LIN28B-AS1-IGF2BP1 binding promotes hepatocellular carcinoma cell progression

IGF2BP1 overexpression promotes hepatocellular carcinoma (HCC) progression. Long non-coding RNA LIN28B-AS1 directly binds to IGF2BP1. In the present study, LIN28B-AS1 and IGF2BP1 expression and their potential functions in HCC cells were tested. Genetic strategies were applied to interfere their expression, and cell survival, proliferation and apoptosis were analyzed. We show that LIN28B-AS1 is expressed in established/primary human HCC cells and HCC tissues. RNA-immunoprecipitation (RIP) and RNA pull-down results confirmed that LIN28B-AS1 directly associated with IGF2BP1 protein in HCC cells. LIN28B-AS1 silencing (by targeted siRNAs) or knockout (KO, by CRISPR-Cas9 method) depleted IGF2BP1-dependent mRNAs (IGF2, Gli1, and Myc), inhibiting HCC cell growth, proliferation, migration, and invasion. Conversely, ectopic overexpression of LIN28B-AS1 upregulated IGF2BP1-dependent mRNAs and promoted HCC cell progression in vitro. Importantly, ectopic IGF2BP1 overexpression failed to rescue LIN28B-AS1-KO HepG2 cells. LIN28B-AS1 siRNA and overexpression were ineffective in IGF2BP1-KO HepG2 cells. In vivo, LIN28B-AS1 KO-HepG2 xenograft tumors grew significantly slower than the control tumors in the nude mice. Taken together, we conclude that LIN28B-AS1 associates with IGF2BP1 to promote human HCC cell progression in vitro and in vivo.

Long non-coding RNAs (LncRNAs) are over 200-nt long non-coding RNA (ncRNA) molecules [10][11][12] . LncRNAs can function as molecular signals, decoys, guides, scaffolds, or enhancers to regulate gene transcription, expression, and functions [10][11][12] . Besides, LncRNAs are actively involved in regulating a number of key cellular behaviors, including genomic imprinting, cell cycle control, cell differentiation, pluripotency maintenance, and development 10-12 . LncRNAs are dysregulated in human HCC, associated with the clinico-pathological features of human HCC [10][11][12][13] . For example, Wang et al. reported that upregulated LncRNA UCA1 promoted HCC progression via inhibition of microRNA-216b 14 . The study by Yuan et al. showed that LncRNA DANCR enhanced HCC's stemness features by directly binding and inhibiting CTNNB1 15 . Quagliata and colleagues have shown that LncRNA HOTTIP/ HOXA13 expression in HCC patients is associated with disease progression and predicts clinical outcome 16 . A very recent study by Wang et al. has discovered a novel cancer-testis specific LncRNA, namely LIN28B-AS1 17 . It is expressed in lung adenocarcinoma and interacts directly with the IGF2BP1 protein 17 . The results of the present study will show that LIN28B-AS1 is vital for IGF2BP1's functions, promoting HCC cell progression in vitro and in vivo.

Culture of established cells
Cultures of the established HepG2 and Huh-7 HCC cell lines as well as HL-7702 human hepatocytes were described in previous studies 18,19 .

Culture of primary human cells
As described 19 , two different human HCC tissues were digested by collagenase I (Sigma). Blood vessel cells, fibroblasts, immune cells, and other non-cancerous cells were abandoned. Primary HCC cells were cultured in the described medium 19 . Two different sets of primary HCC cells were named as "HCC1" and "HCC2". These primary HCC cells are proliferative. The cell doubling time is about 2.5 days for HCC1 cells and 4 days for HCC2 cells. Human primary adult hepatocytes, purchased from the Cell Bank of Fudan University (Shanghai, China), were derived from the liver of a partial hepatectomy patient. Human hepatocytes were cultured in primary cell culture medium 20 . The written-informed consent was obtained from each participant. Experiments and protocols requiring human tissues/cells were approved by the Ethics Board of Soochow University, according to Declaration of Helsinki. Cells in the present study were routinely subjected to mycoplasma and microbial contamination examination for every 3-4 months. Authentication by STR profiling, population doubling time, and cell morphology were checked as well to confirm the genotype.

Human tissues
A total of seven (7) written-informed consent primary HCC patients ("P1-P7", 45-69 years old), administrated at the Affiliated Hospitals of Soochow University, were enrolled. Tumor tissues and the surrounding normal liver tissues were separated carefully by the operating microscopes. The tissues were minced and homogenized by the tissue lysis buffer (Beyotime Biotechnology, Wuxi, China). Written-informed consent was provided by each participant.

RNA extraction and qPCR
RNA extraction and quantitative real time-PCR assay (qPCR, using the ABI Prism 7500 Fast Real-Time PCR system) were performed as described 21 . The melt curve analysis was applied to calculate product melting temperature. GAPDH was tested as the reference gene, using the 2 −ΔΔCt method to quantify target RNAs. The mRNA primers for IGF2, Myc, and Gli1 as well as U6 were provided by Dr. Wang 21 . The primers of LIN28B-AS1 (based on ref. 17 ) were synthesized by Genechem (Shanghai, China). LIN28B-AS1 expression was normalized to U6.

RNA-immunoprecipitation (RIP)
Briefly, following the applied treatment, cells were collected by trypsinization, washed, and incubated with 0.3% formaldehyde and glycine (0.125 M), and cell pellets were re-suspended in the RIP buffer as previously described 22 .
Lysates were then incubated with the anti-IGF2BP1 antibody (Santa Cruz Biotech). IGF2BP1-bound pellets were washed, re-suspended and incubated with three times in cold PBS, and re-suspended and incubated with the proteinase K-containing buffer containing. IGF2BP1bound RNA was isolated. LIN28B-AS1 expression was tested by qPCR.

RNA pull-down
Biotin-labeled full-length LIN28B-AS1 (see ref. 17 ) was transcribed using the described protocol 21 , isolated with the RNeasy Mini kit (Invitrogen). Biotinylated LIN28B-AS1 was dissolved in RNA structure buffer and folded, put on ice immediately, and then transferred to room temperature. For each treatment, 500 μg cleared nuclei lysates of cultured cells were mixed with folded LIN28B-AS1 and Dynabeads MyOne Streptavidin C1 magnetic beads ("Beads", provided by Dr. Wang 21 ). Beads were washed, and the retrieved proteins were tested by Western blotting.

Cell viability assay
Cells were initially seeded into the 96-well tissue culture plates (3 × 10 3 cells per well). MTT (Sigma) assay was performed to test the cell viability. MTT optical density (OD) at 550 nm was recorded.

BrdU incorporation
Cells were seeded into the 24-well tissue culture plates (2 × 10 5 cells per well) at 60% confluence. Cell proliferation was detected using a BrdU incorporation ELISA kit (Cell Signaling, Shanghai, China) after 48 h. Cells were incubated with BrdU (10 μM), with BrdU absorbance value tested at 450 nm.

EdU assay
As previously described 21 the 5-ethynyl-20deoxyuridine (EdU) Apollo-488 Kit (Ribo-Bio, Guangzhou, China) was utilized for the quantification of cell proliferation. Following the applied genetic treatments, HCC cells were cultured for 48 h and stained with EdU (10 μM, 2 h at room temperature). Cell nuclei were costained with DAPI for 10 min, visualized under a fluorescent microscope (Leica).

In vitro cell migration and invasion assays
Corning chambers (12 μm pore, Corning, New York, NY), pre-coated with or without Matrigel (0.5 mg/mL, BD Biosciences, Shanghai, China), were utilized. HCC cells with applied genetic treatments (1 × 10 5 cells of each treatment), starved overnight, were added to the upper chamber, with lower chamber filled with completed medium (with 10% FBS). After 16 h, HCC cells invaded to the lower surface of the chamber were fixed, stained, and counted. Mitomycin (1.5 μg/mL, Sigma) was always added to exclude the influence of cell proliferation 21 .

TUNEL assay of cell apoptosis
HCC cells with the applied genetic treatments were initially seeded into six-well tissue-culture plates (at 1 × 10 5 cells per well), and cultured for 48 h. A TUNEL Kit (Invitrogen Thermo-Fisher, Shanghai, China) was utilized. TUNEL and DAPI dyes were added to HCC cells, visualized under a fluorescent microscope.

Annexin V FACS
Cells with the applied genetic modifications were stained with Annexin V-FITC (15 μg/mL) and PI (15 μg/ mL) (Biyuntian, Wuxi, China), and detected via fluorescence-activated cell sorting (FACS) on a FACS-Calibur machine (BD Biosciences). Annexin V +/+ cells were labeled as the apoptotic cells, and its ratio was recorded.

JC-1 assay of mitochondrial depolarization
Apoptotic cells will often undergo mitochondrial depolarization ("ΔΨ"), and JC-1 dye shall aggregate in mitochondria to form green monomers 23 . HCC cells with the applied genetic treatments were incubated with JC-1 (5 μg/ mL) for 15 min under the dark. JC-1 green fluorescence intensity, at 550 nm, was examined by a fluorescence spectrofluorometer (Titertek Fluoroscan, Germany). The representative JC-1 images, intergrading green and red fluorescence images, were presented as well.

Western blotting
Cells or tumor tissues were harvested via RIPA lysis buffer (Biyuntian, Nanjing, China). Total protein was quantified, mixed with 5× sample buffer, and boiled at 95°C for 5 min. Aliquots of 40 μg lysate proteins per sample were separated by SDS-PAGE gels, and transferred to the PVDF blots, followed by detection with the indicated primary and secondary antibodies. An enhanced chemiluminescence (ECL) detection kit (Amersham, Buckinghamshire, UK) was utilized to visualize the targeted protein bands. Band intensity was quantified by ImageJ software (NIH).

IGF2BP1 KO
HepG2 cells were seeded onto six-well tissue culture plates (1 × 10 5 cells per well). CRISPR/Cas9-IGF2BP1-KO construct (with sgRNA 5′-GAGCACAAGATCTCCTA-CAG-3′, from Dr. Liu 24 ) was transfected to HepG2 cells. FACS sorting of the GFP-positive cells was performed, and cells further cultured for another 10-12 days. Monoclonal cells were then subjected to genotyping of depleted region of IGF2BP1. Two lines of stable HepG2 cells with complete IGF2BP1 KO were established, with IGF2BP1 KO confirmed by Western blotting.

IGF2BP1 overexpression
The recombinant adenovirus encoding the human IGF2BP1 pSUPER-puro construct was provided by Dr. Liu 24 , added to HepG2 cells. Infection was allowed to proceed for 48 h. Expression of IGF2BP1 in the resulting cells was tested by Western blotting.

In vivo tumor growth
As reported 19 , HepG2 cells were injected subcutaneously (s.c.) to the right flanks of female nude mice (6-7 weeks old, 18-19 g). When tumors reached close to 100 mm 3 , mice were randomized into four groups (10 mice per group). Tumor volumes and mice body weights were monitored every 7 days. Tumor volumes were calculated via the formula: (mm 3 ) = (the shortest diameter 2 × the longest diameter)/2. For recording mice body weight, the estimated tumor weight (tumor volume × 1 mg/mm 3 ) was always subtracted from total mice weight. All injections were performed via the described anesthesia method 25 . All animal studies were performed according to the standards of ethical treatment and IACUC of the Second Affiliated Hospital of Soochow University. The protocols of the study were approved by the Ethics Committee (2015-BR021) of the Second Affiliated Hospital of Soochow University.

Statistical analysis
The investigators were always blinded to the group allocation during the experiments of the study. In vitro experiments were repeated at least three times, with similar results obtained. Data were presented as mean ± standard deviation (SD). Statistics were analyzed by one-way ANOVA followed by a Scheffe' and Tukey Test (SPSS 19.0, Chicago, IL). A two-tailed unpaired T test was applied to test significance between two treatment groups (Excel 2007). Significance was chosen as p < 0.05.

LIN28B-AS1 is expressed in human HCC cells and tissues
First, we tested expression of LIN28B-AS1 in human HCC cells. By employing qPCR, we show that LIN28B-AS1 is expressed in established (HepG2 cell line) and primary human HCC cells ("HCC1/2") (Fig. 1a). Conversely, its expression is not detected in established L02 hepatocytes and primary human hepatocytes (Fig. 1a). LIN28B-AS1 expression was also detected in six out of seven human HCC tissues ("T", Fig. 1b). It is however not expressed in all seven normal liver tissues ("N", Fig. 1b). Thus, LIN28B-AS1 is expressed in human HCC cells and tissues.
To test the possible association between LIN28B-AS1 and the IGF2BP1 protein in HCC cells, the LIN28B-AS1 pull-down assay was carried out. Our results show that the IGF2BP1 protein is precipitated with the in vitrotranscribed and biotinylated LIN28B-AS1 in the nuclei of HepG2 cell and primary human HCC cells (Fig. 1c). Additionally, employing a RIP assay, we again confirmed the direct association between endogenous LIN28B-AS1 and the IGF2BP1 protein in HepG2 cells and primary HCC cells (Fig. 1d). Therefore, LIN28B-AS1 directly associates with the IGF2BP1 protein in HCC cells.

LIN28B-AS1 KO inhibits HCC cell progression in vitro
To further support a role of LIN28B-AS1 in HCC cell functions, the CRISPR/Cas9-LIN28B-AS1-KO construct (see "Methods" section) was transfected to HepG2 cells.
Using the same CRISPR/Cas9-LIN28B-AS1-KO construct, we completely depleted LIN28B-AS1 in the primary human HCC cells ("HCC1") (Fig. 3h). Gli1, Myc, Fig. 1 LIN28B-AS1 is expressed in human HCC cells and tissues. Total RNA was extracted from the established/primary human cells, L02 hepatocytes and primary human hepatocytes a, and from a total of seven (7) pairs human HCC tissues ("T") and surrounding normal liver tissues (N") b, LIN28B-AS1 expression was tested by qPCR, and its levels were normalized to U6 a, b. Western blotting of IGF2BP1 protein retrieved by in vitrotranscribed LIN28B-AS1 in HepG2 and primary human HCC cells c. qPCR analyses of LIN28B-AS1 enriched by IGF2BP1 protein in HepG2 and primary human HCC cells d. Data were presented as mean ± standard deviation (SD, n = 5). The experiments were repeated three times, and similar results were obtained. and IGF2 mRNAs were downregulated in LIN28B-AS1 KO HCC1 cells (Fig. 3i). Cell viability and proliferation were inhibited as well (Fig. 3j). Additionally, LIN28B-AS1 KO augmented positive nuclear TUNEL ratio in HCC1 cells (Fig. 3k), indicating apoptosis activation. Collectively, these results show that LIN28B-AS1 KO inhibited human HCC cell survival and proliferation in vitro.

LIN28B-AS1 KO inhibits HepG2 xenograft tumor growth in mice
At last, we tested the potential effect of LIN28B-AS1 on HCC cell growth in vivo. Control HepG2 cells and LIN28B-AS1 KO HepG2 cells ("L1", see Fig. 3) were (see figure on previous page) Fig. 2 LIN28B-AS1 silencing inhibits HCC cell progression in vitro. HepG2 cells were transfected with LIN28B-AS1 siRNA ("si-LIN28B-AS1-S1/ S2", 0.5 μM) or the non-sense control siRNA ("siR-C") for 48 h, expression of LIN28B-AS1 a and listed genes b, c was shown; Cells were further cultured for applied time periods, and cell viability was tested by MTT d, with cell proliferation examined by EdU staining assay e; Cell migration and invasion were tested by "Transwell" and "Matrigel Transwell" assays, respectively f. Cell apoptosis and mitochondrial depolarization were examined by TUNEL staining/Annexin V FACS g and JC-1 staining h assays, respectively. Huh7 cells and primary human HCC cells ("HCC1/2") as well as L02 or primary human hepatocytes ("Hepatocytes", same for all figures) were transfected with si-LIN28B-AS1-S1 ("kd") or non-sense control siRNA ("siR-C") for 48 h. Cells were further cultured for applied time periods, cell survival and proliferation were, respectively, tested by MTT i and BrdU incorporation j assays, with cell apoptosis examined by TUNEL staining k. For EdU-staining assays, five randomly selected views (of each condition) with total 1000 cells were included to calculate EdU/DAPI ratios (same for all figures). For "Transwell" and "Matrigel Transwell" assays, five randomly selected views in each condition were included to calculate the average number of migrated/invaded cells (same for all figures). For all in vitro functional assays, the exact same number of viable cells of different genetic treatment/s were initially seeded onto each well/dish ("Day-0"/"0 h", same for all figures). Listed proteins were quantified and normalized c. "Ctrl" stands for the parental control cells (same for all figures). Data were presented as mean ± standard deviation (SD, n = 5). *p < 0.05 vs. "siR-C" cells. The experiments were repeated three times, and similar results were obtained. Bar = 100 μm. inoculated s.c. to the flanks of the nude mice. Recordings were started when tumor volumes were close to 100 mm 3 ("Day-0"). We show that LIN28B-AS1 KO HepG2 xenografts grew significantly slower than control tumors (Fig. 6a). The estimated daily tumor growth (in mm 3 per day) was also calculated: (volume at Day-42 subtracting volume at Day-0)/42. The results further confirmed that LIN28B-AS1 KO potently inhibited HepG2 tumor growth (Fig. 6b). The mice body weights were not significantly different between two groups (Fig. 6c).

Discussion
LncRNAs are a large class of transcripts with largely unknown biological functions. Existing studies have Total RNA was extracted from the stable HepG2 cells or primary human HCC cells ("HCC1") with CRISPR/Cas9-LIN28B-AS1-KO construct ("KO-LIN28B-AS1") or CRISPR/Cas9 control construct ("CRISPR-C"), expression of LIN28B-AS1 a, h, Gli1, Myc and IGF2 mRNA/protein b, i were tested. Cells were further cultured for applied time periods, cell survival, proliferation, migration, invasion as well as cell apoptosis and mitochondrial depolarization were tested by the assays mentioned, and results were quantified c-g, j, k. Listed proteins were quantified and normalized b. Data were presented as mean ± standard deviation (SD, n = 5). *p < 0.05 vs. "CRISPR-C" cells. The experiments were repeated three times, and similar results were obtained.
implied that LncRNA dysregulation could play a pivotal role in the initiation, tumorigenesis, and progression of HCC [10][11][12][13] . LncRNAs can crosstalk with multiple chromatin, DNA, RNA, and proteins, regulating HCC cell progression via transcriptional and post-transcriptional mechanisms [10][11][12][13] . Hosono et al. in 2017 has first identified a conserved cancer/testis Lnc-RNA, namely THOR (Lnc-THOR). It promoted cancer progression through directly interacting with IGF2BP1 5 . Very recent studies have shown that Lnc-THOR-IGF2BP1 association is important for HCC cell proliferation and migration, as well as liver cancer stem cells expansion 29,30 . Inhibition or disruption Lnc-THOR-IGF2BP1 association potently inhibited human cancer cell progression 29,30 . These results highlight that IGF2BP1-bound LncRNAs should be novel and important therapeutic targets for HCC 29,30 .
IGF2BP1 is one key member of the RNA-binding IGF2BP family proteins. It is essential for mRNA Fig. 4 Ectopic LIN28B-AS1 overexpression promotes human HCC cell progression in vitro. Total RNA was extracted from the stable HepG2 cells with the lentiviral pre-LIN28B-AS1 expression construct ("LV-LIN28B-AS1", two lines, "L1/L2") or empty vector ("LV-Vec"), LIN28B-AS1 a as well as Gli1, Myc, and IGF2 mRNAs b were tested; Listed proteins were tested by Western blotting c. Cells were further cultured for applied time periods, cell survival, and proliferation were tested by MTT d and EdU staining e assays, respectively; Cell migration and invasion were tested by "Transwell" and "Matrigel Transwell" assays, with results quantified f, respectively. Huh7 cells and primary HCC cells ("HCC1/2") were transduced with LV-LIN28B-AS1 or LV-Vec, and stable cells established with puromycin selection. Expression of LIN28B-AS1 was tested g, with cell proliferation and migration examined by EdU incorporation h and "Transwell" assays i, and results were quantified. Listed proteins were quantified and normalized c. Data were presented as mean ± standard deviation (SD, n = 5). *p < 0.05 vs. "LV-Vec" cells. The experiments were repeated three times, and similar results were obtained. stabilization and translation of several key procancerous genes. Our results confirmed that LIN28B-AS1-IGF2BP1 association is essential for IGF2BP1 activity. We show that LIN28B-AS1 is expressed in human HCC cells and tissues. It is however not expressed in human hepatocytes and normal liver tissues. RIP and RNA pull-down assay results confirmed the direct association between LIN28B-AS1 and the IGF2BP1 protein in HCC cells.
Significantly, LIN28B-AS1 siRNA or KO downregulated IGF2BP1-dependent mRNAs (IGF2, Gli1, and Myc), leading to potent inhibition on HCC cell growth, proliferation, migration, and invasion. Conversely, forced overexpression of LIN28B-AS1, by a lentiviral construct, promoted HCC cell progression in vitro. In vivo, LIN28B-AS1 KO-HepG2 tumors grew significantly slower than the control tumors in the nude mice. These results indicate that LIN28B-AS1-IGF2BP1 binding is essential for IGF2BP1's functions in HCC cells.

Conclusion
These results together suggest that LIN28B-AS1 associates with IGF2BP1 to promote human HCC cell progression in vitro and in vivo. LIN28B-AS1 could be a novel and valuable therapeutic target for HCC. Daily tumor growth was calculated as described b; At Day-14, three HepG2 tumors of each group were isolated, expression of LIN28B-AS1 and listed genes were shown d-f. Bars stand for mean ± standard deviation (SD). *p < 0.05 vs. "CRISPR-C" tumors.