL/6 mice were sacrificed and the livers removed. Every three livers were pooled as one sample; therefore, each group was represented by two samples. The total RNA was extracted from the four samples, amplified, and transcribed into fluorescent complementary DNA (cRNA) using Quick Amp Labeling kit (Agilent Technologies, Palo Alto, CA). Labeled samples were hybridized to the Mouse LncRNA Array (4 × 44K, ArrayStar, Rockville, MD), and after the washing steps the arrays were scanned using the Agilent Scanner G2505B. Agilent Feature Extraction Software (v. 10.5.1.1) was used to analyze acquired array images. Quantile normalization and subsequent data processing were performed using the GeneSpringGX v. 11.0 software package (Agilent Technologies). Differentially expressed lncRNAs with statistical significance were identified through Volcano Plot filtering. The threshold we used to screen up- or down-regulated lncRNAs was fold change >1.5 and P < 0.05. Microarray data have been deposited in the National Center for Biotechnology Information (NCBI) Gene Expression Omnibus (GEO) under accession number GSE42185.
RNA Pull-Down Assay.
LncRNA-Dreh and its antisense RNA were in vitro transcribed from vector pSPT19-Dreh and biotin-labeled with the Biotin RNA Labeling Mix (Roche Diagnostics, Indianapolis, IN) and T7/SP6 RNA polymerase (Roche), treated with RNase-free DNase I (Roche), and purified with an RNeasy Mini Kit (Qiagen, Valencia, CA). One milligram of protein from Hepa1-6 cells stably transfected with pcDNA3.1-Dreh extracts was then mixed with 50 pmol of biotinylated RNA, incubated with streptavidin agarose beads (Invitrogen, Carlsbad, CA), and washed. The retrieved proteins were resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), then silver-stained, and the specific bands were excised and analyzed by mass spectrometry.
RNA immunoprecipitation (RIP) experiments were performed using a Magna RIP RNA-Binding Protein Immunoprecipitation Kit (Millipore, Bedford, MA) according to the manufacturer's instructions. Antibody for RIP assays of vimentin (Cell Signaling Technology, Beverly, MA) was diluted 1:1,000. The coprecipitated RNAs were detected by reverse-transcription (RT)-PCR. The gene-specific primers used for detecting Dreh are presented in Supporting Table 2.
For immunocytochemistry analysis, Hepa1-6 cells stably transfected with pcDNA3.1-Dreh or control vector were cultured and fixed on 12 × 12 mm glass slides. After first incubated with rabbit antivimentin antibody (Cell Signaling Technology, 1:100), and then with goat antirabbit IgG (Alexa Fluor 555, Invitrogen, 1:5,000), the slides were mounted by adding DAPI-Fluoromount-G (Southern Biotech, SBA, Birmingham, AL) and examined with a Zeiss axiophot photomicroscope (Carl Zeiss, Oberkochen, Germany).
Expressions of hDREH in HCC patients were compared by the paired-sample t test. The survival curves were calculated using the Kaplan-Meier method, and the differences were analyzed by the log-rank test. The χ2, Fisher exact probability, and Student's t test were used for comparison between groups. Dates are expressed as the mean ± standard deviation (SD) from at least three independent experiments. All P values were two-sided and obtained using SPSS v. 16.0 software (Chicago, IL). P < 0.05 was considered statistically significant.
A description of other methods used in this study is presented in the Supporting Materials and Methods.
Different LncRNA Expression Profiles in the Livers of HBx Transgenic Mice and Wildtype Mice.
Since HBx transgenic mice can simulate the pathogenesis of HBV infection to HCC, it was selected to screen tumor specifically HBx-related lncRNAs. We chose the livers of 20-month-old male mice as experimental samples for microarray analysis. We set a threshold as fold change >1.5 and found that there were 215 up-regulated and 214 down-regulated lncRNAs in the livers of HBx transgenic mice compared with the wildtype mice, which means that the lncRNA expression profiles are different between the two groups (Supporting Table 3). To validate microarray analysis findings, we randomly selected 10 lncRNAs from the differentially expressed lncRNAs with fold change >3 and analyzed their expression by real-time PCR in expanded mouse liver samples. Our data were consistent with the microarray results (Supporting Fig. S1, Supporting Table 4). Thus, our results indicate that there are a series of lncRNAs frequently aberrantly expressed in the livers of HBx transgenic mice and that they may be related to the HBx-induced hepatocarcinogenesis.
LncRNA-Dreh Is Significantly Down-regulated in HBx-Transgenic Mice and Mouse Liver Cells Expressing HBx.
To investigate the relationship between HBx and the aberrantly expressed lncRNAs, we also assessed the expression of some highly conversed lncRNAs with relatively large fold-change in mice of different ages (including 2-month-, 10-month-, and 20-month-old) and sex by real-time PCR (the lncRNAs are list in Supporting Table 5). We found that some lncRNAs were already aberrantly expressed in young (2-month-old) mice, while some other lncRNAs did not changed until 10 months or 20 months old. Among them, we noticed an lncRNA, down-regulated expression by HBx (termed lncRNA-Dreh, NCBI Accession NO. AK050349; UCSC ID uc008dfz) that was significantly down-regulated in the livers of HBx transgenic mice in all three age groups compared with wildtype mice of the same age (at least 10 mice in each group). Furthermore, the difference in the expression levels of this lncRNA between the two types of mice gradually increased with increasing age (Fig. 1A). However, no significant difference was observed in the expression levels in mice of different gender of the same age (data not shown).
Next, to investigate whether HBx alters Dreh expression we measured the levels of Dreh after the transient transfection of pEGFP-HBx and pEGFP-N1 into liver cell lines, including BNL-CL2 and Hepa1-6 cells. We found that Dreh was down-regulated in pEGFP-HBx-transfected cells in comparison with the pEGFP-N1 control group (Fig. 1B).
Inhibition of Dreh Promotes Cell Proliferation and Cell Migration In Vitro.
The frequent down-regulation of lncRNA-Dreh in HBx transgenic mice livers implies that Dreh may have a role in HBV-related hepatocarcinogenesis. To prove this, the effects of reduced expression of Dreh on cell proliferation, apoptosis, cell migration, and invasion were investigated in two mouse liver cell lines: BNL-CL2 and Hepa1-6. We repressed the Dreh expression by RNA interference. The results showed that suppression of cellular Dreh not only enhanced the cell proliferation effect (Fig. 2A,B), but also promoted the migration and invasion activity of liver cancer cells compared with the negative control (Fig. 2E,F). However, Dreh down-regulation had no significant effect on cell apoptosis in vitro (Fig. 2C,D).
LncRNA-Dreh Represses the Growth of Tumor In Vivo.
Next, we performed a rapid amplification of cDNA ends (RACE) analysis to identify the 5′ and 3′ ends of the Dreh transcript. Our results indicate that Dreh is an lncRNA located on mouse chromosome 17 with high conservation in mammals; it has two exons and with no polyadenylate. The transcription start and termination sites and sequences of full-length cDNA of Dreh are presented in Fig. S2.
To determine the effects of Dreh on tumorigenesis in vivo, we subcutaneously injected Hepa1-6 cells stably transfected with pcDNA3.1-Dreh or pcDNA3.1 into nude mice for xenoplantation. We observed that mice injected with cells transfected with pcDNA3.1-Dreh showed significantly decreased tumor growth compared with those injected with cells transfected with pcDNA3.1 (Fig. 3A-C).
Hematoxylin and eosin (H&E) staining of the tumors showed that the degree of cell differentiation was significantly higher in Dreh-transfectant-derived tumors (more orderly cell arrangement in cord-like with obvious blood sinus; Fig. 3E, lower panel) than in the control group (cells were disarranged, irregular in shape with obvious heteromorphism, increased ratio of nucleus/cytoplasm, exhibited heteropyknosis; Fig. 3E, upper panel). These means that the degree of malignancy of Dreh-transfectant-derived tumors is significantly lower than the control group.
Additionally, the biochemical liver function tests showed that serum alanine aminotransferase, aspartate aminotransferase, total bile acid, γ-glutamyl transferase, and lactate dehydrogenase of the Dreh-transfect group were significantly lower than the control group (Fig. 3D). Because these are indicators of liver damage, their down-regulation indicated that lncRNA-Dreh had a protective effect on the liver.
LncRNA-Dreh Inhibits Tumor Metastasis In Vivo.
We next determined whether overexpression of Dreh could inhibit tumor metastasis in vivo. We developed orthotopic liver implanted metastatic models with subcutaneous tumor tissues described before in nude mice. Mice were sacrificed after 6 weeks. We found a dramatic decrease in number of metastatic nodules in the abdominal cavities of mice from the pcDNA3.1-Dreh-transfected group (0 of 5 mice), compared with the pcDNA3.1 control group (3 of 5 mice) (P = 0.038, χ2 test; Fig. 4C), including intrahepatic, abdominal wall, lymph node, and intestinal metastases. Further histological analysis also demonstrated this (Fig. 4A,B).
We also established peripheral intravascular implanted metastatic models by injecting Hepa1-6 cells stably transfected with pcDNA3.1-Dreh or pcDNA3.1 into nude mice through the tail vein. Four weeks after injection, the mice were sacrificed and the lungs were subjected to histological analysis. Our results showed that the mice injected with Dreh-overexpressing cells displayed less definite pulmonary metastasis sites (0 of 5 mice) than the control group (5 of 5 mice) (P = 0.002, χ2 test; Fig. 4C-E). Taken together, these data support an important inhibitory role for Dreh in hepatocellular tumor metastasis in vivo.
LncRNA-Dreh Can Combine with Protein Vimentin and Alter the Vimentin IF Structures.
Recent studies have reported that lncRNAs typically function by binding to specific protein partners, serving key regulatory roles to influence the activity and localization of the proteins they bind.17 Therefore, to investigate whether lncRNA-Dreh functions through this mechanism, we performed an RNA pull-down assay to identify proteins that associated with Dreh (Fig. 5A). Mass spectrometry analysis of the protein band specific to lncRNA-Dreh revealed that protein vimentin was specifically associated with Dreh (Table 1).
Table 1. Mass Spectrometry Analysis of the Proteins Pulled Down by lncRNA-AK050349
Protein Mass (Da)
No. of Peptide
vimentin [Mus musculus]
ATP synthase subunit beta, mitochondrial precursor [Mus musculus]
Vimentin is a type III intermediate filament (IF) and the major cytoskeletal component of mesenchymal cells.18 It is often used as a marker of epithelial-to-mesenchymal transition (EMT) during both normal development and metastatic progression.19, 20 It has been reported that the overexpression of vimentin is associated with HCC metastasis.21 To further validate the association between lncRNA-Dreh and vimentin, we next performed an RIP assay with an antibody against vimentin on Hepa1-6 cellular extracts. Consistently, we observed a significantly higher enrichment level of Dreh with the vimentin antibody compared with the nonspecific IgG control antibody (Fig. 5B).
We then sought to determine the functional relevance of the association between lncRNA-Dreh and vimentin. Vimentin is a protein that is characteristically up-regulated in cells undergoing EMT, and it is dramatically reorganized during cell adhesion and migration.22 Therefore, we hypothesized that Dreh might regulate tumor metastasis by modifying the expression and reorganization of vimentin. To test this hypothesis, we measured the protein expression levels of vimentin by western blot analysis in Hepa1-6 cells with Dreh over- or underexpressed. Our results showed that the expression of vimentin was significantly lower in cells stably transfected with pcDNA3.1-Dreh than with pcDNA3.1 (Fig. 5C), and was significantly increased in the Dreh small interfering RNA (siRNA) group compared with the control siRNA group (Fig. 5D). Both imply that Dreh can inhibit the protein expression of vimentin
We then performed immunofluorescence staining to investigate the effect of Dreh on the vimentin IF structure in Hepa1-6 cells stably transfected with pcDNA3.1-Dreh or pcDNA3.1. The results showed that the vimentin protein was present as helical filament structures extending from the nuclear membrane to the cellular membrane in the Dreh-transfected cells (Fig. 5E), whereas the filament structures were retracted to the nuclear membrane in the control cells, and there were extensive filamentous aggregation and many irregular fragmented aggregated structures in the cytoplasm (Fig. 5F). These changes of cytoskeleton structure can lead to instability of the cells, next to cells shedding to a distance, and finally promote the tumor cell migration. Our results reveal that lncRNA-Dreh can reverse the high-migration phenotype of Hepa1-6 cells by affecting the protein vimentin.
Human Ortholog RNA of Dreh Is Down-regulated in Human HBV-Related HCC Tissues and Could Be an Independent Prognostic Factor for HCC Patient Survival.
Sequence analysis indicated that Dreh was a highly evolutionarily conserved lncRNA in mammals (Fig. 6A); alignment revealed that the murine Dreh lncRNA most likely has a human ortholog RNA, referred to as hDREH, which located on human chromosome 5 (Fig. 6C). We identified the 5′ and 3′ transcription start and termination sites of the hDREH transcript by RACE analysis; the sequences of full-length hDREH are presented in Fig. S2. Analysis of the sequences by ORF Finder failed to predict a protein of more than 46 amino acids, the longest ORF of hDREH was 141bp (Fig. 6B). Moreover, it did not contain a valid Kozak sequence, suggesting the unlikelihood of translation. Thus, the hDREH transcript is consistent with an lncRNA.
We next measured the expression levels of hDREH in 50 pairs of human HBV-related HCC tissues and pair-matched normal liver tissues by real-time PCR and found that the transcript levels of lncRNA-hDREH were significantly down-regulated in HCC tissues compared with the corresponding noncancerous hepatic tissues from the same patient (P < 0.0001, paired-samples t test, Fig. 6D). Further, Kaplan-Meier analysis in 100 HCC patients revealed that lower hDREH expression levels in HCC tissues significantly correlated with a markedly reduced RFS (P = 0.002, log-rank test, Fig. 6E) and OS (P = 0.039, log-rank test, Fig. 6F) in HCC patients. These results together suggest the important roles of hDREH in the pathogenesis of HCC and prognosis of HCC patients.
We also compared hDREH expression with other clinical characteristics and the results showed that there was no direct relation between the expression of lncRNA-hDREH and other clinical characteristics (Supporting Table 6).
HCC is a worldwide disease with a high incidence, especially in Southeast Asia. The 5-year survival rate for HCC is very low, and 600,000 people die of it every year despite advanced therapeutic measures such as surgery, chemotherapy, and radiotherapy.23 Therefore, it is necessary to thoroughly investigate the pathogenesis and develop new targeted treatments for HCC. Alterations in epigenetic modifications are currently considered to be early events in tumorigenesis, which can change the microenvironment to induce a tumor. It has been reported that HBx can promote the malignant transformation of liver cells by various epigenetic modifications, including effects on DNA methylation, histone acetylation, and microRNA (miRNA) regulation.24, 25 However, whether HBx can promote HCC through lncRNA regulation has not been previously discussed. In this study we identified a series of lncRNAs differentially expressed in the livers of HBx transgenic mice compared with wildtype mice. Our results confirmed that HBx can alter the expression of lncRNAs and indicate that HBV carriers may have their specific lncRNA profiles; many of these lncRNAs are aberrantly expressed at a preneoplastic stage during early infection, suggesting that these expression differences may be closely related to tumorigenesis.
We further identified an HBx-down-regulated lncRNA among these differentially expressed lncRNAs, Dreh, which plays a key role in hepatocarcinogenesis, acting as a tumor suppressor, and can inhibit HCC growth and metastasis in vitro and in vivo. An RNA pull-down assay revealed that Dreh could combine with the IF protein vimentin, further repress the expression of vimentin, and change the normal cytoskeleton structure. In addition to maintenance of cell morphology, vimentin is also involved in cell adhesion, migration, proliferation, and signal transduction processes.26 The regulatory role of vimentin in tumor development reveals that vimentin can be used as a new target for anticancer research and therapy. There are some anticancer drugs in currently clinical used that directly affect vimentin, such as silibinin, which can inhibit prostate cancer invasion, motility, and migration by inhibiting the expression of vimentin and matrix metalloproteinase-227; and withaferin A, a tumor inhibitor and antiangiogenic agent, which can also target the protein vimentin.28 Our study enriched the epigenetic mechanisms of liver cancer induced by HBx through lncRNA regulation. The results also indicate that the enhanced expression of Dreh by gene transfer can reverse the malignant phenotype of HCC and suggest a tumor-suppressive role and a potential therapeutic target of Dreh, used alone or in combination with other drugs to treat tumors by affecting vimentin. Therefore, our results provide a strong rationale for developing epigenetic therapies that use synthetic lncRNA drugs to treat HBV-related HCC.
Additionally, we identified a human ortholog gene of Dreh, hDREH, which was noncoding and was down-regulated in HCC tissues compared with the adjacent normal liver tissues. The survival analysis revealed that HCC patients with lower expression of hDREH had a significantly reduced overall postoperative survival and tumor-free survival, which means that the expression of this gene may have considerable potential in predicting the prognosis of HCC and can be used as a prognostic biomarker of HCC. However, many lncRNAs lack strong conservation in general. Unlike messenger RNAs (mRNAs), which have to conserve codon use and prevent frameshift mutations in a single long ORF, selection may conserve only short regions of lncRNAs that are constrained by structure or sequence-specific interactions.29 Because our differentially expressed lncRNAs were derived from the mice microarray data, many of them do not have a homolog gene in the human genome. Thus, our results characterized the important role of the HBx-down-regulated lncRNA-Dreh and its human ortholog RNA in the development of HCC. These findings are an important supplement to the HBV-HCC tumorigenesis and transfer network and provided a new lncRNA target for the prevention and treatment of HCC. However, the roles of other differentially expressed lncRNAs from the array data in hepatocarcinogenesis need further verification and analysis.