H19/miR-194/E2F3 regulating loop promotes gastric cancer growth and metastasis

Background: Gastric cancer (GC) is one of the most frequent malignant digestive tumors and second fatal cancer. This study was to investigate whether lncRNA-H19 can regulate E2F3 expression through competitive binding to microRNA-194 (miR-194), thus regulating GC growth and metastasis. Methods: H19, miR-194, and E2F3 expression levels in GC tissues and cell lines were investigated using quantitative reverse transcriptase-polymerase chain reaction (QRT-PCR). Meanwhile, the mRNA levels of H19 and E2F3 in gastric cancer tissues were also analyzed through the GEPIA web tool. The binding condition of miR-194 with H19 and E2F3 was investigated using a dual-luciferase reporter gene assay. The regulatory effects of H19 on proliferative, migratory, and invasive abilities of AGS cells and SGC-7901 cells were detected by transwell assay and cell counting kit-8 (CCK-8). Genes involved in proliferation, migration, and invasion (PCNA, Vimentin, and N-cadherin) were determined using QRT-PCR and western blot. The regulatory interaction between H19 and miR-194, miR-194, and E2F3 were investigated using rescue experiments. Results: The results revealed that H19 was highly expressed in GC tissues and cell lines than those of controls. Downregulated H19 decreased the proliferation, migration, and invasion of AGS cells and SGC-7901 cells. H19 was demonstrated that being the molecular sponge of miR-194 in regulating the growth of the GC cells. The level of E2F3 expression was also found signicantly higher in GC tissues and cell lines than those of controls. And then, the mimics of miR-194 inhibited the expression of E2F3 in the GC cells. CCK-8 assay showed decreased proliferative ability induced by miR-194 mimics were reversed by E2F3 overexpression. Transwell assays showed decreased migratory and invasive ability induced by miR-194 mimics were reversed by E2F3 overexpression. Conclusions: This study demonstrates that H19 promotes GC growth and metastasis by regulating E2F3 abilities in si-H19-transfected AGS cells or SGC-7901 cells. The expression levels of EMT ‐ associated molecule (Vimentin) in transfected AGS and SGC ‐ 7901 cells were determined using QRT-PCR(E), and western blot (F). The expression levels of EMT ‐ associated molecule (N-cadherin) in transfected AGS and SGC ‐ 7901 cells were determined using QRT-PCR(G) , and western blot (H). Data are presented as mean ± SEM; * p < 0.05.

investigate the mechanism of lncRNAs in tumor progression and to seek novel therapeutic strategies and biomarkers.
In the current study, we recruited several functional assays to determine the role of H19, a highly conservative lncRNA, in GC. Bioinformatics analysis and QRT-PCR assay were used to detect the level of lncRNA-H19 expression in GC. The results showed that the level of lncRNA-H19 expression was signi cantly higher in GC tissues and cell lines than that in control. Bioinformatics databases (http://www.mirdb.org and http://mirtarbase.mbc.nctu.edu.tw/php/index.php) and (http://starbase.sysu.edu.cn/starbase2/) indicated that H19 could directly bind to , and E2F transcription factor 3 (E2F3) as possible target genes for miR-194. This study is to better understand the regulatory role of the H19/miR-194/E2F3 loop in the gastric carcinogenesis.

Sample Information
A total of 30 pairs of GC tissues and para-carcinoma normal tissues were obtained in this study. And GC tissue included 21 metastatic-positive and nine metastatic-free cases. Enrolled patients were pathologically diagnosed as GC.
The Online Database Gene Expression Pro ling (the GEPIA web tool) Based on The Cancer Genome Atlas (TCGA) and the Genotype-Tissue Expression (GTEx) projects, the RNA sequencing expression data related to our project was analyzed using the GEPIA web tool (http://gepia.cancerpku.cn/index. html.).

Cell culture and transfection
The gastric cancer cell lines (SGC-7901, SUN-16, MKN-45, and AGS cells) and the human gastric epithelial cell line GES-1 (American Type Culture Collection, USA) were used in this study. All cells cultured respectively in Dulbecco's Modi ed Eagle Medium (DMEM, Gibco, USA) supplemented with 10% fetal bovine serum (FBS), 100 U/mL penicillin and 100 μg/mL streptomycin (Hyclone, USA) and incubated in a 5% carbon oxide (CO 2 ) incubator at 37°C.
For cell transfection, cells in the logarithmic growth phase were transfected with corresponding constructs when the con uence was up to 80% following the instructions of Lipofectamine2000 (Invitrogen, USA). The culture medium was replaced 6 hours later. H19-small interfering RNA (si-H19), miR-194 mimic, miR-194 inhibitor, pcDNA-E2F3, and negative control (NC) were constructed by Gene Pharma (Shanghai, China).
Cell counting kit-8 (CCK-8) assay For determining the proliferative ability of GC cells, the transfected cells were incubated in 100 μl culture medium each well (6×10 3 cells/well) using 96-well plates. At different set time, the CCK-8 solution (Beyotime, Shanghai, China) (10 μL/well) was added to cells and then incubated at 37°C for 2 hours in the dark. The optical density (OD) value (450 nm) was evaluated by a microplate reader.

Cell migration and invasion assays
Transwell assay was used to investigate the migration and invasion of GC cells. Different group GC cells were put on the upper Matrigel-coated invasion chambers or non-coated migration chambers (BD Biosciences, USA), respectively. 500 μl of DMEM medium containing 10% FBS was put in the lower chamber, and the serum-free medium was put in the upper chamber. The non-invasive cells were wiped off by cotton swabs after 48 hours of incubation. The migrating and invading cells were xed with 95% ethanol, stained with 0.1% crystal violet. The number of migratory and invasive cells in the lower chamber were counted under an inverted microscope (×100). The experiments were independently repeated three times.

Western blotting
The total protein of treated cells was extracted using a radioimmunoprecipitation assay (RIPA) kit (Beyotime, China). The sample of protein was separated by electrophoresis on 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred to polyvinylidene di uoride (PVDF) membrane (Millipore, USA). Followed by the blocks with skimmed milk, the membranes were incubated with primary antibodies (Cell Signaling Technology, USA) overnight at 4°C. After being washed with Tris-buffered Saline with Tween 20 (TBS-T) for three times, the membranes were incubated with secondary antibody at 24°C for 1 hour. The protein blot on the membrane was exposed by chemiluminescence.

Statistical analysis
All numerical data were expressed as the mean ± standard error of mean (SEM), and qualitative data were expressed as n (%). A comparison between the treatment groups of numerical data was analyzed by independent t-tests or 1-way or repeated-measures analysis of variance (ANOVA). Comparison between the treatment groups of qualitative data were analyzed by chi-square tests, and Fisher's exact tests were used for correction if necessary. All p-values were 2-tailed, and p-values less than 0.05 were considered statistically signi cant. For all statistical calculations, p-values were determined using SPSS (version 18.0 for Windows; SPSS, Inc., USA).

H19 was overexpressed in GC tissues and cell lines
Using QRT-PCR assay, the expression level of H19 in GC tissues was found signi cantly higher than that in para-carcinoma normal tissues ( Figure 1A). Consistently, the expression level of H19 was also signi cantly higher in GC cell lines than that in gastric epithelial cell line GES-1 ( Figure 1D). Moreover, the expression level of H19 in metastatic-positive GC tissues was signi cantly higher than that in metastaticfree GC tissues ( Figure 1B). Meanwhile, through the GEPIA web tool, the mRNA levels of H19 was also found signi cantly upregulated in GC tissues, which was consistent with our results ( Figure 1C).

H19 regulated the proliferation, migration, and invasion of GC cells
To further explore the function of H19 in GC cells, H19 was down-regulated by si-H19 in SGC-7901 cells ( Figure 2A) and AGS cells ( Figure 2C). The CCK8 showed that si-H19 signi cantly decreased the proliferative ability of SGC-7901 cells ( Figure 2B) and AGS cells ( Figure 2D). The QRT-PCR and western blot showed that si-H19 signi cantly decreased the expression levels of PCNA of AGS and SGC-7901 cells ( Figure 2E and 2F). Transwell assays were used to detect the cell invasive and migratory abilities in si-H19-transfected GC cells, and si-H19 transfected GC cells exhibited a decreased invasion capacity ( Figure 3A and 3C) and in migratory capacity ( Figure 3B and 3D). The QRT-PCR and western blot showed that si-H19 signi cantly decreased the expression levels of EMT-associated molecules (Vimentin and Ncadherin) of AGS and SGC-7901 cells ( Figure 3E, 3F for Vimentin; Figure 3G, 3H for N-cadherin). Meanwhile, the Overexpression of H19 promotes the PCNA and N-cadherin expression of GC cells in vitro( Figure 4C, 4D for PCNA; and 4E, 4F for N-cadherin). These ndings indicated that H19 promoted GC cell proliferation, migration, and invasion.
H19 was the molecular sponge of miR-194 in regulating the proliferation, migration, and invasion of GC cells The putative binding site between miR-194 and H19 was predicted by bioinformatics (TargetScan, http://www.targetscan.org/) and screened out. H19 WT and H19 MUT plasmids were constructed ( Figure  5A). The level of miR-194 mRNA was signi cantly inhibited in GC cell lines ( Figure 5B). Luciferase reporter gene assay demonstrated that miR-194 overexpression led to a signi cant decrease in luciferase activity of pMIR-H19-WT, but not in pMIR-H19-MUT in AGS cells ( Figure 5C). Moreover, the mRNA level of H19 was remarkably reduced after microRNA-194 mimics transfection in GC cells ( Figure 5D). The above data indicated that H19 was the molecular sponge of miR-194.

E2F3 was the functional target of miR-194 that affect proliferation, migration, and invasion of GC cells
To further study the mechanism of H19/miR-194 in GC cells, bioinformatics analysis (TargetScan, http://www.targetscan.org/) was used to determine the binding sites of miR-194 matched the 3'UTR of E2F3, which was also further con rmed by luciferase reporting assay. H19 WT and H19 MUT plasmids were constructed ( Figure 6A). After 24 hours of transfection, we observed that over-expression of miR-194 led to a signi cant decrease in the Luciferase activity of the E2F3-wild group, but not in the E2F3-mutant group ( Figure 6B). Moreover, the expression level of E2F3 protein was remarkably reduced after miR-194 mimics transfection in GC cells ( Figure 6E). All of these indicated that E2F3 was the direct target of miR-

194.
The level of E2F3 mRNA was signi cantly upregulated in GC cell lines ( Figure 6D). Meanwhile, the levels of E2F3 mRNA were also found signi cantly upregulated in GC tissues by using the GEPIA web tool ( Figure 6B). CCK-8 assay showed decreased proliferative ability induced by miR-194 mimics were reversed by E2F3 overexpression in SGC-7901 cells ( Figure 6F) and AGS cells ( Figure 6G). Transwell assays showed decreased migratory ability ( Figure 6H), and invasive ability ( Figure 6I) induced by miR-194 mimics were reversed by E2F3 overexpression.

Discussion
In the present study, the level of H19 was rstly found signi cantly overexpressed in GC tissues and cell lines. Moreover, miR-194 and E2F3 were demonstrated being the downstream target of H19 and mediated the proliferation, migration, and invasion of GC cells (Figure 7).
Non-coding RNAs, one of the main mechanisms of epigenetics, do not exert the protein-encoding function. Abundant evidences have demonstrated that non-coding RNAs exert a vital role in epigenetic regulation. LncRNAs belong to non-coding RNAs with relatively long chains. It is reported that lncRNA expression is closely related to tumor development [8]. Disordered lncRNAs have an important effect on cellular homeostasis [9]. H19 located on human chromosome 11p15.5. Studies have found that H19 overexpressed in various malignancies, which indicated that H19 overexpression could upregulate the proliferation of tumor cells and promote tumor cell migration and invasion [10,11].
Based on the Starbase database (http://starbase.sysu.edu.cn/starbase2/), miR-194 was manifested being the direct target of H19. Moreover, in our study, the dual-luciferase reporter gene assay demonstrated that H19 could directly bind to miR-194. miR-194 is one of the tumor suppressor miRNAs [12], which explicitly expressed in the gastrointestinal tract, and was aberrantly expressed within diversi ed malignant neoplasms, including GC [13]. Furthermore, our data showed that H19 was the molecular sponge of miR-194 in regulating the proliferation, migration, and invasion of GC, which is consistent with previous researches indicated that miR-194 was down-regulated in cancer [14,15].
Moreover, the Bioinformatics method and luciferase activity assay were performed to con rm that miR-194 directly target at E2F3. Also, western blot results showed that the miR-194 mimics could signi cantly decrease the protein expression level of E2F3 in the GC cells. E2F3 is one of the key members of the E2F family[16], which has a vital role in cell cycle [17] and controls gene expression related to cell cycle during the G1/S transition [18]. E2F3 may function as an oncogene in GC [19,20]. Previous studies have reported that overexpression of E2F3 is a frequent oncogenic event in human tumorigenesis and that this may be regulated by miRNA [21,22]. E2F3 is located on chromosome 6 NC_000006.12 and has two different subtypes, namely E2F3a and E2F3b [23]. Moreover, genes related to cell invasion, including MMP-2, MMP-3, MMP-9, and MMP-13, were downregulated with knockdown of E2F3 [24]. It has been shown that enhanced expression of MMPs may promote not only microglial in ltration of gliomas but also adhesion and migration of glioma cells [25,26]. In this study, the expression level of E2F3 was found signi cantly overexpressed in GC cell lines, which was consistent with the data of GC tissues in the GEPIA web tool (http://gepia.cancer-pku.cn). Moreover, CCK-8 assay showed decreased proliferative ability induced by miR-194 mimics were reversed by E2F3 overexpression in SGC-7901 cells and AGC cells. Transwell assays showed decreased migratory and invasive ability induced by miR-194 mimics were also reversed by E2F3 overexpression. All of these indicated that E2F3 was the functional targets of H19/miR-194 that affect proliferation, migration, and invasion of GC cells.

Conclusions
In conclusion, the major ndings of this study can be summarized as follows: (1) H19 was signi cantly upregulated in GC tissues and cell lines. (2) H19 promoted GC cell proliferation, migration, and invasion.
(3) H19 was the molecular sponge of miR-194 in regulating the proliferation, migration, and invasion of GC cells. (4) H19/miR-675 modulates gastric cancer progression by targeting E2F3. Thus, our study provides valuable clues for understanding the regulatory network of H19 in carcinogenesis and identifying new therapeutic targets for the treatment of gastric cancer.

Declarations
Ethics approval and consent to participate This study was approved by the Research Ethics Committee in Southwest Hospital, and written informed consent of all patients was obtained.

Consent to publication
Not applicable.

Availability of data and materials
All data generated or analysed during this study are included in this published article.

Competing interests
The authors declare that they have no competing interests.

Funding
This work was supported in part by the National Natural Science Foundation of China (No. 81870883) and a clinical innovation program of Southwest Hospital (No. SWH2016YSCXYB-14). The funder(s) did not participate in the design of this study, collecting, analyzing, and interpreting data and writing manuscripts.
Authors' contributions WXQ was the principal investigator and mainly wrote a manuscript. YS, ZD, YHC, and YPL conducted experiments and data analysis. JC designed the experiments and supervised the manuscript. All authors read and approved the nal manuscript.