Long non-coding RNA H19 regulates E2F1 expression by competitively sponging endogenous miR-29a-3p in clear cell renal cell carcinoma

Background Numerous recent studies indicate that the long non-coding RNAs (lncRNAs) are frequently abnormal expressed and take critical roles in many cancers. Renal cell carcinoma is the secondary malignant tumors in the urinary system and has high mortality and morbidity. Around 80% of RCCs is clear cell renal cell carcinoma (ccRCC) and is characterized by high metastasis and relapse rate. However, the clinical significances of lncRNAs in ccRCC are still unknown. Methods The human cancer lncRNA PCR array (Yingbio) was performed to detect the differentially expressed lncRNAs in human ccRCC samples. Real-time PCR (RT-PCR), dual-luciferase assay, RNA binding protein immunoprecipitation (RIP) assay, transwell assay, CCK-8 assay, and western blot were performed to explore the molecular mechanism of lncRNAs in ccRCC cell migration and invasion. Results In this study, lncRNA-H19 was high expressed and negatively correlated with miR-29a-3p in ccRCC. By bioinformatics software, dual-luciferase reporter and RIP assays, we verified that miR-29a-3p was identified as a direct target of lncRNA-H19. RT-PCR and western blot demonstrated that down-regulated lncRNA-H19 could affect the expression of miR-29a-3p targeting E2F1 with competitively binding miR-29a-3p. Furthermore, transwell assays indicated that lncRNA-H19 knockdown inhibited cells migration and invasion, but this effect was attenuated by co-transfection of lncRNA-H19 siRNA and miR-29a-3p inhibitor. Over expression of E2F1 could rescue lncRNA-H19 siRNA induced suppression on cell migration and invasion in ccRCC cells. Conclusions These results show a possible competing endogenous RNAs regulatory network involving lncRNA-H19 regulates E2F1 expression by competitively sponging endogenous miR-29a-3p in ccRCC. This mechanism may contribute to a better understanding of ccRCC pathogenesis, and lncRNA-H19 may be further considered as a potential therapeutic target for ccRCC intervention.

MicroRNAs (miRNAs) are endogenous non-coding RNAs and regulate gene expression by mRNA degradation and translational repression at the post-transcriptional level [36]. Several studies have found that lncRNAs functions as competing endogenous RNAs (ceRNAs) to sponge miRNAs, affecting expression of miRNA targets [37,38]. However, lncRNA-H19 whether functions as ceRNA to regulate expression of targets with binding miRNA has not been reported in ccRCC.
In this study, we hypothesized that lncRNA-H19 might promote ccRCC cells migration and invasion through inhibiting the expression of miR-29a-3p. In this study, we first detected the differentially expressed lncRNAs in human ccRCC samples by the human cancer LncRNA PCR array (Yingbio), and then measured the expression of lncRNA-H19 and miR-29a-3p in tumor tissues from ccRCC patients. Furthermore, the underlying mechanism of lncRNA-H19 in the development of ccRCC was analyzed in vitro. This study might provide a better understanding of ccRCC pathogenesis and a potential therapeutic target for ccRCC intervention.

Ethics statement
This study was conducted based on our protocols approved by the Ethical Committee of the Jinling Hospital of Nanjing University Medical School. All patients signed written informed consent documents prior to this study.

Clinical specimens
Thirty ccRCC tissues and their pair-matched normal cancerous tissues for this study were obtained from the patients diagnosed with ccRCC in Department of Urology, Jinling Hospital (Nanjing, China), between March 2011 and September 2015. Written informed consent was obtained from all patients. All tissue was snap-frozen in liquid nitrogen and stored at − 80 °C until use.

Cell lines, antibodies, inhibitor and plasmid
The human ccRCC cell line (786-O) was purchased from American Type Culture Collection (ATCC). All cells were maintained in RPMI 1640 medium (Genepharma, Shanghai, China) with 10% fetal bovine serum (Invitrogen) at 37 °C with 5% CO 2 [4]. The primary antibodies anti-E2F1, anti-GAPDH and the second antibodies anti-rabbit or anti-mouse IgG were purchased from Santa Cruz Biotechnology (Dallas, TX, USA). The powdery inhibitor of miR-29a-3p and negative control was bought from Genepharma (Shanghai, China). For E2F1 over expression vector, the E2F1 coding sequences were amplified and inserted into the pcDNA-3.1 vector (Invitrogen) using the XhoI and EcoRI restriction sites [39].

Human cancer LncRNA PCR array
Total RNA from tissues was extracted by using Trizol reagent (Invitrogen, Carlsbad, CA, USA) following its manufacturer's specification. Six purified RNA samples were then sent to Yingbio (Shanghai, China) for Human Cancer LncRNA PCR Array. The array contained 6 reference genes and 84 lncRNAs that were associated with cancer. The lncRNAs were collected from the most authoritative databases, including Ensembl, manually curated lncRNA literature sources, RefSeq and UCSC known genes. The detailed processes were: firstly, 1 μg of total RNA was converted to cDNA with RT2 First Strand Kit (Qiagen, 330401) after assessing integrity. Secondly, cDNA was diluted to a total of 100 μL with ddH 2 O, and 1 μL of this diluent was used for each primer set in RT2 lncRNA PCR Array (Yingbio, Shanghai, China). Thirdly, RT-PCR for PCR assay was performed using the RT2 SYBR Green Mastermix (Qiagen, 330401). Only one replicate for each sample and one pair of primer for each LncRNA were performed for Human Cancer LncRNA PCR Array. The parameters for RT-PCR detection: 95 °C, 10 min; 95 °C, 15 s, 60 °C, 1 min (40 cycles). LncRNAs expression was compared according to the CT value, and data were processed using 2 −ΔΔCT method. Differentially expressed lncRNAs with statistical significance (as determined by two-tailed Student's t test < 0.05) were identified through volcano plot filtering.

Western blot
Whole cell extracts were completed by being lysed in RIPA lysis buffer with protease inhibitor (Promega). The protein concentration of each sample was quantified with BCA assay kit (Pierce Biotechnology, Inc.). Each sample was run on 15% SDS-PAGE gel at the same protein level, and then transferred to PVDF membranes (Millipore). After blocking with a 5% no-fatty milk solution in TBS with 0.1% Tween20, the membranes were incubated for 12 h at 4 °C with the antibodies at a suitable dilution (1:1000). The membranes were then incubated with HRPconjugated goat-anti-mouse or goat-anti-rabbit IgG as secondary antibodies for 2 h. The signals were recorded with ECL reagents (Millipore ECL plus kit).

Construction of reporter plasmids and luciferase assays
The method for constructing reporter plasmids has been published elsewhere [40]. HEK293T cells were transfected with the reporter plasmids with Lipofectamine 2000 (Invitrogen, CA, USA). Use of the dual-luciferase reporter assay system (Promega, Madison, WI, USA) to measure luciferase activity has been published elsewhere [41].

RNA binding protein immunoprecipitation (RIP) assay
According with the manufacturer's protocol, RIP assay was performed using the EZ-Magna RIP kit (Millipore, Billerica, MA, USA). 786-O with 85% confluence was scraped off and lysed with the RIP lysis buffer. Whole cell extract was incubated with the RIP buffer containing magnetic beads coated with antihuman argonaute2 (Ago2) antibodies (Millipore), and IgG (Millipore) was used as a negative control (input) [37]. After incubation at 4 °C for 2 h, RT-PCR was performed to detect the enrichment of lncRNA-H19 and miR-29a-3p.

Transwell assay
786-O was removed and suspended in RPMI-1640 medium, after which 1 × 10 5 cells were added into transwell inserts with 8 mm per size for 24-well plates, and the lower chamber added with 200 mL medium supplement. With using a cotton swab, the non-migrated cells of upper chamber was removed after migration, and the cotton filters were fixed by used of 4% paraformaldehyde. The ability of cell migratory was measured with hematoxylin and eosin staining [42].

Cell proliferation (CCK8) assay
Cell proliferation was measured by using a CCK-8 kit (Dojundo, Kumamoto-ken, Japan). 2 × 10 3 786-O cells/ well was seeded in 96-well plates 24 h before experiencing. After respective treatments, the wells received 100 µL culture medium and 10 µL reaction mixture from CCK8 kit. The plates were incubated for 2 h at 37 °C. Then, cell proliferation was estimated by measuring absorbance at 450 nm [43]. All experiments were repeated three times.

Statistical analysis
The statistical analyses were performed using SPSS 19.0 (SPSS Inc., Chicago, IL, USA). Data was presented as mean ± standard deviation (SD) from three independent experiments with each measured in triplicate. The expression differences between ccRCC tissues and adjacent non-tumor tissues were analyzed with paired Student's t test. A value of P < 0.05 was considered to be a statistically significant difference.

LncRNA-H19 is over expressed in ccRCC tissues and ccRCC cells
To explore for important lncRNAs involved in ccRCC, the lncRNAs expression of ccRCC tissues and their pair-matched noncancerous tissues from three ccRCC patients were detected using the Human Cancer LncRNA Array. A total of 84 lncRNAs were detected, among these, ccRCC tissues contained 4 up expressed, 77 equally expressed, and 3 down expressed lncRNAs compared to their pair-matched noncancerous tissues (Fig. 1a). Compared with noncancerous tissues (Fig. 1a), the expression level of lncRNA-H19 was significantly increased in ccRCC tissues (fold change ≥ 2.0, P ≤ 0.05). Besides, the expression of lncRNA-H19 in 30 paired samples of ccRCC tissues and their corresponding non-carcinoma tissues were determined by RT-PCR. As shown in Fig. 1b, the lncRNA-H19 expression was obviously increased in ccRCC tissues compared to their corresponding normal tissues (P < 0.05). Higher expression levels of lncRNA-H19 were detected in three ccRCC cell lines (768-O, ACHN and Caki-1) than HK-2 cell line (Fig. 1c, P < 0.05).
In the cytoplasm, miRNA is one form of miRNA ribonucleoprotein complexes (miRNPs) that also typically contain Ago2 [44,45]. To test whether lncRNA-H19 associates with miRNPs, RIP assay was performed on 786-O using antibodies against Ago2. As shown in Fig. 3c, lncRNA-H19 was preferentially enriched in Ago2-containing miRNPs relative to control IgG immunoprecipitates. Similarly, miR-29a-3p was detected at a high level greater than that of control anti-IgG. Thus, lncRNA-H19 was present in Ago2-containing miRNPs through association with miRNA-29a-3p, consistent with our bioinformatic analysis and luciferase assays.

LncRNA-H19 modulated expression of endogenous miR-29a-3p targets E2F1
E2F1 was a well known oncogene that is targeted by miR-29a-3p in osteosarcoma [46]. We next measured whether lncRNA-H19 modulated E2F1 expression by targeting miR-29a-3p in 786-O cells. As present in Fig. 4a, b, compared with control group, E2F1 expression at the mRNA and protein levels were down-regulated after lncRNA-H19 silencing and up-regulated by transfecting with miR-29a-3p inhibitors in 786-O cells. However, the Fig. 2 Identification of potential lncRNA-H19 targeting miRNAs. a A cohort of six potential miRNAs that could interact with lncRNA-H19 was predicted by starbase 2.0 and Miranda (http://starbase.sysu.edu.cn/ and http://www.microrna.org/). b RT-PCR was performed to confirm the interaction between lncRNA-H19 and the six potential miRNAs. c Dual-luciferase assay was performed to confirm the interaction between lncRNA-H19 and the six potential miRNAs. d Pearson's correlation was used for correlation analysis between the expression of lncRNA-H19 mRNA and miR-29a-3p mRNA. e Pearson's correlation was used for correlation analysis between the expression of lncRNA-H19 mRNA and E2F1 mRNA Fig. 3 MiR-29a-3p was one target of lncRNA-H19. a Representation the lncRNA-H19 and miR-29a-3p binding site by miRanda (http://www.microrna.org/). b Dual-luciferase assay was performed to confirm the interaction between lncRNA-H19 and miR-29a-3p. c LncRNA-H19 mRNA level was detected in the substrate of RIP assay by RT-PCR. All data were represented as the mean ± SD from three independent experiments, *P < 0.05 modulating effects of lncRNA-H19 on E2F1 were diminished in the group of co-transfecting with miR-29a-3p inhibitor and siRNA-H19. E2F1 plays a vital role in cell proliferation, it is unclear whether lncRNA H19 has effects on proliferation of clear cell renal cell carcinoma cells. As present in Fig. 5, 786-O cells were plated in a 24-well plate and transfected with si-H19, and cell proliferation was significantly reduced at the different indicated time points using a CCK-8 assay (P < 0.05).

E2F1 promoted migration and invasion in ccRCC cells
To investigate the functional roles of E2F1 on ccRCC cells, we measured the expression levels of E2F1 in ccRCC tissues and their matched normal tissues through RT-PCR. Compared with normal tissues, we found that E2F1 was up-regulated in ccRCC tissues (Fig. 6a). Compared to control group, the number of migration and invasive cells was markedly decreased after E2F1 silencing (Fig. 6c, d).

LncRNA-H19/miR-29a-3p/E2F1 axis on cell migration and invasion in ccRCC cells
We continued to explore the effects of lncRNA-H19/ miR-29a-3p/E2F1 axis on migration and invasion in ccRCC cells. Initially, compared to control group, knockdown of lncRNA-H19 could decrease the cell number of migration and invasion in 786-O, but was increased by co-transfected with miR-29a-3p inhibitor and siRNA-H19 (Fig. 7a-d). Moreover, we continued to explore that whether over expression of E2F1 could rescue lncRNA-H19 siRNA induced suppression on cell migration and invasion in ccRCC cells. As shown in Fig. 8, compared to control group, knockdown of lncRNA-H19 could decrease the cell number of migration and invasion in 786-O, but was increased by co-transfected with over expression of E2F1 and siRNA-H19 (Fig. 8a-d).

Discussion
Clear cell renal cell carcinoma is one of the most important human urological diseases in developed and in developing countries. At present, ccRCC patients have a limited option of clinical therapeutic and poor prognosis. Therefore, it is very urgent to study the biological characteristics of ccRCC and find novel molecular markers for cancer prevention and prognosis. In this study, we identified a popular lncRNA-H19 which is over expressed in ccRCC. Our results showed that lncRNA-H19 promotes ccRCC migration and invasion by affecting E2F1 expression with competitively sponging endogenous miR-29a-3p.
LncRNAs are important players in cancer development and emerging in various fundamental biological processes [31][32][33]. LncRNA HOTAIR is highly expressed in gallbladder cancer tissues and leads to tumor metastases [34,35]. Linc-POU3F3 is up-regulation in esophageal squamous cell carcinoma samples and promotes tumor development [36]. LncRNA CCAL can increase colorectal cancer cell progression with activating Wnt/βcatenin pathway [38]. In recent studies, lncRNA-H19 revealed two opposite function according to tumor types: one effect of lncRNA-H19 is as an oncogene in breast cancer [37], colorectal cancer [27], glioblastoma [38], and ovarian cancer [36]; the other one effect of lncRNA-H19 is as a tumor suppressor in hepatocellular carcinoma [39], nephroblastoma [41] and prostate cancer [40]. In this present study, we found that lncRNA-H19 functions as an oncogene in ccRCC and promoted cell migration and invasion by up regulating the E2F1 expression in ccRCC cells. Based on this result, our further revealed that lncRNA-H19/miR-29a-3p/E2F1 might be a possible ceRNA regulatory network in ccRCC.
E2F1 is the abbreviation of E2F transcription factor 1, and it encodes the protein which plays a crucial regulation role in controlling cell cycle progression and activating tumor suppressor proteins [47,48]. Altered expression of E2F1 has been reported in many types of human cancer in conjunction with worse patient survival [49][50][51]. E2F1 can be used as an effective inducer of cancerous tumor through directly activating transcription of MDM2 and subsequent stimulation of p53-dependent manner [52]. In this study, our results indicated that down regulation of E2F1 might inhibit migration and invasion in ccRCC cells, and the up regulation of lncRNA-H19 was sufficient to increase the expression of E2F1and promote ccRCC migration and invasion. E2F1 plays a vital role in cell proliferation, it is unclear whether lncRNA H19 has effects on proliferation of clear cell renal cell carcinoma cells. In this study, 786-O cells were plated in a 24-well plate, transfected with si-H19 and measured cell proliferation by CCK-8 assay. Our The protein expression of E2F1 after knockdown was detected in ccRCC cells by western blot. c, d Cells migration and invasion ability and cells number were detected after E2F1 knockdown in ccRCC cells (P < 0.05). All data were represented as the mean ± SD from three independent experiments, **P < 0.01 result was consistent with other study [31] and cell proliferation of clear cell renal cell carcinoma cells was significantly reduced at the different indicated time points with si-H19.
It is worth noting that the relationship between lncRNA-H19 and E2F1 depends on miR-29a-3p. Our results proposed that lncRNA-H19 functioned as a ceRNA for miR-29a-3p to regulate E2F1 epigenetically. At the post-transcriptional regulation, lncRNA-H19 could encode the placental specific expression of miR-675 in gestational time point with targeting insulin-like growth factor 1 receptor [53]. Another studies indicated that lncRNA-H19 might work on TGF-β signaling [54] and tumor suppressor RB [55] by miR-675. We found that silenced the expression of lncRNA-H19 could decrease the relative luciferase of miR-29a-3p (Fig. 2c), and the expression of lncRNA-H19 was significant negatively correlation with miR-29a-3p in ccRCC tissues (Fig. 2d). The expression of lncRNA-H19 was not changed by miR-29a-3p mimic (Fig. 3b), which indicated that lncRNA-H19 was the upstream of miR-29a-3p. Our results suggests a potential regulation pathway that involving lncRNA H19 in both molecular and biological aspects in ccRCC.