Long terminal repeat (LTR) retrotransposons are a major class of transposable elements, accounting for 8.67% of the human genome. LTRs can serve as regulatory sequences and drive transcription of tissue or cancer-specific transcripts. However, the role of these LTR-activated transcripts, especially long non-coding RNAs (lncRNA), in cancer development remains largely unexplored. Here, we identified a novel lncRNA derived from MER52A retrotransposons (lncMER52A) that was exclusively expressed in hepatocellular carcinoma (HCC). HCC patients with higher lncMER52A had advanced TNM stage, less differentiated tumors, and shorter overall survival. LncMER52A promoted invasion and metastasis of HCC cells in vitro and in vivo. Mechanistically, lncMER52A stabilized p120-catenin and triggered the activation of Rho GTPase downstream of p120-catenin. Furthermore, we found that chromatin accessibility was crucial for the expression of lncMER52A. In addition, YY1 transcription factor bound to the cryptic MER52A LTR promoter and drove lncMER52A transcription in HCC. In conclusion, we identified an LTR-activated lncMER52A, which promoted the progression of HCC cells via stabilizing p120-catenin and activating p120-ctn/Rac1/Cdc42 axis. LncMER52A could serve as biomarker and therapeutic target for patients with HCC.

Significance:

A novel long noncoding RNA lncMER52 modulates cell migration and invasion via posttranslational control of p120-catenin protein stability.

Hepatocellular carcinoma (HCC) is the most frequent subtype of liver cancer and the fourth leading cause of cancer-related deaths (1, 2). Metastasis and recurrence are the biggest challenges in the treatment of HCC and are responsible for the poor prognosis of patients with HCC (3). It is important to discover the underlying molecular mechanisms of HCC metastasis and recurrence.

Transposable elements (TE) can drive the activation of oncogenes in cancer, including protein coding genes and noncoding RNAs (4–6). Long terminal repeats (LTR) are a major class of TEs, accounting for 8.67% of the human genome and flanked full-length ERVs (7, 8). During human evolution, the vast majority of endogenous retroviruses (ERV) have lost their internal domain encoding genes and the ability to transpose (6). LTRs often still contain functional regulatory elements, such as promoters, enhancers, transcription factor-binding sites and polyadenylation sites (9, 10). LTRs are normally silenced by a number of epigenetic mechanisms, such as DNA methylation, histone modifications, and RNA interference (11–13). Once epigenetic regulation is disrupted, LTRs are massively transcribed in primordial germ cells, oocytes, induced pluripotent stem (iPS) cells and cancer cells (14–16). For example, lincRNA family HPATs (human pluripotency-associated transcripts) are specifically expressed in pluripotent hESCs. Of which, HPAT5 is derived from human LTR8/HERV-H and regulates pluripotency during human preimplantation development (17). Hashimoto and colleagues (18) used CAGE to map transcription start sites in human HCC with focus on ncRNAs distant from protein-coding genes. They found that a large proportion of the distal ncRNAs were LTR-derived and were more than 10-fold upregulated in the vast majority of HCC samples. Furthermore, HCC patients with high LTR activity mostly have a viral etiology, are less differentiated and have a higher risk of recurrence. LTRs occur in biased positions and orientations within long noncoding RNAs (lncRNA), particularly at their transcription start sites, consequently placing their LTRs in the appropriate position to promote transcription (10). However, the function and molecular mechanism of LTR-derived lncRNAs in HCC are largely unknown.

Here, we identify a novel lncRNA lncMER52A (liver cancer-specific oncogenic long noncoding RNA transcribed by MER52A LTR retrotransposon of the ERV1 class) that is not expressed in most normal tissues but is widely expressed in HCC samples. LncMER52A promotes invasion and metastasis of HCC cells by modulating the p120-catenin/Rac1/Cdc42 signaling pathway.

Human Tissues

A total of 120 HCC tissues and matched adjacent nontumorous liver tissues were obtained from the Fudan University Shanghai Cancer Center. 16 pairs of human primary HCC and matched adjacent non-tumorous liver tissues for lncMER52A and p120-catenin correlation analysis were also collected from Fudan University Shanghai Cancer Center. This study was approved by the Clinical Research Ethics Committee of Fudan University Shanghai Cancer Center, China. Written informed consents were obtained from all patients. This study was conducted in accordance with the ethical guidelines of the Declaration of Helsinki.

Analysis on RNA-seq data

The RNA-seq datasets of 10 paired HCC and matched nontumor tissues were downloaded from GSE101432. The raw data were quantified by FastQC and were individually aligned using 2-pass mapping of STAR software (19), which was consistent with the GDC (Genomic Data Commons; ref. 20). The alignment result was provided as input for StringTie (21), and GENECODE_v22 was used as a reference to assemble the transcript. All the unique transcripts fulfilled the following criteria: The transcript was only expressed in HCC and the expression frequency of the transcript was in more than 30% HCC samples. lncRNAs with the transcription start site (TSS) that overlapped with LTR within ±1,000 bp were considered as LTR-derived transcripts by bedtools analysis (v2.25.0; ref. 22). The LTR sequences were downloaded from UCSC Table Brower (GENECODE_v22). The protein coding potential was predicted by the Coding Potential Calculator (23).

Cell culture

HEK293T cells and HepG2-C3A were obtained from the ATCC in 2014. Huh7, SNU449, MHCC97L, and MHCCLM3 cell lines were purchased from the Shanghai Cell Bank Type Culture Collection (Shanghai, Chinese Academy of Sciences, China) in 2016. All cell lines were authenticated by short tandem repeat profiling. Mycoplasma contamination was regularly examined using the Lookout Mycoplasma PCR Detection Kit (Sigma-Aldrich). Cells were maintained in DMEM medium supplemented with 10% FBS, 100 mg/mL penicillin and 100 U/mL streptomycin.

5′- and 3′-RACE assay

We performed 5′-RACE and 3′-RACE analyses to determine the full-length of lncMER52A using a SMARTer RACE cDNA Amplification Kit (Clontech) according to the manufacturer's instructions. The gene-specific PCR primers used for the RACE analyses are provided in Supplementary Table S1.

Transfection of cell lines

siRNAs and negative control siRNA were designed and synthesized by RiboBio (RiboBio Biotechnology). Cells were transfected with siRNAs using Oligofectamine transfection reagent (Invitrogen) according to the manufacturer's instructions. Forty-eight hours after transfection, the cells were harvested for further analysis. The sequences of the siRNAs were listed in Supplementary Table S2. The plasmids were transfected into HCC cells using Hieff Trans Liposomal Transfection Reagent (Yeasen Biotech) according to the manufacturer's instructions. The primers used for cloning were listed in Supplementary Tables S2 and S3.

CRISPR/Cas9-mediated genome editing

The desired Cas9-cutting site in the lncMER52A promoter genomic region was selected from http://www.e-crisp.org/E-CRISP/. The short guide RNA (sgRNA) was designed, and the sequences of highly scored sgRNAs were presented in Supplementary Table S3. The three candidate sequences (sgRNA#1-#3) were cloned into a vector (lenti-gRNA-puro). The combination of paired sgRNAs was used to achieve the best efficiency. The “#1+#2”-sgRNA pair and the “#1+#3”-sgRNA pair were identified to be effective in the knockout system and were called sgRNA (1+2) and sgRNA (1+3), respectively. For the CRISPR–Cas9-mediated genome editing, lentivirus was collected 48 hours after cotransfection of the lenti-sgRNA1/lenti-sgRNA2 or lenti-sgRNA1/lenti-sgRNA3 together with lenti-cas9-blast vectors into HEK293T cells using the Lipofectamine 2000 transfection reagent (Invitrogen). The target cells were infected with lentivirus plus 6 μg/mL polybrene (Sigma-Aldrich) for 24 hours and then treated with 4 μg/mL puromycin and 4 μg/mL blasticidin (InvivoGen) for more than 10 days before initiating the experiments.

In vitro transcription and translation assay

In vitro transcription and translation assay was performed as previously described (24) using TnT Quick Coupled Transcription/Translation Kit (Promega) and Transcend Non-Radioactive Translation Detection System (Promega). Briefly, 2.0 μg of circular plasmid DNA containing a T7 promoter and lncMER52A sequence was added to the TnT Quick Master Mix and incubated for 90 minutes at 30°C. The reaction produced biotinylated proteins if the inserted sequence had coding ability. Then the products were subjected to SDS-PAGE and electro-blotting and visualized via binding of streptavidin–horseradish peroxidase (streptavidin–HRP) and chemiluminescence detection.

Subcellular fractionation

Cytoplasmic and nuclear RNA isolations were performed with a PARIS Kit (Life Technologies) following the manufacturer's instructions.

RNA-pull down assays

Biotin-labeled RNAs were in vitro transcribed with the Biotin RNA Labeling Mix (Roche, Inc.) and T7 RNA polymerase (New England Biolabs, Inc.). Biotin-labeled RNAs were incubated with streptavidin magnetic beads (Thermo Fisher Scientific, Inc.) for 30 minutes. Then, streptavidin magnetic beads were incubated with cell lysates overnight. Proteins binding to the biotin-labeled RNAs were subjected to Western blot analysis. The primers of full-length or truncated lncMER52A used for the RNA pull-down assays were provided in Supplementary Table S4.

Western blot analysis

Proteins were subjected to SDS-PAGE and transferred to the nitrocellulose membranes (GE). After blocked by nonfat milk, the membrane was incubated with primary antibodies, followed by the incubation of the secondary antibodies. The full scans of the Western blot in the main figures are provided in the Supplementary Data S1.

Northern blotting

Northern blots were performed with digoxigenin (DIG)-labeled anti-sense probes (DIG Northern Starter Kit, Roche) using the NorthernMax Kit from Ambion (Thermo Fisher Scientific, Inc.). Briefly, total RNA was extracted from HCC cells using TRIzol (Invitrogen), followed by electrophoresis on a formaldehyde denaturing agarose gel. Samples were transferred to a positively charged nylon membrane (GE). Membranes were cross-linked under ultraviolet light for 1 minute, prehybridized for 1 hour at 68°C and hybridized overnight at 68°C with the DIG-labeled lncMER52A probes. DIG signals were detected with anti-digoxigenin-AP (Roche). The gene-specific primers used for Northern blotting analysis are provided in Supplementary Table S5.

RNA immunoprecipitation

RNA immunoprecipitation (RIP) experiments were performed using a Magna RIP RNA-Binding Protein Immunoprecipitation Kit (Millipore) according to the manufacturer's instructions. The antibodies for the RIP assays included p120-catenin (BD Biosciences), Smad2, Smad3, β-catenin, TWIST, ZO-1, Sox2, Snail1, Slug, ZEB-1, N-cadherin and E-cadherin (Cell Signaling Technology). The primers used in qPCR are listed in the Supplementary Table S6.

Chromatin immunoprecipitation

HCC cells or HCC tissues were washed twice with 1 × PBS before harvest and were cross-linked with 1% formaldehyde for 10 minutes at room temperature. The fixation was stopped by the addition of glycine to a final concentration of 0.125 mol/L for 5 minutes at room temperature. DNA was sheared in SimpleChIP Chromatin IP buffers (Cell Signaling Technology). Then, the samples were transferred to new tubes, and 1% of lysate was removed as the input. Protein G was incubated with 5 μg primary antibodies (anti-YY1, anti-H3K4me3 or anti-H3K27ac) or 5 μg IgG for 30 minutes, washed with washing buffer and incubated in cell lysis buffer overnight at 4°C. After washing two times with washing buffer, chromatin fragments were harvested using the elution buffer. The samples were de-crosslinked and examined using qPCR. The primers used in the chromatin immunoprecipitation (ChIP) experiments were listed in Supplementary Table S7.

Luciferase reporter assays

The lncMER52A promoter was cloned into the Firefly luciferase reporter vector pGL3-Promoter (Promega). Cells were transfected with the plasmid using Hieff Trans Liposomal Transfection Reagent (Yeasen Biotech). The pRL-TK vector (Promega) was cotransfected as an internal control. Luciferase activities were measured using the Dual-Luciferase Reporter Assay System (Promega).

Small G-protein activation assays

The activity of RhoA/Rac1/Cdc42 was measured using a RhoA/Rac1/Cdc42 Activation Assay Combo Biochem Kit (Cytoskeleton, Inc.) according to the manufacturer's instructions.

Transwell and wound-healing assays

Cell migration and invasion assays were performed in a Transwell chamber. Briefly, cells in FBS-free medium (1 × 10⁁5) were added to the top chamber with the bottom chamber containing 500 μL DMEM with 10% FBS. After 24 to 48 hours incubation, the membranes were washed, fixed, and stained with crystal violet. In invasion assays, the top chamber was coated with Matrigel before the addition of cells. For wound-healing assays, cells were placed into 6-well plates and cultured until reaching 100% confluence. An artificial scratch was created using a 200 μL pipette tip. The cells were cultured in serum-free medium. At 0, 24, and 48 hours, images were captured in the same field under magnification.

Immunofluorescence assay

Cells were grown on polylysine-precoated glass cover slips and incubated overnight at 37°C. Cells were fixed with 4% paraformaldehyde (PFA) for 20 minutes at room temperature and permeabilized with 1% Triton-X 100 in PBS for 20 minutes. Then cells were blocked with Immunol Staining Blocking Buffer (Beyotime Biotechnology) and incubated with FITC-phalloidin (1:300, Yeasen Biotech, Shanghai, 40735ES75) at 4°C overnight in the dark. After washing in PBS, nuclei were stained with DAPI. Finally, the slides were washed with PBS, and the cover slips were mounted with an anti-fade Mounting Medium (Beyotime Biotechnology). Images were obtained with a Leica fluorescence microscope (Leica).

In vivo assays

To investigate the effect of lncMER52A in vivo, Huh7 cells infected with the Ctrl-Cas9, lncMER52A-sgRNA(1+2) and lncMER52A-sgRNA(1+3) were harvested at the exponential growth stage and injected into the lateral tail veins of nude mice (0.2 mL cell suspension containing 2 × 10⁁6 cells for each mouse). After 6 weeks, mice were sacrificed, and the resected lung tissues were fixed in 10% neutral PB-buffered formalin. The fixed samples were embedded in paraffin and stained with hematoxylin and eosin. The numbers of metastatic loci in the lung were counted. All animal studies were conducted in accordance with relevant guidelines and regulations and were approved by the Animal Care and Use Committee of Fudan University, and animal care was in accordance with institutional guidelines.

Statistical analysis

Each experiment was repeated at least three times. Two-tailed Student t test was used for statistical analysis with GraphPad Prism 6 software, and P < 0.05 was considered significant (*, P < 0.05; **, P < 0.01; and ***, P < 0.001). For histograms, data are shown as the means ± SEM. Two-tailed Student t test was also used for statistical analysis with GraphPad Prism 6. The Kaplan–Meier method was used to estimate the overall survival of the subgroups as determined by the expression level of lncMER52A (The high/low expression of lncMER52A was calculated by the ROC curve). Univariate and multivariate regression analyses of Cohort was used to analyze the risk factors for HCC prognosis. The way of univariate and multivariate regression analyses was based on Cox proportional hazards model.

lncMER52A, an LTR-derived lncRNA, is specifically expressed in HCC

To identify specially expressed lncRNAs derived by LTR in HCC, we analyzed RNA-seq datasets from 10 paired HCC tumor tissues and matched nontumor (NT) tissues. A total of 728 novel HCC-specific transcripts derived from LTRs were observed. 464 of these HCC-specific transcripts were predicted to be noncoding RNAs with low coding potential (Fig. 1A; Supplementary Data S2). Among these lncRNAs, lncMER52A was most overlapped with LTR element (Fig. 1B). As expected, lncMER52A was only expressed in HCC tissues and not in nontumorous liver tissues and normal liver (Fig. 1C). Moreover, lncMER52A was not expressed in most human normal tissues, except for testis and placenta (Fig. 1D). We found that HCC patients with increased lncMER52A had an advanced TNM stage, less differentiated tumors and shorter overall survival (Fig. 1E; Supplementary Table S8). In addition, univariate and multivariate (Fig. 1F) regression analyses demonstrated that lncMER52A was an independent predictor for the survival of patients with HCC. LncMER52A was located on human chromosome 4 and had not been previously annotated in the National Center for Biotechnology Information database (NCBI; Supplementary Fig. S1A). One exon of lncMER52A was shown to be derived from MER52A ERV1, with the promoter within the 5′LTR of this element annotated as MER52A. The cytoplasmic/nuclear fractionation experiment indicated that lncMER52A was predominantly located in the cytoplasm of HCC cells (Supplementary Fig. S1B). 5′ and 3′ RACE assays and Northern blot assays showed that the dominant lncMER52A transcript was 828 nucleotides in HCC cells (Supplementary Fig. S1C and S1D). The complete sequence of lncMER52A was shown in Supplementary Fig. S1D. The CPAT (Coding-Potential Assessment Tool) and Coding Potential Calculator (Supplementary Fig. S1E) algorithm indicated that lncMER52A had no protein coding potential. In consistent with the algorithm prediction, in vitro transcription and translation assay showed that neither the sense nor the antisense transcript of lncMER52A could encode protein, confirming that lncMER52A was a bona fide noncoding RNA (Supplementary Fig. S1F).

Figure 1.

LncMER52A was uniquely expressed in HCC. A, Screening flowchart of HCC-specific, LTR-derived lncRNAs. B, Plot depicting number of bases overlapping with LTR of TSS (±1,000 bp) of candidate lncRNAs (y-axis) identified in HCC. C, Relative expression of lncMER52A in HCC tissues in comparison with adjacent noncancerous tissues and normal liver tissues. D, Expression of lncMER52A RNA in human normal tissue panel. E, Kaplan–Meier plots showed the association between lncMER52A expression and overall survival in patients with HCC. F, Univariate analysis and multivariate analysis showed that lncMER52A was an independent marker for HCC prognosis. The bars correspond to 95% confidence intervals.

Figure 1.

LncMER52A was uniquely expressed in HCC. A, Screening flowchart of HCC-specific, LTR-derived lncRNAs. B, Plot depicting number of bases overlapping with LTR of TSS (±1,000 bp) of candidate lncRNAs (y-axis) identified in HCC. C, Relative expression of lncMER52A in HCC tissues in comparison with adjacent noncancerous tissues and normal liver tissues. D, Expression of lncMER52A RNA in human normal tissue panel. E, Kaplan–Meier plots showed the association between lncMER52A expression and overall survival in patients with HCC. F, Univariate analysis and multivariate analysis showed that lncMER52A was an independent marker for HCC prognosis. The bars correspond to 95% confidence intervals.

Close modal

LncMER52A promotes migration, invasion, and metastasis of HCC cells

To investigate the role of lncMER52A in HCC, lncMER52A was knocked down via siRNA and CRISPR-Cas9 technology in Huh7 and MHCCLM3 cells, which had higher levels of lncMER52A (Supplementary Fig. S1G). To validate the specificity of lncMER52A knockdown/knockout, downregulation of lncMER52A in Huh7 and MHCCLM3 cells was validated using qRT-PCR (Supplementary Fig. S2A and S2B). In addition, we also detected the expression of MER52A-derived lnc00355 and lncMER52A neighboring gene EPH5A after lncMER52A knockdown/knockout and the results showed that lncMER52A knockdown or knockout had no effect on the expression of lnc00355 and EPH5A (Supplementary Fig. S2C and S2D). Next, we determined the effects of lncMER52A on HCC cells. Both Transwell assays and wound-healing assays showed that lncMER52A knockdown/knockout significantly reduced the migration and invasion of Huh7 and MHCCLM3 (Fig. 2A and B; Supplementary Fig. S2E). In addition, we transfected sgRNA(1+2), sgRNA(1+3) or si-LncMER52A-1, si-LncMER52A-2 into HeLa cells that did not express lncMER52A. The result showed that lncMER52A knockdown/knockout had no effect on the migration of HeLa cells (Supplementary Fig. S2F), which validated the specificity of lncMER52A knockdown. We also constructed lncMER52A overexpressed cell lines in SNU449 and HepG2-C3A, which had lower level of lncMER52A (Supplementary Figs. S1G and S2G). Overexpression of lncMER52A accelerated the migration and invasion of SNU449 and HepG2-C3A cells (Fig. 2C and D; Supplementary Fig. S2H). However, lncMER52A had no effect on colony formation or proliferation of HCC cells (Supplementary Fig. S2I and S2J). Furthermore, murine tumor xenografts derived from stably lncMER52A knockout Huh7 cells exhibited a markedly reduced experimental metastasis ability compared with control cells (Fig. 2E; Supplementary Fig. S2K). Together, these results indicated that lncMER52A promoted the invasion and metastasis of HCC cells in vitro and in vivo.

Figure 2.

LncMER52A promoted HCC cells migration and invasion in vitro and in vivo. A and B, Migration and invasion of Huh7 and HCCLM3 cells were suppressed following lncMER52A knockdown by siRNAs (A) and by the CRISPR/Cas9 knockout technology (B). C and D, LncMER52A overexpression promoted migration and invasion of HepG2-C3A and SNU449 cells. E, LncMER52A knockout inhibited the invasion and metastasis of HCC to lung in vivo. Representative images of hematoxylin and eosin–stained metastatic loci from lungs in each group, together with the numbers of tumor nodules are shown. Data are shown as means ± SEM. Two-tailed Student t test was used for statistical analysis. *, P < 0.05; **, P < 0.01; ***, P < 0.001; P < 0.05 was considered significant.

Figure 2.

LncMER52A promoted HCC cells migration and invasion in vitro and in vivo. A and B, Migration and invasion of Huh7 and HCCLM3 cells were suppressed following lncMER52A knockdown by siRNAs (A) and by the CRISPR/Cas9 knockout technology (B). C and D, LncMER52A overexpression promoted migration and invasion of HepG2-C3A and SNU449 cells. E, LncMER52A knockout inhibited the invasion and metastasis of HCC to lung in vivo. Representative images of hematoxylin and eosin–stained metastatic loci from lungs in each group, together with the numbers of tumor nodules are shown. Data are shown as means ± SEM. Two-tailed Student t test was used for statistical analysis. *, P < 0.05; **, P < 0.01; ***, P < 0.001; P < 0.05 was considered significant.

Close modal

LncMER52A regulates the epithelial–mesenchymal transition signaling pathway

When we overexpressed lncMER52A in HepG2-C3A cells, HepG2-C3A cells exhibited significant changes in cell morphology (Fig. 3A). Overexpression of lncMER52A produced more spindle-shaped or fibroblast-like cells than cells transfected with the control vector. These morphologies were characteristic of mesenchymal cell types (Fig. 3A). Moreover, GO pathway analysis (https://david.ncifcrf.gov/) revealed that silencing lncMER52A affected a list of genes associated with extracellular matrix organization and the epithelial–mesenchymal transition (EMT) signaling pathway (Fig. 3B). We determined the expression of differentiated expressed genes associated with EMT process (Supplementary Data S3), and the result showed that lncMER52A did affect the expression of these genes (Fig. 3C). Therefore, we hypothesized that lncMER52A might regulate the EMT signaling pathway in HCC. Moreover, immunofluorescence experiments revealed that lncMER52A impaired F-actin (cytoskeletal marker) rearrangements (Fig. 3D; Supplementary Fig. S2L). Meanwhile, immunoblot analysis demonstrated that lncMER52A overexpression increased the protein and of mesenchymal makers, such as N-cadherin, vimentin and slug, but decreased the levels of epithelial makers, such as E-cadherin and claudin-1 (Fig. 3E), whereas si-lncMER52A decreased the expression of N-cadherin, vimentin, and slug but increased the expression of E-cadherin and claudin-1 (Fig. 3E). These results indicated that lncMER52A promoted EMT process of HCC cells.

Figure 3.

LncMER52A regulated the EMT signaling pathway. A, HepG2-C3A cells adopted EMT-related cell morphology following lncMER52A overexpression. B, GO analyses listing the top fourteen biological processes regulated by lncMER52A. C, mRNA levels of EMT markers in HCC cells were analyzed by quantitative RT-PCR after lncMER52A knockdown or overexpression. D, Immunofluorescence staining of F-actin in HCC cells after lncMER52A knockdown or overexpression. E, Protein levels of EMT markers in HCC cells were analyzed by Western blot after lncMER52A knockdown or overexpression. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, not significant.

Figure 3.

LncMER52A regulated the EMT signaling pathway. A, HepG2-C3A cells adopted EMT-related cell morphology following lncMER52A overexpression. B, GO analyses listing the top fourteen biological processes regulated by lncMER52A. C, mRNA levels of EMT markers in HCC cells were analyzed by quantitative RT-PCR after lncMER52A knockdown or overexpression. D, Immunofluorescence staining of F-actin in HCC cells after lncMER52A knockdown or overexpression. E, Protein levels of EMT markers in HCC cells were analyzed by Western blot after lncMER52A knockdown or overexpression. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, not significant.

Close modal

LncMER52A interacts with p120-catenin in HCC cells

Because lncMER52A regulated the EMT signaling pathway, we next investigated whether lncMER52A bound to EMT-related proteins. First, we performed a RIP assay to identify EMT-related proteins that might bind to lncMER52A (Fig. 4A). We found that p120-catenin (also called CTNND1), a member of the Armadillo protein family, immunoprecipitated with lncMER52A in Huh7 cells, whereas the negative control lncRNAs (MALAT1, UCA1, and PVT1) exhibited no association with p120-catenin (Fig. 4B). Next, we performed a biotin-labeled RNA pull-down assay and found that biotin-labeled lncMER52A pulled down the p120-catenin protein (Fig. 4C). Furthermore, we constructed a series of truncated lncMER52As based on its secondary structure (http://rna.tbi.univie.ac.at; Fig. 4D). Of note, the 54–364 nt fragment of lncMER52A was necessary for lncMER52A to bind the p120-catenin protein (Fig. 4D). In addition, we constructed 3 × FLAG-tagged full-length or truncated p120-catenin and performed RIP assays. The results showed that the deletion of the Armadillo repeat domain of p120-catenin significantly abolished the association between p120-catenin and lncMER52A (Fig. 4E). Together, these results indicated that lncMER52A specifically bound to p120-catenin in HCC cells.

Figure 4.

LncMER52A interacted with p120-catenin in HCC cells. A, LncMER52A was enriched in p120-catenin immunoprecipitates in HCC cells. RIP assays using EMT-related proteins were performed in Huh7 cells and qRT-PCR was used to determine the enrichment of lncMER52A. Data represented as fold enrichment over IgG group. Error bars represent the SEM for three biological replicates. B, lncMER52A was specially binding to p120-catenin. RIP assays using antibody against p120-catenin, followed by qRT-PCR to determine the enrichment of lncMER52A and several control lncRNAs. C, RNA pulldown assays indicated that lncMER52A pulled down p120-catenin. D, Mapping analysis of p120-catenin binding domain on lncMER52A. Secondary structure of lncMER52A was analyzed by RNA-fold web server (http://rna.tbi.univie.ac.at). E, Deletion mapping for the domains of p120-catenin that bound to lncMER52A. Error bars represent the SEM for three biological replicates in A, B, and E. ns, not significant.

Figure 4.

LncMER52A interacted with p120-catenin in HCC cells. A, LncMER52A was enriched in p120-catenin immunoprecipitates in HCC cells. RIP assays using EMT-related proteins were performed in Huh7 cells and qRT-PCR was used to determine the enrichment of lncMER52A. Data represented as fold enrichment over IgG group. Error bars represent the SEM for three biological replicates. B, lncMER52A was specially binding to p120-catenin. RIP assays using antibody against p120-catenin, followed by qRT-PCR to determine the enrichment of lncMER52A and several control lncRNAs. C, RNA pulldown assays indicated that lncMER52A pulled down p120-catenin. D, Mapping analysis of p120-catenin binding domain on lncMER52A. Secondary structure of lncMER52A was analyzed by RNA-fold web server (http://rna.tbi.univie.ac.at). E, Deletion mapping for the domains of p120-catenin that bound to lncMER52A. Error bars represent the SEM for three biological replicates in A, B, and E. ns, not significant.

Close modal

LncMER52A blocks ubiquitination/proteasome-dependent p120-catenin degradation

Next, we explored whether lncMER52A affected p120-catenin expression in HCC cells. We first examined the correlation of lncMER52A RNA and p120-catenin protein in 16 pairs of HCC samples. We found that the expression of lncMER52A was positively correlated with p120-catenin protein in HCC samples (Supplementary Fig. S3A–S3C). Then we determined the expression of p120-catenin after lncMER52A overexpression or knockdown. The results showed that lncMER52A overexpression increased the protein level of p120-catenin but had no effect on the mRNA level of p120-catenin in HepG2-C3A cells (Fig. 5A; Supplementary Fig. S3D). Consistent with these results, silencing lncMER52A reduced the protein level of p120-catenin and had no effect on the mRNA level of p120-catenin (Fig. 5A; Supplementary Fig. S3E). Moreover, the half-life of p120-catenin in lncMER52A-overexpressing cells was longer than that in control cells after treatment with cycloheximide, whereas the half-life of p120-catenin in lncMER52A-knockdown cells was shorter than that in control cells after treatment with cycloheximide (Fig. 5B). These results indicated that lncMER52A affected p120-catenin protein stability. p120-catenin was regulated by the ubiquitination/proteasome-dependent pathway (25). Following treatment with MG132 (proteasome inhibitor), p120-catenin was stabilized either in lncMER52A-overexpressing or lncMER52A-knockdown cells (Fig. 5C). Next, p120-catenin and HA-tagged Ub were cotransfected into lncMER52A-overexpressing HepG2-C3A cells or control cells treated with MG132, respectively. Immunoprecipitation assays followed by Western blot analysis indicated that the ubiquitination levels of p120-catenin significantly decreased in lncMER52A-overexpressing cells (Fig. 5D). These results demonstrated that lncMER52A inhibited ubiquitin/proteasome-dependent p120-catenin degradation. β-TrCP1 is an E3 ligase involved in p120-catenin ubiquitination (25). We performed immunoprecipitation assays and found that p120-catenin interacted with β-TrCP1 in HCC cells (Fig. 5E). Moreover, β-TrCP1 overexpression significantly reduced the protein levels of p120-catenin (Fig. 5E). Following treatment with MG132, ubiquitinated p120-catenin accumulated in β-TrCP1–overexpressing cells (Supplementary Fig. S4A). Intriguingly, lncMER52A overexpression significantly inhibited the interaction between β-TrCP1 and p120-catenin, whereas lncMER52A knockdown notably increased this association in HCC cells (Supplementary Fig. S4B; Fig. 5F). We concluded that lncMER52A could promote the stability of p120-catenin by blocking its ubiquitination–proteasome degradation.

Figure 5.

LncMER52A blocked ubiquitination/proteasome-dependent p120-catenin degradation. A, Immunoblotting for the protein levels of p120-catenin after lncMER52A knockdown or overexpression. β-Actin served as the internal control. B, lncMER52A regulated p120-catenin stability. Huh7 cells and HepG2-C3A cells were treated with cycloheximide (CHX) at 50 μg/mL and were harvested for Western blot assays at the indicated time points. C, LncMER52A knockdown or overexpressed cells were treated with MG132 (20 μmol/L) for 12 hours, then immunoblotting for p120-catenin levels was performed. D, Ub-HA and p120-ctn-1A-3xflag were transfected into HepG2-C3A cells together with PCDH or PCDH-lncMER52A. The cells were treated with MG132 for 6 hours before lysis. Immunoprecipitations were performed with anti-flag antibody, and Western blot analyses were performed with anti-HA (top) or anti-flag. p120-ctn (Ub) indicated ubiquitinated p120-catenin. The bottom panel (p120-ctn) indicates efficiency of immunoprecipitations. E, β-TrCP1 could bind and decrease the protein of p120-ctn. F, lncMER52A knockdown increased the association between p120-ctn and β-TrCP1. p120-ctn-1A-3xflag plasmid and β-TrCP1-Myc plasmid were cotransfected into si-NC– or si-lncMER52A–transfected Huh7 cells. The cells were treated with MG132 for 6 hours before lysis and β-actin served as the loading control.

Figure 5.

LncMER52A blocked ubiquitination/proteasome-dependent p120-catenin degradation. A, Immunoblotting for the protein levels of p120-catenin after lncMER52A knockdown or overexpression. β-Actin served as the internal control. B, lncMER52A regulated p120-catenin stability. Huh7 cells and HepG2-C3A cells were treated with cycloheximide (CHX) at 50 μg/mL and were harvested for Western blot assays at the indicated time points. C, LncMER52A knockdown or overexpressed cells were treated with MG132 (20 μmol/L) for 12 hours, then immunoblotting for p120-catenin levels was performed. D, Ub-HA and p120-ctn-1A-3xflag were transfected into HepG2-C3A cells together with PCDH or PCDH-lncMER52A. The cells were treated with MG132 for 6 hours before lysis. Immunoprecipitations were performed with anti-flag antibody, and Western blot analyses were performed with anti-HA (top) or anti-flag. p120-ctn (Ub) indicated ubiquitinated p120-catenin. The bottom panel (p120-ctn) indicates efficiency of immunoprecipitations. E, β-TrCP1 could bind and decrease the protein of p120-ctn. F, lncMER52A knockdown increased the association between p120-ctn and β-TrCP1. p120-ctn-1A-3xflag plasmid and β-TrCP1-Myc plasmid were cotransfected into si-NC– or si-lncMER52A–transfected Huh7 cells. The cells were treated with MG132 for 6 hours before lysis and β-actin served as the loading control.

Close modal

LncMER52A regulates GTPase activity of proteins downstream of p120-catenin in HCC

p120-catenin promotes cell motility by modulating the activities of Rho GTPases (e.g., RhoA, Rac1 and Cdc42; refs. 26–30). Tang and colleagues (31) observed that the protein levels of p120-catenin were markedly higher in HCC tissues than in normal liver tissues by IHC. Moreover, p120-catenin overexpression was significantly correlated with distant metastasis in HCC samples. We first investigated the biological function of p120-catenin in HCC cells. Knockdown and overexpression of p120-catenin were validated by qRT-PCR and Western blot analysis (Supplementary Fig. S5A and S5B). The results showed that p120-catenin promoted the migration and invasion of HCC cells (Supplementary Fig. S5C and S5D). Wound-healing assays also showed that knockdown p120-catenin inhibited the migration of HCC cells (Supplementary Fig. S5E). Immunoblot analysis demonstrated that p120-catenin promoted the expression of mesenchymal markers but decreased the expression of epithelial markers (Supplementary Fig. S5F). Immunofluorescent experiments revealed that p120-catenin could affect F-actin rearrangements (Supplementary Fig. S5G and S5H). p120-catenin overexpression in HepG2-C3A cells induced a dramatic branching phenotype (Supplementary Fig. S5H).

Many reports have indicated that p120-catenin induces the activation of Rac1 and Cdc42, which are essential for cell migration and invasion (27, 32, 33).We next investigated whether lncMER52A affected Rac1 and Cdc42 activation. We examined the activity of Rho GTPases in control cells and lncMER52A-overexpressing cells. As shown in Fig. 6A, lncMER52A overexpression resulted in a significant upregulation of p120-catenin, which induced Rac1 and Cdc42 activation. By contrast, Rac1 and Cdc42 activity were reduced in lncMER52A knockdown cells (Fig. 6B). The specific small-molecule inhibitors, EPOH-016 and ML141 could inhibit Rac1 and Cdc42 activity, respectively (34–37). We found that these inhibitors could significantly block the cell migration and invasion induced by lncMER52A (Fig. 6C). In addition, immunofluorescent experiments also showed that the F-actin rearrangements induced by lncMER52A were abrogated by treatment with EPOH-016 and ML141(Fig. 6D; Supplementary Fig. S5I). These results indicated that lncMER52A exerted its biological function through regulating the activities of GTPase proteins downstream of p120-catenin. Moreover, Transwell assays showed that lncMER52A overexpression promoted HCC cell migration and invasion, whereas these effects could be partially inhibited by p120-catenin knockdown (Fig. 6E). Collectively, these data strongly indicated that lncMER52A regulated HCC cell migration and invasion though the p120-ctn/Rac1/Cdc42 axis.

Figure 6.

LncMER52A regulated HCC cell migration and invasion though the p120-ctn/Rac1/Cdc42 axis. A and B, Total protein and protein with GTPase activity were determined by performing rhotekin and PAK GST-pulldown assays in HepG2-C3A cells and Huh7 cells with lncMER52A overexpression or knockdown, respectively. C, Migration and invasion of HepG2-C3A cells stably overexpressing lncMER52A was inhibited following treatment with the Rac1 inhibitor EPOH-016(4 μmol/L) or Cdc42 inhibitor ML141(20 μmol/L). D, Immunofluorescence staining of F-actin after lncMER52A overexpression or following treatment with EPOH-016 or ML141. E, p120-ctn silencing suppressed the migration and invasion induced by lncMER52A overexpression. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 6.

LncMER52A regulated HCC cell migration and invasion though the p120-ctn/Rac1/Cdc42 axis. A and B, Total protein and protein with GTPase activity were determined by performing rhotekin and PAK GST-pulldown assays in HepG2-C3A cells and Huh7 cells with lncMER52A overexpression or knockdown, respectively. C, Migration and invasion of HepG2-C3A cells stably overexpressing lncMER52A was inhibited following treatment with the Rac1 inhibitor EPOH-016(4 μmol/L) or Cdc42 inhibitor ML141(20 μmol/L). D, Immunofluorescence staining of F-actin after lncMER52A overexpression or following treatment with EPOH-016 or ML141. E, p120-ctn silencing suppressed the migration and invasion induced by lncMER52A overexpression. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Close modal

The activation of lncMER52A is tightly regulated by chromatin accessibility and transcription factors YY1

Next, we elucidated the mechanism of the specific expression of lncMER52A in HCC. The reactivation of LTR usually is regulated by epigenetic regulation, such as DNA methylation, histone modifications and RNA interference (11–13). We first examined the H3K4me3 and H3K27ac conditions of the lncMER52A promoter in MHCC97L and SNU449 cells. As expected, MHCC97L cells with high expression levels of lncMER52A had high enrichment and overlaps of the H3K4me3 and H3K27ac peaks in the promoter region of lncMER52A, whereas SNU449 cells with a low expression level of lncMER52A had no H3K4me3 and H3K27ac peaks in the promoter region of lncMER52A (Fig. 7A). Next, ChIP assays indicated that the HCC cell lines and HCC samples with higher levels of lncMER52A had increased enrichment of H3K4me3 and H3K27ac in the promoter region of lncMER52A (Fig. 7B; Supplementary Fig. S6A and S6B). We thus concluded that the active histone modifications in the lncMER52A promoter were a hallmark of aberrant activation of lncMER52A in HCC.

Figure 7.

Cryptic MER52A LTR promoter activation was associated with active histone modifications and binding of transcription factors in liver cancer. A, ChIP-seq from the HCC cell lines MHCC97L and SNU449 aligned to the hg38 genome with Tophat 2.0.8 using standard parameters and showed enrichment of H3K4me3 and H3K27ac at the promoter of lncMER52A. B, ChIP assays detected the level of H3K4me3 and H3K27ac at the promoter of lncMER52A in HCC cells and tissues. C, Identification of core promoter region in the lncMER52A promoter. Left, diagrams of the full-length 1.78kb lncMER52A promoter and the truncated promoter fragments; right, luciferase activity of the full-length or truncated promoter in HEK-293T cells. D, Knockdown YY1 inhibited the mRNA expression of lncMER52A and luciferase activity of lncMER52A promoter. E, YY1 promoted lncMER52A expression and activity of lncMER52A promoter. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, not significant.

Figure 7.

Cryptic MER52A LTR promoter activation was associated with active histone modifications and binding of transcription factors in liver cancer. A, ChIP-seq from the HCC cell lines MHCC97L and SNU449 aligned to the hg38 genome with Tophat 2.0.8 using standard parameters and showed enrichment of H3K4me3 and H3K27ac at the promoter of lncMER52A. B, ChIP assays detected the level of H3K4me3 and H3K27ac at the promoter of lncMER52A in HCC cells and tissues. C, Identification of core promoter region in the lncMER52A promoter. Left, diagrams of the full-length 1.78kb lncMER52A promoter and the truncated promoter fragments; right, luciferase activity of the full-length or truncated promoter in HEK-293T cells. D, Knockdown YY1 inhibited the mRNA expression of lncMER52A and luciferase activity of lncMER52A promoter. E, YY1 promoted lncMER52A expression and activity of lncMER52A promoter. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, not significant.

Close modal

Next, we inserted the full-length (−1000 ∼ +780 nt of Transcription Start Sites) lncMER52A promoter and truncated promoter fragments into a luciferase reporter vector. The luciferase activity of the −100 ∼ +95 nt promoter region was obviously higher than that of the other promoter fragments (Fig. 7C). Analysis of the core promoter region in UCSC revealed that the −100 ∼ +95 nt region had multiple binding sites for various transcription factors (Fig. 7D). We knocked down these transcription factors and analyzed the expression levels and promoter activity of lncMER52A. The results showed that knockdown YY1 or REST significantly decreased the promoter activities and expression of lncMER52A (Fig. 7D), but only YY1 overexpression induced the expression of lncMER52A (Fig. 7E). Interestingly, ChIP assays showed that YY1 was only enriched in the promoter region of lncMER52A in HCC cell lines with higher expression of lncMER52A (Fig. 7E). Taken together, these data indicated that the chromatin accessibility controlled the transcription of lncMER52A in cell lines and tissues. Once the promoter region was opened, YY1 was able to bind to the promoter of lncMER52A and activated the transcription of lncMER52A.

TEs can induce broad oncogene expression in human cancers, and LTRs are the most enriched TEs in all re-activated TEs across multiple cancer types (4). However, the onco-exaptation of LTR-derived lncRNAs in HCC remain largely unexplored. Here, we analyzed cancer-specific and LTR-derived lncRNAs in HCC. These lncRNAs contain a high proportion of LTR-derived sequences and were specifically expressed in HCC cells. LTRs are usually silenced in somatic cells, but they are usually reactivated in cancers (38–40). In our study, one of the LTR-derived lncRNAs, lncMER52A, was specifically expressed in HCC and had no expression in most normal tissues or adjacent nontumor liver tissues, except for testis and placenta. These results were consistent with a previous report that showed that LTRs were specifically expressed in early embryonic development, germ cells, and pluripotent stem cells (18). Interestingly, the expression of lncMER52A was closely correlated with the clinical features of patients with HCC. HCC patients with higher lncMER52A developed an advanced TNM stage, less differentiated tumors and showed a shorter overall survival. Furthermore, lncMER52A significantly promoted the invasion and metastasis of HCC cells in vitro and in vivo. These results demonstrated that lncMER52A was an oncogenic lncRNA in HCC and promoted the progression of HCC. Our results revealed an additional type of onco-exaptation event in which LTR-derived lncRNAs promoted the progression of HCC. It was intriguing to propose that lncMER52A may be a potential target to treat patients with HCC because it was only expressed in HCC and not in most normal tissues. Moreover, our results revealed that lncMER52A could act as an independent predictor for HCC patient survival, which indicated that lncMER52A also was a promising biomarker for HCC prognosis.

To dissect the mechanism by which lncMER52A promoted invasion and metastasis, we found that lncMER52A bound to p120-catenin and blocks its degradation. p120-catenin could either act as an oncogene or tumor suppressor in different tumors. In HCC, p120-catenin was aberrantly upregulated and significantly correlated with distant metastasis in HCC samples (31). We also found that p120-catenin was upregulated in HCC tissues and promoted the migration and invasion of HCC cells. Moreover, lncMER52A induced Rac1 and Cdc42 activation, while Rac1 and Cdc42 inhibitors could block the promotion of lncMER52A on migration and invasion of HCC cells. These findings demonstrated that lncMER52A promoted HCC cell invasion and metastasis through the p120-catenin/Rac1/Cdc42 axis. The regulation of p120-catenin protein remain largely unexplored. Previous studies demonstrated that abundance of p120-catenin could be post-transcriptionally by microRNAs (41, 42). In this study, we identified that lncMER52A participated in the proteasome degradation of p120-catenin, which provides a novel molecular mechanism for p120-catenin stabilization.

Accumulating evidence had revealed that the epigenetic reactivation of LTRs was due to DNA methylation, histone modifications, and RNA interference (11–13). For example, Lynch and colleagues (43) found that 58% of H3K27ac peaks and 32% of H3K4me3 peaks overlapped TEs, including LTRs. Hashimoto and colleagues (18) found that active histone marks (H3K4me3 and H3K27ac) and active transcription factors were strongly enriched around LTR-derived ncRNAs TSS (18). In our study, we found that lncMER52A expression was correlated with the enrichment levels of the active histones in its promoter in HCC cell lines and tissues. The transcription of lncMER52A in HCC was driven by MER52A LTR–derived promoter. MER52A LTR was reported to be abnormally activated in endometrial cancer through demethylation (44). Zhang and colleagues (44) report 5 MER52A LTRs that was hypomethylated in endometrial cancer. In our study, we found 2 MER52A LTR–derived lncRNAs that were abnormally activated in HCC (Supplementary Data S2). These results indicated that MER52A LTRs were frequently activated in cancers. However, different MER52A LTR sites were activated in various cancers, respectively. This suggests that MER52A LTR could induce cancer-specific transcripts and provided insight into how the cancer-specific transcripts were produced. Moreover, we found that YY1 contributed to the transcription of lncMER52A in HCC cells. Therefore, it was clear that lncMER52A was tissue-specific expression in HCC that tightly regulated by active histone modifications (H3K4me3 and H3K27ac) and transcription factors (YY1). The abnormally activation of lncMER52A in HCC cells and the promotion of lncMER52A on HCC progression might explain why lncMER52A could act as an independent marker for HCC prognosis.

No potential conflicts of interest were disclosed.

Conception and design: Y. Wu, S. Huang, X. He, L. Liang

Development of methodology: Y. Wu, L. Liang

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): Y. Wu, Y. Zhao, L. Wang, L. Liang

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): Y. Wu, J. Zhao, L. Liang

Writing, review, and/or revision of the manuscript: Y. Wu, S. Huang, L. Liang

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): Y. Wu, Y. Zhao, L. Huan, Y. Zhou, L. Xu, Z. Hu, Y. Liu, Z. Chen, L. Wang, L. Liang

Study supervision: Y. Wu, X. He, L. Liang

This work was supported by grants from National Natural Science Foundation of China to L. Liang (81672366 and 81972253) and the State Key Project on Infectious Diseases of China to L. Wu (2018ZX10723204).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Bray
F
,
Ferlay
J
,
Soerjomataram
I
,
Siegel
RL
,
Torre
LA
,
Jemal
A
. 
Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries
.
CA Cancer J Clin
2018
;
68
:
394
424
.
2.
Visvader
JE
. 
Cells of origin in cancer
.
Nature
2011
;
469
:
314
22
.
3.
Bruix
J
,
Gores
GJ
,
Mazzaferro
V
. 
Hepatocellular carcinoma: clinical frontiers and perspectives
.
Gut
2014
;
63
:
844
55
.
4.
Jang
HS
,
Shah
NM
,
Du
AY
,
Dailey
ZZ
,
Pehrsson
EC
,
Godoy
PM
, et al
Transposable elements drive widespread expression of oncogenes in human cancers
.
Nat Genet
2019
;
51
:
611
7
.
5.
Howard
G
,
Eiges
R
,
Gaudet
F
,
Jaenisch
R
,
Eden
A
. 
Activation and transposition of endogenous retroviral elements in hypomethylation induced tumors in mice
.
Oncogene
2008
;
27
:
404
8
.
6.
Mourier
T
,
Willerslev
E
. 
Retrotransposons and non-protein coding RNAs
.
Brief Funct Genomic Proteomic
2009
;
8
:
493
501
.
7.
Bannert
N
,
Kurth
R
. 
The evolutionary dynamics of human endogenous retroviral families
.
Annu Rev Genomics Hum Genet
2006
;
7
:
149
73
.
8.
Waterston
RH
,
Lindblad-Toh
K
,
Birney
E
,
Rogers
J
,
Abril
JF
,
Agarwal
P
. 
Institutionen FMS, Faculty OS, Göteborgs U, Gothenburg U. Initial sequencing and comparative analysis of the mouse genome
.
Nature
2002
;
420
:
520
62
.
9.
Mager
DL
,
Medstrand
P
. 
Retroviral repeat sequences
.
In:
Cooper
D
,
editor
.
Nature encyclopedia of the human genome
.
Hampshire (United Kingdom)
:
Macmillan Publishers
; 
2003
.
p
.
57
63
.
10.
Katoh
I
,
Kurata
S
. 
Association of endogenous retroviruses and long terminal repeats with human disorders
.
Front Oncol
2013
;
3
:
234
.
11.
Mita
P
,
Boeke
JD
. 
How retrotransposons shape genome regulation
.
Curr Opin Genet Dev
2016
;
37
:
90
100
.
12.
Castro-Diaz
N
,
Friedli
M
,
Trono
D
. 
Drawing a fine line on endogenous retroelement activity
.
Mob Genet Elements
2015
;
5
:
1
6
.
13.
Wolf
G
,
Greenberg
D
,
Macfarlan
TS
. 
Spotting the enemy within: Targeted silencing of foreign DNA in mammalian genomes by the Krüppel-associated box zinc finger protein family
.
Mobile DNA
2015
;
6
:17.
14.
Ehrlich
M
. 
DNA methylation in cancer: too much, but also too little
.
Oncogene
2002
;
21
:
5400
13
.
15.
Wild
L
,
Flanagan
JM
. 
Genome-wide hypomethylation in cancer may be a passive consequence of transformation
.
Biochim Biophys Acta
2010
;
1806
:
50
7
.
16.
Peaston
AE
,
Evsikov
AV
,
Graber
JH
,
de Vries
WN
,
Holbrook
AE
,
Solter
D
, et al
Retrotransposons regulate host genes in mouse oocytes and preimplantation embryos
.
Dev Cell
2004
;
7
:
597
606
.
17.
Durruthy-Durruthy
J
,
Sebastiano
V
,
Wossidlo
M
,
Cepeda
D
,
Cui
J
,
Grow
EJ
, et al
The primate-specific noncoding RNA HPAT5 regulates pluripotency during human preimplantation development and nuclear reprogramming
.
Nat Genet
2016
;
48
:
44
52
.
18.
Hashimoto
K
,
Suzuki
AM
,
Dos Santos
A
,
Desterke
C
,
Collino
A
,
Ghisletti
S
, et al
CAGE profiling of ncRNAs in hepatocellular carcinoma reveals widespread activation of retroviral LTR promoters in virus-induced tumors
.
Genome Res
2015
;
25
:
1812
24
.
19.
Dobin
A
,
Davis
CA
,
Schlesinger
F
,
Drenkow
J
,
Zaleski
C
,
Jha
S
, et al
STAR: ultrafast universal RNA-seq aligner
.
Bioinformatics
2013
;
29
:
15
21
.
20.
Li
S
,
Hu
Z
,
Zhao
Y
,
Huang
S
,
He
X
. 
Transcriptome-wide analysis reveals the landscape of aberrant alternative splicing events in liver cancer
.
Hepatology
2019
;
69
:
359
75
.
21.
Pertea
M
,
Pertea
GM
,
Antonescu
CM
,
Chang
T
,
Mendell
JT
, et al
StringTie enables improved reconstruction of a transcriptome from RNA-seq reads
.
Nat Biotechnol
2015
;
33
:
290
5
.
22.
Quinlan
AR
,
Hall
IM
. 
BEDTools: a flexible suite of utilities for comparing genomic features
.
Bioinformatics
2010
;
26
:
841
2
.
23.
Kang
Y
,
Yang
D
,
Kong
L
,
Hou
M
,
Meng
Y
,
Wei
L
, et al
CPC2: a fast and accurate coding potential calculator based on sequence intrinsic features
.
Nucleic Acids Res
2017
;
45
:
W12
6
.
24.
Li
Z
,
Zhang
J
,
Liu
X
,
Li
S
,
Wang
Q
,
Di
Chen
, et al
The LINC01138 drives malignancies via activating arginine methyltransferase 5 in hepatocellular carcinoma
.
Nat Commun
2018
;
9
:
1572
.
25.
Hong
JY
,
Park
JI
,
Cho
K
,
Gu
D
,
Ji
H
,
Artandi
SE
, et al
Shared molecular mechanisms regulate multiple catenin proteins: canonical Wnt signals and components modulate p120-catenin isoform-1 and additional p120 subfamily members
.
J Cell Sci
2010
;
123
:
4351
65
.
26.
Yanagisawa
M
,
Huveldt
D
,
Kreinest
P
,
Lohse
CM
,
Cheville
JC
,
Parker
AS
, et al
A p120 catenin isoform switch affects rho activity, induces tumor cell invasion, and predicts metastatic disease
.
J Biol Chem
2008
;
283
:
18344
54
.
27.
Mariner
DJ
,
Zheng
Y
,
Reynolds
AB
,
Crawford
HC
,
Thoreson
MA
,
Anastasiadis
PZ
, et al
Inhibition of RhoA by p120 catenin
.
Nat Cell Biol
2000
;
2
:
637
44
.
28.
Zhang
L
,
Gallup
M
,
Zlock
L
,
Finkbeiner
WE
,
McNamara
NA
. 
Rac1 and Cdc42 differentially modulate cigarette smoke–induced airway cell migration through p120-catenin–dependent and–independent pathways
.
Am J Pathol
2013
;
182
:
1986
95
.
29.
Menke
A
,
Giehl
K
. 
Regulation of adherens junctions by Rho GTPases and p120-catenin
.
Arch Biochem Biophys
2012
;
524
:
48
55
.
30.
Anastasiadis
PZ
,
Reynolds
AB
. 
Regulation of Rho GTPases by p120-catenin
.
Curr Opin Cell Biol
2001
;
13
:
604
10
.
31.
Tang
B
,
Tang
F
,
Wang
Z
,
Qi
G
,
Liang
X
,
Li
B
, et al
Overexpression of CTNND1 in hepatocellular carcinoma promotes carcinous characters through activation of Wnt/β-catenin signaling
.
J Exp Clin Cancer Res
2016
;
35
:
82
.
32.
Epifano
C
,
Megias
D
,
Perez-Moreno
M
. 
p120-catenin differentially regulates cell migration by Rho-dependent intracellular and secreted signals
.
EMBO Rep
2014
;
15
:
592
600
.
33.
Yanagisawa
M
,
Anastasiadis
PZ
. 
p120 catenin is essential for mesenchymal cadherin–mediated regulation of cell motility and invasiveness
.
J Cell Biol
2006
;
174
:
1087
96
.
34.
Montalvo-Ortiz
BL
,
Castillo-Pichardo
L
,
Hernandez
E
,
Humphries-Bickley
T
,
De la Mota-Peynado
A
,
Cubano
LA
, et al
Characterization of EHop-016, novel small-molecule inhibitor of Rac GTPase
.
J Biol Chem
2012
;
287
:
13228
38
.
35.
Dharmawardhane
S
,
Hernandez
E
,
Vlaar
C
. 
Development of EHop-016: a small molecule inhibitor of Rac
.
Enzymes
2013
;
33 Pt A
:
117
46
.
36.
Chen
HY
,
Yang
YM
,
Stevens
BM
,
Noble
M
. 
Inhibition of redox/Fyn/c-Cbl pathway function by Cdc42 controls tumour initiation capacity and tamoxifen sensitivity in basal-like breast cancer cells
.
EMBO Mol Med
2013
;
5
:
723
36
.
37.
Hanin
G
,
Shenhar-Tsarfaty
S
,
Yayon
N
,
Yau
YH
,
Bennett
ER
,
Sklan
EH
, et al
Competing targets of microRNA-608 affect anxiety and hypertension
.
Hum Mol Genet
2014
;
23
:
4569
80
.
38.
Timp
W
,
Feinberg
AP
. 
Cancer as a dysregulated epigenome allowing cellular growth advantage at the expense of the host
.
Nat Rev Cancer
2013
;
13
:
497
510
.
39.
Berdasco
M
,
Esteller
M
. 
Aberrant epigenetic landscape in cancer: how cellular identity goes awry
.
Dev Cell
2010
;
19
:
698
711
.
40.
Skulte
KA
,
Phan
L
,
Clark
SJ
,
Taberlay
PC
. 
Chromatin remodeler mutations in human cancers: epigenetic implications
.
Epigenomics
2014
;
6
:
397
414
.
41.
Wang
Y
,
Liu
C
,
Luo
M
,
Zhang
Z
,
Gong
J
,
Li
J
, et al
Chemotherapy-induced miRNA-29c/Catenin-δ signaling suppresses metastasis in gastric cancer
.
Cancer Res
2015
;
75
:
1332
44
.
42.
Hamada
S
,
Satoh
K
,
Miura
S
,
Hirota
M
,
Kanno
A
,
Masamune
A
, et al
miR-197 induces epithelial-mesenchymal transition in pancreatic cancer cells by targeting p120 catenin
.
J Cell Physiol
2013
;
228
:
1255
63
.
43.
Lynch
VJ
,
Nnamani
MC
,
Kapusta
A
,
Brayer
K
,
Plaza
SL
,
Mazur
EC
, et al
Ancient transposable elements transformed the uterine regulatory landscape and transcriptome during the evolution of mammalian pregnancy
.
Cell Rep
2015
;
10
:
551
61
.
44.
Zhang
B
,
Xing
X
,
Li
J
,
Lowdon
RF
,
Zhou
Y
,
Lin
N
, et al
Comparative DNA methylome analysis of endometrial carcinoma reveals complex and distinct deregulation of cancer promoters and enhancers
.
BMC Genomics
2014
;
15
:
868
.