KB-68A7.1 Inhibits Hepatocellular Carcinoma Development Through Binding to NSD1 and Suppressing Wnt/β-Catenin Signalling

Hepatocellular carcinoma (HCC) is one of the most lethal malignancies with extremely poor prognosis. Therefore, revealing the critical molecules involved in HCC progression and prognosis is urgently needed. In this study, through combining public dataset and our cohort, we found a novel prognosis-related long non-coding RNA KB-68A7.1 in HCC. KB-68A7.1 was lowly expressed in HCC, whose low expression was associated with large tumour size, aggressive clinical characteristic, and poor survival. Gain- and loss-of-function assays demonstrated that KB-68A7.1 restricted HCC cellular proliferation, induced HCC cellular apoptosis, and suppressed HCC cellular migration and invasion in vitro. Xenograft assays demonstrated that KB-68A7.1 suppressed HCC tumour growth and metastasis in vivo. These functional assays suggested KB-68A7.1 as a tumour suppressor in HCC. Histone methyltransferase nuclear receptor binding SET domain-containing protein 1 (NSD1) was found to bind to KB-68A7.1. KB-68A7.1 was mainly distributed in the cytoplasm. The binding of KB-68A7.1 to NSD1 sequestrated NSD1 in the cytoplasm, leading to the reduction in nuclear NSD1 level. Through decreasing nuclear NSD1 level, KB-68A7.1 reduced di-methylation of histone H3 at lysine 36 (H3K36me2) and increased tri-methylation of histone H3 at lysine 27 (H3K27me3) at the promoter of WNT10B, a target of NSD1. Thus, KB-68A7.1 repressed WNT10B transcription. The expression of WNT10B was negatively correlated with that of KB-68A7.1 in HCC tissues. Through repressing WNT10B, KB-68A7.1 further repressed Wnt/β-catenin signalling. Functional rescue assays showed that overexpression of WNT10B reversed the tumour-suppressive roles of KB-68A7.1, whereas the oncogenic roles of KB-68A7.1 depletion were abolished by Wnt/β-catenin signalling inhibitor. Overall, this study identified KB-68A7.1 as a lowly expressed and prognosis-related lncRNA in HCC, which suppressed HCC progression through binding to NSD1 and repressing Wnt/β-catenin signalling.


INTRODUCTION
Liver cancer is the six most common malignancy and the second leading cause of cancer-related deaths worldwide (1). The relative high mortality indicates that liver cancer is a much malignant cancer, with a 5-year survival of only about 18% (2). Hepatocellular carcinoma (HCC) is the major histological subtype of liver cancer (3). Although surgical resection and some targeted therapies have been applied to fight against HCC, most HCCs remain to have quick progression, which leads to poor prognosis (4)(5)(6). Further elucidating the underling molecular mechanisms driving HCC progression would be of clinical importance to develop more efficient therapies against HCC.
Although some genomic mutations have been found in HCC, more genes with aberrant expression accumulate in HCC (7)(8)(9). Epigenetic modulation is one of the critical mechanisms contributing to the dysregulated expression of genes (10)(11)(12). As one of the epigenetic modulation manners, histone methylation modification exerts multiple effects on gene transcription (13)(14)(15). Methylation modification at some specific lysine activates gene transcription, and whereas methylation modification at other lysine represses gene transcription, such as the classical repressive histone marker tri-methylation of histone H3 at lysine 27 (H3K27me3) (16)(17)(18). In our previous study, we found an HCC-related histone lysine methyltransferase NSD1 (nuclear receptor binding SET domaincontaining protein 1), which catalyses di-methylation of histone H3 at lysine 36 (H3K36me2) (19). Through inducing H3K36me2, NSD1 further decreased H3K27me3 at the promoter of its target genes (19,20). Our previous study identified WNT10B as a critical downstream target of NSD1 (19). Through upregulating WNT10B expression, NSD1 activates Wnt/b-catenin signalling and exerts oncogenic roles in HCC (19). Except NSD1, other epigenetic modification enzymes are also involved in the progression of HCC (21)(22)(23). However, the factors regulating the effects of epigenetic modification enzymes in HCC need further investigation.
To search the lncRNAs contributing to HCC progression, we analyzed The Cancer Genome Atlas (TCGA) liver hepatocellular carcinoma (LIHC) dataset and identified a novel HCC prognosis-related lncRNA KB-68A7.1. KB-68A7.1 has 644 nucleotides in length and is mainly localized in the cytoplasm. The Ensembl ID of KB-68A7.1-encoding gene is ENSG00000274225.1, which is located at chromosome 21q22. 3. In this study, we studied the expression pattern, clinical significances, biological roles, and regulatory mechanisms of KB-68A7.1 in HCC. Our finding revealed that KB-68A7.1 is lowly expressed in HCC, and its low expression is correlated with poor survival. KB-68A7.1 has tumour-suppressive roles via binding to NSD1, sequestrating NSD1 in the cytoplasm, and repressing Wnt/b-catenin signalling.

Clinical Samples
Ninety pairs of HCC tissues and adjacent non-cancerous liver tissues were obtained from HCC patients who received surgical excision at Union Hospital, Tongji Medical College, Huazhong University of Science and Technology (Wuhan, China). All samples were frozen in liquid nitrogen and stored in −80°C freezer until use. The study was performed following the ethical principles for medical research involving human subjects of the Helsinki Declaration and approved by the Ethic Committee of Union Hospital, Tongji Medical College, Huazhong University of Science and Technology (Wuhan, China). Written informed consents were acquired from all patients.

Cell Culture and Grouping
Human immortalized liver cell line (THLE-3) and human HCC cell lines (SNU-398 and SK-HEP-1) were obtained from the American Type Culture Collection (ATCC). Human HCC cell line (HuH-7) was obtained from Chinese Academy of Science Cell Bank (Shanghai, China). SNU-398 and HuH-7 cells were both epithelial, and SK-HEP-1 cell was endothelial. THLE-3 was cultured using BEGM Bullet Kit (Lonza, Walkersville, MD, USA) in accordance with the provided protocol. SNU-398, SK-HEP-1, and HuH-7 were cultured in Roswell Park Memorial Institute (RPMI) 1640 medium (Invitrogen, Carlsbad, CA USA), Eagle's Minimum Essential Medium (Invitrogen), and Dulbecco's modified Eagle's medium (Invitrogen), respectively, all of which were supplemented with 10% foetal bovine serum (FBS, Invitrogen). Cells were maintained at 37°C in an incubator with 5% CO 2 .

Cell Proliferation Assays
Cellular proliferation ability was assessed by the Cell Counting Kit-8 (CCK-8) and 5-ethynyl-2′-deoxyuridine (EdU) incorporation assays. CCK-8 assay was performed as we previously described (19). Briefly, 5,000 indicated HCC cells were seeded into 96-well plate. After culture for 72 h, cellular proliferation ability was measured using CCK-8 reagent (Dojindo Laboratories, Kumamoto, Japan), which was indicated by the optical density value at 450 nm. EdU incorporation assay was performed using the Cell-Light EdU Apollo 567 In Vitro Kit (RiboBio, Guangzhou, China) as previously described (43).

Cell Apoptosis Assays
Cellular apoptosis was assessed by terminal deoxynucleotidyl transferase (TdT)-mediated dUTP nick end labelling (TUNEL) and caspase-3 activity assays. TUNEL assay was performed using the One Step TUNEL Apoptosis Assay Kit (Beyotime Biotechnology, Shanghai, China). Caspase-3 activity assay was performed using the Caspase-3 Activity Assay Kit (Cell Signalling Technology, Danvers, MA, USA) following the provided protocol.

Cell Migration and Invasion Assays
Cellular migration and invasion abilities were assessed by Transwell migration assay and Transwell invasion assay, respectively, as we previously described (19). Briefly, indicated HCC cells re-suspended in FBS-free medium were added to the upper chamber of Transwell inserts (pore size, 8 mm; Corning, NY, USA) coated with or without Matrigel. Medium containing 20% FBS were added to the lower chamber. After culture for 24 h, the cells in the upper chamber were removed. The cells on the outer layer of the membranes were fixed, stained, photographed, and counted.

Xenografts Experiments
Xenograft experiments were performed with the approval of the Institutional Animal Care and Use Committee of Union Hospital, Tongji Medical College, Huazhong University of Science and Technology (Wuhan, China). Five-week-old male athymic BALB/c nude mice were used. A total of 2 × 10 6 indicated cells were subcutaneously inoculated into the flanks of mice. Subcutaneous xenograft volumes were measured every 7 days using a calliper and calculated by the formula V = 0.5 × a × b 2 (a, the long diameter; b, the short diameter). On the 21st day after inoculation, subcutaneous xenografts were resected and weighed. Subcutaneous xenografts were subjected to immunohistochemistry (IHC) staining using the primary antibodies against Ki67 (ab15580, 1:200, Abcam, Cambridge, MA, USA), PCNA (ab92552, 1:200, Abcam), or cleaved caspase-3 (#9661, 1;200, Cell Signalling Technology). Furthermore, subcutaneous xenografts were also subjected to TUNEL assays using the One Step TUNEL Apoptosis Assay Kit (Beyotime Biotechnology) following the provided instruction. A total of 2 × 10 6 indicated cells were intrasplenically injected into the mice to construct liver metastasis model. On the 35th day after injection, the livers were resected and subjected to haematoxylin-eosin (H&E) staining.

RNA Fluorescence In Situ Hybridization
For in situ detection of KB-68A7.1, the probes against KB-68A7.1 were designed and synthesized by Advanced Cell Diagnostics (Newark, CA, USA). The hybridization and fluorescence detection were conducted using the RNAscope Fluorescent Multiplex Detection Kit (Advanced Cell Diagnostics) following the provided manual. The subcellular localization of KB-68A7.1 was detected using the Leica Confocal Microscope TCS SP8 (Leica, Wetzlar, Germany).

Purification of Cytoplasmic and Nuclear RNA
Cytoplasmic and nuclear RNA was purified using the PARIS Kit (Thermo Scientific) in accordance with the provided manual. The purified RNA was measured using reverse transcription qPCR (RT-qPCR) as above described.

RNA-Protein Pull-Down Assay
KB-68A7.1 full-length sequences were PCR amplified with the primers 5′-GCTCTAGATAGACCACAGGGAGGCCTG-3′ (sense) and 5′-GGGGTACCACTTGTCAATCGAGATTTTAA-3′ (antisense). The products were further subcloned into the XbaI and KpnI sites of pSPT19 vector (Roche, Basel, Switzerland) to construct KB-68A7.1 in vitro transcription vector. KB-68A7.1 was in vitro transcribed from this constructed vector using T7 RNA polymerase (Roche). After purification, KB-68A7.1 RNA was end-labelled with desthiobiotin using the Pierce RNA 3′ End Desthiobiotinylation Kit (Thermo Scientific). Pull-down reactions were further conducted using the desthiobiotinylated RNA and the Pierce Magnetic RNA-Protein Pull-Down Kit (Thermo Scientific) following the provided manual. The enriched proteins were detected by Western blot.

Dual Luciferase Reporter Assay
Wnt/b-catenin reporter TOPflash (Addgene, Watertown, MA, USA), which expresses Firefly luciferase, was co-transfected with pRL-TK (Promega, Madison, WI, USA), which expresses Renilla luciferase into indicated cells using Lipofectamine 3000 (Invitrogen). Forty-eight hours later, the luciferase activities were detected by the Dual-Luciferase Reporter Assay System (Promega) in accordance with the provided manual. Renilla luciferase activity was employed as internal control.

Statistical Analysis
Statistical analyses were conducted using GraphPad Prism v6.0 (San Diego, CA, USA). As shown in figures and table legends, Wilcoxon matched-pairs signed-rank test, log-rank test, one-way ANOVA followed by Dunnett's multiple comparisons test, twotailed unpaired t-test, Spearman correlation analysis, and Pearson chi-square test were performed. p < 0.05 was set as the threshold for statistical significance.

Low Expression of KB-68A7.1 Was Associated With Poor Survival in HCC
The expression and clinical relevance of KB-68A7.1 in HCC were assessed via investigating TCGA liver hepatocellular carcinoma (LIHC) dataset using the GEPIA (http://gepia.cancer-pku.cn/ index.html) (44). The results revealed that KB-68A7.1 was lowlier expressed in HCC tissues than that in normal liver tissues ( Figure 1A). The TCGA LIHC dataset also presented that low expression of KB-68A7.1 was associated with short overall survival ( Figure 1B). Furthermore, the expression and clinical relevance of KB-68A7.1 were assessed in HCC tissues obtained in our hospital. The results once again revealed that KB-68A7.1 was lowlier expressed in HCC tissues than that in matched adjacent liver tissues ( Figure 1C). Encouragingly, Kaplan-Meier survival analysis once again revealed that low expression of KB-68A7.1 was associated with short overall survival in our HCC cohort ( Figure 1D). Correlation analyses between KB-68A7.1 expression levels and clinicopathological characteristics revealed that low expression of KB-68A7.1 was associated with large tumour size, encapsulation incomplete, vascular invasion, and advanced clinical stage ( Table 1). The expression of KB-68A7.1 in immortalized liver cell line THLE-3 and HCC cell lines SNU-398, HuH-7, and SK-HEP-1 were assessed using RT-qPCR. We found that the expression of KB-68A7.1 was lower in HCC cell lines than that in immortalized liver cell line ( Figure 1E). Taken together, these results demonstrated that KB-68A7.1 was lowly expressed in HCC, and its low expression was associated with large tumour size, aggressive clinical characteristics, and poor survival.

KB-68A7.1 Suppressed HCC Tumour Growth and Metastasis
The roles of KB-68A7.1 in HCC were further explored using xenograft models in vivo. SNU-398 cells with KB-68A7.1 stable overexpression, or control were subcutaneously injected into nude mice. The subcutaneous tumour growth curve showed that the xenografts formed by SNU-398 cells with KB-68A7.1 overexpression grew much slower than those formed by control SNU-398 cells ( Figure 4A). On the 21st day after injection, the subcutaneous xenografts were resected and weighed, which once again showed that ectopic expression of KB-68A7.1 suppressed tumour growth in vivo ( Figure 4B). The resected xenografts were subjected to proliferation marker Ki67 and PCNA IHC staining, which presented that KB-68A7.1 repressed cellular proliferation in vivo (Figures 4C, D). Cellular apoptosis was assessed by TUNEL assay and cleaved caspase-3 IHC staining, which presented that KB-68A7.1 promoted cellular apoptosis in vivo ( Figures 4E, F). To explore the roles of KB-68A7.1 in HCC metastasis in vivo, SNU-398 cells with KB-68A7.1 stable overexpression or control were intrasplenically injected into nude mice to construct the liver metastasis model, which presented that SNU-398 cells with KB-68A7.1 overexpression formed much fewer and smaller liver metastatic nodules than control SNU-398 cells ( Figures 4G-I). Collectively, these results demonstrated that KB-68A7.1 restricted HCC tumour growth and metastasis in vivo.

KB-68A7.1 Bound to NSD1 and Reduced the Nuclear Distribution of NSD1
To explore the mechanisms mediating the roles of KB-68A7.1 in HCC, we first assessed the subcellular distribution of KB-68A7.1 in SNU-398 cells using RNA fluorescence in situ hybridization (FISH), which demonstrated that KB-68A7.1 was mostly distributed in the cytoplasm ( Figure 5A). Cytoplasmic and nuclear RNA purification, followed by RT-qPCR once again showed the cytoplasmic distribution of KB-68A7.1 in SNU-398 cells ( Figure 5B). One of the important mechanisms of cytoplasmic lncRNAs is to bind proteins. Thus, the online in silico tool RNA-Protein Interaction Prediction (RPISeq) (http:// pridb.gdcb.iastate.edu/RPISeq/) was used to predict the proteins bound by KB-68A7.1. Intriguingly, histone lysine methyltransferase NSD1 was predicted to interact with KB-68A7.1 with RF score of 0.85 and SVM score of 0.9. RIP assay revealed that KB-68A7.1 was specifically enriched by NSD1 primary antibody ( Figure 5C). RNA-protein pull-down assay using RNA end-labeled with desthiobiotin revealed that NSD1 was specifically enriched by KB-68A7.1 ( Figure 5D). Therefore, the RIP assay and RNA-protein pull-down assay both confirmed the interaction between KB-68A7.1 and NSD1. Considering the cytoplasmic location of KB-68A7.1, we next investigated whether KB-68A7.1 regulates nuclear distribution of NSD1. Nuclear and cytoplasmic proteins were isolated from SNU-398 cells with KB-68A7.1 stable overexpression or depletion, followed by Western blot to assess nuclear and cytoplasmic NSD1 levels. The results found that ectopic expression of KB-68A7.1 downregulated nuclear NSD1 level and increased cytoplasmic NSD1 level ( Figure 5E). Silencing of KB-68A7.1 increased nuclear NSD1 level and reduced cytoplasmic NSD1 level ( Figure 5F). IF assays once again confirmed that SNU-398 cells with KB-68A7.1 overexpression had reduced nuclear NSD1 level, and SNU-398 cells with KB-68A7.1 depletion had increased nuclear NSD1 level Our previous study has found that NSD1 induced H3K36me2 at WNT10B promoter and further repressed H3K27me3 at WNT10B promoter, leading to transcription activation of WNT10B (19).
Therefore, we next investigated whether KB-68A7.1 modulates WNT10B. ChIP assays were undertaken in SNU-398 cells with KB-68A7.1 stable overexpression or depletion to assess H3K36me2 and H3K27me3 levels at WNT10B promoter, which found that ectopic expression of KB-68A7.1 reduced H3K36me2 level and increased H3K27me3 level at WNT10B promoter ( Figure 6A), while silencing of KB-68A7.1 increased H3K36me2 level and reduced H3K27me3 level at WNT10B promoter ( Figure 6B). Consistent with the changes in histone methylation, WNT10B  Figures S2B, C). To investigate the relevance between KB-68A7.1 and WNT10B in vivo, WNT10B expression was measured in the same HCC tissues used in Figure 1C, and the result presented that the expression of WNT10B was inversely correlated with that of KB-68A7.1 in HCC tissues ( Figure 6G). The inverse correlation between WNT10B and KB-68A7.1 expression was further found in TCGA LIHC dataset ( Figure 6H), analysed by the online in silico tool ENCORI (https://starbase.sysu.edu.cn/) (45). Taken together, these results demonstrated that conversely with NSD1, KB-68A7.1 repressed WNT10B transcription.

KB-68A7.1 Repressed Wnt/b-Catenin Signalling
WNT10B is well known to activate Wnt/b-catenin signalling (46,47). Therefore, we further explored the potential effects of KB-68A7.1 on Wnt/b-catenin signalling. Wnt/b-catenin reporter TOPflash luciferase activities were assessed in SNU-398 cells with KB-68A7.1 stable overexpression or depletion, which presented that ectopic expression of KB-68A7.1 reduced TOPflash luciferase activity, while silencing of KB-68A7.1 increased TOPflash luciferase activity ( Figures 7A, B). Ectopic expression of KB-68A7.1 also reduced TOPflash luciferase activity in HuH-7 cells (Supplementary Figure S2D). Figures S2E, F). Nuclear b-catenin level was decreased in  (Figures 8A, B). Caspase-3 activity assay and TUNEL assay demonstrated that ectopic expression of WNT10B reversed the induced cellular apoptosis caused by KB-68A7.1 overexpression (Figures 8C, D). Transwell migration and invasion assays demonstrated that ectopic expression of WNT10B reversed the reduced cellular migration and invasion caused by KB-68A7.1 overexpression (Figures 8E, F). Moreover, WNT10B was also overexpressed in HuH-7 cells, CCK-8 assay, EdU incorporation assay, caspase-3 activity assay, TUNEL assay, and Transwell migration and invasion assays, all demonstrated that ectopic expression of WNT10B partially reversed the tumour-suppressive roles of KB-68A7.1 in HuH-7 cells (Supplementary Figure S3). To further explore whether the roles of KB-68A7.1 in HCC were dependent on Wnt/ b-catenin signalling, SK-HEP-1 cells with KB-68A7.1 stable depletion were treated with Wnt/b-catenin signalling inhibitor ICG-001. CCK-8 and EdU incorporation assays demonstrated that treatment with ICG-001 reversed the increased cellular proliferation ability caused by KB-68A7.1 depletion (Figures 9A, B). Caspase-3 activity assay and TUNEL assay demonstrated that treatment with ICG-001 reversed the reduced cellular apoptosis caused by KB-68A7.1 depletion ( Figures 9C, D). Transwell migration and invasion assays demonstrated that treatment with ICG-001 reversed the increased cellular migration and invasion caused by KB-68A7.1 depletion (Figures 9E, F). Taken together, these results demonstrated that the roles of KB-68A7.1 in HCC were dependent on the WNT10B/ Wnt/b-catenin signalling axis.

DISCUSSION
HCC is one of the most lethal malignancies, only next to pancreatic cancer (3). HCC has quick progression and therefore poor prognosis (48). Revealing the critical molecules driving HCC development is of clinical importance. In this study,  we identified a novel HCC-related lncRNA KB-68A7.1, which represses HCC development. Both public dataset and our own cohort found that KB-68A7.1 was lowly expressed in HCC tissues compared with liver tissues, and its low expression was correlated with poor overall survival of HCC patients. Except KB-68A7.1, other lncRNAs were also previously reported to be correlated with prognosis of HCC patients, such as lncRNA-ATB, PVT1, CASC9, and GPC3-AS1 (37,(49)(50)(51). The combination of these lncRNAs would be more efficient for There are various mechanisms of action of lncRNAs (39). Routinely, nuclear lncRNAs bind epigenetic modification complex, change genomic localization of the complex, and therefore modulate the epigenetic modification and transcription of target genes (37). Cytoplasmic lncRNAs could bind microRNAs, relieve the repressive roles of microRNAs on their targets, and therefore increase the expressions of these targets (49). Cytoplasmic lncRNAs could also directly bind mRNAs and regulate their stability and/or translation (52). Furthermore, cytoplasmic lncRNAs may bind proteins and modulate the stability and/or function of the interacted proteins (53). Here, we first confirmed that KB-68A7.1 was mainly distributed in the cytoplasm. Intriguingly, here we found that cytoplasmic KB-68A7.1 directly bound to NSD1. NSD1 is a histone lysine methyltransferase, which exerts function in nucleus (20). The binding of KB-68A7.1 to NSD1 sequestrated NSD1 in the cytoplasm, leading to the reduction in nuclear NSD1 level. Thus, the binding of KB-68A7.1 to NSD1 also attenuated the functions of NSD1. As a chromatin modifier, NSD1 has various roles in different pathophysiological processes (54,55). Through catalyzing H3K36me2, NSD1 further modulates H3K27me3, DNA methylation, and/or acetylation of histone H3 at lysine 27 (H3K27ac), leading to activation or suppression of gene expression (20,54). In HCC, we have previously identified WNT10B as a critical downstream target of NSD1, which mediated the oncogenic roles of NSD1 in HCC (19). In this study, we further found that through binding to and sequestrating NSD1 in the cytoplasm, KB-68A7.1 repressed NSD1-induced H3K36me2 increasing and H3K27me3 reduction at WNT10B promoter, leading to the repression of  In conclusion, these data demonstrated that KB-68A7.1 is lowly expressed in HCC and its low expression is correlated with poor survival of HCC patients. KB-68A7.1 functions as a tumour suppressor via binding to and sequestrating NSD1 in the cytoplasm, reducing WNT10B transcription, and repressing Wnt/b-catenin signalling. This study suggested KB-68A7.1 as a potential target for HCC prognostic prediction and targeted therapy.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding author.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by the Ethic Committee of Union Hospital, Tongji Medical College, Huazhong University of Science and Technology. The patients/participants provided their written informed consent to participate in this study. The animal study was reviewed and approved by the Institutional Animal Care and Use Committee of Union Hospital, Tongji Medical College, Huazhong University of Science and Technology.   overexpression or control was measured by transwell invasion assay. Scale bars = 100 µm. Results are shown as mean ± SD based on three independent experiments. *p < 0.05, **p < 0.01, ns, not significant by one-way ANOVA followed by Dunnett's multiple comparisons test.