Circular RNA circ_0000423 promotes gastric cancer cell proliferation, migration and invasion via the microR-582-3p/Disheveled-Axin domain containing 1 axis

ABSTRACT For humans, gastric cancer (GC) is a common malignancy. Multiple circular RNAs (circRNAs) have been confirmed to be important cancer-promoting or tumor-suppressive factors. The present study discusses the roles and mechanisms of circ_0000423 in GC development. In this study, circ_0000423 expression in GC patient tissue samples and cell lines was detected via quantitative real-time polymerase chain reaction. Disheveled-Axin domain containing 1 (DIXDC1) expression in GC cells was examined via Western blot. Besides, cell counting kit-8 was utilized for detecting GC cell viability. GC cell migration and invasion were examined through Transwell assays. Bioinformatics and dual-luciferase reporter gene assays were employed to verify the regulatory relationships between microRNA-582-3p (miR-582-3p) and circ_0000423 or DIXDC1. In the present study, we demonstrated that circ_0000423 was highly expressed in GC. Circ_0000423 knockdown suppressed GC cell viability, migration and invasion. Moreover, miR-582-3p was confirmed as a direct target of circ_0000423, and an upstream regulator of DIXDC1. MiR-582-3p inhibition or DIXDC1 overexpression could reverse the above-mentioned effects of knocking down circ_0000423 on GC cells. In conclusion, circ_0000423 facilitates GC progression by modulating the miR-582-3p/DIXDC1 axis.


Introduction
Gastric cancer (GC), as the fifth most common carcinoma worldwide, ranks third among the causes of cancer-related death [1]. It is estimated that more than 1 million patients are diagnosed with GC every year, and half of them are from China [2,3]. GC is associated with multiple pathological factors, and both environmental and genetic factors have a role in its etiology; despite dramatic advances in surgery and adjuvant therapy for GC, since over 80% of GC patients are diagnosed with advanced cancer, the mortality of GC still is very high: the 5-year overall survival rate is less than 30% due to frequent tumor recurrence and metastasis [4][5][6]. In this context, seeking new diagnostic biomarkers and therapy targets for GC becomes urgent.
Non-coding RNA (ncRNA) has become a hot spot in cancer research. As a newly discovered class of ncRNA, circular RNAs (circRNAs) are widely expressed in eukaryotes; they are formed by non-canonical splicing and have a covalently closed loop with neither 3ʹ poly(A) tail nor 5ʹ end [7,8]. The special structure of circRNAs makes it relatively resistant to degradation mediated by exonuclease [8,9]. CircRNAs are strongly linked to the development of various diseases, especially cancer [10]. For instance, circ_0007294 expression is markedly increased in triple-negative breast carcinoma, and its high expression is strongly associated with lymph node metastasis and advanced clinical stage [11]; circ_0101432 inhibits hepatocellular cancer cell apoptosis through targeting miR-622 and miR-1258, and up-regulating mitogen-activated protein kinase 1 expression [12]. In the present study, bioinformatics analysis suggested that circ_0000423 was highly expressed in GC tissues. Circ_0000423 is formed by reverse splicing of protein phosphatase 1 regulatory subunit 12A (PPP1R12A) mRNA. Reportedly, circ_0000423 expression is up-regulated in colon carcinoma tissues, and high circ_0000423 expression is strongly related to the short survival time of patients [13]. Nonetheless, the specific role of circ_0000423 in GC is still not fully elucidated, and further exploration is needed.
We hypothesized that circ_0000423 played an important role in GC progression. The current study was aimed at exploring the role and mechanism of circ_0000423 in GC, providing a theoretical basis for clarifying the mechanism of the progression of this deadly disease.

Clinical samples and ethics statement
A total of 43 GC patients (38-72 years old, 28 males and 15 females) who received gastrectomy at the First Affiliated Hospital of Zhengzhou University from 2016 to 2019 were enrolled in the study. Based on tumor, lymph node metastasis and tumor node metastasis (TNM) grading according to the International Union Against Cancer (7th edition), there were 14 cases at stage I, 10 cases at stage II, 19 cases at stage III. The tumorous and adjacent nontumorous tissues (5 cm away from the tumor) were surgically removed and frozen with liquid nitrogen and then kept at −80°C. None of the participants had undergone chemotherapy or radiotherapy before tissue collection. The study was endorsed by the Ethics Committee of First Affiliated Hospital of Zhengzhou University (approval no. 201606032), and we obtained the written informed consent of each patient. The present study was carried out following the Declaration of Helsinki. Gibco, Grand Island, NY, USA) containing 100 U/ ml penicillin, 0.1 mg/ml streptomycin and 10% fetal bovine serum (FBS, Gibco, Grand Island, NY, USA) in 5% CO 2 at 37°C.

CircRNA expression profile analysis
From the Gene Expression Omnibus (GEO) (http://www.ncbi.nlm.nih.gov/geo) database, we downloaded three datasets (GSE78092, GSE89143, and GSE141977) with GC circRNA expression profiles. To screen out differentially expressed circRNAs, the GEO2R online software was utilized to analyze them.

Quantitative real-time polymerase chain reaction (qRT-PCR) and RNase R treatment
A TRIzol kit (Invitrogen, Carlsbad, CA, USA) was utilized for extracting the total RNA, and the PrimeScript™ RT Reagent Kit with gDNA Eraser (Takara Bio Inc., Shiga, Japan) was employed to reversely transcribe RNA into complementary DNA (cDNA). qRT-PCR was implemented with a SYBR® Premix Ex Taq™ II kit (Takara Bio Inc., Shiga, Japan) in Applied Biosystems 7500 Realtime PCR System (Applied Biosystems, Inc. Carlsbad, CA, USA). GAPDH and U6 were used as the internal reference for genes and miRNAs expression, respectively. For RNase R treatment, 1 μg of RNA was incubated with 1 U/μg RNase R (Epicenter, Madison, Wisconsin, USA) for 20 min at 37°C, and then reverse transcription and PCR was performed. Primer sequences are shown in Table 1.

Cell counting kit-8 (CCK-8) assay
After the transfection with the plasmids or oligonucleotides, MGC-803 and AGS cells (2 × 10 3 cells/well in a 96-well plate) were cultured for 0, 24, 48, and 72 h. Next, each well was added with CCK-8 reagent (10 μL; Dojindo, Tokyo, Japan), followed by the incubation for 2 h. After that, the absorbance of the cells in each well at 450 nm was detected with a microplate reader (Bio-Rad, Hercules, CA, USA) [14].

Transwell assays
Using the Transwell chamber (24-well; Corning Incorporated, Corning, NY, USA), cell migration and invasion were measured via Transwell assays [15]. The processes of migration assay and invasion assay were similar. For invasion assay, however, the filter of upper compartment was precoated with Matrigel (BD Bioscience, San Jose, CA, USA). GC cells (1 × 10 4 cells/well) cultured in FBS-free medium were transferred to the top compartment, and 10% FBS-containing DMEM was injected into the lower compartment. The cells were cultured for 24 h, and the cells on the top membrane surface were removed using cotton swabs, and the cells on the lower surface were fixed with methanol and stained with crystal violet (0.25%; Sigma, St. Louis, MO, USA). In three random fields under the microscope, the numbers of migrated or invaded cells were analyzed.

Western blot
To obtain the total protein, protease inhibitorscontaining radioimmunoprecipitation assay (RIPA) lysis buffer was applied to lyse cells. After being separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), the protein was transferred to a polyvinylidene difluoride (PVDF) membrane (Amersham Bioscience, Piscataway, NJ). Subsequently, the membrane was blocked in 5% skim milk and then incubated with primary antibodies for DIXDC1 (ab226210, Abcam, Cambridge, MA, USA) and GAPDH (ab181602, Abcam, Cambridge, MA, USA) overnight at 4°C. After Table 1. Primer sequences used in this study.
that, the membranes were incubated with secondary antibodies for 1 h at room temperature. A chemiluminescence detection system (Beyotime Biotechnology, Shanghai, China) was adopted to develop the protein bands.

Immunohistochemistry (IHC)
GC tissue and adjacent normal tissue were fixed in 4% paraformaldehyde for 24 h and embedded in paraffin. Serial sections of 4-μm thick were used to conducted IHC assay [16]. Briefly, the sections were dewaxed, hydrated, and incubated with primary antibody at 4°C overnight in a wet box. The next day, the sections were incubated with secondary antibodies for 2 h. Then, the diaminobenzidine (DAB) system was used to develop the color, and the sections were counterstained with hematoxylin. Finally, the stained section images were captured using a microscope.

Statistical analysis
All experiments were conducted in triplicate. The tool for statistical analysis was GraphPad Prism 8.0 (GraphPad Software, Inc., San Diego, CA, USA). All data were represented by the 'mean ± standard deviation'. Independent-sample t-test was conducted to compare the measurement data between the two groups. One-way analysis of variance (ANOVA) with Tukey's post-hoc test was conducted to compare the data among more groups. Correlation analysis was conducted through Pearson's correlation coefficient analysis. A difference was of statistical significance when P < 0.05.

Results
In this study, we investigated the expression pattern, function and molecular mechanism of circ_0000423 in GC progression through bioinformatics, function-gain and function-loss experiments. It was revealed that, circ_0000423, which is highly expressed in GC, promotes the GC cell proliferation, migration and invasion via regulating microR-582-3p (miR-582-3p)/Disheveled-Axin domain containing 1 (DIXDC1) axis.

Circ_0000423 is high-expressed in GC tissues and cell lines
First of all, the microarray data of GSE78092, GSE89143 and GSE141977 downloaded from the GEO database were analyzed, and in each dataset, there were multiple up-regulated circRNAs. However, circ_0000423, was the only circRNA which was up-regulated in GC tissues in all of the three GEO datasets (Figure 1a). Hence, circ_0000423 was selected for further research. Circ_0000423 is formed by head-to-tail splicing of PPP1R12A transcript (Figure 1b). Then, qRT-PCR was employed to examine circ_0000423 expression in 43 pairs of GC and normal para-tumorous tissues, and it was found that circ_0000423 expression was elevated in GC tissues as against paracancerous tissues (Figure 1c). Additionally, the amplification products of PCR were evaluated by Sanger sequencing, and the result validated the circular structure of circ_0000423 ( Figure S1) . In comparison to GES-1 cells, circ_0000423 expression was markedly up-regulated in HGC-27, MGC-803, AGS and KN-45 cells (Figure 1d). Moreover, the stability of circ_0000423 in AGS and MGC-803 cells was further verified using RNase R exonuclease. RNase R treatment could degrade GAPDH, yet it did not affect circ_0000423 (Figure 1e). These findings suggested that up-regulated circ_0000423 expression might be implicated in GC development.

Circ_0000423 knockdown inhibits AGS and MGC-803 cell multiplication, migration and invasion
To ascertain circ_0000423's role in GC development, AGS and MGC-803 cells with higher circ_0000423 expression, were transfected with si-circ_0000423#1 and si-circ_0000423#2 as well as si-NC to construct cell models with low circ_0000423 expression. Knockdown efficiency of si-circ_0000423#1 was higher, so it was used in subsequent experiments (Figure2a). Subsequently, we investigated the impacts that circ_0000423 had on GC cell growth, migration and invasion. CCK-8 assay showed that knocking down circ_0000423 inhibited AGS and MGC-803 cell viability (Figure2b). Transwell analysis suggested that circ_0000423 knockdown significantly inhibited the migration and invasion of MGC-803 and AGS cells (Figure2c-d).

The circ_0000423/miR-582-3p axis partakes in modulating the biological behaviors of GC cells
To pinpoint the role of circ_0000423/miR-582-3p axis in GC development, MGC-803 and AGS cells were transfected with si-NC, si-circ_0000423 or si-circ_0000423 + miR-582-3p inhibitor (Figure 4a). It was revealed miR-582-3p inhibition attenuated the inhibiting effects of knocking down circ_0000423 on the growth, migration and invasion of MGC-803 and AGS cells (Figure 4b-d). These data suggested that the circ_0000423/miR-582-3p axis participated in modulating the biological behaviors of GC cells.

Discussion
GC is one of the most common malignancies of the digestive system; due to its insidious early symptom, high aggressiveness, and resistance to chemotherapy and radiotherapy, GC has a relatively high mortality [19]. Therefore, exploring the biological mechanism underlying GC occurrence and development and identifying more biomarkers will help to develop better treatment strategies.
CircRNAs are vital in the tumorigenesis and progression of multiple cancers. For instance, in GC tissues, some circRNAs are differentially expressed and partake in regulating the malignancy of cancer cells [20]. For example, circ_006100 expression is enhanced in GC tissues, and its high expression is closely associated with poor differentiation of tumor tissues, high TNM stage and lymph node metastasis [21]. Circ-RPL15 regulates miR-502-3p/OLFM4/STAT3 axis to regulate the malignant phenotype of GC cells, thus facilitating GC progression; abnormal circ-RPL15 expression can predict the GC patient's poor prognosis [22]. Circ_0004872 expression is remarkably down-regulated in GC tissues, and circ_0004872 overexpression inhibits GC cell growth and metastasis in vitro and in vivo [23]. Circ_0001546 modulates ATM/Chk2/p53 pathway via repressing miR-421, thereby inhibiting GC cell proliferation and resistance to oxaliplatin [24]. Our study found that circ_0000423 was highly expressed in GC tissues and cells, and circ_0000423 knockdown restrained GC cell proliferation, migration and invasion. To the best of our knowledge, this is the first work to investigate the role of circ_0000423 in GC.
Emerging evidence shows that circRNA has multiple binding sites for miRNAs and may act as a miRNA sponge, thereby modulating the translation process of mRNA [25,26]. This competitive endogenous RNA (ceRNA) mechanism is reported to participate in regulating the pathogenesis of many human diseases, including GC [27,28]. MiR-582-3p has tumor-suppressive properties in several human malignancies. For example, it represses the growth and cell cycle progression of acute myeloid leukemia cells via inhibiting cyclin B2 [29]. MiR-582-3p suppresses the malignancy of prostate cancer cells through regulating TGF-β pathway [30]. Reportedly, in GC, miR-582-3p is low-expressed, and the transfection of miR-582-3p mimics inhibits GC cell growth, migration and invasion [31], which is in line with the findings of our study. Moreover, herein, it was confirmed that circ_0000423 directly targeted miR-582-3p to repress its expression, which helps better understand the mechanism of miR-582-3p dysregulation in GC.
Disheveled-Axin domain containing 1 (DIXDC1) is a new type of Disheveled-Axin domain protein [32]. There is growing evidence suggesting that DIXDC1 has carcinogenic effects in many cancers, for example, prostate cancer, glioma, non-small cell lung carcinoma and acute myeloid leukemia [33][34][35][36][37]. Reportedly, DIXDC1 is high-expressed in GC The binding sites between DIXDC1 mRNA 3ʹUTR and miR-582-3p were predicted by StarBase. (d) The luciferase activity of cells cotransfected with miR-582-3p mimics and DIXDC1-WT reporter or DIXDC1-MUT1/2/1&2 reporter was measured by dual-luciferase reporter gene assay. (e) Western blot was applied to detect the effects of circ_0000423 and miR-582-3p on DIXDC1 expression.NS: differences were not statistically significant; *P < 0.05, **P < 0.01 and ***P < 0.001. and is linked to advanced TNM stage, lymph node metastasis and poor prognosis [38]. By upregulating MMPs expression, down-regulating E-cadherin expression and enhancing the nuclear accumulation of β-catenin to activate the WNT signaling, DIXDC1 participates in promoting GC cell invasion and metastasis [38,39]. Reportedly, DIXDC1 is a target gene of miR-154 in GC cells [40]. In this work, DIXDC1 was also identified as a target gene of miR-582-3p. Additionally, it was revealed that circ_0000423 promoted DIXDC1 expression in GC cells. These findings suggest that circ_0000423 regulates DIXDC1 expression via serving as a ceRNA of miR-582-3p in GC cells.
Many circRNAs have been used as diagnostic markers for cancers to predict the prognosis of the patients [41,42]. Due to the limitation of the sample number in the present work, the potential of circ_0000423 as a prognostic biomarker for GC has not been evaluated. In the following work, more patients should be recruited to solve these issues. In addition, the biological function of circ_0000423/miR-582-3p/DIXDC1 axis in GC development needs to be further validated with in-vivo models in the following studies.

Conclusion
To sum up, our study reports that circ_0000423 expression is enhanced in GC, and circ_0000423 can promote GC cell proliferation, migration and invasion through modulating miR-582-3p/ DIXDC1 axis. These findings provide a novel explanation of GC progression.

Data availability statement
The data used to support the findings of this study are available from the corresponding author upon request.

Ethics statement
Our study has been approved by the Ethics Review Board of the First Affiliated Hospital of Zhengzhou University.

Highlight
• Circ_0000423 expression is up-regulated in gastric cancer tissues and cells. • Circ_0000423 promotes gastric cancer cell viability, migration and invasion. • Circ_0000423 acts as a competing endogenous RNA (ceRNA) of miR-582-3p to regulate DIXDC1 expression.