HCC and the corresponding normal liver tissues were collected from 72 patients who were diagnosed with HCC and underwent curative surgery at the First Affiliated Hospital of Jinzhou Medical University. Besides, the patients were not subjected to other treatment prior to the surgery. Apart from that, informed consent was obtained from all patients, and the study was carried out according to the ethical guidelines of the Declaration of Helsinki and was approved by the Ethics Committee of the First Affiliated Hospital of Jinzhou Medical University.

Human normal hepatocyte THLE-2 cells and HCC cell lines, including Hep3B, Huh7, MHCC97h, SK-Hep-1, PLC/PRF/5, and HCCLM3, were obtained from the Institute of Biochemistry and Cell Biology of the Chinese Academy of Sciences (Shanghai, China). Then, these cells were cultured in DMEM medium (Gibco) with 10% fetal bovine serum (Gibco). Besides, siRNAs were purchased from GenePharma Company (Shanghai, China), and then they were transfected into cells using Lipofectamine™ RNAiMAX Transfection Reagent (Thermo) according to the manufacturer’s instructions. After that, 48 h later, the cells were used for further experiments. As for the target sequences of siRNAs, they were as follows: siALKBH5: GGCTCATCCTTACGTAGTT; siYTHDF2: GCTCTGGATATAGTAGCAATT; siUBE2T: GCTGACATATCCTCAGAATTT.

The lentiviral particles expressing scramble control or CASC11 shRNA (GenePharma, Shanghai, China) were used to knock down CASC11 expression in Hep3B cells, while those expressing the empty vector or CASC11 or mutant CASC11 (GenePharma, Shanghai, China) were infected into Huh7 cells. Apart from that, it should be mentioned that the target sequence of shRNAs was shown as follows: scramble shRNA (shCon): TTCTCCGAACGTGTCACGT; shCASC11-1: TGCAGAAGGTCCGAAGAAA; shCASC11-2: GGTTCAGAGGTGACTATTC. Furthermore, the stable cells were selected using 2 μg/ml puromycin for 2 weeks.

The total RNA from tissues or cells was extracted using Trizol reagent (Invitrogen) according to the standard protocol, while the cDNA was synthesized using PrimeScript™ 1st Strand cDNA Synthesis Kit (Takara). In addition, qRT-PCR was carried out on ABI StepOne Plus System (Applied Biosystems) using a standard protocol from the SYBR^® Green Premix Pro Taq HS qPCR Kit (Accurate Biology).

Firstly, 2.0×10³ cells were seeded into a 96-well plate. After that, 10 μl CCK-8 reagent was added into each well at different time points and then incubated for 1 h. Finally, the optical density value was measured at 450 nm using Multiskan™ FC System (Thermo Scientific).

The cell migration and invasion assay was carried out using Transwell chambers (8 μm pore size) without or with Matrigel (BD Bioscience), respectively. To be specific, cells suspending in the 200 μl serum-free DMEM medium were plated into the upper chamber, and DMEM supplemented with 10% FBS was placed into the lower chamber. After 24 h, the cells that passed through the membrane to the lower surface were fixed by 4% paraformaldehyde for half an hour, stained by 0.5% crystal violet for 10 min, and then counted under a microscope (Zeiss).

Cells were treated with the RNA synthesis inhibitor, Actinomycin D (2 μg/ml). At different time points, the RNA from cells was extracted, and then the loss of UBE2T mRNA was detected by qRT-PCR.

Proteins were extracted by a RIPA lysis buffer (Beyotime Company) and quantified by a bicinchoninic acid (BCA) protein quantification kit (Thermo). Besides, protein lysates were separated using SDS-PAGE and transferred onto a PVDF membrane (Millipore) that then was blocked with 5% non-fat milk and incubated with specific primary antibodies overnight at 4°C. After washing, the membrane was incubated with their corresponding second antibodies (Jackson), followed by the bands being visualized by the Immobilon Western Chemilum HRP Substrate (Millipore). As for anti-UBE2T, anti-ALKBH5, anti-FTO, anti-YTHDF2, and anti-GAPDH antibodies, they were used and purchased from Abcam.

The full-length transcript of UBE2T was cloned and inserted into pmirGLO plasmid (Promega). Apart from that, the mutant UBE2T reporter plasmid was constructed by replacing the adenosine bases within the m⁶A consensus sequences to cytosine. Then, wild-type and mutant UBE2T reporter plasmid was transfected into cells. After 48 h, cells were assayed with Dual-Luciferase™ Reporter (DLR™) Assay Systems. Furthermore, luciferase activity was measured and calculated as the relative ratio of firefly luciferase activity to renilla luciferase activity.

For detection of the interaction between CASC11 and UBE2T mRNA, MS2bs (MS2-binding protein)-MS2bp (MS2-binding sequences)-based RIP was performed according to the previous studies (9, 17). Detailedly, cells were transfected with pLV-MS2, pLV-CASC11-MS2, or pLV-CASC11-mut-MS2, and pMS2-GFP (Addgene). After 48 h, cells were used to perform the RIP assay via a GFP antibody (Roach) and the Magna RIP RNA-Binding Protein Immunoprecipitation Kit (Millipore) according to the manufacturer’s instruction. Besides, the RNA pulled down by CASC11 was detected using qRT-PCR or employed for RNA sequencing (BGI Group, Guangdong, China). For identifying the interaction of CASC11 or UBE2T mRNA with ALKBH5, FTO or YTHDF2, cells were used to conduct RIP under the help of the Magna RIP RNA-Binding Protein Immunoprecipitation Kit (Millipore) according to the manufacturer’s instructions. Here, it should be mentioned that anti-ALKBH5, anti-FTO, anti-YTHDF2, and negative control IgG antibodies were purchased from Abcam.

The RNA pull-down assay was performed as previously described (9). In brief, wild-type or mutant CASC11 was transcribed in vitro, respectively, and biotin-labeled with the Biotin RNA Labeling Mix (Roche) and T7 RNA polymerase (Roche), treated with RNase-free DNase I (Roche), and purified with the RNeasy Mini Kit (Qiagen). Apart from that, 1 mg of whole-cell lysate from cells was incubated with 3 ug of purified biotinylated transcripts for 1 h at 25°C, and complexes were isolated with streptavidin agarose beads (Invitrogen). Moreover, the RNA present in the pull-down material was detected by qRT-PCR analysis.

For measuring the m⁶A modification of UBE2T mRNA, the MeRIP assay was carried out using Magna MeRIP™ m⁶A Kit (Millipore) following the manufacturer’s instructions. The antibody used in this experiment was as follows: m⁶A (Abcam); IgG (Millipore).

Male athymic BALB/c nude mice (4–6 weeks old) were maintained under specific pathogen-free conditions. All animal work was conducted according to national guidelines and approved by the ethical committee of the First Affiliated Hospital of Jinzhou Medical University. For the subcutaneous implantation model, 5×10⁶ control and CASC11 overexpressing Huh7 cells were suspended in the 100 μl serum-free DMEM medium and then injected subcutaneously. Then, the tumor volumes were measured. After 4 weeks, mice were sacrificed and the tumors were weighed. Other than that, for the metastatic model, 1×10⁵ control and CASC11 overexpressing Huh7 cells were suspended in the 50 μl serum-free DMEM medium, followed by being injected into the tail vein of nude mice. After 8 weeks, mice were sacrificed, and the lung tissues were isolated and used for H&E staining. Furthermore, the metastatic nodules in lung tissues were counted under a microscope (Zeiss).

Statistical analysis was conducted using SPSS 16.0 or Prism, while data were analyzed by the two-tailed student’s t test between two groups or by one-way ANOVA followed by Bonferroni test for multiple comparison. Besides, the relationship between CASC11 expression and prognosis of HCC patients was determined using Kaplan-Meier survival analysis and the log-rank test, whereas the correlation between CASC11 and UBE2T expression was measured using by Pearson correlation analysis, when a p-value less than 0.05 was considered to be statistically significant.

To determine the expression pattern of CASC11, firstly, the CASC11 expression in 72 pairs of HCC and matched normal liver tissues was detected. The results showed that HCC tissues exhibited higher CASC11 expression compared with normal liver tissues ( Figure 1A ). Then, the correlation between CASC11 and clinicopathological features of HCC patients was analyzed. To be specific, the patients were divided into CASC11-high and CASC11-low groups based on the median value of CASC11 expression in HCC tissues. As displayed in Table 1 , CASC11 expression was significantly associated with the tumor grade and metastasis, while Kaplan-Meier method and the log-rank test demonstrated that high CASC11 expression was correlated with the poor overall survival rate of HCC patients to some extent ( Figure 1B ). Furthermore, the expression of CASC11 between HCC cell lines and human normal hepatocyte THLE-2 cells was also tested. Similarly, CASC11 expression was much higher in HCC cell lines than that in THLE-2 cells ( Figure 1C ). Thus, the data suggest that CASC11 may function as an oncogene in HCC.

The effect of CASC11 on HCC cell proliferation was determined. Among the HCC cell lines ( Figure 1C ), Hep3B cells displayed the highest level of CASC11, while Huh7 cells expressed the lowest CASC11 level. Therefore, Hep3B cells were chosen for knockdown experiments and Huh7 cells were selected for overexpression experiments, respectively. In addition, the knockdown or overexpression efficiency of CASC11 was validated by qRT-PCR ( Figures 2A, B ), whereas the CCK-8 and colony formation assays were performed, and then it was found that knockdown of CASC11 obviously decreased the proliferation rate of Hep3B cells, while overexpression of CASC11 accelerated the Huh7 cell proliferation ( Figures 2C–F ). Furthermore, to confirm the promotion of CASC11 in HCC proliferation in vivo, the control and CASC11-overexpressed Huh7 cells were subcutaneously injected into nude mice, and then the tumor growth was examined. The results showed that the Huh7-CASC11 group exhibited larger tumor volume and weight than the control group ( Figures 2G–I ). Taken together, these findings suggest that CASC11 promotes HCC cell proliferation both in vitro and in vivo.

The correlation between CASC11 expression and metastasis implied that CASC11 may affect the metastatic ability of HCC cells. To validate this idea, Transwell assays were carried out, and the results demonstrated that knockdown of CASC11 repressed the migration and invasion of Hep3B cells ( Figure 3A ). Conversely, the migratory and invasive capacity was dramatically enhanced by CASC11 overexpression in Huh7 cells ( Figure 3B ). Furthermore, the tail vein injection model was established to prove the pro-metastatic effect of CASC11 in vivo. In fact, the CASC11-overexpressed group was featured with more metastatic nodules in lungs than the control group ( Figure 3C ). Collectively, our data reveal a pro-metastatic function of CASC11 in HCC cells in vitro and in vivo.

Then, the underlying mechanism of CASC11 regulating HCC growth and metastasis was investigated. Firstly, the cellular distribution of CASC11 in HCC cells was tested, and it was observed that CASC11 was mainly located in the cytoplasm in both Hep3B and Huh7 cells ( Figure 4A ). Besides, the cytoplasm could associate with mRNA, resulting in the stabilization of target mRNA (9). However, whether CASC11 functions in this manner remains unknown. Apart from that, a MS2bs-MS2bp-based RIP assay was also performed to pull down endogenous mRNAs associated with CASC11 ( Figure 4B ), and the retrieved RNA was sequenced. In terms of ubiquitin-conjugating enzyme E2T (UBE2T), it was the most enriched transcript pulled down by CASC11, and has been reported to contribute to tumor growth and metastasis (18). In addition, the highly complementary region between CASC11 and UBE2T mRNA was identified using BLAST (http://blast.ncbi.nlm.nih.gov/) ( Figure 4B ), whereas the binding sites in CASC11 were mutated (named CASC11-Mut). Furthermore, to confirm the direct interaction between CASC11 and UBE2T mRNA, the MS2bs-MS2bp-based RIP assay was performed, and the results demonstrated that the UBE2T mRNA was noticeably enriched by CASC11 in both Huh7 and Hep3B cells than by the empty vector, IgG and CASC11-Mut ( Figure 4C ). Beyond that, the results of the RNA pull-down assay further confirmed the interaction of CASC11 with UBE2T mRNA ( Figure 4D ).

The effect of CASC11 on UBE2T expression was detected by the western blot and qRT-PCR. Then, it was observed that CASC11 knockdown significantly decreased both mRNA and protein level of UBE2T in Hep3B cells ( Figures 4E, F ), while wild-type CASC11, but not mutant CASC11, upregulated UBE2T expression in Huh7 cells ( Figures 4G, H ). To test whether CASC11 regulates the stability of UBE2T mRNA, HCC cells with CASC11 knockdown or overexpression were treated with Actinomycin D to block new RNA synthesis, before the loss of UBE2T mRNA was measured in different time points. After that, it was known that the overexpression of wild-type CASC11 instead of its mutant suppressed the degradation of UBE2T mRNA ( Figure 4I ), whereas silence of CASC11 shortened the half-life of UBE2T mRNA ( Figure 4J ). To sum up, these findings reveal that CASC11 increases UBE2T expression through enhancing the stability of UBE2T mRNA, which depends on the association with UBE2T mRNA.

Then, the mechanism of CASC11-mediated stability of UBE2T mRNA was explored. Increasing evidence has demonstrated that N⁶-methyladenosine (m⁶A) modification is involved in the mRNA decay process (19, 20). To testify whether CASC11 modulates UBE2T mRNA stability through this manner, the MeRIP assay was performed. Then, it was found that depletion of CASC11 expression increased the m⁶A level of UBE2T mRNA ( Figure 5A ) to a great degree, while overexpression of wild-type CASC11, instead of mutant CASC11, obviously decreased m⁶A modification of UBE2T mRNA ( Figure 5B ), implying that CASC11 might recruit RNA demethylase to UBE2T mRNA when the m⁶A demethylases include FTO and ALKBH5. In this case, to validate which demethylase was involved in this process, RIP assays using FTO or ALKBH5 antibody were carried out. After that, it was discovered that CASC11 could only be observably enriched by ALKBH5 antibody in both Huh7 and Hep3B cells ( Figure 5C ). Moreover, CASC11 knockdown repressed the association between ALKBH5 and UBE2T mRNA ( Figure 5D ), while the interaction of them was enhanced by upregulation of wild-type CASC11, but not mutant CASC11 ( Figure 5E ). Additionally, siRNA targeting ALKBH5 3’UTR (siALKBH5) was transfected into CASC11-overexpressed Huh7 cells, whereas the CASC11-mediated upregulation of UBE2T was attenuated by ALKBH5 siRNAs. Besides, restoration of ALKBH5 expression rather than catalytic-inactive mutant ALKBH5 (H204A) reversed this effect ( Figure 5F ).

To further address the m⁶A modification of UBE2T mRNA, both wild-type and mutant UBE2T reporters were constructed. Then, to construct the mutant form of UBE2T, the adenosine bases in m⁶A consensus sequences (i.e., RRACH) were replaced by cytosine, which abolished m⁶A modification. Apart from that, the wild-type or mutant UBE2T-fused reporter was transfected into Huh7 cells. When compared to the wild-type, mutation on the m⁶A consensus sequences increased the luciferase activity of UBE2T. Moreover, luciferase activity of the wild-type UBE2T-fused reporter was significantly increased after CASC11 overexpression or ALKBH5. However, upregulation of CASC11 or ALKBH5 did not affect the luciferase activity of the mutant UBE2T-fused reporter ( Figure 5G ). These results prove that CASC11 strengthens UBE2T mRNA stability via ALKBH5-mediated m⁶A demethylation.

YTH N⁶-methyladenosine RNA binding protein 2 (YTHDF2), the m⁶A reader protein, selectively recognizes and binds m⁶A-containing mRNA for the purpose of promoting mRNA degradation. Then, the effect of CASC11 on the interaction between YTHDF2 and UBE2T mRNA was confirmed. YTHDF2 could associate with UBE2T mRNA, which was enhanced by CASC11 knockdown in Hep3B cells, but suppressed by CASC11 overexpression in Huh7 cells ( Figures 6A, B ). Furthermore, siRNAs targeting YTHDF2 were transfected into Hep3B cells with CASC11 knockdown. Besides, depletion of YTHDF2 expression upregulated UBE2T expression, which was inhibited by CASC11 knockdown ( Figures 6C, D ). Together, our findings show that CASC11-mediated m⁶A demethylation enhances UBE2T expression through suppressing the YTHDF2-dependent mRNA degradation.

Finally, rescue experiments were carried out to examine whether CASC11 promotes malignant phenotypes of HCC cells through UBE2T. Knockdown of UBE2T significantly abolished the proliferation, migration, and invasion, which was enhanced by CASC11 overexpression in Huh7 cells ( Figures 7A–C ). Besides, the pathological association between CASC11 and UBE2T was also examined in 72 HCC tissues, and after that, the statistical analysis revealed a positive correlation of CASC11 with UBE2T mRNA expression ( Figure 7D ), supporting that CASC11 functions as an oncogene via regulation of UBE2T in HCC.