LncRNA ZFPM2-AS1 Promotes Metastasis and Epithelial-mesenchymal Transition of Hepatocellular Carcinoma via Targeting miR-3612/ADAM15 Axis

Background: Long non-coding RNAs (lncRNAs) have obtained growing attention due to their potential effects as novel regulators in various tumors. This study aimed to investigate the expression and roles of lncRNA ZFPM2-AS1 in the progression of hepatocellular carcinoma (HCC). Methods: Transwell was used to determine migration and invasion of HCC cells in vitro. The lung metastasis mouse model was established to detect tumor metastasis of HCC in vivo. The direct binding of miR-3612 to 3'UTR of DAM15 was conrmed by luciferase reporter assay. The expression of ZFPM2-AS1 and miR-3612 in HCC specimens and cell lines were detected by real-time PCR. The correlation among ZFPM2-AS1 and miR-3612 were disclosed by a dual-luciferase reporter assay, RIP assay and biotin pull-down assay. Results: In present study, we found that ZFPM2-AS1 was up-regulated in HCC tissues and cells and its upregulation was associated with TNM stage, vascular invasion, and poor prognosis of HCC patients. Functionally, gain- and loss-of-function experiments indicated that ZFPM2-AS1 promoted cell migration, invasion and EMT progress in vitro and in vivo. ZFPM2-AS1 could function as a competing endogenous RNA (ceRNA) by sponging miR-3612 in HCC cells. Mechanically, miR-3612 inhibited HCC metastasis and alternation of miR-3612 reversed the promotive effects of ZFPM2-AS1 on HCC cells. In addition, we conrmed that ADAM15 was a direct target of miR-3612 in HCC and mediated the biological effects of miR-3612 and ZFPM2-AS1 in HCC. Curcumin, an active derivative from turmeric, exerts its anticancer effects through ZFPM2-AS1/miR-3612/ADAM15 pathway. Our data identied ZFPM2-AS1 as a novel oncogenic lncRNA and correlated malignant clinical outcomes in HCC patients. Conclusions: ZFPM2-AS1 performed as oncogenic role via targeting miR-3612 and subsequently promoted ADAM15 expression in HCC. Our results revealed that ZFPM2-AS1 could be a potential prognostic biomarker and therapeutic target for HCC. carcinoma; AFP, TNM, tumor-node-metastasis. domain; lncRNA, Long non-coding RNA; qRT-PCR, real-time quantitative reverse transcription polymerase chain reaction; IHC, immunohistochemistry; IF, Immunouorescence; RIP, RNA immunoprecipitation. 3'-UTR: 3'-untranslated region; AFP: Alpha-fetoprotein; TNM: Tumor-node-metastasis; H&E: Hematoxylin and eosin; EMT: Epithelial-mesenchymal transition


Background
Hepatocellular carcinoma (HCC) is one of the most frequently diagnosed cancers and a leading cause of cancer-related deaths worldwide, especially in South-Eastern Asia and Africa, causing nearly 745,000 deaths every year [1,2]. Although the diagnosis and treatment technology of HCC have been improved in recent decades, including surgical resection, tumor ablation and transarterial chemoembolization, the clinical outcomes of HCC patients remain unsatisfactory because of the high rate of postoperative recurrence and metastasis [3]. Therefore, it is urgent to elucidate the molecular mechanisms underlying HCC progression and identify novel biomarkers for the diagnosis and treatment for HCC patients.
Long non-coding RNAs (lncRNAs) are a group of non-coding RNA molecules over 200 nucleotides in length, which is characterized by limited protein coding potential [4]. Abnormal expression of lncRNAs play a key role in cancer progression and carcinogenesis through various mechanisms, such as transcriptional regulation, chromatin remodeling, histone modi cation and mRNA splicing and stability [5][6][7]. Zinc nger protein multitype 2 antisense RNA 1 (ZFPM2-AS1) is a novel lncRNA which has not received much attention. Previous studies con rmed that lncRNA ZFPM2-AS1 promoted lung adenocarcinoma progression by interacting with UPF1 to destabilize ZFPM2 [8]. LncRNA ZFPM2-AS1 enhances the malignancy of cervical cancer by functioning as a molecular sponge of miR-511-3p and consequently increasing FGFR2 expression [9]. ZFPM2-AS1 facilitated cell growth in esophageal squamous cell carcinoma via up-regulating TRAF4 [10]. LncRNA ZFPM2-AS1 promotes the tumorigenesis of renal cell cancer via targeting miR-137 [11]. However, the role of ZFPM2-AS1 in HCC remained largely unknown.
In this study, we con rmed for the rst time that lncRNA ZFPM2-AS1 was up-regulated in HCC tissues and cell lines. High lncRNA ZFPM2-AS1 was associated with poor clinical features, including advanced TNM stage and vascular invasion, and worse prognostic survival. Gain-and loss-of-function experiment showed that ZFPM2-AS1 regulated migration, invasion, and epithelial-mesenchymal transition of HCC in vitro and in vivo. Furthermore, we explore whether ZFPM2-AS1 can regulate the expression of ADAM15 by regulating miR-3612 expression and affect the effects of HCC. Our results suggest that ZFPM2-AS1 exerts a critical role in HCC progression and might be a new molecular target for the treatment of HCC.

HCC tissues and cell culture
We collected HCC tissues and corresponding adjacent non-tumor tissues from our department. All patients received no therapy before surgery. All patients signed the informed consent. The patients' clinicopathological issues and demographic details were described in Table 1  qRT-PCR qRT-PCR was performed as reported previously [12][13][14]. TRIzol Reagent (Invitrogen) was used to extract total RNA from cells. PrimeScript RT Master Mix or TaqMan MicroRNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA) was used to reverse total RNA into cDNA. qRT-PCR was performed by SYBR Premix Ex Taq (Takara) and specifc primers. The results were calculated by the 2 -ΔΔCt method. GAPDH and U6 was utilized as endogenous control.

Western blot analysis
Detailed assays were performed as previously reported [15][16][17]. Brie y, RIPA lysis buffer (Invitrogen) was used to extract total proteins. After examining the protein concentration by BCA Kit (Beyotime technology), 10% SDS-PAGE was isolated the proteins and transferred to PVDF membranes. After blocking with 5% milk, membranes were cultured in primary antibodies (Cell signaling technology) and subsequently probed with secondary antibody. Then the signals were determined using chemiluminescence system (Thermo Fisher Scienti c).

Transwell assay
The capacity of cells migration and invasion was evaluated via transwell assay. The upper chamber was pre-coated with BioCoat Matrigel (BD Biosciences, Franklin Lakes, NJ, USA) for invasion assay, whereas migration assay was not pre-coated with BioCoat Matrigel. Subsequently, 200 μl cells were seeded into upper chamber which mixed with serum-free medium, 500 μl DMEM medium containing 10% FBS was added into the lower chamber. After 24 hours incubation, cells on the upper chamber surface were removed using cotton swabs. Cells on the lower surface were xed in 4% paraformaldehyde and stained with 0.5% crystal violet. Finally, NIS Elements image software (Nikon, Tokyo, Japan) was used to detect the number of migrating and invading cells.

Luciferase reporter assay
The experiments were performed similar as our previous studies [18]. The 3′UTR of ZFPM2-AS1 or ADAM15 wild type (wt) or mutant (mt) sequence was inserted in pmirGLO dual-luciferase vector (Genecopoeia, Guangzhou, China). Then the cells were co-transfected with miR-3612 or control mimics by Lipofectamine 2000 (Invitrogen) into HCC cells. Dual-Luciferase Reporter Assay System (Promega) was used to analyze the luciferase activity.

RIP (RNA immunoprecipitation) assay
We used the EZ-Magna RIP Kit (Millipore) to conduct the assays. The cells were lysed in RIP lysis buffer and incubated with magnetic beads of anti-Ago2 antibody (Millipore). Then the mix were incubated with proteinase K and the immunoprecipitated RNA was examined to real-time PCR analysis.
In vivo metastasis assay 4-6-week-old male BALB/c nude mice (Centre of Laboratory Animals, The Medical College of Xi'an Jiaotong University, Xi'an, China) were randomized into two groups (n=5). Transfected cells (1×10 6 ) were injected into the tail veins for the establishment of the pulmonary metastatic model. Mice were sacri ced 3 weeks post-injection and examined microscopically by H&E staining for the development of lung metastatic foci. Animals were housed in cages under standard conditions. All in vivo protocols were approved by the Institutional Animal Care and Use Committee of Xi'an Jiaotong University.

Statistical analysis
All data were expressed as the mean ± SD and carried out by GraphPad Prism 6.0 (San Diego, USA). All experiments were performed at least three times. Survival analysis was plotted by Kaplan-Meier method and log-rank test. The relationship between PICSAR and clinical features was evaluated by Chi square test. The variance was analyzed by calculated by a Student's two-side t test. Difference with P < 0.05 was considered as signi cant.

Results
Up-regulation of ZFPM2-AS1 in HCC was associated with poor prognosis The analysis of data from Gene Expression Pro ling Interactive Analysis (GEPIA) (Fig.1A) and ENCORI Pan-Cancer Analysis platform ( Fig.1B) described the expression of ZFPM2-AS1 was higher in the HCC tissues compared to non-tumor tissues. Moreover, the data from our cohort was consistent with this (P<0.05, Fig. 1C). Moreover, our data showed that ZFPM2-AS1 was up-regulated in a group of HCC cells compared to normal hepatic cell L02 (P<0.05, Fig. 1D). To clarify the clinical importance of ZFPM2-AS1 in HCC patients, we divided the patients into high or low subgroups according to the median value. As shown in Table 1, high ZFPM2-AS1 was associated with TNM stage (P=0.003) and vascular invasion (P=0.033). Kaplan-Meier survival analysis revealed that high ZFPM2-AS1 patients had a worse overall survival (OS) and disease-free survival (DFS) than patients with low ZFPM2-AS1 level (P=0.0006, 0.0002, respectively, Fig.1E). Moreover, the data from GEPIA (Fig.1F) and ENCORI (Fig.1G) also con rmed that ZFPM2-AS1 expression was dramatically correlated with OS and DFS. Taken together, these data suggest that ZFPM2-AS1 was up-regulated in HCC, and this up-regulation may play an important role in the progression of HCC.
EMT has been identi ed as a key process in the initiation of metastasis progression of HCC. To elucidate the potential role of ZFPM2-AS1 in HCC metastasis, we performed Western blot to con rm that ZFPM2-AS1 overexpression decreased the epithelial marker E-cadherin and increased the mesenchymal marker N-cadherin and Vimentin expression in Hep3B cells (P<0.05, Fig. 2D). By contrast, ZFPM2-AS1 knockdown showed opposite effects on EMT (P<0.05, Fig. 2D). Moreover, we used tail vain injection to construct lung metastasis model. The data showed that ZFPM2-AS1 overexpression signi cantly promoted the number of lung metastasis of Hep3B cells whereas ZFPM2-AS1 knockdown reduce lung metastasis amounts (P<0.05, Fig.2E). In conclusion, these data suggested that ZFPM2-AS1 could regulate HCC cell migration, invasion and EMT of HCC.
miR-3612 mediates the effects of ZFPM2-AS1 on migration, invasion and EMT of HCC cells To explore the effects of miR-3612 in HCC, we con rmed that miR-3612 was down-regulated in HCC cells compared with L02 cells (P<0.05, Fig.4A). We transfected HCCLM3 cells by lentivirus and knocked down miR-3612 in Hep3B cells with different endogenous miR-3612 expression to investigate the biological effects on HCC (P<0.05, Fig.4B). MiR-3612 suppressed migration, invasion and EMT of HCC cells (P<0.05, Fig.4C-E).
To investigate whether miR-3612 mediated the function of ZFPM2-AS1 in HCC cells, we co-transfected sh-ZFPM2-AS1 with miR-3612 inhibitors into HCC cells. The rescue experiments demonstrated that miR-3612 overexpression vectors impaired the promotion effect of ZFPM2-AS1 on migration and invasion (P<0.05, Fig.4F-G). Moreover, miR-3612 inhibitors reversed the inhibitory effect of sh-ZFPM2-AS1 on migration and invasion (P<0.05, Fig.4F-G). In conclusion, these data suggest that ZFPM2-AS1 promoted HCC migration and invasion partly by repressing miR-3612 function.
Curcumin attenuates HCC migration and invasion via ZFPM2-AS1/miR-3612/ADAM15 signaling pathway The above data clarify the molecular mechanisms of ZFPM2-AS1-mediated biological progress in HCC cells. To develop new agents to treat HCC, we found Curcumin, a yellow-colored dietary avor from the plant (Curcuma longa), has been used in Chinses traditional medicine to treat various diseases, such as hepatic brosis, cancers [20][21][22]. Curcumin is a hydrophobic polyphenol derived from turmeric (Fig.8A). To con rm the inhibitory effects on HCC cells, we treated HCC cells with different concentrations of curcumin and cell viability was detected by MTT assays. As shown in Fig.8B, curcumin treatment signi cantly inhibited cell viability in a time-and dose-dependent manner. We choose 20 μM for subsequent experiment. Curcumin obviously inhibited HCC cell migration, invasion and EMT progress (P<0.05, Fig.8C-D). Moreover, curcumin signi cantly inhibited ZFPM2-AS1, ADAM15 expression while increased miR-3612 expression (P<0.05, Fig.8E-G). Collectively, these data demonstrated that curcumin performed an anti-cancer effects on HCC cells.

Discussion
Accumulating evidence indicated that lncRNAs have the signi cant function in the carcinogenesis and the progression of malignancies during these years [23]. Therefore, it's crucial to investigate the regulatory characteristics of lncRNAs in the pathogenesis of HCC. In this study, we con rmed for the rst time that lncRNA ZFPM2-AS1 was signi cantly up-regulated in HCC tissues and cell lines. Moreover, ZFPM2-AS1 expression was signi cantly associated with TNM stage and vascular invasion of HCC. These data suggest that ZFPM2-AS1 is an oncogene in HCC and plays a critical role in the progression of HCC.
Previous studies reported that lncRNA ZFPM2-AS1 was identi ed as a diagnostic and prognosis marker in cancers. ZFPM2-AS1 attenuated p53 pathway and promoted gastric carcinogenesis by stabilizing MIF [24]. ZFPM2-AS1 was involved in lung adenocarcinoma via miR-511-3p/AFF4 pathway [25]. ZFPM2-AS1 promoted proliferation via miR-18b-5p/VMA21 axis in lung adenocarcinoma [26]. In this study, we demonstrated that ZFPM2-AS1 promoted cell migration, invasion and EMT progress by gain-and loss-of function experiment in vitro and in vivo. Previous studies reported that aberrant lncRNAs act as ceRNAs for miRNAs to modulate tumor development and progression [27]. In this study, we con rmed miR-3612 was remarkably down-regulated and negatively correlated with ZFPM2-AS1 in HCC tissues. Moreover, bioinformatics analysis, luciferase reporter assay, biotin pull-down assay and RIP assay all de ned that miR-3612 was a target of ZFPM2-AS1 in HCC cells. In addition, we demonstrated that miR-3612 exerted its suppressive effects on migration and invasion of HCC cells. MiR-3612 mediated the biological function of ZFPM2-AS1 on HCC cells. Conclusively, ZFPM2-AS1 acted as miR-3612 sponge.
It has been con rmed that miRNAs could recognize and directly bind to their target mRNA 3' untranslated regions (3'-UTR) and consequently leading to the degradation or translation inhibition of mRNAs [28]. ADAM15, is a member of a family of catalytically active disintegrin membrane metalloproteinases that function as molecular signaling switches, shed membrane growth factors and cleave and inactivate cell adhesion molecules [29]. ADAM15 has been reported to be overexpressed in numerous malignancies including melanoma, breast cancer [30,31]. Here, we con rmed that ADAM15 is targeted by miR-3612 and overexpressed in HCC. Additionally, overexpressed ADAM15 recovered the effect of miR-3612 and ZFPM2-AS1 on HCC cell migration and invasion. Taken together, these results demonstrated that ZFPM2-AS1 exert an oncogene role via miR-3612/ADAM15 axis in HCC.
Finally, we disclosed that curcumin, an active component of turmeric, displays various pharmacological activities [32]. The promising role of curcumin against different diseases is widely publicized [33].
Curcumin not only has attracted much attention for its anticancer properties, but also has been regarded to be safe in clinical trials owing to its low toxicity and good tolerance to human. Here, we demonstrated that curcumin inhibited HCC cell migration and invasion via regulating ZFPM2-AS1/miR-3612/ADAM15 axis.
In conclusion, we demonstrated that ZFPM2-AS1 was up-regulated in HCC, and could promote cell migration, invasion and EMT progression of HCC cells via ZFPM2-AS1/miR-3612/ADAM15 axis, which could be a valuable and promising therapeutic target for HCC.

Conclusion
To conclude, we reported for the rst time that ZFPM2-AS1 was up-regulated in HCC tissues and cells. Its overexpression was associated with malignant clinical features and unfavorable prognosis. ZFPM2-AS1 facilitates HCC cell migration, invasion and EMT progress in vitro and in vivo. miR-3612 was identi ed as not only a target but also a functional mediator of ZFPM2-AS1 in HCC cells. ADAM15 was a direct target gene of miR-3612 and mediated its biological effects. Curcumin, as an inhibitor of HCC through regulating ZFPM2-AS1/miR-3612/ADAM15 pathway, provides novel mechanism for cancer therapy. In conclusion, ZFPM2-AS1/miR-3612/ADAM15 axis promoted cell migration and invasion of HCC. This nding will improve understanding of mechanism involved in cancer progression and provide novel targets for the molecular treatment of HCC.        Modulation of ADAM15 partially abolishes ZFPM2-AS1 or miR-3612-mediated cellular processes in HCC. (A) miR-3612-overexpressing or ZFPM2-AS1-suppressive HCCLM3 cells that were transfected with empty vector (EV) or ADAM15 overexpression plasmid were subjected to western blot for ADAM15. miR-3612suppressive or ZFPM2-AS1-overexpressing Hep3B cells that were transfected with scrambled shRNA or ADAM15 shRNA were subjected to western blot for ADAM15. (B-D) ADAM15 restoration abrogated the effects of miR-3612 overexpression or ZFPM2-AS1 knockdown on cell migration, invasion and EMT of HCCLM3 cells. ADAM15 knockdown reversed the promotive effects of miR-3612 knockdown or ZFPM2-AS1 overexpression in Hep3B cells. *P < 0.05.