LINC01023 Promotes the Hepatoblastoma Tumorigenesis via miR-378a-5p/WNT3 Axis

Hepatoblastoma is the most common type of hepatic tumors occurring in children between 0 and 5 years. And the exact pathophysiology of the disease is still mysterious. Accumulating studies on LncRNA have shown its pivotal role in the development and progression of distinct human cancers. However, the role of LINC01023 in hepatoblastoma is unknown. The relative expression of LINC01023, miR-378a-5p, and Wnt3 on hepatoblastoma tissue and cell lines was determined by quantitative polymerase chain reaction (qRT-PCR). The effect of LINC01023 downregulation and upregulation on cell proliferation, colony formation and apoptosis activities in HUH6 and HepG2 Cells was assessed by CKK8, clonogenic and flow cytometry analysis, respectively. Dual luciferase, RNA immunoprecipitation (RIP), and RNA pull-down were performed to confirm the interaction between LINC01023 and miR-378a-5p. Similarly, Dual luciferase assay was performed to confirmed the interaction between Wnt3 and miR-378a-5p. The xenograft tumorgenicity test was performed to elucidate the tumorgenicity potential of LINC01023. LINC01023 was significantly upregulated in hepatoblastoma tissue and cell lines rather than in adjacent normal hepatic tissue and QSG7701 cell lines. LINC01023 silencing attenuated cell proliferation, colony formation and increased cell apoptosis. Conversely, LINC01023 upregulation results in significant increase in cell proliferation, and colony formation activities however, a significant reduction in apoptosis activity was reported. Interaction between the LINC01023 and WNT3 was confirmed by dual luciferase assay. Xenograft animal tumorgenicity test confirmed the in-vivo tumorigenesis potential of LINC01203. To the best of our knowledge, this study is the first study demonstrating the role of LINC01023 in hepatoblastoma tumorigenesis through the LINC01023/miR-378a-5p/Wnt3 axis. It could be a potential therapeutic target and a prognostic biomarker in hepatoblastoma.


Introduction
Hepatoblastoma is a benign tumor occurring in children aged between 0 and 5 years. It accounts for only 1% of pediatric tumors however, 70-80% of pediatric hepatic tumors [1][2][3][4]. The exact pathophysiology of the disease is still mysterious; however, it has been associated with numerous genetic disorders, including Beckwith-Wiedemann syndrome, Familial adenomatous polyposis (FAP), and Trisomy 18 [5][6][7][8]. The annual incidence rate is estimated approximately 1.2 to 1.5 per million population [4,9]. Due to rarity of diseases, there are numerous challenges in developing and advancements Ramesh Bhandari & Imaran Ibrahim Shaikh Equally Contributed. 1 3 in therapeutic targets drugs as well as exploring the disease pathophysiology.
The Wnt/β-catenin signaling pathway is the most pivotal in embryogenesis, organogenesis and liver regeneration [10,11]. However, accumulating studies suggest a frequent dysregulation Wnt/β-catenin in a variety of human cancers, including hepatocellular carcinoma and hepatoblastoma [12,13]. A distinct wingless secreted glycoprotein molecule, including Wnt1, Wnt2, Wnt3 and others, has been found to be abnormally secreted to activate the Wnt/β-catenin signaling pathway in distinct human cancers [14][15][16]. However, the role of Wnt3 and the Wnt/β-catenin signaling pathway in hepatoblastoma has yet to be elucidated. Thereby, exploration of the role of Wnt3 and the Wnt/β-catenin signaling pathway helps in understanding the underlying molecular mechanisms and pathophysiology of hepatoblastoma as well as helps in identification of new therapeutic targets for the treatment of hepatoblastoma patients [3,17,18].
Long intergenic non-coding RNA (LincRNA) are subtypes of non-coding RNAs with a nucleotide length greater than 200 bp and are known to regulate gene expression at the post-transcriptional or epigenetic level. They play a significant role in regulating the development and progression of various human cancers, and also affecting tumor cell proliferation, apoptosis, and migration [19][20][21] Accumulating studies on LincRNAs have shown their involvement in the development and progression of distinct human cancers and diseases, including hepatoblastoma [22]. For insane LINC00662 promotes the hepatocellular carcinoma via Wnt/β-catenin pathway [23]. Long intergenic non-coding RNA1023 (LINC01023), a subtype of noncoding RNA found on human chromosome 5, was found to be significantly upregulated in gliomas [24]. However, the expression profile, molecular mechanisms and biological functions of LINC01023 in hepatoblastoma have not yet been investigated.
MicroRNAs (miRNAs) are small, endogenous, noncoding RNAs with 18-25 nucleotide lengths. They play a significant role in regulating the expression of messenger RNAs by interacting with the 3'-UTRs of the binding sites of target genes and interfering with their biological function. Increasing studies on miRNAs have shown deregulation (upregulation/downregulation) of miRNAs has been associated with the development and progression of various human cancers, including hepatoblastoma [25][26][27]. Recent studies on miR-378a-5p have shown that it suppresses colorectal cancer (CRC) and renal cell carcinoma (RCC), which has been associated with a patient's prognosis [28,29]. In contrast, upregulation of miR-378a-5p has been associated with the development and progression of melanoma and breast cancer [30,31]. However, the function of miR-378a-5p in hepatocellular carcinoma and hepatoblastoma is unknown and needs to be elucidated.
In this study, we investigated the possible potential role of the LINC01023/miR-378a-5p/Wnt3 axis in the development and progression of hepatoblastoma tumorigenesis and its correlation with clinicopathological characteristics. Initially, we determined the differential expression of LINC01023 in hepatoblastoma tissues and cells and found that it was significantly upregulated on the both hepatoblastoma primary tissues and cell lines. Next, we investigated the effect of up and down-regulation of LINC01023 on the biological functional activities on the hepatoblastoma cell lines. A significant reduction in cell proliferation and colony forming activity was reported in the LINC01023 knockdown HUH6 and HepG2. Conversely, a significant increase in cell proliferation and colony forming activities was observed in LINC01023 overexpressed HUH6 and HepG2 cells. Similarly, significantly low cellular apoptosis activity was observed in LINC01023 upregulated hepatoblastoma (HUH6 and HepG2) cells, while, a significantly higher  To  explore the possible underlying mechanisms, we applied  RegRNA, RNAhybrid, star base and Target Scan online bioinformatics software to identify the possible downstream  targeting miR-378a-5p for LINC01023 and upstream Wnt3 target genes for miR-378a-5p. The possible association between LINC01023 and miR-378a-5p and the interaction between the miR-378a-5p and Wnt3 was confirmed by dual luciferase activity. The dual-luciferase assay showed a strong interaction between LINC01023 and miR-378a-5p and similarly, miR-378a-5p and Wnt3 interaction. In-vivo xenograft tumorgenicity test to demonstrate the oncogenic potential of LINC01023. Based on the above findings of our study revealed that LINC01023 promotes hepatoblastoma tumorigenesis by sponging miR-378a-5p and upregulating the Wnt3 expression. And it could also be a potential therapeutic target for the treatment of patients with hepatoblastoma.

Patients, clinical specimen study setting, and duration
A total of 40 pairs of primary hepatoblastoma tissue and normal hepatic tissue were obtained from pediatric patients who underwent hepatic surgery at the Department of Pediatric hepatobiliary surgery, Shanghai Tenth Peoples hospitals of Tongji University of Medicine, Shanghai, China.

Criteria for patient's selection
Inclusion criteria: (1) All the patients confirmed for hepatoblastoma tumors following the histopathological examination of hematoxylin and eosin-stained hepatic tumor tissue were included. Exclusion criteria: (1) Patients underwent preoperative treatments including the radiation therapy or chemotherapy prior to surgery. (2) Patients with underwent the multiple surgery and having distinct underlying medical conditions. Excised hepatoblastoma tissue and normal hepatic tissue were immediately transferred to liquid nitrogen and carried to a laboratory for further processing. This study was approved by the Institutional Research Ethical committee. And, written and verbal consent was taken from all the participating patients. Patients with hepatoblastoma were classified into high (n = 22) and low (n = 22) LINC01023 expression groups according to the median expression of LINC01023 in hepatoblastoma tissue and the detailed clinicopathological characteristics of patients are presented in Table 1.

HB cell lines, cell culture and transfections
Hepatoblastoma (HepG2 and HUH6) and normal hepatic (QSG7701) cell lines were purchased from Chinese Academy of Medical Sciences, Shanghai. HepG2, HUH6 and QSG7701 were cultured in MEM, DMEM and RPMI-1640 cell culture medium respectively. All culture media were supplemented with 10% (Fetal Bovine Serum) FBS and contained 100 UI Penicillin/Streptomycin. All the cell culture plates were incubated in a humidified CO 2 incubator at 37°C. Short interfering RNAs for LINC01023 (si-LINC01023#1, si-NC), miR-378a-5p mimics, miR-378a-5p inhibitors and their negative controls were synthesized by Gene Pharma (Shanghai, China). pCDNA-LINC01023, pCDNA Empty vector and short hairpin RNA (sh-LINC01023#1) sh-NC were synthesized and purchased from Keli Biotechnology, Shanghai, China. The sequences of siRNAs and miRNAs oligonucleotides are listed in supplementary File S2. Transfection of hepatoblastoma (HUH6 and HepG2) cell lines with siRNAs, shRNAs, PCDA-LINC01023 plasmid, miR-378a-5p mimics/inhibitors and their corresponding negative controls was carried out according to the manufacturer's protocols. Briefly, hepatoblastoma cell lines (HepG2 and HUH6) were seeded on 6 well plates and incubated in a 5% CO 2 incubator to allow for confluent growth of 30-40% (siRNAs/miRNAs) and 70-80% (sh-RNAs) on the next day. Transfection of HUH6 and HepG2 cells with LINC01023 siRNA/shRNA/miR-NAs/PcDNA-LINC01023 and their negative controls was performed using Lipofectamine 2000 (Invitrogen; Thermo Fisher, Scientific Inc). Cells harvested for 48 h after transfected were collected for further analysis. Cell transfection efficacy of was investigated by qRT-PCR and western blot analysis.

RNA extraction and quantitative reverse transcriptase polymerase chain reaction (qRT-PCR)
Total RNA and miRNA from primary hepatoblastoma tissues, and cell lines were extracted using Trizol reagent (Invitrogen, Carlsbad, CA, USA). mRNA and miRNA were reversed transcribed into cDNA using Prime Script RT-PCR reagents and miR-X Tm miRNA First-Strand Synthesis kit (Takara Biotechnology). Quantitative Polymerase Chain Reaction (qRT-PCR) was performed for quantitative estimation of mRNAs and miRNA in cDNA using the Kappa SYBR Green reagent (Kapa Japan) on ABI 7900 HT. GAPDH and U6 were used as the endogenous controls for quantitative analysis of mRNAs and miRNA, respectively. Relative expression of mRNAs and miRNAs was determined by using the 2 ∆∆CT formula. The primer sequences 1 3 of LINC01023, miRNA, WNT3, GAPDH and U6 are listed as supplementary file S1.

Cell proliferation assay
Cell Counting Kit-8 assay (CCK-8, Beyotime, Jiangsu China) performed to investigate the effect of LINC01023/ miR-378a-5p knockdown or overexpression on cell proliferation activity of HUH6 and HepG2 cells. Briefly, HUH6 and HepG2 cells transfected with the siRNA/shRNA of LINC01023 and miR-378a-5p mimics/inhibitors were seeded in 96 wells plate at a density of 1 × 10 3 cells/well. 96 wells plates was further incubated at 37 °C in a humidified CO 2 incubator. After 24, 48, 72, 96 and 120 h of seeding HUH6 and HepG2 cells, 10 μl CCK8 reagent was added to each well and incubated at 37 °C in a CO 2 incubator for 2 h, followed by the measurement of the absorbance value colorimetric reaction of each well using a SpectraMax multiplate microplate reader at 450 nm (Molecular Diagnostic Devices, USA). Each experiment was repeated three times independently.

Clonogenic assay
Hepatoblastoma (HUH6 and HepG2) cells transfected with the siRNA/shRNA of LINC01023 and miR-378a-5p mimics/ inhibitors were seeded at a density of 1 × 10 3 cell/well in a 6 well plate and incubated at 37 °C in a humidified CO 2 incubator for 14 days. On every fourth day old culture medium of each well replenished with the new culture medium. At the end of the 14 days, the culture medium from each well was removed and washed three times with 1 PBS. Cell colonies were then fixed with 4% paraformaldehyde (PFA) for 15 min and stained with 0.1% crystal violet for 15 min. After staining, each well was washed with tap water and air dried. Image of colonies on six well plates were captured and using Image J software the number of colonies in each well counted.

Hepatoblastoma cellular apoptosis
The effect of LINC01023 depletion and upregulation on apoptosis activity of HUH6 and HepG2 was determined by flow cytometry analysis. Briefly, 48 h after transfection, HepG2 and HuH6 were collected and centrifuged at 1000 RPM in a 4 °C centrifuged. The cell pellets were suspended in 1 × Annexin V binding buffer. Finally, the HUH6 and HepG2 cell suspensions were distributed at a density of 1 × 10 5 in differentially labelled tubes. 50 μl Annexin-V-FITIC staining solution was added to each tube and incubated for 20-30 min at room temperature. Then 250 μl PI-Annexin V solution was added to each tube and incubated for 5 min at room temperature in the absence of light. Finally, cell apoptosis was assessed using flow cytometry (FACSCanto TM II, BD Biosciences) and the ratio/percentage of apoptosis activity was determined using the flow Jo software.

Western blotting
The proteins from LINC01023 siRNA/shRNA and miR-378a-5p mimics/inhibitors transfected HUH6 and HepG2 cells were extracted using RIPA cell lysis buffer (Beyotime, Jiangsu, China). The protein concentration in the cell lysate was quantified using the BCA protein estimation kit (Beyotime, Jiangsu, China). An equal amount of protein lysate 80 μg was separated by 10% SDS-PAGE gel electrophoresis at 80-120 V for 60-120 min. The separated proteins on the SDS-PAGE gel were electrophoretic transferred to the nitrocellulose membrane (Millipore, Billerica, MA, USA). Nonspecific binding of antibodies to the separated proteins in the nitrocellulose membrane was blocked by treating and incubating of the nitrocellulose membrane with 5% BSA blocking solution for 1 h uniform rocking rotator. The nitrocellulose membranes were washed 3 times in PBST, then the membranes were incubated with primary antibodies WNT3 (Rabbit, Abcam) and p-actin (Rabbit, Abcam) overnight at 4 °C. Subsequently, the nitrocellulose membrane was washed 4 times with PBST and incubated with horse radish peroxidase conjugated goat anti-rabbit antibody (ab6721) for 2 h at room temperature. Finally, membranes were washed 2-3 times with PBST solution and protein bands on the membrane were detected using the Enhanced Chemiluminescence (ECL) System reagent kit in the AmershamTM A600 chemiluminescent film scanner. β−Actin was used as an endogenous control.

Subcellular localization of LINCRNA
Nuclear and cytoplasmic fractional distribution of LINC01023 in HUH6 and HepG2 was extracted using the PARIS™ kit (Invitrogen) according to the manufacture guidelines. Relative fractional expression of LINC010203 was determined by qRT-PCR. GAPDH and U6 was used as cytoplasmic control and nuclear control, respectively.

RNA immunoprecipitation (RIP) assay
The RIP assay was performed using EZ-Magana RNA immunoprecipitation (RIP) kit (Millipore, USA). Briefly, hepatoblastoma (HepG2 and HUH6) cells were transfected with miR-378a-5p mimics and miR-mimics NC was collected after trypsinization. HuH6 and HepG2 cells were completely lysed with RIP lysis buffer containing 0.1 mM PMSF and 1% protease inhibitor cocktail on ice. After 10 min, the cell lysate is centrifuged at 12,000 RPM for 20 min at 4 °C. The supernatant solution is collected and followed by incubation with magnetic beads conjugated with Argounated two antibodies (anti-Ago2) and IgG antibodies (Abcam, USA) for 4-6 h at 4 °C. The immunoprecipitated RNA on the magnetic beads was collected after the elution and its concentration was measured by nanodrop spectrophotometry. The enrichment analysis of LINC01023 & miR-378a-5p immunoprecipitated RNA was determined by qRT-PCR.

Dual-Luciferase reporter assay
A dual luciferase reporter assay was performed to validate the interaction between LINC01023 and miR-378a-5p, and the interaction between miR-378a-5p and WNT3. (a) Validation of the interaction between LINC01023 and miR-378a-5p: Briefly HUH6 and HepG2 cells were seeded on 12 well plates and transfected with pmirGLO-LINC01023-WT vector and MUT vector (Sangoon Biotech, Shanghai, China) along with miR-378a-5p mimics and miRNA-Nc using Lipofectamine 2000 (Invitrogen). (b) Validation of interaction between WNT3 and miR-378a-5p mimics/inhibitors. Briefly, HUH6 and HepG2 cells were co-transfected with pmirGLO-WNT-WT/Mutant vector and miR-378a-5p mimics/inhibitors using Lipofectamine 2000. At 48 h after post-transfection HUH6 and HepG2 cells were lysed with passive lysis buffer and luciferase activity in the cell lysate was measured using a dual-luciferase assay kit (Promega, Madison, WI, USA). Luciferase activity was normalized with Renilla activity.

Xenograft tumorgenicity test
The in-vivo tumorgenicity potential of LINC01023 was assessed with the inoculation of stably transfected sh-LINC01023 #1, or sh-NC HuH6 into BALB/c nude mice. A total of 12 female BALB/c nude mice of aged 4-6 weeks were purchased and randomly assigned into two groups sh-Nc and sh-hLINC01023#1 group. 3 × 10 5 stably transfected HUH6 (lv-shNC & lv-shLINC01023) was injected subcutaneously into the posterior right flank of each BALB/c nude mouse. Tumor growth was monitored and tumor volume was measured in every 4 days using the formula: Volume (mm 3 ) = length × width/2. After 24 days, all the mice of sh-Nc and sh-LINC01023#1 group were killed by cervical dislocation. All the animal experiments were carried out in accordance to the guidelines of the care and use of laboratory animals issued by the National Institute of Health (NIH).

Bioinformatics
The online bioinformatic software Reg RNA was used to identify LINC01023 targeting down-stream miRNAs [32]. Subsequently, the online bioinformatic software Target-scan was used to identify the miR-378a-5p targeting gene [33,34]. In addition, the online bioinformatic online software RNA hybrid was used to identify predictive putative 3′-UTR binding sites of LINC01023 to miR-378a-5p and similarly of miR-378a-5p to WNT3 [35].

Statically analysis
The results of this studies were presented as mean ± SD. Statistical data were analyzed using GraphPad software (Ver. Prism 8, La Jolla, CA, USA). Student T-test was used to compare differences between two individual groups, and a one-way ANOVA was used to compare difference between more than two groups. Spearman correlation test was performed to demonstrate the association between two distinct groups. The following p-values indicate statistical significance: *p ≥ 0.5, **p ≥ 0.01, ***p ≥ 0.001, and ****p ≥ 0.0001.

LINC01023 upregulated in hepatoblastoma tissues and cell lines and associated hepatoblastoma prognosis
An initial, preliminary screening of LINC01023 expression on the hepatoblastoma tissue and (HUH6 & HepG2) cell lines by qRT-PCR revealed the significantly high expression of LINC01023 hepatoblastoma tissue and cell lines (HUH6 and HepG2) compared to normal hepatic tissue and (QSG7701) normal hepatic cell lines ( Fig. 1a and b). Next, to efficacy of interference (sh-LINC01023 #1, or sh-NC)/ overexpression plasmid (pCDNA-LINC01023 or Empty vector) was evaluated. Overexpression of LINC01023 in HUH6 and HepG2 reported the remarkably high expression in HUH6 and HepG2 (Fig. 1c). Conversely, knockdown of LINC01023 resulting in a significant depletion in the LINC01023 expression in HUH6 and HepG2 cells (Fig. 1d). Furthermore, to determine the prognostic potential of LINC01023 in hepatoblastoma patients we divided 42 hepatoblastoma patients into two major groups, based on the relative expression of LINC01023 in hepatoblastoma tumor tissue, (a) high LINC01023 expression group (N = 21) and (b) low LINC01023 expression groups (N = 21). The Kaplan Meir survival curve analysis was performed to determine the impact of (low/high) LINC01023 expression on the overall survival outcomes of hepatoblastoma patients. This showed that patients with high LINC01023 expression had poor survival rate compared to hepatoblastoma patients with low LINC01023 expression (Fig. 1e). In addition, we performed a correlation test to determine the correlation in between the LINC01023 expression and clinicopathological characteristics. As shown in Table 1, LINC01023 expression significantly correlated with the histological types of hepatoblastoma tumors and tumor staging (PRETEXT), but no significant correlation was found with other clinicopathological characteristics, such as age, sex, tumor size, alpha fetoprotein (AFP) and so on. These findings suggest that LINC01023 expression is significantly upregulated in hepatoblastoma tumors and significantly correlates with clinicopathological characteristics and survival in HB patients. These findings emphasize the potential value of LINC0102 as a predictive biomarker for outcomes HB. Fig. 2 Variation in LINC01023 expression level impact on cell proliferation, colony formation and apoptosis activities of hepatoblastoma cell lines. a-f Cell proliferation and colony formation activity on LINC01023 knocked down/downregulated HepG2 and HUH6 cell determined by CCK8 and clonogenic assay. g-l CCK8 and clonogenic assay was performed to determine the impacts of LINC01023 overexpressed/upregulated in HUH6 and HepG2 Cells proliferation and colony formation activity. m Apoptosis activity on HUH6 and HepG2 cells transfected with si-LINC01023#1 and si-NC determined by flow cytometry analysis. n Apoptotic activity changes in HUH6 and HepG2 cells transfected with pCDNA-LINC01023 and Empty vector/NC determined by Flow cytometry analysis ◂

Variation in LINC01023 expression level impacts the biological functional activities of the hepatoblastoma cells
Initially, we evaluated the knockdown efficiency of chemically synthesized LINC01023 siRNAs/shRNAs in HUH6 and HepG2 was evaluated and found significant knockdown efficiency (Fig. 1c). Next, we investigated the impacts of LINC01023 knockdown and overexpression on HUH6 and HepG2 cells proliferation, colony formation, and apoptosis activities. Hepatoblastoma (HUH6 and HepG2) cell knock down with si/sh-LINC01023#1, showed significantly reduced cell proliferation and colony formation activities compared to si/sh-NC (Fig. 2a-f). In contrast, HUH6 and HepG2 cells transfected with the pCDNA-LINC01023 overexpression plasmid showed a remarkably high cell proliferation and colony formation activity ( Fig. 2g and l). Inadditional, we investigated the effect on cellular apoptosis activity. Significantly high apoptosis activity was observed in stably LINC01023 (sh-LINC01023#1) knockdown HUH6 and HepG2 cells compared to sh-NC (Fig. 2n). Meanwhile, significantly low apoptosis activity was observed in LINC01023 overexpressed/upregulated HUH6 and HepG2 cells (Fig. 2n). These findings confirm the tumorigenic potency of LINC01023 and strongly suggest its involvement in the development and progression of hepatoblastoma tumors in humans.

LINC01023 directly targets miR-378a-5p and interferes it's activity
Initially, we applied the RegRNA online bioinformatics software to identify miR-378a-5p as the downstream miRNA target of LINC01023. Next, we examine the relative expression of the miR-378a-5p in hepatoblastoma tissues and cells, and found that miR-378a-5p expression in hepatoblastoma primary tissue and cell lines was significantly lower than that in adjacent normal hepatic tissue and cell lines (QSG7701) (Fig. 3a and b).
Accumulating studies on LincRNAs have reported most of cytoplasmic LincRNAs functions as endogenous RNAs (ceRNAs) or molecular sponges, and that bind to miRNAs, interfering the normal biological and molecular function of miRNAs [32][33][34]. We next investigated the subcellular fractional distribution of LINC01023 in HUH6 and HepG2. Subcellular fractional analysis showed that a relatively high proportion of LINC01023 localized in the cytoplasm and less in the nucleus of HUH6 and HepG2 cells (Fig. 3c and  d). Similarly, we identified the putative 3′-UTR binding sites of LINC01023 on miR-378a-5p using the RNAhyrbid online software, (Fig. 3g). In addition, we examine the effects of LINC01023 downregulation and upregulation on the relative expression of miR-378a-5p. As shown in Fig. 3e, a significantly high miR-378a-5p expression was observed in LINC01023 knock down HUH6 and HepG2 cells. In contrast, LINC01023 upregulated/overexpressed HUH6 and HepG2 cells showed significantly low miR-378a-5p expression (Fig. 3f).
In addition, we synthesized pmirGLO reporter wild Type LINC01023 (LINC01023-WT) and mutant Type LINC01023 (LINC1023-MUT) vectors from Gene pharma Shanghai, China to perform the dual luciferase assay to validate the interaction LINC01023 and miR-378a-5p. The dual luciferase reporter assay showed significantly reduced/low luciferase activity in wild-typeLINC01023 (LINCO1023-WT) and miR-378a-5p mimics co-transfected HUH6 and HepG2 cells, whereas no significant differences in luciferase activity were detected in mutant-type LINC01023 (LINC01023-MUT) and miR-378a-5p mimics co-transfected HUH6 and HepG2 cells. It indicates that LINC01023 directly binds to the 3'-UTR of miR-378a-5p and further suppresses its activity/expression. Further, we performed the RIP assay to investigate the interaction between miR-378a-5p and LINC01023. Significantly high enrichment of LINC01023 and miR-378a-5p was observed in AGO2 precipitated RNA in-contrast to IgG co-precipitated RNA (Fig. 3j and k). In addition, we performed the RNA biotin pull down assay to confirm the interaction between LINC01023 and miR-378a-5p and found a significant enrichment of LINC01023 and miR-378a-5p expression in antisense biotin precipitated compared to sense biotin precipitated RNA (Fig. 3l and m). Eventually, we performed Spearman's correlation test to elucidate the correlation between LINC01023 and miR-378a-5p and found that they were negatively correlated (Fig. 3n). In conclusion, LINC01023 directly interacts with miR-378a-5p and regulates the activity of miR-378a-5p in HB cell lines.  -378a-5p. a, b Relative expression of miR-378a-5p in hepatoblastoma tissue/cells and adjacent normal hepatic tissue/ (QSG7701) cell lines determined by qRt-PCR. c, d Fractional distribution of the LINC01023 in HUH6 and HepG2 cell, respectively, determined by qRT-PCR. e The relative expression of miR-378a-5p in stably sh-LINC01023 or sh-NC transfected HUH6 and HepG2 cells. f miR-378a-5p expression in LINC01023 overexpressed plasmid (pCDNA-LINC01023 or Empty vector) transfected HUH6 and HepG2. g Schematic diagrammatic illustration of possible predicated 3'-UTR binding sites of LINC01023 in miR-378a-5p. h, i The luciferase activities in HUH6 and HepG2 cells co-transfected with miR-378a-5p mimics and miR-Nc mimics and LINC01023 WT and LINC01023 MUT reporter vector determined by dual luciferase assay. j, k RIP assay demonstrated the enrichment of LINC01023 and miR-378a-5p in Ago2 precipitated RNA than IgG (Nc) antibody precipitated RNA. l, m RNA biotin pull down to demonstrate the LINC01023 and miR-378a-5p level interaction. n The Spearman's correlation analysis was performed to analyzed the correlation between LINC01023 and miR-378a-5p ◂ miR-378a-5p suppressed hepatoblastoma tumor development and progression miR-378a-5p was significantly downregulated in hepatoblastoma tissue and cell lines in compared to normal hepatic tissues and cell lines (QSG7701) (Fig. 3a and b). Next, we investigated the effect of miR-378a-5p knockdown and overexpression in the relative expression of miR-378a-5p in HUH6 and HepG2 cells. As shown in Fig. 4a, miR-378a-5p expression was significantly upregulated in miR-378a-5p mimics transfected HUH6 and HepG2 cells meanwhile a significantly low miR-378a-5p expression was reported in miR-378a-5p inhibitors transfected HUH6 and HepG2 cells (Fig. 4a). Furthermore, we investigated the consequences of miR-378a-5p down and upregulation of miR-378a-5p on the proliferative activity of HUH6 and HepG2 cells. As shown in Fig. 4b and c, miR-378a-5p upregulation in HUH6 and HepG2 cells resulting in significant reduction cell proliferation activity. On the contrary, miR-378a-5p downregulation/ depletion on HUH6 and HepG2 cells resulting in significant enhancement in cell proliferative activity (Fig. 4b and c). In additional, we investigated the effects down and up-regulation of miR-378a-5p on cell apoptosis activity of HUH6 and HepG2 cells. As shown in Fig. 4d, miR-378a-5p upregulated Fig. 4 miR-378a-5p downregulates the hepatoblastoma tumors development and progression. a The relative expression of miR-378a-5p in hepatoblastoma (HUH6 and HepG2) cells transfected with miR-378a-5p mimics/inhibitors and their negative controls determined by qRT-PCR. b, c Cell proliferation activity of miR-378a-5p mimics/inhibitors or miR-NC transfected HUH6 and HepG2 determined by CCK8 assay. d, e Cell apoptosis activity in HUH6 and HepG2 cells transfected with miR-378a-5p mimics, miR-378a-5p inhibitors or miR-NC was determined by flow cytometric analysis. f Tumors retrieved from miR-Nc mimics and miR-378a-5p mimics transfected HUH6 injected mice BALB/c Nude mice groups. g, h Tumor volume, tumor weight and miR-378a-5p expression in miR-378a-5p mimics group and miR-mimics-Nc group HUH6 and HepG2 cells, showed significantly high cellular apoptosis activity. Meanwhile, miR-378a-5p downregulated HUH6 and HepG2 cells showed significantly reduced cellular apoptosis (Fig. 4e). In addition, we performed the xenograft tumorgenicity assay to demonstrate the in-vivo tumor suppressor activity of miR-378a-5p. In this experiment, we divided the BALB/c nude mice into two groups (miR mimics-NC groups and miR-378a-5p group) and each group consisted of three (3) BALB/c nude mice. miRmimics-NC and miR-378a-5p mimics transfected HUH6 cell suspensions was injected into the respective groups and observed the development of tumors in the mice was observed for 24 days. After 24 days, the mice were scarified and the tumors were isolated. miR-378a-5p group mice showed a significant inhibition in tumor growth, size and weight compared to miR-mimics-NC groups of mice ( Fig. 4f  and h). Significantly lower expression of miR-378a-5p was observed in mice tumors retrieved from miR-378a-5p mimics groups than miR-Nc groups (Fig. 4I). These finding confirmed the tumor suppressor function of miR-378a-5p in hepatoblastoma.

miR-378a-5p interacts with Wnt3 and interferes it's activity
To elucidate the underlying mechanism by which miR-378a-5p regulates hepatoblastoma cell growth, we initially identified upstream target gene candidates using Target Scan online bioinformatics software (Fig. 5a). Next, we determined the relative expression of Wnt3 in hepatoblastoma tissues and cell lines and found that it was significantly upregulated (Fig. 5b and c). Furthermore, we investigated the effects of up and down regulation of miR-378a-5p on relative Wnt3 mRNA and protein expression. qRT-PCR and western blot analysis showed a significantly low Wnt3 mRNA and protein expression on miR-378a-5p was overexpressed in HUH6 and HepG2 cells (Fig. 5d and e). In contrast, significantly high Wnt3 mRNA and protein expression was reported in miR-378a-5p inhibitors transfected HUH6 and HepG2 cells (Fig. 5f-g).
Next, we identified the putative 3′-UTR binding sites of miR-378a-5p on Wnt3 using online bioinformatics software RNA-hybrid (Fig. 5h). Furthermore, we performed the dual luciferase reporter assay to determine the interaction between miR-378a-5p and WNT3. Significantly reduced dual luciferase activity was reported on Wnt3 wild type reporter vector (WNT3-WT) and miR-378a-5p mimics co-transfected noted HUH6 and HepG2 compared to Wnt3 mutant type Wnt3 (WNT3-MUT) and miR-378a-5p mimics co-transfected HUH6 and HepG2 cells. At the same time, significantly high luciferase activity was observed in HUH6 and HepG2 cells co-transfected with Wnt3 mutant reporter vectors (WNT3-MUT) and miR-378a-5p inhibitors (Fig. 5i). Finally, we performed the Spearman's correlation test to demonstrate the correlation between miR-378a-5p and Wnt3 expression and found that they were negatively correlated (Fig. 5j). In conclusion, miR-378a directly interacts with Wnt3 and downregulates its activity.

LINC01023 upregulates the Wnt3a secretion and stimulates the Wnt/β-catenin signaling pathway
To investigate the possible underlying mechanisms by which LINC01023 regulates proliferation and metastasis of hepatoblastoma cells. Initially, we performed the qRT-PCR and western blot analysis to elucidate the effects of LINC01023 down-regulation and up-regulation in target genes including Wnt3, and ß-catenin associated with the Wnt/β-catenin signaling pathway. LINC01023 knockdown in HUH6 and HepG2 showed the significant reduction inWnt3 and β-catenin mRNA and protein expression (Fig. 6a-e). Meanwhile, a remarkably high expression of Wnt3 and β-catenin mRNA and protein expression were reported in LINC01203 overexpressed HUH6 and HepG2 cells (Fig. 6f-j). Further, to elucidate the functional role of LINC01023 in hepatoblastoma (HUH6 and HepG2) cell growth, we functionally downregulated and upregulated LINC01023 expression in HUH6 and HepG2 and examined the cell proliferation, clonogenic and apoptosis activities. As mentioned in the earlier experiment, LINC01023 knockdown resulting in a significant reduction in HUH6 and HepG2 cell proliferation, colony formation and enhanced cell apoptosis (Fig. 2a-f). Conversely, upregulation of LINC01023 significantly increased in cell proliferation, clonogenic activity and inhibited cellular apoptosis (Fig. 2g-l). Thus, these findings confirmed that LINC01023 promoted the development and progression of hepatoblastoma by regulating the Wnt/β-catenin signaling pathway.

miR-378a-5p/WNT3 is involved in LINC01023 tumorigenesis activity in hepatoblastoma
To investigate the involvement of miR-378a-5p/WNT3 in LINC01023 in hepatoblastoma tumorigenesis, we co-transfected HUH6 and HepG2 cells with pCDNA-LINC01023 and miR-378a-5p mimics and examined the relative Wnt3 mRNA and protein expression levels, cell proliferation activity, cellular apoptosis and xenograft tumorigenicity assay.
HUH6 and HepG2 cells co-transfected with pcDNA-LINC01023 and miR-378a-5p mimics showed significantly lower Wnt3 mRNA and protein expression than HUH6 and HepG2 cells transfected with pCDNA-LINC01023 (Fig. 7a-c). Next, we examined the effects of co-transfection of pCDNA-LINC01023 and miR-378a-5p mimics on HUH6 and HepG2 cells and found a significant reduction in cell proliferation activity compared to cell proliferation activity of HUH6 and HepG2 cells transfected with pCDNA-LINC01023 (Fig. 7e). In addition, we also examined the effect of co-transfection of pCDNA-LINC01023 and miR-378a-5p mimics cellular apoptosis activity on HUH6 and HepG2 cell apoptosis activity, and revealed significant rise in HUH6 and HepG2 cell among HUH6 and HepG2 co-transfected with PcDNA-LINC01023 and miR-378a-5p mimics compared to HUH6 and HepG2 cells transfected with PcDNA-LINC01023 overexpression plasmid (Fig. 7f).
Next, the in-vivo tumorgenicity assay was performed with the injection of stably sh-NC, sh-pCDNA-LINC01023 and pCDNA LINC01023 + miR-378a-5p mimics transfected HUH6 cell suspensions in the respective groups (sh-NC group, sh-pCDNA-LINC01023 group and pCDNA. LINC01023 + miR-378a-5p mimics group of mice. Xenograft tumorgenicity were significantly smaller size, volume and low weight tumors was observed in the pCDNA LINC0102 + miR-378a-5p mimics group of BALB/C nude mice when compared to those that had developed in the pCDNA-LINC01023 group of BALB/c mice (Fig. 7g-i). Similarly, qRT-PCR analysis from RNA extracted from the xenograft mice tumor tissues showed a significantly reduced mRNA expression of LINC01023, Wnt3 in pCDNA-LIC01023 + miR-378a-5p group of mice compared to pCDNA-LINNC01023 group of mice. At the same time, a significant increase in miR-378a-5p expression was reported in tissue isolated from pCDNA-LIC01023 + miR-378a-5p group of mice compared with the pCDNA-LINNC01023 group of mice (Fig. 7j). These findings showed that cotransfection of miR-378a-5p mimics attenuates the activity of LINC01023 activity. And strongly suggest that miR-378a-5p/WNT3 involved in the activity of LINC01023 in tumorigenesis development.

LINC01023 silencing attenuated tumor growth in BALB/C nude mice in-vivo
The xenograft tumorigenicity test was performed to elucidate the in-vivo tumorgenicity potential of LINC01023. HUH6 cells stably knockdown with sh-LINC01023#1 and sh-NC was injected w subcutaneously in sh-LINC01023 and sh-NC group BALB/c nude mice. As shown in the Fig. 8a, qRT-PCR analysis showed sh-LINC01023 a significant knockdown efficiency in HUH6. As shown in Fig. 8b, xenograft tumorgenicity test showed a significant attenuation in tumor growth in the sh-LINC01023 group of BALB/C nude mice compared to sh-NC group of BALB/C nude mice. Next, we discovered that tumor volume and tumor weight was significantly low on tumors retrieved from sh-LINC01023 group of BALB/C nude mice compared to sh-NC group of BALB/c nude mice (Fig. 8c and d). Next, we measured the relative expression of LINC01023 and WNT3 in tumor tissue retracted from the sh-LINC01023 and sh-NC groups of mice and found significantly low expression of LINC01023 and Wnt3 in sh-LINC01023 group than in the sh-NC groups ( Fig. 8e and f). However, significantly high expression of miR-378a-5p expression was observed in in tumor tissue isolated form sh-LINC01023 group in contrast to sh-NC groups of BALB/C nude mice. Altogether, LINC01023 is potent oncogenes with a potential tumorgenicity.

Discussion
Hepatoblastoma is the most prevalent hepatic tumor occurring in children between aged 0-5 years [35]. Every year, the number of people diagnosed with hepatoblastoma are increasing. However, the exact cause and pathophysiology of the diseases are unknown. Similarly, due to the lack of reliable diagnostic tools and techniques which delays in the early diagnosis, treatment and management of patients with hepatoblastoma [25,36].
Long intergenic non-coding RNAs, previously referred to as transcriptome noise, that have no biological functions [25,37]. However, a n increasing study on lincRNA suggest that they are involved in the development and progression of distinct human cancers. It also functions as potential biomarkers in the early diagnosis of distinct human cancers [19,[38][39][40]. Thus, exploration of the underlying molecular mechanisms of lincRNAs in hepatoblastoma could provide therapeutic strategies for hepatoblastoma. Considering its clinical importance in molecular biology, in the present study we identified LINC01023.
f Cellular apoptosis activity of HepG2 and HUH6 cell transfected with Empty vector (NC), pcDNA-LINC01023 plasmid, and cotransfected with pCDNA-LINC01023 + miR-378a-5p mimics. g Tumor retrieved from Nc group, pCDNA-LINC01023 group, and pCDNA-LINC01023 + miR-378a-mimics groups. h, i Tumor volume and tumor weight in distinct group of BALB/c nude mice. j Relative expression of LINC01023, Wnt3 and miR-378a-5p in the tumors tissue isolated from Empty vector (NC), pCDNA-LINC01023 and pCDNA-LINC01023 + miR-378a-5p mimics groups of BALB/c mice 1 3 reported that LINC00284 promotes the colorectal cancer by sponging miR-27a [39]. Zhang et al. reported Linc00485 promotes lungs cancer by miR-298/c-Myc axis [39,40]. A previous study conducted by Yu Mingjun et al. reported a significant upregulation of LINC01023 in glioma, as well as its role in the development and progression of glioma via the activation of IGF1R/AKT signaling pathway [24]. However, the role of LINC01023 in hepatoblastoma tumorigenesis need to be elucidated. Here in this study, we reported a significant upregulation of LINC01023 in hepatoblastoma tissue and cell lines which is consistent with previous studies on glioma. In addition, we reported that patients with higher LINC01023 expression have a worse prognosis. Furthermore, in vitro findings suggest LINC01023 silencing in hepatoblastoma cells resulting in significant reduction in the cell proliferative, and clonogenic activity, whereas LINC01023 overexpression in hepatoblastoma cell lines increased both the cell proliferation and clonogenic activities. In vivo tumorigenesis studies showed that LINC01023 silencing remarkably inhibited tumor growth in sh-LINC01023 group mice in contrast to the sh-NC group of BALB/c Fig. 8 Xenograft tumorgenicity test to demonstrate the in-vivo tumorgenicity potential of LINC01023. a Estimation of LINC01023 expression in stably (sh-LINC01023#1 or sh-Nc) knockdown HUH6 cell to validate the knockdown efficiency and was determined by qRT-PCR. b Tumors isolated from the sh-Nc and sh-LINC01023. c, d Tumor volume and tumor weight measured in sh-NC and sh-LINC01023 group of BALB/c mice. e-g Relative expression of LINC01023, Wnt3 and miR-378a-5p in tumor tissue isolated from sh-NC and sh-LINC01023 group of BALB/c nude mice. Results were expressed in ± SD. *p ≤ 0.05, **p ≤ 0.01, and ***p ≤ 0.001 mice. These findings were consistent with the previous study of LINC01023 in glioma. These findings strongly suggest that LINC01023 plays a crucial role in the development and progression of hepatoblastoma tumorigenesis. And to our acknowledgement it is the first study to explore the role of LINC01023 in the progression and development of hepatoblastoma [39].
miR-378a-5p, a well-known potent tumor suppressor in distinct human cancers including colorectal cancer, hepatocellular carcinoma, renal cell carcinoma and so on was found significantly downregulated in hepatoblastoma tissue and cell lines and its expression is negatively correlated with hepatoblastoma progression [28,29,42]. Therefore, our hypothesis is that LINC01023 promotes hepatoblastoma tumorigenesis by directly sponging/ to miR-378a-5p and inhibiting the tumor suppressor activity of miR-378a-5p. The online bioinformatics software RNA hybrid predicts the 3′-UTR putative binding sites of LINC01023 on miR-378a-5p. Dual luciferases, RIP, RNA pull-down and qRT-PCR analysis confirmed the interaction between LINC01023 and miR-378a-5p. A Dual luciferase reporter assay showed the reduced luciferase activity on Wild type LINC01023 (LINC01023-WT) reporter and miR-378a-5p mimics co-transfected hepatoblastoma cells.
Next, the miR-378a-5p expression level was significantly reduced in LINC10203 upregulated hepatoblastoma cells, while being significantly increased in LINC01023 knockdown hepatoblastoma cells. These findings indicate LINC01023 promotes hepatoblastoma tumorigenesis by directly binding to miR-378a-5p.
Wnt signaling is a highly conserved pathway and plays an important role in embryonic development, cellular proliferation, apoptosis and tissue homeostasis [43]. Wingless wnt3 secretory glycoproteins receptors ligands play an important role in Wnt/ β − catenin pathway activation. Abnormal secretion and unusual ligation of Wnt3 proteins to LRP5 receptors has been reported in distinct human cancers, thus promoting the cell growth, tumor progression, invasion and metastasis of distinct human cancers [15,44,45]. Kim et al. reported Wnt3 overexpression is associated with hepatocellular carcinoma development [7]. Nie et al. reported that Wnt3 silencing inhibits cell proliferation and invasion in colorectal cancer [8]. Our study findings showed that miR-378a-5p targets the upstream Wnt3 genes and interfering normal function of the Wnt3. miR-378a-5p interaction with the Wnt3 was confirmed by a dual luciferase assay. HUH6 and HepG2 co-transfected with Wnt3 Wild type reporter vector (Wnt3-WT vector) and miR-378a-5p mimics showed the significantly reduced luciferase activity. Similarly, the Spearman's correlation between miR-378a-5p and Wnt3 expression reported a negative correlation. Meanwhile, LINC01023 and Wnt3 expression are positively correlated. These findings strongly indicate that LINC01023 enhances Wnt3 expression and regulates cell proliferation, apoptosis by competitively binding with miR-378a-5p in hepatoblastoma. Thereby, aberrant expression of LINC01023 may serve a potent oncogene and promotes the hepatoblastoma tumorigenesis. (1) Due to the rarity of this disease, the number of patients included in this study is less. Higher the patient's participants in study reproduce the better results. (2) Immuno fluorescent (IF) staining was not performed for the confirmation of interaction between LINC01023 and Wnt3 (3) The role of LINC01023 in chemoresistance has not been studied and needed to explore.

Conclusions
LINC01023 is significantly upregulated in hepatoblastoma and promotes cell proliferation, in hepatoblastoma. LINC01023 acts the ceRNAs/ miR-sponges for miR-378a-5p. MiR-378a-5p is potent tumor suppressor and downregulate the up-stream Wnt3 target gene in hepatoblastoma. Similarly, LINC01023 upregulates the wingless Wnt3 secretion and which in turns activates the Wnt/β − catenin pathway. Thus, LINC01023, Wnt3 and miR-378a-5p axis which in turns promotes hepatoblastoma tumorigenesis. These study findings may help to understand the underlying molecular mechanisms for the initiation and progression of hepatoblastoma tumors. The LINC01023/miR-378a-5p/ Wnt3 axis is a potent therapeutic target for treating patients and an axis for regulating hepatoblastoma tumorigenesis.
University of medicine (Approval no:2016-STPH-1012863). All the Animals experiments were carried out following the approval by Animal protection and utilization committed of Tongji Hospital (No:TJ2016279778664034STH). We declare that all the patients were participated in this study were aware about the purpose and content of this research. A verbal and written consents were taken prior to their participant in this study.
Consents of publication Not applicable.