Long noncoding RNA FOXD2‐AS1 aggravates hepatocellular carcinoma tumorigenesis by regulating the miR‐206/MAP3K1 axis

Abstract LncRNAs play crucial roles in the development of various cancers including hepatocellular carcinoma (HCC). Nevertheless, the function of the long noncoding RNA (lncRNA) FOXD2‐AS1 in HCC is still poorly understood. In this study, we focused on the role of FOXD2‐AS1 in HCC. We found that FOXD2‐AS1 was significantly upregulated in HCC cells in comparison to normal human liver cells, LO2. In this study, we also demonstrated that miR‐206 expression was greatly reduced in HCC cells. Furthermore, the inhibition of FOXD2‐AS1 repressed HCC cell proliferation, enhanced cell apoptosis, and restrained cell invasion and migration. The knockdown of FOXD2‐AS1 elevated miR‐206 expression, and we validated an interaction between these RNAs. Additionally, miR‐206 mimics inhibited HCC development while miR‐206 mimics had the opposite effect. MAP kinase 1 (MAP3K1) was predicted to be a target of miR‐206. We discovered that FOXD2‐AS1 modulated MAP3K1 expression by sponging miR‐206 in MHCC‐97L and HepG2 cells. Finally, our in vivo experiments validated that the knockdown of FOXD2‐AS1 inhibited HCC progression by modulating the miR‐206/MAP3K1 axis. In conclusion, this work implies FOXD2‐AS1 accelerates HCC progression through sponging miR‐206 and regulating MAP3K1 expression.

LncRNAs are a class of RNA transcripts with more than 200 nucleotides and no protein encoding potential. 6 Accumulating evidence indicates lncRNAs are closely associated with multiple biological processes 7 and are implicated in numerous cancers. For example, lncRNA TUG1 is elevated in HCC, which regulates HCC cell growth and apoptosis by repressing KLF2. 8 LncRNA UFC1 can interact with HuR and increase β-catenin levels in HCC cells. 9 Additionally, lncRNA HULC can modulate lipid metabolism in hepatoma cells through targeting RXRA. 10 LncRNA DANCR promotes the stemness of HCC cells by repressing CTNNB1. 11 Previous studies have reported that FOXD2-AS1 lncRNA can serve as a significant oncogene in several cancers, including lung cancer and gastric cancer. 12,13 Moreover, a recent study demonstrated FOXD2-AS1 as a prognostic biomarker for HCC. 14 However, the functional role and the underlying mechanism of FOXD2-AS1 in HCC development remain unknown.
MicroRNAs are small RNAs involved in many pathological processes through regulating target genes. [15][16][17] Studies have well established microRNA as crucial participators in HCC. For instance, miR-195 can inhibit HCC metastasis by inhibiting YAP. 18 Decreased miR-125a level can indicate poor survival of HCC patients. 19 miR-30a-5p can suppress HCC proliferation and invasion by targeting FOXA1. 20 miR-18a can trigger HCC proliferation and metastasis by targeting KLF4. 21 Previous studies have found that miR-206 can act as a tumor suppressor in multiple cancers. It can repress renal cell cancer growth by targeting GAK, 22 and suppress glioblastoma progression by regulating BCL-2. 23 In HCC, the expression of miR-206 is significantly reduced, and it is also a potential diagnostic biomarker. 24 However, the potential function of miR-206 in HCC remains poorly understood.
We found that FOXD2-AS1 expression was upregulated in HCC cells. Depletion of FOXD2-AS1 restrained HCC cell growth. Meanwhile, miR-206 was predicted as a target of FOXD2-AS1, and MAP3K1 was predicted as a target of miR-206. Thus, we investigated the functional relationship of FOXD2-AS1, miR-206, and MAP3K1, and hypothesized that FOXD2-AS1 can regulate HCC progression by modulating miR-206 and MAP3K1 expression.

| Clinical samples
Clinical samples were obtained from HCC patients at Renmin Hospital of Wuhan University. All subjects provided written informed consent. The study was approved by the Institutional Ethics Committee of Renmin Hospital of Wuhan University and conducted according to the Helsinki Declaration and government policies. Blood samples were disposed within 2 hours. Then, serum was obtained by centrifugation (1200 × g, 10 minutes) at 4°C. We maintained the serum samples in liquid nitrogen for RNA extraction.

| Lentiviral transfection
Full-length FOXD2-AS1 cDNA was amplified from HCC cells. Sequences were constructed with shLuc as a negative control (NC). Short hairpin RNAs (shRNA) competing against FOXD2-AS1 were designed by Shanghai GenePharma Co., Ltd. shRNAs were subcloned into lentiviral plasmids and infected with lentiviral packaging plasmids to produce two lentiviruses. Cells with FOXD2-AS1 downregulation were incubated with 800 μg/mL G418 at a multiplicity of infection (MOI) of 1-3. Cells were selected 48 hours later by cultivation in a culture medium supplemented with 10 μg/mL blasticidin for 7 days. Surviving cells were reseeded using fresh culture medium.

| EdU assay
EdU proliferation experiments were performed with a EdU kit (Millipore, Billerica, MA, US) to assess cell proliferation. Cells were incubated with 50 μmol/L EdU for 2 hours, and EdU-positive cells were observed using Apollo and DAPI staining.

| Apoptosis
The PI/Annexin V-FITC kit (BioLegend) was used for double apoptotic staining. All cells were digested using trypsin. The cells were then centrifuged, collected, and washed with cold PBS. PI and Annexin V-FITC (5 µL each) were used and the cells were incubated for 15 minutes without light. Samples were analyzed using a BD FACS Calibur Cytometer (BD Bioscience).

| Transwell migration and invasion assay
For the Transwell migration assay, 1 × 10 5 /mL cells were added to the Transwell upper chambers without Matrigel. To evaluate cell invasion, Matrigel-coated Transwell chambers were utilized. Precoated Matrigel and cells were added to the upper chambers. Serum-free DMEM medium was added in 8-μm pore Transwell plates (Corning, MA, USA). Next, cells in the upper chamber were removed using a cotton swab. Cells in the lower chamber were fixed using methanol and stained with 1% crystal violet.

| qRT-PCR
TRIzol regent (Thermo Fisher Scientific) was utilized to extract total RNA, which was preserved at −80°C. RNA concentrations were assessed using a biological spectrometer. cDNA synthesis was carried out using the PrimerScript RT Master kit from Takara Biotechnology (Dalian, China). FOXD2-AS1, miR-206, and MAP3K1 levels were quantified using RT-qPCR by employing SYBR Green PCR Master mix (Roche, Mannheim, Germany) and an Applied Biosystems 7900 Real-Time PCR System. Changes in FOXD2-AS1 levels in the serum were calculated using 2 −ΔCt . Relative gene expression was calculated using 2 −ΔΔCt . The related primer sequences were shown in Table 1.

| Western blot
Proteins were separated using a 10% SDS-PAGE gel and then, the proteins were transferred to PVDF membranes (Bio-Rad). The PVDF membranes were washed using TBST containing 5% nonfat milk for 1 hours and then with antibodies against MAP3K1 and GAPDH (1:1000, CST, MA, USA) overnight at 4°C. The next day, secondary antibodies (1:2000, CST, MA, USA) were incubated with the membranes for 2 hours. ECL solution (Thermo Fisher Scientific) was used to develop the protein bands.

| RNA immunoprecipitation assay
RNA immunoprecipitation (RIP) was conducted using the Magna RIP™ RNA-Binding Protein Immunoprecipitation Kit (Millipore). Cells were lysed in RIP lysis buffer. A quantity of 100 μL of cell extract was coimmunoprecipitated in RIP buffer with magnetic beads conjugated to an anti-Ago2 antibody or normal mouse IgG.

| Animal studies
Twelve nude mice (6-8 weeks, male) were obtained from the Shanghai Experimental Animal Center of the Chinese Academy of Sciences and the mice were maintained at the Experimental Animal Centre in a pathogen-free facility. HepG2 cells, in which FOXD2-AS1 was stably silenced, were injected into the flank area subcutaneously. One month later, the mice were sacrificed by cervical dislocation. All experiments were approved and supervised by the Animal Welfare and Ethics Committee of Renmin Hospital of Wuhan University.

| Statistical analysis
Statistical analysis was done using GraphPad Prism 6.0. Statistical tests consisted of Student's t tests and one-way analysis of variance, where appropriate. Differences were considered statistically significant at P < .05.

| FOXD2-AS1 level is increased and miR-206 expression is decreased in HCC
We first examined the level of FOXD2-AS1 in the serum of HCC patients. FOXD2-AS1 expression was markedly elevated in HCC patients compared to that in healthy controls ( Figure 1A). Next, FOXD2-AS1 expression was examined in 60 pairs of HCC tumors and adjacent nontumor tissues. Consistently, we observed that FOXD2-AS1 was also significantly upregulated in tumor tissues ( Figure 1B). In addition, miR-206 expression was decreased in the serum of HCC patients and HCC tumor tissues compared to that of the controls ( Figure 1C,D). We also examined the differences in FOXD2-AS1 and miR-206 expression in HCC cells compared to those in LO2 cells and observed that FOXD2-AS1 expression was increased in HCC cells compared to that in LO2 cells ( Figure 1E). In contrast, miR-206 expression was downregulated in HCC cells ( Figure 1F). We also analyzed the clinicopathological characteristics of the 60 HCC patients ( Table 2). Analysis of the plasma samples of the patients indicated FOXD2-AS1 and miR-206 were negatively correlated with HCC carcinogenesis (Table 3).

growth, migration, and invasion
Next, to determine whether FOXD2-AS1 affected HCC cell proliferation, CCK8 assay was performed. Hepatocellular carcinoma cells were infected with sh-NC, sh-FOXD2-AS1, sh01, sh02, or sh03 for 48 hours. qRT-PCR analysis indicated that sh02 exhibited the strongest effect on FOXD2-AS1 levels ( Figure 2A). Hepatocellular carcinoma cell survival was repressed in response to decreased FOXD2-AS1 expression in HCC cells ( Figure 2B). EdU assays proved that cell proliferation was inhibited by LV-shFOXD2-AS1 ( Figure 2C). In addition, the knockdown of FOXD2-AS1 strongly triggered apoptosis of HepG2 and MHCC-97L cells ( Figure 2D). The migration and invasion capacity of HCC cells was restrained by LV-shFOXD2-AS1 ( Figure 2E,F). These findings indicate that loss of FOXD2-AS1 inhibits HCC cell growth, migration, and invasion.

| MAP3K1 is a target of miR-206
MAP3K1 was also predicted as a target of miR-206, and a schematic of their interaction was described in Figure 4A. WT-MAP3K1 and MUT-MAP3K1 were shown in Figure 4B. Cotransfection of WT-MAP3K1 with miR-206 markedly suppressed reporter activity ( Figure 4C). MAP3K1 mRNA expression was strongly induced in HCC cells compared to that in LO2 cells ( Figure 4D). We then showed that MAP3K1 was inhibited by miR-206 mimics ( Figure 4E,F). Next, as shown in Figure 4G,H MAP3K1 expression was inhibited by LV-shFOXD2-AS1 infection in HCC cells. These data suggest that FOXD2-AS1 can modulate HCC progression by targeting miR-206/MAP3K1.

| Silencing of FOXD2-AS1 inhibits HCC progression in vivo
Subsequently, an in vivo assay was employed to confirm the effect of FOXD2-AS1 on HCC progression. Six mice were injected in the front dorsum with parental HepG2 cells and the other six were injected with HepG2 cells infected with LV-shFOXD2-AS1. Inhibition of FOXD2-AS1 profoundly inhibited tumor growth ( Figure 5A). HE staining and IHC showed Ki-67 was depressed by LV-shFOXD2-AS1 ( Figure 5B). Loss of FOXD2-AS1 repressed MAP3K1 through sponging miR-206 ( Figure 5C-F). Thus, the downregulation of FOXD2-AS1 inhibits HCC development in vivo.

| FOXD2-AS1 is a biomarker for HCC diagnostic prediction
Kaplan-Meier analysis was conducted to identify the correlation between FOXD2-AS1 and HCC patient survival. We detected the expression level of FOXD2-AS1 in HCC serum in all patients. FOXD2-AS1 were high expressed in HCC with an AUC of 0.866 ( Figure 6). These results suggest that FOXD2-AS1 is an appropriate biomarker for HCC diagnostic prediction.

| DISCUSSION
Many lncRNAs have been reported as effective cancer biomarkers. In this study, we found that in HCC cells, lncRNA FOXD2-AS1 expression was significantly elevated, whereas miR-206 expression was downregulated. Additionally, silencing FOXD2-AS1 suppressed HCC progression, and miR-206 mimics repressed HCC progression in vitro. FOXD2-AS1 negatively regulated miR-206 expression, and MAP3K1 was predicted a target of miR-206. FOXD2-AS1 acts as an oncogene in various cancers. FOXD2-AS1 promoted nasopharyngeal cancer by regulating miR-363-5p 25 and acted as an oncogene in colorectal cancer by modulating Notch. 26 Roles for FOXD2-AS1 in HCC progression have recently been reported. For instance, FOXD2-AS1 acted as a ceRNA against miR-150-5p and reversed sorafenib resistance in HCC. 27 FOXD2-AS1 exerted an oncogenic role in HCC by silencing CDKN1B and EZH2. 28 In addition, FOXD2-AS1 can also contribute to HCC development by targeting miR-206. 29 Here, we validated the FOXD2-AS1 level was increased in HCC cells and found that the knockdown of FOXD2-AS1 markedly restrained HCC cell growth, migration, and invasion. FOXD2-AS1 expression can be a novel biomarker to predict HCC diagnosis.
LncRNAs can act as microRNA sponges in several cancers. MALAT1 triggered the progression of HCC by sponging miR-146b-5p. 30 XIST was able to modulate PTEN through sponging miR-181a, which promoted HCC progression. 31 Moreover, TUSC7 sponged miR-10a in HCC. 32 miR-206 was predicted to be a FOXD2-AS1 target in our study. In previous studies, miR-206 has been reported to inhibit HCC by regulating CDK9. 33 And miR-206 overexpression can repress HCC development. 34 Furthermore, we observed that FOXD2-AS1 knockdown increased miR-206 levels. A direct interaction between FOXD2-AS1 and miR-206 was established using RIP and RNA pull-down assays.
MAPKs are significant mediators of cell physiology. 35,36 MAP3K1 belongs to the MAP3K family, 37 and studies have shown that MAP3K1 impacted several cellular processes in tumors. 38,39 miR-451 can inhibit esophageal carcinoma cell proliferation by suppressing MAP3K1. 40 miR-302 increased the sensitivity of breast cancer cells to adriamycin through targeting MAP3K1. 41 MAP3K1 was a target of miR-206, and interestingly, miR-206 inhibited HCC development by targeting MAP3K1. The interaction between miR-206 and MAP3K1 was validated using a dual luciferase reporter assay. We also found that FOXD2-AS1 F I G U R E 6 FOXD2-AS1 is a biomarker for HCC diagnostic prediction. ROC curve analysis was performed to evaluate the diagnostic value of FOXD2-AS1 in HCC