Circular RNA ITCH has inhibitory effect on ESCC by suppressing the Wnt/β-catenin pathway.

Circular RNAs with exonic sequences represent a special form of non-coding RNAs, discovered by analyzing a handful of transcribed genes. It has been observed that circular RNAs function as microRNA sponges. In the present study, we investigated whether the expression of circular RNAs is altered during the development of esophageal squamous cell carcinoma (ESCC). Using a TaqMan-based reverse transcriptase polymerase chain reaction assay, the relationship between cir-ITCH and ESCC was analyzed in a total of 684 ESCC and paired adjacent non-tumor tissue samples from eastern and southern China. We found that cir-ITCH expression was usually low in ESCC compared to the peritumoral tissue. The functional relevance of cir-ITCH was further examined by biochemical assays. As sponge of miR-7, miR-17, and miR-214, cir-ITCH might increase the level of ITCH. ITCH hyper expression promotes ubiquitination and degradation of phosphorylated Dvl2, thereby inhibiting the Wnt/β-catenin pathway. These results indicate that cir-ITCH may have an inhibitory effect on ESCC by regulating the Wnt pathway.


INTRODUCTION
Esophageal cancer is the eighth most common cancer worldwide and sixth most common cause of cancer death [1]. One of the main subtypes is esophageal squamous cell carcinoma (ESCC), which is a malignancy that arises from esophageal epithelial cells [2]. Cancer is widely regarded as a genetic disease, and ESCC is no exception, but the molecular and genetic basis of esophageal carcinogenesis remains largely unknown [3,4].
High-throughput RNA sequencing (RNA-Seq), an emerging method to study the RNA regulation mechanism in the whole genome, has been able to detect circular RNA [5]. Circular RNA, in general does not encode protein, but can occur in any genomic region; 85% of circular RNAs are aligned in sense orientation to known protein-coding genes, and they span 1-5 exons [6]. The existence of circular RNA was proposed for several years in early research, for example, the circular testisdetermining gene, SRY [7]. The most well-known circular RNA is CDR1, coding for cerebellar degeneration-related protein 1, which has been observed in all domains of life, but overall, circular RNAs are considered extremely rare in nature [8]. Recently, circular RNAs were proposed to harbor microRNAs (miRNAs), and were found to be enriched with functional miRNA binding sites [6,9]. Mature miRNAs always play an important regulatory role in cell growth, proliferation, differentiation, and cell death. Following database analysis of the study by Memczak et al. on circular RNA, we found that cir-ITCH spanned several exons of the E3 ubiquitin (Ub) protein ligase (ITCH). Moreover, both cir-ITCH and the 3'-untranslated region (UTR) of ITCH shared some miRNAs binding sites, which might suggest its role as a miRNA sponge [6,10]. ITCH belongs to the Nedd4-like E3 family and typically contains 4 WW domains known to associate with PPxY-containing targets [11]. The targets of ITCH are usually associated with tumor formation and chemosensitivity [12]. A study has shown that ITCH degrades the phosphorylated form of disheveled (Dvl) via the proteasome pathway, thus inhibiting the action of the canonical Wnt pathway [13]. The Wnt/β-catenin pathway plays a role in the carcinogenesis of many cancer subtypes including hepatocellular carcinoma, pancreatic cancer, ovarian carcinoma, and ESCC [14][15][16][17]. A previous study has demonstrated that circular RNA has anti-cancer effects in malignant melanoma cell lines [18]. However, there are no reported studies on the functional roles of circular RNA in ESCC.
In this study, we hypothesized that cir-ITCH might influence the expression level of ITCH and may be involved in ESCC development. To address this hypothesis, we conducted this study to delineate any cir-ITCH transcriptional aberrations between ESCC and paired adjacent non-neoplastic tissues obtained from two distinct eastern and southern Chinese centers.

Identification of cir-ITCH
RT-PCR assays were used in this study to verify the circular form of ITCH. We designed two sets of primers for ITCH: a divergent set that was expected to amplify only the circular form, and an opposite-directed set to amplify the linear forms. Our results showed that the circular form was amplified using the divergent primers ( Figure 1A). cDNA and genomic DNA were used as templates, and as expected, no amplification was observed Oncotarget 6003 www.impactjournals.com/oncotarget with the divergent primers on genomic DNA. GAPDH was used as a linear control ( Figure 1A).

Expression of cir-ITCH in ESCC tissues
A TaqMan-based qRT-PCR assay was used for the divergent primer set to determine the levels of ITCH in 358 and 326 paired ESCC samples and matched noncancerous tissues from eastern (Suzhou) and southern Chinese centers (Guangzhou), respectively, cir-ITCH was expressed at a higher level in approximately 70.1% (251/358 patients in the Suzhou cohort) and 71.2% (232/326 patients in the Guangzhou cohort) of the ESCC adjacent tissues compared to matched ESCC tissues ( Figure 1B).

Characterization of cir-ITCH in ESCC cells
To study circular RNA at the cellular level, and to test the cyclization mechanism of RNA, we constructed a vector-based system expressing cir-ITCH. The constructed plasmids were transiently transfected into Eca-109 and TE-1 cells.
Random and oligo-dT primers were used in the reverse transcription experiments. We expected that circular products would be depleted in the poly-(A)enriched samples, in contrast to the linear products. When the oligo-dT primers were used, the relative expression of linear ITCH was significantly higher than that of circular ITCH (5.8-fold in Eca-109 cells and 10-fold in TE-1 cells; Figure 1C) [18].
To confirm further the circular characteristics of cir-ITCH, we used the enzyme RNase R, a highly processive 3' to 5' exoribonuclease that does not act on circular RNAs [19,20]. As expected, in contrast to the linear control RNAs (10-fold in Eca-109 cells and 11-fold in TE-1 cells), the circular RNA was resistant to RNase R treatment ( Figure 1D).

miRanda
(http://www.microrna.org/) and TargetScan/TargetScanS (http://www.targetscan.org/) software were used to predict the binding sites for human microRNA within cir-ITCH and the 3'-UTR region of ITCH. The results of both showed that miR-216b, miR-17, miR-214, miR-7, and miR-128 could bind to the 3'-UTR of ITCH and cir-ITCH. The sequence of the predicted 5 microRNAs binding sites were presented in Supplementary Table 1. The 5 miRNAs and one of the reporter constructs were transiently co-transfected into ESCC cells, and subsequently, luciferase activity was assessed. The results of the empty vector construct and miR-17 showed that luciferase activity was significantly reduced in a concentration-dependent manner in Eca-109 cells (Figure 2A), and this trend was also present with mimics of miR-7, miR-214, miR-216b, and miR-128 in the control cells ( We next investigated cir-ITCH stability. Eca-109 and TE-1 cells transfected with the plasmid construct were treated with actinomycin D, a transcription inhibitor, in the presence of 1 or 40 pmol of miR-17, miR-7, and miR-214, and total RNA was harvested at indicated time points. There was little change in cir-ITCH levels in Eca-109 cells following incubation with actinomycin D for 1-3 hours, while the associated control remained level at 20-30%; the former cir-ITCH levels were significantly higher than the latter (P < 0.01). We repeated these experiments in TE-1 cells with the same results ( Figure 2B).

Correlation analysis of cir-ITCH and ITCH in ESCC
We evaluated the expression of cir-ITCH in a randomly selected cohort of 30 pairs of ESCC adjuvant non-cancerous tissues from Suzhou. The results showed that patients with higher cir-ITCH expression levels in ESCC tissues had a substantial up-regulation of linear ITCH. Furthermore, we tested the mRNA expression www.impactjournals.com/oncotarget levels of linear ITCH in constructed cir-ITCH cells with stable high expression, and the results showed that the linear ITCH positively correlated with levels of cir-ITCH (R 2 = 0.32, P < 0.01; Figure 3A).

cir-ITCH is involved in the regulation of Wnt Signaling in vivo
Previous studies using the yeast two-hybrid system and co-immunoprecipitation assays have shown that ITCH is a novel Dvl-interacting protein. The ITCH protein can ubiquitinate the phosphorylated form of Dvl and promote its degradation, thereby inhibiting canonical Wnt signaling [21]. To confirm whether cir-ITCH regulates the Wnt/β-catenin signaling pathway in ESCC cells, we used a β-catenin/T-cell factor (TCF)-responsive luciferase reporter assay. In Eca-109 cells, we used an anti-Dvl2 antibody to test endogenous Dvl2 levels. Transient ITCH hyperexpression reduced the intensity of Dvl2, but the effect was not so obvious in the control cells. A western blot analysis was performed to determine β-catenin levels in cells with ITCH hyperexpression, and as shown in Figure 3B, there was an obvious decrease in β-catenin levels. We overexpressed ITCH in Eca-109 cells and evaluated its effects using the TOP-Flash reporter system. As shown in Figure 3C, overexpression of ITCH significantly suppressed relative TCF transcriptional activity. Then, we investigated the effect of ITCH on the expression of endogenous Wnt target genes like c-Myc, in cells transfected with cir-ITCH, and discovered that the Wnt target genes were further suppressed ( Figure 3D).

cir-ITCH regulates the cell cycle of ESCC cells
We investigated the effects of cir-ITCH on cell cycle using fluorescence-activated cell sorting (FACS) analysis of PI-stained ESCC cells. The representative results of the cell cycle distribution in empty vector and cir-ITCH transfected Eca-109, and TE-1 cells in the presence of miR-17 are shown in Figure 4A. It is evident that overexpression of cir-ITCH resulted in a statistically significant accumulation of ESCC cells in the G1 phase (Eca-109 cells: 14.4% increase, P = 0.007; TE-1 cells: 9.6% increase, P = 0.007) accompanied by a decrease
Consistently, colony formation assays performed at 2 weeks showed that cells with cir-ITCH-overexpression had reduced colony-forming ability compared to the control groups (Eca-109 cells: 31.25% decrease; TE-1 cells: 32.25% decrease; Figure 4C). Thus, both assays show that cir-ITCH modulates cell proliferation.

cir-ITCH suppresses tumor growth
As shown in Figure 4D, Eca-109 and TE-1 cells with up-regulated cir-ITCH (Eca-109-cir-ITCH and TE-1-cir-ITCH cells) were subcutaneously injected into the back flank of nude mice.

DISCUSSION
Through a series of functional experiments, we identified a cir-ITCH circular RNA whose role had not yet been elucidated. Here, we demonstrated that cir-ITCH acts as a miRNA sponge, increases the level of ITCH, and provokes ubiquitin-mediated Dvl2 degradation, which inhibits canonical Wnt signaling. These results illustrate the basic interaction between circular RNA, miRNA, and protein in cells, and alteration of this fine regulation may contribute to cancer initiation and progression.
Cir-ITCH is located on chromosome 20q11.22 on the plus strand, is aligned in a sense orientation to the known protein-coding gene ITCH, and spans exons 6-13. ITCH is a member of the E3 ubiquitin ligases that regulate protein stability and immunological responses, as well as cancer progression [22]. ITCH was initially identified when disruption of the ITCH gene induced a fatal autoimmune inflammatory condition [23]. Latterly, ITCH was found to be crucial in the control of proteasome degradation of several important substrates such as p63, p73, Notch1, and Dvl2 [21,[24][25][26]; all of which are usually associated with tumor formation and chemosensitivity.
The protein-coding function of messenger RNAs can be suppressed by the binding of short miRNA sequences, or miRNAs can guide the effector protein Argonaute to the mRNAs of coding genes to repress their protein production [27][28][29]. Previous experiments using ectopic expression of miRNA indicated that it could bind to the 3'-UTR of ITCH to decrease its expression [30]. Memczak et al. (2013) and Hansen et al. (2011) proposed that CDR1, a circular RNA, harbored a miR-7 binding site. This striking feature suggested that circular RNA possibly functioned as a miRNA sponge [6,9,31]. In our study, we found that the cir-ITCH harbors many miRNA binding sites that could bind to the 3'-UTR of ITCH, including those for miR-7, miR-17, and miR-214. Furthermore, these miRNAs have often been associated with cancer. Elevated miR-7 expression has been described in a variety of tumor types including ESCC and has been implicated in oncogenesis, classification, and cancer progression [32,33]. miR-214 is aberrantly expressed in several human tumors, for example, it is upregulated in ovarian cancer and gastric cancer [34,35]. miR-216b is a tumor suppressor in nasopharyngeal cancer [36]. miR-17, which is frequently up-regulated in ESCC, has been widely studied, and it is a potential prognostic biomarker for ESCC [37]. Collectively, cir-ITCH is involved in regulating miRNAs that are associated with cancer.
cir-ITCH acts as a miRNA sponge and increases the level of ITCH. ITCH plays an important role in the Wnt/β-catenin pathway and its activation is thought to contribute to the development of some human cancers [38,39]. Initially, it was reported that activated Wnt/β-catenin signaling was defined by β-catenin nuclear expression and associated with breast cancer [40]. Recent studies have identified that Wnt/β-catenin signaling activation is found in hepatocellular carcinoma [16]. The Wnt/β-catenin pathway has been very well described, as it is important in cancer research. It is mediated by Dvl2, which functions as an essential scaffold protein, bridging the receptors and downstream signaling components [41,42]. Phosphorylation of Dvl2 is required for Wnt signaling, and is crucial to regulate its stability and activity for proper signal transduction. According to previous research, ITCH could promote the ubiquitination and degradation of phosphorylated Dvl2, and therefore, inhibit canonical Wnt signaling [21]. Our data demonstrated that overexpressed cir-ITCH was able to suppress phosphorylated Dvl2 and inhibit Wnt/β-catenin signaling in ESCC cells. The β-catenin/TCF-responsive luciferase reporter assay was used to examine whether a single gene regulates the Wnt/ β-catenin signaling pathway. In our study, overexpression of cir-ITCH significantly suppressed relative TCF transcriptional activity.
The oncogene c-Myc is a target gene in canonical Wnt signaling; it is frequently overexpressed in many cancers and has a crucial role in cell growth, reproduction, apoptosis, and differentiation [43]. Knockdown of c-Myc leads to cell-cycle arrest and limits malignant cancer cell growth. Our cell-cycle results showed the same trend and indicated that cir-ITCH affects the expression of c-Myc. Therefore, we suggest that cir-ITCH may have an antitumor function in ESCC and suppress the canonical Wnt pathway. Based on these findings, we are able to conclude that cir-ITCH has an antitumor role by controlling miRNA activity, which increases the concentration of ITCH, and results in suppression of the canonical Wnt pathway by degradation of phosphorylated Dvl2, and inhibition of c-Myc expression, which may prevent oncogenesis ( Figure 5).
In summary, the current study represents an analysis of 2-stage clinical ESCC samples for the circular RNA, cir-ITCH. The results show that there is low expression of cir-ITCH in ESCC tumor tissues, and that cir-ITCH reduces cell viability, and arrests proliferation in ESCC cells. The inhibition effects of cir-ITCH may be related to its cooperation with Dvl2 to suppress the Wnt/β-catenin pathway. Overall, our data support the assumption that cir-ITCH acts as a miRNA sponge and increases expression of the miRNA target gene ITCH. The availability of circularizing the miRNA sponge in cells is a candidate for new strategies for RNA-based cancer diagnosis and therapy.

Cell culture and animals
Human esophageal carcinoma cancer cell lines Eca-109 and TE-1 were purchased from the Cell Bank of Type Culture Collection of the Chinese Academy of Sciences, Shanghai Institute of Cell Biology, and were passaged for less than 6 months. Cells were maintained in RPMI-1640 medium (Gibco BRL, Gaithersburg, MD), and supplemented with 10% heat-inactivated fetal bovine serum (Gibco BRL), 2 mM L-glutamine, 100 U/ mL penicillin, and 100 U/mL streptomycin. Culture plates were then incubated at 37°C in an atmosphere of 5% CO 2 .
Female BALB/c nude mice that were 4-5 weeks old were purchased from the Shanghai Laboratory Animal Center at the Chinese Academy of Sciences (Shanghai, China). All mouse experiments were performed according to the Laboratory Animal Center of Soochow University's guidance.

Circular RNA plasmid construction
We constructed the circular RNA plasmid used in this study. Human cir-ITCH cDNA was synthesized by GeneWiz (Suzhou, China) and cloned into pLVX-IRESneo (Clontech Laboratories Inc., San Francisco, CA, USA). We used a 996 bp DNA fragment corresponding to exons 6-13 of the ITCH gene, and added 1 kb upstream and 200 bp downstream to the nonlinear splice sites. Also, an 800-bp DNA stretch was added upstream of the splice acceptor site and inserted downstream in the reverse orientation [9]. The resulting construct (pLVX-cir-ITCH) was verified by direct sequencing.

Lentiviral production and transduction
The cDNA sequence of cir-ITCH was synthesized by the Genewiz Company (Suzhou, China) and then cloned into the lentiviral expression vector pLVX-IRES-neo (Clontech Laboratories Inc., San Francisco, CA, USA). Lentiviral production and transduction were conducted by following previously published procedures [44].
Oncotarget 6010 www.impactjournals.com/oncotarget RNA extraction and real-time quantitative polymerase chain reaction Total RNA was isolated from cells and tissues using the TRIzol® reagent (Invitrogen) according to the manufacturer's instructions. Esophageal tissue samples from newly diagnosed ESCC patients were immediately placed in liquid nitrogen and then stored at -80°C before analysis. cDNA was synthesized from total RNA using SuperScript III® (Invitrogen) according to the supplied protocol. The relative gene expression of cir-ITCH was determined using the ABI Prism 7500 sequence detection system (Applied Biosystems, Foster City, CA, USA), which is based on the TaqMan method. GAPDH was used as an internal standard control, and all reactions were performed in triplicate [45]. The primers used for polymerase chain reaction (PCR) amplification are listed in Table 2.

RNase R digestion
The RNase R digestion reaction was performed following previously published procedures. The digestion and precipitation reactions were repeated twice with a ratio of 3 U enzyme/1 mg RNA [8].

Transient transfections and luciferase assays
Eca-109 and TE-1 cells were seeded in 24-well plates (1 × 10 5 cells per well) and cultured to 60-70% confluence before transfection; then, cells were transfected with 800 ng of the reporter plasmids described above using Lipofectamine 2000 (Invitrogen, CA, USA). Cells were co-transfected with 0, 1 or 40 pmol of miR-216b, miR-17, miR-214, miR-7, and miR-128 mimics according to the manufacturer's instructions [46]. cir-ITCH hyperexpression in cells was evaluated with the same test. Each group included 6 replicates, and independent triplicate experiments were performed. After transfection for 24 h, the cells were collected using 100 μL passive buffer, and Renilla luciferase activity was detected using the Dual-Luciferase Reporter Assay System (Promega), and a TD-20/20 illuminometer (Turner Biosystems, Sunnyvale, CA); the results were normalized against the activity of the Renilla luciferase gene.

Actinomycin D assay
TE-1 and Eca-109 cells were seeded at 5 × 10 4 cells per well in 10 mm 24-multiwell plates. Sixteen hours later, the cells were transiently transfected using Lipofectamine 2000 (Invitrogen) and co-transfected with 1 or 40 pmol of miRNA mimics (Ambion) with or without 40 pmol of miRNA inhibitor, as indicated, for 24 h; then the cells were exposed to actinomycin D (Sigma, St Louis, MO) for 1, 2, and 3 h. The cells were harvested and the stability of the cir-ITCH mRNA was analyzed using quantitative reverse transcriptase PCR (qRT-PCR). Actinomycin D was used at concentrations of 2 mg/L.

Western blotting
Protein lysates from ESCC tissues and cells were subjected to western blot analysis according to standard protocols as previously described [47]. Antibodies recognizing ITCH, Dvl2 and β-Actin were from Cell Signaling. Antibodies recognizing β-catenin and Wnt3a were from Abcam.

Cell viability assay
In 96-well, flat-bottomed plates (BD Biosciences, Bedford, MA), 100 μL of Eca-109 and TE-1 cell suspensions (10,000 cells per mL) were aliquoted into each well. After transfection, as described previously for the actinomycin D assay, and 1, 2, and 3 days of cultivation, cell viability was measured by the Cell Counting Kit-8 (CCK-8) system (Dojindo Laboratory, Kumamoto, Japan) according to the manufacturer's instructions. Briefly, 10 μL of CCK-8 solution was added to each well, the plates were incubated at 37°C for 1 h, and the absorbance of each well was read at 450 nm using a microplate reader (MRX; Dynex Technologies, West Sussex, United Kingdom). There were 6 replicates for each group, and the experiments were repeated at least 3 times.

Colony formation assay
ESCC cells with stable cir-ITCH overexpression and controls (80 cells) were seeded into a 60-mm plate. After incubation for 2 weeks at 37°C in a 5% humidified CO 2 atmosphere , colonies (>50 cells per colony) were stained with Giemsa, counted, and photographed. Each experiment was performed in triplicate.

Cell cycle analysis
For cell cycle analysis, ESCC cells with stable cir-ITCH overexpression were stained with 0.5 mL propidium iodide (PI) staining solution, and cellular DNA content was analyzed by flow cytometry (FACScalibur; BD Biosciences).