The samples examined in this study were from liver (n = 80), lung (n = 36), kidney (n = 34), oral (n = 28), and colon (n = 15) cancers. Tumor and adjacent normal tissues were collected from patients who underwent surgery at China Medical University Hospital. All samples were immediately snap-frozen in liquid nitrogen and stored at −80°C. The study was approved by the Research Ethics Committee of China Medical University Hospital (CMUH102-REC1-037).

Total RNA was isolated using a NucleoSpin^® RNA Kit (Macherey Nagel, Düren, Germany) and reverse transcribed into complementary DNA (cDNA) using a high-capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA). Quantitative PCR (qPCR) was performed using the IQ2 TaqMan Probe System (Biogenesis, Taiwan) on a LightCycler 480 instrument (Roche Diagnostics, Mannheim, Germany) in accordance with the manufacturer’s instructions. The 2^−ΔΔCt method was used to calculate the relative gene expression normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH). Primer and probe sequences are listed in Supplementary Table 1 .

Proteins were extracted using a cell lysis solution [20 mM Tris/HCl, pH 7.5, 150 mM NaCl, 1 mM ethylenediaminetetraacetic acid (EDTA), 1% Nonidet P-40, 1% sodium deoxycholate, 2.5 mM sodium pyrophosphate, 1 mM b-glycerophosphate, 1 mM Na₃VO₄, and 1 μg·ml⁻¹ leupeptin] and separated by sodium dodecyl sulfate/polyacrylamide gel electrophoresis (SDS/PAGE). After proteins were transferred onto polyvinylidene difluoride membranes (Millipore), the membranes were blocked with 5% bovine serum albumin (BSA) (Santa Cruz Biotechnology) and then exposed at 4°C overnight to the anti-c-MYB (Abcam, Cambridge, MA, USA) and anti-GAPDH (GeneTex, Inc., Taiwan), followed by horseradish peroxidase-conjugated secondary antibody for detection by an ECL chemiluminescence detection system (GE Healthcare, Pittsburgh, PA, USA).

The HCC cell lines (Huh7 and HepG2) were maintained in our laboratory and cultured in Dulbecco’s modified Eagle’s medium (GIBCO BRL, Gaithersburg, MD, USA) supplemented with 10% fetal bovine serum (GIBCO BRL), 100 U/ml penicillin, and 100 mg/ml streptomycin in a humidified incubator containing 5% CO₂ at 37°C. We used short tandem repeats to authenticate HCC cell lines that were used in this study. Primary human hepatocytes (HH) was from nearby normal parts of HCC tissues used as normal liver control.

For NTT knockdown, short hairpin RNA (shRNA) (5′-CAGAGCTGACTACACGGAGTGTT-3′) and negative control shRNA (5′-TTCTCCGAACGTGTCACGTTT-3′) were synthesized by MDBio (Taipei, Taiwan). The NTT and negative control shRNAs were constructed in pSUPER.neo (OligoEngine, Seattle, WA, USA). Huh7 and HepG2 cells were transiently transfected with shRNA or negative control using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA), and a stable clone was selected by G418. The transfection efficiency was determined by qPCR.

For ATF3 knockdown, shATF3 was obtained from the National RNAi Core Facility at Academia Sinica (Taiwan). The shRNA sequence was used for ATF3 knockdown (5′-GCTGAACTGAAGGCTCAGATT-3′).

Cytoplasmic and nuclear cell lysates were isolated using the NE-PER Nuclear and Cytoplasmic Extraction Kit (Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer’s instructions followed by TRIzol RNA extraction. The nuclear fraction was determined by qPCR.

A wound-healing assay was used to assess cell migration. Huh7 and HepG2 cells were seeded and grown in 24-well plates until a monolayer formed. The cells were then scratched using a 200-μl plastic tip to create a straight wound. Wound gaps were photographed and analyzed using the free TScratch software.

Cell proliferation was assessed using a trypan blue exclusion assay. Cells were plated in 96-well plates at 1 × 10⁵ cells per well and incubated for 24 or 48 h. After incubation, the number of cells was determined using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay.

RNA-binding protein immunoprecipitation (RIP) was conducted using the RNA ChIP-IT kit (Active Motif, Carlsbad, CA, USA) according to the manufacturer’s instructions. Briefly, after cross-linking, cells were lysed, and the lysates were incubated with protein G agarose beads conjugated with specific antibodies (anti-EZH2, anti-EED, anti-MLL1, anti-PML, or anti-H3H4me3) or control immunoglobulin G (IgG). Then, the beads were incubated with proteinase K, and the purified RNA was subjected to qRT-PCR.

DNA ChIP was performed using the ChIP-IT kit (Active Motif). Cells were treated with formaldehyde and incubated for 10 min to generate DNA protein cross-links. The cell lysates were sonicated to obtain 200–500-bp chromatin fragments and immunoprecipitated with a specific antibody (anti-PML, anti-H3K4me3, anti-EZH2, or anti-ATF3) or IgG as the control. Precipitated chromatin DNA was recovered and analyzed by qPCR. The MYB promoter primer sequences were 5′-CCTAGCCAAACAGCCTATGAA-3′ (forward) and 5′-TGGAGACGGGGAAATTAGG-3′ (reverse). The NTT promoter primer sequences were 5′-CACCCACATGGTAGACAGGA-3′ (position 5350 forward) and 5′-CCCAGCTCCCAGAAGATACA-3′ (reverse).

Cells were harvested by directly culturing on cover slips. After fixation in 1% formaldehyde, the slides were inserted into solution A [80% formamide, 10% 1× saline sodium citrate (SSC) and 10% H₂O] at room temperature for 2 min. Then, they were quenched in ice-cold 70% ethanol for 5 min to fix the slides. The RNA fluorescence in situ hybridization (FISH) locked nucleic acid (LNA) probe (corresponding to the sequence of U54776.1 from 10,024 to 10,044) was purchased from Exiqon Inc. (MA, USA) and labeled in orange. The slides were dehydrated, air dried, and preheated to 37°C for 2 min before probe addition. The slides were incubated overnight at 37°C in a humid chamber in an incubator. After RNA FISH washing, the slides were postfixed prior to DNA FISH in 4% paraformaldehyde for 15 min at room temperature and rinsed once with phosphate-buffered saline (PBS). The slides were then denatured in 70% formamide and 2× SSC at 80°C for 10 min. The slides were dehydrated before the denatured probes were added for overnight hybridization at 42°C in a dark humid environment. DNA FISH probe were purchased from Cytocell (Cat. No. LPH016). The red 183-kb MYB probe covers the entire MYB gene; the probe mix also contains a green control probe for the chromosome 6 centromere probe (D6Z1). Then, the slides were washed twice in 2× SSC at 45°C, and 4′,6-diamidino-2-phenylindole (DAPI) was added to this slides. Images were obtained using an E600 microscope (Nikon, Tokyo, Japan) with CytoVision software (Applied Imaging, Santa Clara, CA, USA).

Formalin-fixed, paraffin-embedded HCC and paired non-tumor tissue sections (1–2 μm) on poly-l-lysine-coated slides were deparaffinized. Before hybridization, the slides were incubated with 10% pepsin for 30 min at 37°C and then heated in denaturation solution at 72°C for 3 min. The RNA FISH probe (corresponding to the sequence of U54776.1 from 13,559 to 13,620) was labeled with Cy3 dyes using the Label IT labeling kit (Mirus, Madison, WI, USA). Hybridization with the denatured NTT probe was performed in a humid chamber at 37°C overnight. Then, the slides were washed at 45°C in washing buffer for 5 min. Finally, the slides were counterstained with DAPI and analyzed by fluorescence microscopy.

BALB/c athymic nude mice (male, 4–6 weeks old) were purchased from the National Laboratory Animal Breeding and Research Center, Taiwan. To establish a HCC xenograft model, 1 × 10⁷ scramble or shNTT-Huh7 cells were suspended in 100 ml PBS and inoculated subcutaneously into the flanks of eight nude mice (left: shNTT; right: scramble). All animal experiments were performed in accordance with the guidelines set by the Institutional Animal Care and Use Committee (IACUC) of China Medical University (CMU). All animals were housed in the Laboratory Animal Center of CMU under a 12 h light/dark (08:00/20:00) cycle with free access to food and water. The mice were sacrificed using CO₂, and the tissues were subsequently harvested. All breeding and subsequent use of animals in this study, including sacrifice, was approved by the IACUC of CMU. The IACUC approval number was 102-203-N.

Fragments of DNA from the NTT promoter region were obtained through PCR amplification and cloned into the pGL3-basic reporter vector (Promega Corporation). The constructs were transfected into Huh7 cells (1 × 10⁵) using Lipofectamine 2000 (Thermo Fisher Scientific, Inc.). The cells were lysed and assayed for luciferase activity using Steady-Glo Luciferase assay system (Promega Corporation) according to the manufacturer’s protocol at 24 h posttransfection.

Student’s t-test was used to test for differences in mean values between two groups. Correlations between the NTT level and clinicopathological characteristics of the HCC patients were analyzed with the chi-square test. Survival analysis was performed using the Kaplan–Meier method, followed by the log-rank test. The level of statistical significance was set at p < 0.05. All statistical analyses were performed using GraphPad Prism software (ver. 8.0.2; GraphPad Software, Inc., La Jolla, CA, USA).

We explored NTT expression in liver (HCC), lung (adenocarcinoma), kidney (renal cell carcinoma), oral (squamous cell carcinoma), and colon (adenocarcinoma) cancers. NTT expression was higher in cancer than in normal tissue in oral cancer but lower in cancer compared with normal tissues in the other types of cancer ( Figure 1A ). Our data showed that NTT was dysregulated in cancers. NTT showed the highest differences in HCC compared to its surrounding normal tissue, in respect to lung, kidney, oral, and colon cancers. Next, we investigated NTT expression in normal hepatocytes and HCC cell lines (Huh7 and HepG2). We found that the NTT expression in normal hepatocytes was higher than in Huh7 and HepG2 ( Figure 1B ), similar to the findings for HCC and nearby tissues. FISH confirmed that NTT expression was lower in tumor cells than in non-tumor cells ( Figure 1C ). Based on these results, we focused on NTT in HCC and the HCC cell lines Huh7 and HepG2 in the following experiments.

Using the mean NTT expression level in the 80 HCC patients, 62 patients (77.5%) were classified into a low-NTT expression group, and the remaining 18 patients (22.5%) were classified into a high-NTT expression group. Further survival analysis revealed that higher NTT expression was correlated with a shorter survival time in patients with HCC (p < 0.0001, Figure 1D ). Moreover, there were significant correlations between NTT expression and age, tumor size, and tumor stage ( Supplementary Table 2 ). In the multivariate analysis, which incorporated independent prognostic factors of gender, age, differentiation, tumor size, and stage, we found that NTT was not an independent predictor for overall survival (p = 0.059) ( Supplementary Table 3 ).

lncRNA usually plays an important role in the regulation of genes nearby. We measured the expression of genes near NTT in HCC tissues. The genes were as follows: CTGF, STX7, MYB, BCLAF1, IFNGR1, TNFAIP3, and HIVEP2 ( Supplementary Figure 1 ). The expression of CTGF and MYB was significantly higher in HCC tissues than that in the adjacent non-tumor tissues (p < 0.05, Figure 2A ), while the expression of BCLAF1, IFNGR1, and HIVEP2 was significantly lower in HCC tissues than in adjacent non-tumor tissues (p < 0.05, Figure 2A ). The high MYB levels in tumors predicted a poor clinical outcome (p = 0.014, Figure 2B ), similar to the results for NTT expression. Based on these results, we explored the relationship between MYB and NTT, and the regulation mechanism.

First, we analyzed the correlation between the expressions of NTT and MYB in human tissues ( Figure 2C ). We found that NTT-positive regulation of MYB expression in 55% (44/80) of the HCC tissues, which including both NTT and MYB expressions were increased in 17 HCC tissues and both expressions were decreased in 27 HCC tissues ( Figure 2D ). From these results, we suggested that NTT plays like an oncogene.

However, NTT negative regulation of MYB expression in 45% (36/80) of the HCC tissues, which contained NTT expression, was decreased, and MYB expression was increased in 35 HCC tissues; NTT expression was increased and MYB expression was decreased in one HCC tissue ( Figure 2D ).

No differences were observed in the survival rates between patients with NTT positive and negative regulation of MYB expression (p = 0.796) ( Supplementary Figure 2A ). Further survival analysis revealed that there was significant correlation between higher and lower expression of both NTT and MYB (p = 0.000) ( Supplementary Figure 2B ). Survival is lower among both NTT- and MYB-upregulated patients compared with both NTT- and MYB-downregulated patients ( Supplementary Figures 2B, D ).

We also found that MYB play a determining role for HCC survival, overexpression of MYB resulted in poor survival, and downexpression of MYB correlated with good survival ( Figure 2D ). The expression of NTT could not determine patient’s survival directly. These data support the context-dependent role of NTT in liver tumorigenesis. Western bot analysis was also done to confirm the MYB protein expression in HCC tissue samples, the results of representative cases showed that NTT overexpression upregulated the expression of MYB, and NTT downexpression downregulated the expression of MYB in the paired normal and tumor tissues ( Figure 2E ).

We further explored the biological role of NTT in HCC cell lines. The shNTT plasmid was transfected into HCC Huh7 and HepG2 cells, which effectively knocked down the expression of NTT ( Figure 3A ). Knockdown of NTT significantly reduced the number of migratory cells in Huh7 and HepG2 cells compared with the sh-ctrl transfected cells ( Figure 3B ).

To test whether NTT levels affect cell proliferation, viable cells were counted using the MTT assay after Huh7 cells were transfected with shRNA for 48 h. We found no difference in cell proliferation between the Huh7 cells transiently transfected with sh-ctrl and shNTT ( Supplementary Figure 3 ).

Next, we investigated how NTT regulates the MYB gene. NTT expression in Huh7 and HepG2 were knocked down by shRNA, and the relative MYB mRNA level was analyzed ( Figure 4A ). Transfection of shNTT in Huh7 and HepG2 led to decreased MYB expression.

Since it has been reported that lncRNAs may bind to polycomb-repressive complex 2 (PRC2), which consists of EZH2, EED, and SUZ12, to repress the expression of downstream genes, we examined whether NTT binds to the MYB promoter via interaction with PRC2. An RIP assay showed that NTT could bind EZH2 and EED ( Figure 4B ). Furthermore, we found that NTT not only bound to PRC2 but also bound to the activated complex (MLL1, PML, and H3K4me3) to activate gene transcription ( Figure 4B ). A similar result was observed with the HepG2 cells ( Figure 4B ).

DNA ChIP showed that the activated complex binds to positions 706–778 of the MYB promoter, and activated complex binding became undetectable or reduced following NTT knockdown ( Figure 4C ), suggesting that NTT enhances MYB expression by interacting with the activated complex binding to the MYB promoter. In agreement with the above results, we also found that NTT knockdown reduced the binding of PML and MLL1 to the MYB promoter ( Figure 4C ).

To further confirm the localization of NTT RNA transcript with MYB gene area, we used DNA (MYB genomic probe) FISH to detect the genomic location of MYB and RNA (NTT 21 base pair LNA probe) FISH to detect the NTT RNA transcript. Colocalization of NTT transcripts with the MYB gene was revealed ( Figure 4D ). These results suggest that NTT modulates nearby genes, such as MYB, via RNA transcripts and related protein complexes.

To explore the molecular mechanism underlying MYB-induced tumor promotion and metastasis, genes associated with tumor progression and metastasis were examined in sh-ctrl and shNTT cells by real-time quantitative PCR (RT-qPCR). Bcl-xL, cyclinD1, and VEGF are associated with cell proliferation, the cell cycle, and cell migration/invasion, respectively. All three were downregulated following NTT knockdown ( Figure 4E ).

Examining the NTT promoter sequence, we identified the potential binding motif of ATF3, a key transcription factor in HCC cell lines. To verify if ATF3 binds to the NTT promoter and regulates NTT expression, we performed DNA ChIP and shRNA knockdown of ATF3. ATF3 binding was detected at position 5350 of the NTT promoter in Huh7 and HepG2 ( Figure 5A ), and ATF3 knockdown resulted in decreased NTT and its downstream genes expression ( Figure 5B ).

We have performed the luciferase reporter assays to determine the NTT promoter activity in ATF3-knockdown cells, and the results showed that the luciferase activity was decreased after knockdown ATF3 ( Figure 5C ).

We constructed stable cell lines that knocked down NTT or sh-ctrl control to verify the effect of NTT on tumorigenicity. From the in vivo experiments, we could not define NTT is an oncogene or tumor suppressor. Therefore, we examined NTT and MYB expression of eight mice tumor samples. The results showed both NTT and MYB overexpression result in larger tumor size than control in three-fourths of the mice, and NTT downexpression with MYB overexpression may result in either larger or smaller tumor or no change size than control (one-third with small, one-third with larger, and one-third with no definite change than control), and one mouse with both downexpression of NTT and MYB showed slight size change than the control ( Supplementary Figure 4 ). These data are consistent with the findings of the human tissues experiments and suggested a context-dependent function of NTT for the regulation of MYB expression in in HCC development.