Long noncoding RNA TINCR facilitates hepatocellular carcinoma progression and dampens chemosensitivity to oxaliplatin by regulating the miR-195-3p/ST6GAL1/NF-κB pathway

The tumors and normal liver tissues were obtained from HCC patients who received hepatectomy in Sun Yat-sen University Cancer Center (SYSUCC). Detailed clinicopathological data were collected according to patient medical records. The long-term follow-up interval was from June 2010 to December 2020. This study was conducted according to the ethical guidelines of the 1975 Declaration of Helsinki, and approved by the Institutional Ethical Review Board of the SYSUCC (B2019–057-01).

The human HCC cell lines (HepG2, HuH7, MHCC-LM3, MHCC-97H, MHCC-97 L, PLC/PRF/5, SK-HEP-1, Hep3B, SMMC-7721, SNU-449) were cultured in DMEM (Invitrogen) supplemented with 10% fetal bovine serum (Gibco) in an incubator sustaining an atmosphere of 5% CO2 and 37 °C. All the cell lines were purchased from the Cell Lines Service (Cellcook Biotech Co., Ltd., Guangzhou, China).

Total RNA was purified using Trizol (Invitrogen) or RNA quick purification kit (ESscience, China). Reverse transcription was performed using miRNA 1st Strand cDNA Synthesis Kit (Vazyme) with Bulge-Loop miRNA-specific RT primers (RiboBio) for miRNA or random primers (Promega) for mRNA and lncRNA. The primers used in qRT-PCR are listed in Supplementary Table S1. The relative expression level was compared with that of β-actin and fold changes were calculated using the 2^-△△ct method.

Lipofectamine 3000, RNAiMAX, or Opti-MEM I reagents (Invitrogen) were used for transient transfection. The siRNA targeting TINCR, ST6GAL1 were synthesized and purchased from GenePharma. The sequence of TINCR, ST6GAL1, and nonsensical fragment were synthesized and cloned into pcDNA3.1 by Umine Biotechnology Co., LTD (Guangzhou). The mimics and inhibitors of miR-195-3p were designed and supplied by RiboBio. Stably transfected cell lines were constructed and purchased from Obio Technology Corp. All the sequence is listed in Table S2. Conventionally, transfection systems were co-cultured with targeted cells for 12 h, and then changed to normal medium.

Transfected cells (2 × 10³) were seeded in 96-well plates, and cell viability was detected every 24 h for 5 days using cell counting kit-8 (cck-8) (Glpbio). In the colony formation experiment, transfected cells (4 × 10²) were seeded in 6-well plates with and incubated for 14 days, and then fixed, stained, and counted. In addition, the cells were incubated with oxaliplatin (0, 0.5, 1, 2, 4, 8, 16, and 32 μg/ml) for 72 h and then tested for cell viability. In apoptosis assay, the cells were incubated with oxaliplatin at the right concentration for 24 h and then collected for flow cytometry assay.

Transfected cells were digested with 0.25% trypsin and washed twice with PBS, adjusting to a concentration of 1 × 10.⁶ Next, 10 μl of Annexin V-FITC and 10 μl propidium staining solution (ESscience, China) were mixed and added to a 0.1 ml cell suspension and incubated at room temperature for 5 min in the dark. Apoptosis was detected by flow cytometry (FACSVantageSE, BD, USA).

Transwell chambers (Corning) coated without or with matrigel (Corning) were prepared for cell migration and invasion, respectively. Cells (1 × 10⁵) suspended in 200 μl of serum-free medium were seeded in the upper chamber, while 600 μl medium mixed with 10% FBS was added to the lower chamber. After 24 h, the cells that migrated/invaded to the lower surface of the membrane were observed and counted.

Cells were collected and lysed in ice-cold RIPA lysis buffer (Sigma) containing protease and phosphatase inhibitor (Selleck) for 15 min. The supernatant was collected after centrifugation at 14000×g for 15 min. The protein concentration was determined using a BCA protein quantification kit (Beyotime). Then, 30 μg of total protein was added to a 10% SDS-PAGE gel and transferred to a 0.25-μm PVDF membrane (Bio-Rad). The PVDF membrane was blocked with 5% skim milk for 1 h at 37 °C. Next, the membrane was incubated with primary antibodies against GAPDH (Abcam, ab8245), ST6GAL1 (Proteintech, 14,355–1-AP), IκBα (Cell Signaling Technology, no.4814), p-IκBα (Cell Signaling Technology, no.2859), p65(Cell Signaling Technology, no.3031), p-p65(Cell Signaling Technology, no.3033) overnight at 4 °C. After incubation with horseradish peroxidase (HRP)-labeled secondary antibodies at 37 °C for 30 min, the bands were detected using enhanced chemiluminescence (GBCBIO, China) with the ChemiDoc MP Imaging System (Bio-Rad).

The wild-type or mutant fragment of TINCR or ST6GAL1 containing the predicting binding sequence of miR-195-3p was subcloned into a psiCHECK2 Dual-luciferase vector (Promega). To verify the target genes of miRNAs, HCC cells were co-transfected with luciferase reporter plasmids and miR-195-3p mimics or negative control. Luciferase activity was measured using the dual-luciferase reporter assay system (Promega). Renilla luciferase expressed by pRL-PGK (Promega) was used as an internal control to correct for differences in both transfection and harvest efficiency.

The biotin-labeled miRNA pull-down assay was performed using the RNA Pulldown Kit (BersinBio, Guangzhou, China). Briefly, cells were transfected with biotin-labeled miR-195-3p and miR-ctrl (RiboBio, 50 nM), and lysed after 48 h incubation. Simultaneously, 25 μg streptavidin magnetic beads were mixed with sample lysates and incubated with rotation overnight at 4 °C. Beads were then washed to remove unbound materials. RNA was eluted, isolated, and subjected to qRT-PCR analysis.

BALB/c nude mice (4–5 weeks old, female) were provided by Guangdong Animal Experiment Center. For tumor growth model, sh-TINCR and scrambled control cells (4 × 10⁶) were resuspended in 100 μl PBS containing 10% matrigel (Corning), and then subcutaneously injected into either side of the posterior flank of the same mouse. When tumor volume reached approximately 100 mm³, the mice were randomly divided into four groups (n = 7), and intraperitoneally injected with normal 5% glucose solution or oxaliplatin (Selleck, 3 mg/kg) every 3 days. Tumors were measured every 3 days from the start of dosing. On day 24, mice were sacrificed and tumors were detached for further measurement and immunohistochemistry detection. All animal experiments complied with the 1978 National Institutes of Health guide for the care and use of laboratory animals. The animal procedures were approved by the Institutional Animal Care and Use Committee of SYSUCC (L102042019110H).

Paraffin-embedded tumor tissues were cut as 5-μm sections and processed for IHC. Tissue sections were prepared for antigen retrieval using microwave treatment in citrate buffer (pH 8.0, Beyotime) and then incubated with anti-ST6GAL1 antibody (Proteintech, 14355–1-AP) overnight at 4 °C. Immunostaining was performed using the Envision Inspection System with diaminobenzidine as substrate (DAKO Cytomation, Glostrup, Denmark).

To identify lncRNAs that regulate HCC progression, we performed a full transcriptome sequence on 5 pairs of tumor and normal tissues from 5 HCC patients. 49 lncRNAs with different expressions were found (fold change > 2, FDR < 0.05, and P < 0.01; Fig. 1A), of which TINCR was one of the significant lncRNAs upregulated in HCC. In addition, we reconfirmed that TINCR expression was significantly higher in 50 HCC tissues compared to 50 normal tissues by qRT-PCR (Fig. 1B-C; Fig. S1A, P < 0.001).

We further investigated the prognostic value of TINCR with 139 HCC frozen HCC tissues using qRT-PCR. According to the median value of TINCR expression, the patients were divided into high and low expression groups. The result showed that patients in the high TINCR group were associated with worse overall survival (Fig. 1D-E, P < 0.001) and recurrence-free survival (Fig. 1F, P < 0.026). In addition, combined with the clinical data of the patients, we surprisingly found that TINCR expression was not related to those generally recognized risk factors, including tumor size, number, vascular invasion, degree of differentiation (Table. S3). However, the univariate and multivariate analysis suggested that TINCR expression was an independent prognostic factor, which further confirmed the unique biological regulatory function of TINCR (Table. S4–5).

We then constructed a prognostic model integrating albumin-bilirubin grade, tumor size, macrovascular invasion, and TINCR expression, and HCC patients were divided into low, intermediate, and high-risk groups (Table. S6). Survival plots showed significant stratification in the overall and recurrence-free survival among the three groups (Fig. 1G-H, both P < 0.001).

The expression of TINCR was evaluated in various common HCC cell lines using qRT-PCR (Fig. S1B). Given the generally low expression value of TINCR in HCC cell lines, HuH7 and HepG2, with relatively higher expression of TINCR compared with other cell lines, were selected as candidates for further functional exploration. We transfected HuH7 and HepG2 cells with two siRNAs against TINCR (si-TINCR 1# and si-TINCR 2#) and TINCR-overexpressing plasmid (pcDNA3.1-TINCR). The transfection efficiency was detected using qRT-PCR (Fig. 2A-B). CCK-8 and colony formation assays showed significant inhibition of proliferation of HCC cells transfected with si-TINCR compared to those transfected with scrambled control (Fig. 2C and E, all P < 0.001), similarly the opposite effects were observed in HCC cells transfected with pcDNA3.1-TINCR compared to those transfected with empty vector (Fig. 2D and F, P < 0.05). In addition, silencing TINCR inhibited HCC cells migration and invasion (Fig. 2G, all P < 0.01), while overexpressing TINCR enhanced those cell behaviours (Fig. 2H, all P < 0.05). Moreover, we examined the role of TINCR in the regulation of oxaliplatin sensitivity. Under different concentration gradients of oxaliplatin, CCK-8 and apoptosis assays revealed that silencing TINCR could significantly enhance the sensitivity of HCC cells to oxaliplatin (Fig. 3A-D; Fig. S1C, all P < 0.01). Furthermore, the opposite effect was verified in HCC cells with overexpression (Fig. 3E-H; Fig. S1D, all P < 0.05). These findings indicated that TINCR acts as an oncogenic lncRNA to promote proliferation, invasion, and migration, and induces sensitivity to oxaliplatin in HCC.

To figure out the underlying regulatory mechanism, we firstly explored the predictive location of TINCR in LncLocator (http://​www.​csbio.​sjtu.​edu.​cn/​bioinf/​lncLocator). TINCR was presumed to be located mainly in the cytoplasm (Fig. 4A). QRT-PCR analysis of TINCR expression in the nucleus and cytoplasm further confirmed this conclusion (Fig. 4B). Based on these results, we hypothesized that TINCR functions as a ceRNA. To find the binding miRNAs, we performed the sequencing of the miRNA of HepG2 with TINCR knockdown compared with control treatment and found the miR-195-3p was one of the significantly increased miRNAs with the most maximum fold change value (Fig. 4C). Next, dual-luciferase reporter assays showed that overexpression of miR-195-3p reduced luciferase activity of the wild-type TINCR gene fragment, however, not the mutant TINCR vector (Fig. 4D-E, P < 0.001). We further verified the opposite expression value of TINCR and miR-195-3p through qRT-PCR in HCC cells. The results showed that miR-195-3p was upregulated with suppression of TINCR, while downregulated with overexpression of TINCR (Fig. 4F-G, all P < 0.001). In contrast, TINCR was downregulated when miR-195-3p was overexpressed (Fig. 4H, P < 0.001). RNA pulldown assay showed a conspicuous elevated ratio of TINCR expression to total input in HCC cells transfected with biotin-labeled miR-195-3p compared with that in the control (Fig. 4I, P < 0.01). Finally, findings from the TCGA database and Starbase (http://​starbase.​sysu.​edu.​cn/​index.​php) indicated that miR-195-3p was negatively correlated with TINCR, and might serve as a tumor suppressor molecule with an opposite trend in clinicopathological characteristics and survival prediction to TINCR (Fig. 4J-K; Fig. S2A). The above results suggested that the carcinogenic effect of TINCR is partially mediated by the negative regulation of miR-195-3p.

To find the downstream target protein of miR-195-3p, we screened the target genes with a cumulative weighted context score less than − 0.1 in TargetScan (http://​www.​targetscan.​org/​). Among these predicted target genes, ST6GAL1 was the probable candidate gene, since it exhibited an upward trend according to the sequencing of mRNA of HepG2 with TINCR knockdown compared with control treatment (Fig. S2G), and was reported to be closely related to tumor progression and drug resistance in several studies [19]. We found that ST6GAL1 expression was affected by the change of miR-195-3p (Fig. 5A-B, P < 0.01). To validate this potential interaction, dual-luciferase reporter assays showed that overexpression of miR-195-3p reduced luciferase activity of the wild-type ST6GAL1 gene fragment, but not the mutant ST6GAL1 vector (Fig. 5C-D, p < 0.001). These results indicated that ST6GAL1 was the target gene of miR-195-3p.

In addition, western blotting revealed that the expression of ST6GAL1 was positively correlated with the expression of TINCR (Fig. 5E-F, P < 0.01), which was further validated using the GEO data set (GSE54236, Fig. 5G). To find the potential signaling pathway, GSEA analysis was performed based on the TCGA data set. The result showed significant enrichment of high TINCR expression on gene sets related to several classical pathways, including NF-κB pathway (Fig. S2B-C), while some typical related pathways were excluded after exploration (Fig. S2H). To verify this, results of further experiments revealed that ST6GAL1 was decreased after silencing TINCRs protein levels and the phosphorylation levels of IκBα (p- IκBα) and p65 (p-p65). The effects could be partially reversed by inhibition of miR-195-3p (Fig. 5H). In contrast, TINCR overexpression increased ST6GAL1 protein levels and the phosphorylation levels of IκBα (p- IκBα) and p65 (p-p65), which were partially reversed by overexpression of miR-195-3p (Fig. 5I). Based on these findings, we proposed that TINCR absorbed miR-195-3p to upregulate ST6GAL1, and activate the NF-κB pathway.

To confirm the TINCR-mediated function of ST6GAL1, we transfected HuH7 and HepG2 cells with ST6GAL1-overexpression vector or empty vector on basis of TINCR knockdown. Cell function assays revealed that restoration of ST6GAL1 expression partially rescued the reduced ability of proliferation, invasion, migration (Fig. 6A-E; Fig. S2D-E, P < 0.01), as well as oxaliplatin resistance (Fig. 6F; Fig. S2F, P < 0.05). Besides, restoration of ST6GAL1 also partially make up for the weakened phosphorylation levels of IκBα (p- IκBα) and p65 (p-p65), whereas overexpression of ST6GAL1 exhibited the opposite effect (Fig. 6G-H). In general, these finding further illustrated that TINCR promoted HCC cells progression and resistance to oxaliplatin via the TINCR/miR-195-3p/ST6GAL1/ NF-κB signaling pathway (Fig. 6I).

To verify the oncogenic function of TINCR in vivo, nude mice were subcutaneously inoculated with TINCR silencing and control stable-transfected HuH7 cells to build xenograft mouse models. The degree of TINCR knockdown was confirmed by qRT-PCR (Fig. 7A). The growth of tumor size and weight in sh-TINCR group were strongly suppressed compared with that in scramble control group. Moreover, tumors in sh-TINCR group were significantly more sensitive to oxaliplatin treatment compared with that in ctrl-TINCR group (Fig. 7B-D, P < 0.01). Synchronously, ST6GAL1 expression in tumor tissues was reduced with the silencing of TINCR (Fig. 7E-F, P < 0.05). These data indicated that silencing TINCR could restrain HCC cell proliferation and oxaliplatin resistance in vivo.