Sage Journals: Discover world-class research

Abstract

BACKGROUND:

Circular RNAs (circRNAs) have been found playing important roles in regulating cancer progression. Human circRNA microarray was performed to screen for abnormally expressed circRNA in gastric cancer tissues. In this study, we are aimed to investigate the relationship between a new circular RNA named hsa_circ_0000520 and gastric cancer development.

METHODS:

The hsa_circ_0000520 levels were detected by quantitative reverse transcription-polymerase chain reaction (qRT-PCR) in gastric tissue, cell and plasma levels, respectively. Then, the association between the expression level of hsa_circ_0000520 and the clinicopathological features of patients with gastric cancer was further analyzed. Finally, a network of hsa_circ_0000520-miRNA-mRNA interactions was predicated.

RESULTS:

In this study, hsa_circ_0000520 was first found to be significantly down-regulated in gastric cancer tissues, plasma and gastric cancer cell lines compared with control cases. Clinicopathological features showed that hsa_circ_0000520 level in GC tissues was negatively associated with TNM stage and in GC plasma linked with CEA expression. Finally, a total of 9 miRNAs and 9 candidate mRNA were predicted to have an interaction with hsa_circ_0000520.

CONCLUSIONS:

We first identified that hsa_circ_0000520 was significantly down-regulated in gastric cancer. Our study indicated hsa_circ_0000520 might serve as a novel biomarker for gastric cancer and is involved in gastric carcinoma development.

Keywords

Gastric cancer hsa_circ_00000520 diagnosis miRNA mRNA

1. Introduction

Gastric cancer (GC), a genetic disease involving multi-step changes in the genome, remains the primary factor in male mortality in developing countries with its incidence continuously increasing in recent years [1, 2]. In spite of improved techniques in diagnosis, such as magnifying endoscopy with narrow band imaging, and in treatment including target therapy, most patients are diagnosed in the advanced stage and have poor prognosis. Thus, improving the diagnosis and treatment of GC is a major strategy for reducing the mortality of gastric cancer.

Circular RNAs (circRNAs), a new class of noncoding RNAs, were identified in the early 1990s as transcripts with scrambled exon order and continued to be reported for a number of transcripts over the following two decades [3, 4]. Unlike linear RNAs that are terminated with 5’caps and 3’tails, circRNAs form covalently closed loop structures with neither 5’-3’ polarities nor polyadenylated tails [5]. Owing to the development of high-throughput sequencing techniques and bioinformatics analysis programs, researchers have discovered thousands of circRNAs in mammalian cells and may be involved in atherosclerotic vascular disease risk, neurological disorders, prion diseases as well as carcinomas [6, 7, 8, 9]. Evidences are arising that some circRNAs might regulate microRNA (miRNA) function as microRNA sponges and play a significant role in transcriptional control in cancers [10]. For example, Zhong et al., reported that circTCF25 binded to miR-103a-3p/miR-107 and potentially contributed to the up-regulation of thirteen targets about cell proliferation, migration and invasion, which suggested that circTCF25 could become a new potential marker for this cancer [11].

The present study utilized human circRNA microarray analysis to screen circRNA expression profiles in gastric cancer tissues and discovered a significantly down-expressed circRNA called hsa_circ_0000520(GSE100170 in GEO database). Based on the microarray findings, a series of clinical validation experiments were performed to explore the role of hsa_circ_ 0000520 in gastric cancer from tissue, cell and plasma levels, respectively. In addition, a prediction and annotation of hsa_circ_0000520 molecular function with was miRNAs or mRNAs was performed on relevant database (circinteractome and MIRDB).

2. Materials and methods

2.1 Patients and samples

A total of 56 pairs of gastric cancer tissues and corresponding samples were obtained from patients with gastric cancer who underwent surgery during 2015–2016 at Nanjing Hospital affiliated with the Nanjing Medical University, People’s Republic of China. Before the operation, the patients have not received radiotherapy or chemotherapy. These tissue specimens were immediately stored in RNA-fixer Reagent after removal from patients’ stomachs and were kept at $-$ 80 ${}^{\circ}$ C in a refrigerator until using. Following the principle, the paired adjacent non-tumor tissues were confirmed to have no tumor cells through pathological analysis, and were located 5 cm away from the edge of the gastric cancer. According to the tumor-node-metastasis (TNM) (V.8.2016) staging system of the International Union against Cancer, all tumors were staged accurately.

Peripheral blood (3 ml) of 45 gastric cancer patients were obtained from before the operation and then the plasma was isolated. Following the age-and-gender matched criterion, fresh normal plasma samples were collected from 17 healthy people at Nanjing Hospital, People’s Republic of China in February 2017. All peripheral blood samples were used ethylenediaminetetraacetic acid (EDTA) as the anti-coagulant.

The study was approved by the Ethics Committee of Nanjing First Hospital, Nanjing Medical University. Before they participated in the research, written informed consent was obtained from all patients.

2.2 Cell culture procedures

Human GC cell lines MKN-45, BGC-823, MGC-803 and AGS were established from samples extracted from human GC patients by Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences (Shanghai, China). The human gastric epithelial cell line GES-1 was obtained from the Cancer Institute and Hospital of the Chinese Academy of Medical Sciences (Beijing, China). All cells were cultured in a humidified 5% CO ${}_{2}$ incubator at 37 ${}^{\circ}$ C with RPMI-1640 containing 10% FBS with 100 units/mL penicillin and 100 mg/mL streptomycin.

Figure 1.

Differentially expressed circRNAs between GC tissues and adjacent nontumorous tissues. The result shows distinguishable circRNA expression profiling among samples (G for GC and C for control adjacent nontumorous tissues). Each column represents the expression profile of a tissue sample, and each row corresponds to a circRNA. “Red” indicates higher expression level, and “green” indicates lower expression level.

2.3 RNA isolation, reverse transcription and quantitative real-time polymerase chain reaction

Total RNA from paired tissues were extracted by us- ing TRIzol reagent (Thermo Fisher Scientific, Waltham, MA, USA) and total RNA in plasma was extracted by TIANamp Virus RNA Kit following the manufacturer’s instructions. All complementary DNAs (cDNAs) were obtained with the Prime-Script™ RT-PCR kit (Takara, Dalian, People’s Republic of China) following the manufacturer’s instructions. The sequences of hsa_circ_0000520 divergent primers were as follows: Forward: 5’-GTCTGAGACTAGGGCCAGAGG C-3’and Reverse: 5’-GACATGGGAGTGGAGTGAC AGG-3’. The primers’ sequences of glyceraldehyde-3-phosphate dehydrogenase (GAPDH), as a control, were 5’-TCGACAGTCAGCCGCATCTTCTTT-3’ and 5’-ACCAAATCCGTTGACTCCGACCTT-3’. These primers were synthesized by Primer Bank. Determining the reproducibility of the qRT-PCR method for detecting hsa_circ_0000520 levels in gastric cancer tissue or plasma, we performed qRT-PCR three times to measure differences between experimental batches. Then, the correlation of their 2 ${}^{-\Delta\rm Ct}$ value was determined.

2.4 Statistical analysis

All statistical analyses were performed with Statistical Package for the Social Sciences Version 19 (IBM Corporation, Armonk, NY, USA). The differences in levels of hsa_circ_0000520 between gastric cancer tissues and paired adjacent nontumorous tissues and between paired gastric cancer patient plasma samples and normal ones were assessed using the t test for paired data. A receiver operating characteristic (ROC) curve was established to evaluate its diagnostic value. For all results, $P<$ 0.05 was considered to be statistically significant.

3. Results

3.1 CircRNA expression profile in gastric cancer tissues

A total of 5 gastric cancer tissues and their matched non-gastric cancer (non-GC) tissues were collected and screened for dysregulated circRNA using human circRNA microarray. A total of 713 circRNAs were differentially expressed in GC tissues vs. non-GC tissues among all the candidate circRNAs detected in both GC and non-GC tissues according to the T-test. 191/713 circRNAs were significantly up-regulated in GC tissues, whereas 522/713 circRNAs were significantly down-regulated in GC tissues (Fig. 1). These dysregulated circRNAs may play an important role for tumor genesis and development.

3.2 Hsa_circ_0000520 was identified to be signifi- cantly down-regulated in gastric cancer tissues

CircRNA microarray revealed the differences in the expression of circRNA in gastric cancer tissue compared to adjacent noncancerous tissue. These dysregulated circRNA may provide a better understanding of the role of circRNA in gastric cancer tumorigenesis. Among screened down-regulated circRNA, hsa_circ_0000520 ranked third and interestingly found to be reported exhibiting significantly different expression in hepatocellular carcinoma in previous study [12]. Therefore, hsa_circ_0000520 was selected as a target for investigating the relationship with gastric cancer. To further test the expression level of the dysregulated circRNA hsa_circ_0000520, quantitative RT-PCR was used to measure its expression in 56 paired primary cancerous and adjacent noncancerous tissues from GC patients. The results indicated that hsa_circ_0000520 expression was significantly down-regulated than those in corresponding nontumorous tissues ( $P=$ 0.0374, Fig. 2).

Figure 2.

The expression levels of hsa_circ_0000520 expression in tumor and adjacent normal tissues in 56 GC patients.

Figure 3.

The expression levels of hsa_circ_0000520 expression in gastric cancer plasma in 45 GC patients and 17 healthy controls.

3.3 Expression of hsa_circ_0000520 in gastric cancer plasma

Hsa_circ_0000520 had been tested to be down-regulated in gastric cancer tissues. To investigate this further, the expression of hsa_circ_0000520 in 45 GC plasma was measured. Fresh normal plasma samples from 17 healthy individuals were as controlled. The results indicated that hsa_circ_0000520 expression was down-regulated in gastric cancer plasma compared with normal cases ( $P<$ 0.0001, Fig. 3).

3.4 Expression of hsa_circ_0000520 in gastric cancer cell lines

Hsa_circ_0000520 expression was also examined by qRT-PCR in four human gastric cancer cell lines and a nontumorigenic gastric epithelial cell line GES-1. This experiment showed that the levels of hsa_circ_ 0000520 in five human gastric cancer cell lines, MKN-45, BGC-823, MGC-803 and AGS, were lower than the levels in the GES-1 human normal gastric epithelial cell line (Fig. 4).

Figure 4.

Hsa_circ_0000520 expression in gastric cancer cells. * means $P<$ 0.05, *** means $P<$ 0.0001.

Figure 5.

ROC curve of hsa_circ_0000520.

Table 1

The relationship of hsa_circ_0000520 expression levels (2 ${}^{-\Delta\rm Ct}$ ) in gastric cancer tissues with clinicopathological factors of patients with gastric cancer

Variables	No. of patients (%)	Mean $\pm$ SD	$P$
Age (year)			0.290
$\geqslant$ 60	44	0.000109 $\pm$ 0.0001778
$<$ 60	12	0.000054 $\pm$ 0.000024
Gender			0.663
Female	27	0.000107 $\pm$ 0.000205
Male	29	0.000088 $\pm$ 0.000103
Diameter			0.185
$\geqslant$ 3 (cm)	37	0.0001170 $\pm$ 0.0001920
$<$ 3 (cm)	19	0.0000570 $\pm$ 0.0000320
Differentiation			0.355
Low/middle	48	0.000105 $\pm$ 0.000170
High	8	0.000048 $\pm$ 0.000036
Lymphatic metastasis			0.254
Yes	36	0.000079 $\pm$ 0.000097
No	20	0.000130 $\pm$ 0.000233
TNM stage			0.030*
I $+$ IIA	25	0.0001480 $\pm$ 0.000216
IIB $+$ III $+$ IV	31	0.000056 $\pm$ 0.000073
AFP			0.293
High	6	0.000032 $\pm$ 0.000018
Normal	50	0.000105 $\pm$ 0.000167
CEA			0.535
High	9	0.000128 $\pm$ 0.000159
Normal	47	0.000091 $\pm$ 0.000160

Note: * $P<$ 0.05.

3.5 Potential diagnostic values of hsa_circ_0000520 in gastric cancer

To evaluate the potential diagnostic value, the ROC curve has been used (Fig. 5). The larger the area under the ROC curve, the higher diagnostic value. We found that in tissues, the area under the ROC curve (AUC) was 0.6129. The sensitivity and specificity were 53.57% and 85.71%, respectively. For plasma, the AUC was 0.8967 and the sensitivity and specificity were 82.35% and 84.44%, respectively.

Table 2
The relationship of hsa_circ_0000520 expression levels (2 ${}^{-\Delta\rm Ct}$ ) in gastric cancer plasma with clinicopathological factors of patients with gastric cancer

Variables	No. of patients (%)	Mean $\pm$ SD	$P$
Age (year)			0.292
$\geqslant$ 60	832	0.221140 $\pm$ 0.170240
$<$ 60	13	0.291679 $\pm$ 0.264332
Gender			0.597
Female	20	0.223482 $\pm$ 0.192366
Male	25	0.255946 $\pm$ 0.210959
Diameter			0.914
$\geqslant$ 3 (cm)	37	0.243052 $\pm$ 0.213318
$<$ 3 (cm)	8	0.234423 $\pm$ 0.143546
Differentiation			0.207
Low/middle	26	0.274179 $\pm$ 0.191645
High	19	0.196824 $\pm$ 0.210687
Lymphatic metastasis			0.267
Yes	41	0.252031 $\pm$ 0.207989
No	4	0.133763 $\pm$ 0.022614
TNM stage			0.842
I $+$ IIA	15	0.250141 $\pm$ 0.158478
IIB $+$ III $+$ IV	30	0.237207 $\pm$ 0.222028
AFP			0.348
High	5	0.160912 $\pm$ 0.116676
Normal	40	0.251594 $\pm$ 0.208283
CEA			0.033*
High	19	0.167182 $\pm$ 0.130332
Normal	26	0.295840 $\pm$ 0.295840

Note: * $P<$ 0.05.

3.6 Association between hsa_circ_0000520 expression and clinical factors in gastric cancer

A pathological feature of 56 primary GC and nontumor samples was analyzed. As shown in Table 1, abnormal expression of hsa_circ_0000520 level was not aberrantly associated with age, gender, diameter, differentiation, lymph node metastasis or common blood tumor markers in patients with GC. However, down-expression of hsa_circ_0000520 level was negatively associated with TNM stage ( $P=$ 0.030).

Plasma of 45 gastric cancer patients were obtained from before the operation was isolated and fresh normal plasma samples were collected from 17 healthy people. In the plasma of patients with gastric cancer (Table 2), hsa_circ_0000520 levels were associated with CEA ( $P=$ 0.033). The results showed that CEA positive patients had lower hsa_circ_0000520 levels. Nonetheless, we did not find any association between hsa_circ_0000520 and other clinicopathological features such as age, gender, diameter, differentiation, lymphatic metastasis, TNM stage or other common blood tumor markers.

3.7 Prediction and annotation of hsa_circ_0000520 targeted miRNA and mRNA network

It has recently come to light that circRNAs play critical roles in fine-tuning the level of miRNA-mediated regulation of gene expression by sequestering the miRNAs. To identify the miRNAs that bind to hsa_circ_ 0000520, we performed a circRNA-miRNA interaction network prediction on circinteractome database. A total of 9 highest-ranking candidate miRNAs (hsa-miR-556-5p, hsa-miR-521, hsa-miR-1204, hsa-miR-146b-3p, hsa-miR-512-5p, hsa-miR-663b, hsa-miR-1258, hsa-miR-1233, hsa-miR-1296) and 200 corresponding target mRNAs were predicted to have an interaction with hsa_circ_0000520 in this study. Among them, several miRNAs (hsa-miR-512-5p, hsa-miR-663b, hsa-miR-1258, hsa-miR-1233, hsa-miR-1296) have been previously investigated and confirmed involved in suppressing or promoting the formation of gastric cancer [12, 13, 14, 15, 16]. However, to the best of our knowledge, hsa-miR-556-5p, hsa-miR-521, hsa-miR-1204, hsa-miR-146b-3p have never been researched.

It has been shown that RNA-binding proteins (RBPs) that can bind to circRNAs [17]. Some circRNAs can store, sort, or localize RBPs, and probably regulate the function of RBPs by acting as competing elements, in the same way as they modulate miRNA activity. Our prediction showed that hsa_circ_0000520 was able to have a binding with 9 candidate mRNA (AGO1, AGO2, AGO3, Fus, FMRP, CAPRIN1, DGCR8, PTB, HuR) according the rankings (Fig. 6). These prediction results revealed the unique structure of hsa_circ_ 0000520 may also play an important role in the assembly of RNA or RBP complexes.

Figure 6.

The predicted hsa_circ_0000520 targeted miRNA and mRNA network.

4. Discussion

CircRNAs, a recently discovered RNA type, were thought to be the products of transcription errors. Thanks to the advancements in high-throughput sequencing technologies and bioinformatics, circRNAs were found to be broadly expressed and perform regulation in atherosclerotic vascular disease, neurological disorders, prion diseases and carcinomas. Recent works have suggested that circRNAs may play important roles in the initiation and development of cancers and could potentially become new biomarkers for cancers [18, 19].

More attention was focused on the potential functions of circRNAs. However, their clinical diagnostic value remains largely unknown. In this study, hsa_circ _0000520 was first found to be significantly down-regulated in gastric cancer tissues, plasma and gastric cancer cell lines compared with control cases. Clinicpathological features showed that down-expression of hsa_circ_0000520 level was negatively associated with TNM stage. The expression of hsa_circ_0000520 in patients with GC with I $+$ IIA TNM stage was higher than that in patients with IIB $+$ III $+$ IV stage, which suggested this indicator plays an inhibiting role in the occurrence and development of gastric cancer. In the plasma of patients with gastric cancer, hsa_circ_0000520 levels were found to be associated with CEA and patients with positive CEA expression tended to express lower hsa_circ_0000520 levels. The ROC is a comprehensive index used to reflect the sensitivity and specificity of continuous variables. We found that in tissues, the AUC was 0.6129. The sensitivity and specificity were 53.57% and 85.71%, respectively. For plasma, the AUC was 0.8967. The sensitivity and specificity were 82.35% and 84.44%, respectively. As a result, our study indicated that hsa_circ_0000520 had a higher sensitivity and specificity in the screening of plasma and might serve as a novel biomarker for gastric cancer. Due to the limited number of samples of gastric cancer tissues and plasma, we only analyzed hsa_circ_0000520 expression levels in 56 paired gastric cancer tissues and 45 preoperative GC and 17 healthy plasma samples. More samples should be analyzed in the future.

The current studies have provided evidences that circRNAs could function as microRNA sponges or potent competing endogenous RNA molecules, thereby influencing the posttranscriptional actions of microRNAs as suppressors of the translation in recent literature [20]. Thomas B. Hansen and others recently uncovered the function of one such circRNA, ciRS-7 efficiently tethers miR-7, resulting in reduced miR-7 activity and increased levels of miR-7-targeted transcripts, consistent with the miRNA sponge and ceRNA hypothesis [21]. As a result, the circ-miRNA axis, regardless of promotion or suppression, have significant effects on various pathways in human diseases and is worthy more thorough study. Therefore, we supposed that the carcinogenic mechanisms of hsa_circ_0000520 may occur through their miRNA-mediated effects on gene expression in gastric cancer as well. Annotation and function prediction revealed a total of 9 miRNAs and 200 corresponding target mRNAs were predicted to have an interaction with hsa_circ_0000520. It has been shown that RNA-binding proteins can bind to circRNAs. Therefore, we identified 9 candidate mRNA binding targets which can interact with hsa_circ_0000520 in this study. Experimental identification and characterization of their associated molecules, such as miRNAs or proteins, are suggested in future. Taken together, we propose the hypothesis that the circRNA-microRNA-mRNA axis or circRNA-mRNA may be the possible mechanism promoting the development of gastric cancer, but further studies are needed for this mechanism.

5. Conclusion

In summary, we first identified that hsa_circ_00005 20 was significantly down-regulated in gastric cancer. Our study indicated hsa_circ_0000520 might serve as a novel biomarker for gastric cancer and is involved in gastric carcinoma development. However, the detailed molecular mechanism by which this circRNA contributes to GC proliferation, invasion, and metastasis requires further research. In the future, we believe that by further studying the functions of circRNAs, we could improve our understanding of the mechanisms of disease associated with circRNAs and improve the diagnosis and prevention of circRNA-associated diseases.

Footnotes

Acknowledgments

This study was supported by the Development of Medical Science and Technology Foundation of Nanjing (grant No. ZKX14035) to Professor Hongyong Cao.

Conflict of interest

The authors report no conflicts of interest in this work.

References

Siegel

R.L.

Miller

K.D.

and Jemal

, Cancer statistics 2016, CA Cancer J Clin 66(1) (2016), 7–30.

Ferlay

Soerjomataram

Dikshit

et al., Cancer incidence and mortality worldwide: Sources methods and major patterns in GLOBOCAN 2012, International Journal of Cancer 136 (2015), E359–E386.

Capel

Swain

Nicolis

et al., Circular transcripts of the testis-determing gene sry in adult mouse testis, Cell 73 (1993), 1019–1030.

Guo

J.U.

Agarwal

Guo

et al., Expanded identification and characterization of mammalian circular RNAs, Genome Biol 15 (2014), 409.

Zheng

Bao

Guo

et al., Circular RNA profiling reveals an abundant circHIPK3 that regulates cell growth by sponging multiple miRNAs, Nature Communications 7 (2016), 11215.

Yang

Zhou

et al., Circular RNAs in cancer: Novel insights into origins properties functions and implications, Am J Cancer Res 5 (2015), 472–480.

Szabo

Morey

Palpant

N.J.

et al., Statistically based splicing detection reveals neural enrichment and tissue-specific induction of circular RNA during human fetal development, Genome Biology 16(1) (2015), 126.

Vausort

Salgado-Somoza

Zhang

et al., Myocardial infarction-associated circular RNA predicting left ventricular dysfunction, Journal of the American College of Cardiology 68(11) (2016), 1247–1248.

Ghosal

Das

Sen

et al., Circ2Traits: A comprehensive database for circular RNA potentially associated with disease and traits, Front Genet 4 (2013), 283.

10.

Xie

Ren

Xin

et al., Emerging roles of circRNA 001569 targeting miR-145 in the proliferation and invasion of colorectal cancer, Oncotarget 3 (2016), 26680–26691.

11.

Zhong

and Chen

, Screening differential circular RNA expression profiles reveals the regulatory role of circTCF25-miR-103a-3p/miR-107-CDK6 pathway in bladder carcinoma, Sci Rep 6 (2016), 30919.

12.

Shang

Liu

et al., Comprehensive circular RNA profiling reveals that hsa_circ_0005075 a new circular RNA biomarker is involved in hepatocellular crcinoma development, Medicine 95(22) (2016), e3811.

13.

Saito

Suzuki

Tsugawa

et al., Chromatin remodeling at alu repeats by epigenetic treatment activates silenced microRNA-512-5p with downregulation of mcl-1 in human gastric cancer cells, Oncogene 28(30) (2009), 2738–2744.

14.

F.Q.

Y.J.

et al., Differential microRNA expression in signet-ring cell carcinoma compared with tubular adenocarcinoma of human gastric cancer, Genetics and Molecular Research 14(1) (2015), 739–747.

15.

Shi

Chen

Sun

et al., MicroRNA-1258: An invasion and metastasis regulator that targets heparanase in gastric cancer, Oncology Letters 13(5) (2017), 3739–3745.

16.

Zhang

Z.Z.

Wang

C.J.

Niu

et al., Analysis of plasma microRNAs to identifying early diagnostic molecule for gastric cancer, International Journal of Clinical and Experimental Medicine 8(3) (2015), 3700–3706.

17.

Shan

Wen

Zhu

et al., miR 1296-5p inhibits the migration and invasion of gastric cancer cells by repressing ERBB2 expression, PloS One 12(1) (2017), e0170298.

18.

Ashwal-Fluss

Meyer

Pamudurti

N.R.

et al., circRNA biogenesis competes with pre-mRNA splicing, Mol Cell 56 (2014), 55–66.

19.

Lin

Deng

et al., Profiling and bioinformatics analyses reveal differential circular RNA expression in radioresistant esophageal cancer cells, Journal of Translational Medicine 14(1) (2016), 225.

20.

Gong

Sun

et al., The circular RNA cdr1as act as an oncogene in hepatocellular carcinoma through targeting miR-7 expression, Plos One 11(7) (2016), e0158347.

21.

Thomas

L.F.

and Saetrom

, Circular RNAs are depleted of poly-morphisms at microRNA binding sites, Bioinformatics 30 (2014), 2243–2246.

22.

Hansen

T.B.

Jensen

T.I.

Clausen

B.H.

et al., Natural RNA circles function as efficient microRNA sponges, Nature 495 (2013), 384–388.