hsa_circ_0003176 Suppresses the Progression of Non-Small-Cell Lung Cancer via Regulating miR-182-5p/RBM5 Axis

Background Non-small-cell lung cancer (NSCLC) is one of the major diseases that threaten human health, and there is still no fundamental treatment method. Emerging evidences suggested that circRNAs might be an effective target to treatment NSCLC. However, the roles and detailed mechanisms of hsa_circ_0003176 in NSCLC still not clear. Methods hsa_circ_0003176 was identified from GSE101684 and GSE112214 datasets of Gene Expression Omnibus (GEO) database. The expression of hsa_circ_0003176 was detected by RT-qPCR in NSCLC tissues, paired adjacent nontumor tissues, and cell lines. RNA fluorescence in situ hybridization and nuclear and cytoplasmic RNA fractionation analysis was used to detect the subcellular localization of hsa_circ_0003176 in H1299 and A549 cells. Dual-luciferase reporter and RNA pull-down assay were used to confirm the regulatory of miR-182-5p to hsa_circ_0003176 and RBM5. The roles of hsa_circ_0003176 in NSCLC progression was evaluated both in vitro by CCK-8 assay, colony formation assay, wound-healing assay, and matrigel transwell assay and in vivo by the subcutaneous xenograft nude mouse experiment and lung metastasis nude mouse experiment. In addition, RNA pull down and luciferase reporter assays were carried out to investigate the interaction between hsa_circ_0003176 or RBM5 and miR-182-5p. Results Our results indicated that hsa_circ_0003176 showed typical characteristic of circRNAs, which was downregulated in both NSCLC tissues and cell lines. Functionally, overexpression of hsa_circ_0003176 suppressed the proliferation, migration, and invasion of NSCLC cells in vitro and inhibited NSCLC growth and metastasis in vivo. Furthermore, we found that hsa_circ_0003176 acts as sponge of miR-182-5p to regulate RBM5 expression. Further, in vitro rescue experiments demonstrated that hsa_circ_0003176 suppressed the proliferation, migration, and invasion of NSCLC cells by regulating miR-182-5p/RBM5 axis. Conclusion We demonstrated that hsa_circ_0003176 suppressed the NSCLC progression via regulating miR-182-5p/RBM5 axis. These data indicated that hsa_circ_0003176 might be a novel molecular target for NSCLC treatment.


Introduction
Lung cancer is the malignant tumor, which presented the highest morbidity and mortality all over the world [1]. Based on the pathological types, it is divided into small-cell lung cancer (SCLC) and non-small-cell lung cancer (NSCLC). Among them, NSCLC is characterized by slow growth and division, late metastasis and spread, and low malignancy, accounting of about 85% of the total number of lung cancers [2]. More than half of the patients are in the middle and late stages when they are diagnosed, losing the best time for treatment [3]. In recent years, important breakthroughs have been made in the treatment of NSCLC, which has developed from traditional surgery, chemotherapy, and radiotherapy to the era of precise molecular targeted therapy and immunotherapy [4]. However, the 2-year relative survival rate of NSCLC patients reached only 42% [5]. Therefore, in-depth clarifying the molecular mechanisms of the occurrence and development of NSCLC is crucial for the development of new antitumor drugs and the improvement of prognosis.
Circular RNAs (circRNAs) are a new class of noncoding RNAs and widely exist in the cytoplasm of eukaryotic cells [6]. circRNAs have a unique closed-loop structure and play a crucial role in regulating gene transcription, which can either directly affect protein expression, or act as ceRNAs to regulate target gene expression by sponging miRNAs [7]. Accumulating evidence has suggested that circRNAs not only play an important biological role in the development of organisms but also play a critical role in the diagnosis and treatment of different kinds of diseases [8,9]. With the deepening of NSCLC research, increasing circRNAs proved to contribute to different processes of NSCLC. For example, Sun et al. identified 8 key circRNAs by comprehensive bioinformatics analysis with three gene expression microarrays [10]. Fu et al. indicated that hsa_circRNA_012515 was significantly increased in NSCLC patients and cells, indicating that it might be a potential biomarker in the prognosis of NSCLC [11]. Liang et al. showed that circRNA_103615 promoted NSCLC cisplatin resistance and tumor progression by regulating ABCB1 [12]. circNDUFB2 activated antitumor immunity that inhibited NSCLC progression by regulating IGF2BPs expression [13]. hsa_circ_0000429 was highly expressed in the NSCLC tissues and cell lines, and overexpression promotes NSCLC progression by elevating SMADD expression by sponging miR-1197 [14]. These findings suggested that exploring the roles and molecular mechanism of circRNAs in NSCLC is helpful for the identification of new molecular targets and the implementation of new treatment options for NSCLC.
Although current studies have demonstrated that cir-cRNAs play an important role in the pathogenesis of NSCLC, the biological functions and regulatory mechanisms of most circRNAs in NSCLC remain still unknown and need to be further explored. Hereby, we identified that hsa_circ_0003176 (a circRNA derived from TBX5) was significantly downregulated in NSCLC tissues compared with match adjacent nontumorous tissues through analyzing the public lung cancer dataset in the GEO database and further to explored the biological function and underlying molecular mechanism of hsa_circ_ 0003176 in the development of NSCLC. Our study not only disclosed a novel mechanism of NSCLC development but also provided a new molecular target for NSCLC treatment.

Clinical Specimen.
A total of 10 NSCLC tissues and 10 corresponding adjacent nontumor tissues were collected from the University of Hong Kong-Shenzhen Hospital (Shenzhen, China). These tissue specimens were reviewed and approved by the Committees for Ethical Review of Research at the University of Hong Kong-Shenzhen Hospital (approval number: [2020]171).

Quantitative
2.13. RNA Pull-Down Assay. The biotin-labeled probes for hsa_circ_0003176 were designed and synthesized by Sangon Biotech (Shanghai, China). The binding relationship between hsa_circ_0003176 and miR-182-5p was performed by the RNA pull-down assay. Briefly, the cells with were fixed, lysed, and centrifuged, and then, the supernatants were collected. Subsequently, the supernatants were incubated with the biotin-labeled probes for hsa_circ_0003176 overnight at room temperature. After that, the formaldehyde crosslinking was reversed by the lysis buffer and proteinase K. RNA was extracted by TRIzol reagent (Invitrogen, 15596-026). The supernatants were used as input, and the expression of hsa_ circ_0003176 and miR-182-5p was detected by RT-qPCR.
2.14. In Vivo Experiments. Male BALB/c nude mice (SPF, 4 weeks old) were purchased at Zhuhai Baishitong Biotechnology Co., Ltd (License No.: SCXK (Guangdong) 2020-0051, Animal Certificate No. 44822700003250). After one week of adaptive feeding, subcutaneous xenograft experiments (n = 6 mice/ group) and lung metastasis experiment (n = 5 mice/group) were carried out in nude mice. First, we constructed A549 cells with stable overexpressing hsa_circ_0003176. Briefly, the A549 cells were pretransfected with overexpressing hsa_circ_0003176 lentivirus, and then, the overexpressing hsa_circ_0003176 A549 cells was filtered with puromycin (MCE, USA) to get stable A549 cells. The A549 cells transfected with negative control lentivirus acted as negative control. The transfection efficiency was determined by RT-qPCR. For subcutaneous xenograft experiments, a total of 2 × 10 6 A549 stable cells with or without overexpressing hsa_circ_0003176 were subcutaneously injected into the right armpit of male nude mice. The tumor volume and the weight of the nude mice were measured on days 3, 7, 14, 21, and 28 after subcutaneous xenograft, and tumor volume was calculated according to the formula volume = length × width2 × 0:5. At the end of feeding, mice were sacrificed with the tumor removed for weighing and histological analysis. For lung metastasis experiment, the nude mice were injected tail vein with a total of 2 × 10 6 A549 stable cells with or without overexpressing hsa_circ_0003176. At the end of feeding (4 weeks), mice were sacrificed with the lung tissues removed for histological analysis. All the animal experiments were approved by the Ethics Committee of the University of Hong Kong-Shenzhen Hospital (approval number: HTSW201090924).

Histological Analysis.
The collected tissues were routinely dehydrated, embedded, and cut into 3 μm sections. For hematoxylin-eosin (H&E) staining, the sections were stained with hematoxylin and differentiation with hydrochloric acid alcohol. And then, ammonia water was used to immerse the sections for 10-30 s to return to the blue. And then, the sections were stained with eosin solution (Servicebio, G1004) and dehydrated with alcohol, xylene permeabilization and neutral gum sealing. After that, observation was performed under the light microscope (Leica, Cat. DMI1).
For Ki-67 immunohistochemistry, primary antibody anti-Ki67 (Proteintech, 21309-1-AP) was used to incubate the sections. After that, the sections were incubated with secondary antibody anti-rabbit horseradish peroxidase-conjugated antibody for 45 min. And then, the slides were stained using 3,3 ′ diaminobenzidine (DAB) solution (Beyotime, P0202). Images were taken and analyzed with the light microscope (Leica, Cat. DMI1) and Image J software (U. S. National Institutes of Health, Bethesda, Maryland, USA), respectively. The expression level of Ki-67 was calculated as average optical density ð AODÞ = integrated option density ðIODÞ/area.

Statistical Analysis.
Statistical analyses in the present study were detected using SPSS 22.0 software (IBM) and GraphPad Prism 8. Each study has at least three biological replicates. Data are shown as means ± standard deviation ðSDÞ. Pearson correlation coefficient analysis was used to evaluate correlations between genes. Differences between two groups and more than two groups were analyzed by Student's t-tests and one-way ANOVA assay. A value of p < 0:05 indicated a statistically significant difference.

hsa_circ_0003176 Was Downregulated in NSCLC.
In order to investigate the role of hsa_circ_0003176 in NSCLC, we used GSE101684 containing four NSCLC lung samples and three matched adjacent normal samples (https://www .ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE101684) and GSE112214 datasets containing three NSCLC lung samples and three matched adjacent normal samples (https://www .ncbi.nlm.nih.gov/geo/query/acc.cgi) to analyze the hsa_circ_ 0003176 expression in NSCLC tissues and matched adjacent normal tissues. The result showed that hsa_circ_0003176 was significantly downregulated in NSCLC tissues compared with matched adjacent normal tissues (Figure 1(a)). In order to further confirm the expression of hsa_circ_0003176 in NSCLC, the expression of hsa_circ_0003176 was detected by RT-qPCR in NSCLC tissues, matched adjacent normal tissue, and cell lines. The results indicated that the expression of hsa_ circ_0003176 was greatly downregulated in NSCLC lung tissues compared with matched adjacent normal tissues (Figure 1(b)). Meanwhile, we found that the expression of hsa_circ_0003176 was also significantly reduced in lung cancer cell lines (H1299, H446, 95-D, and A549) compared with 16HBE cells (Figure 1(c)). Further ROC curve indicated that the AUC reached at 98% with higher sensitivity and specificity (Figure 1(d)), suggesting that hsa_circ_0003176 has the certain diagnostic value for NSCLC. Furthermore, the data from circBase and circRNADb databases showed that hsa_circ_ 0003176 was transcribed from TBX5, which localized on     (Figure 1(e)). Furthermore, the target segment of hsa_circ_0003176 was amplified by divergent primers in cDNA regardless of RNase R treatment. On contrast, no target segment was observed amplified by convergent primers in cDNA when treatment with RNase R and amplified by divergent primers in gDNA (Figure 1(e)). In addition, FISH assay indicated that hsa_circ_0003176 was predominately localized to the cytoplasm of A549 and H1299 cells, and nuclear and cytoplasmic RNA fractionation analysis also confirmed the results (Figure 1(f)). These results suggested that hsa_circ_0003176 was downregulated in NSCLC and mainly located in the cytoplasm of NSCLC cells.

hsa_circ_0003176
Suppressed the Proliferation, Migration, and Invasion of NSCLC Cells. In order to explore the biological function of hsa_circ_0003176 on the progression of NSCLC, gain function of hsa_circ_0003176 was performed in A549 and H1299 cells by transfection with overexpressing hsa_circ_0003176 plasmid. RT-qPCR analysis showed that the expression of hsa_circ_0003176 was significantly upregulated in the hsa_circ_0003176 group compared with the mock group (Figure 2(a)). CCK-8 and colony formation assay indicated that the proliferation of both A549 and H1299 cells was inhibited when overexpression of hsa_circ_0003176 (Figures 2(b) and 2(c)). Transwell assay showed that the invasive cell number of both A549 and H1299 cells was notably reduced in the hsa_ circ_0003176 group compared with that in the mock group (Figure 2(d)). Wound-healing assay showed that the wound closure of both A549 and H1299 cells was dramatically decreased in the overexpressing hsa_circ_0003176 group compared with that in the mock group (Figure 2(e)). These data suggested that hsa_circ_0003176 suppressed the proliferation, migration, and invasion of NSCLC cells.

hsa_circ_0003176 Inhibited NSCLC Growth and
Metastasis In Vivo. In order to further investigate the effect of hsa_circ_0003176 on NSCLC growth in vivo, a total of 2 × 10 6 A549 stable cells with or without overexpressing hsa_circ_ 0003176 were subcutaneously injected into the right armpit of male nude mice. The result indicated that no significant difference was observed in body weight of nude mice between overexpressing hsa_circ_0003176 and mock groups during the experiment (Figure 3(a)). However, the tumor volume was significantly reduced in the overexpressing hsa_circ_0003176 group compared with that in the mock group (Figures 3(b)  and 3(c)). And the tumor weight was also greatly reduced in the overexpressing hsa_circ_0003176 group compared with the mock group at the end of the experiment (Figure 3(d)). H&E staining showed that overexpressing hsa_circ_0003176 significantly reduced the inflammatory infiltration of tumor tissues compared with the mock group (Figure 3(e)). Ki-67 immunohistochemical staining assay showed that the Ki-67 expression was significantly downregulated by overexpressing hsa_circ_0003176 (Figures 3(e) and 3(f)), suggesting that overexpression of hsa_circ_0003176 inhibited NSCLC cell proliferation. In addition, to explore the effect of hsa_circ_0003176 on NSCLC metastasis in vivo, the nude mice were injected tail vein with a total of 2 × 10 6 A549 stable cells with or without overexpressing hsa_circ_0003176. The result showed that the number of tumor metastasis on the surface of lung tissues in the overexpressing hsa_circ_0003176 group was significantly lower than that in the mock group (Figure 3(g)). H&E staining confirmed that the number of lung metastasis nodules was dramatically reduced in the overexpressing hsa_circ_0003176 group compared with the mock group (Figure 3(h)). These results indicated that overexpression of hsa_circ_0003176 inhibited NSCLC growth and metastasis.
3.4. hsa_circ_0003176 Suppressed the Proliferation, Migration, and Invasion of NSCLC Cells via Sponging miR-182-5p. Several studies have proved that circRNAs affected miRNA activity by acting as miRNA "sponge" [15,16]. To explore regulatory functions of hsa_circ_0003176 at posttranscriptional level, we used CircInteractome to predict the potential binding miRNA of hsa_circ_0003176. We found that miR-182-5p showed the highest binding score (Table S1). Previous studies have indicated that miR-182-5p is an oncogenic gene in non-small-cell lung cancer [17,18]. RT-qPCR results also indicated that the expression of miR-182-5p was significantly upregulated in both NSCLC cell lines and tumor tissues compared with their corresponding controls (Figures 4(a) and 4(b)). Pearson correlation coefficient analysis showed that the expression of hsa_circ_0003176 and miR-182-5p was negatively correlated in 10 NSCLC tissues (Figure 4(c)). RNA pull-down assay showed that a specific enrichment of hsa_ circ_0003176 and miR-182-5p was detected by qRT-PCR in the hsa_circ_0003176 probe group compared with control probe in A549 cells and H1299 cells (Figure 4(d)), suggesting the interaction between hsa_circ_0003176 and miR-182-5p in NSCLC cells. Moreover, dual-luciferase reporter assays were used to verify direct interaction between hsa_circ_0003176 and miR-182-5p. Cotransfection cells with hsa_circ_0003176-WT vector and miR-182-5p mimics led to significantly reduced relative luciferase activity, while co-transfection with hsa_circ_0003176-MUT vector and miR-182-5p mimics induced nonsignificant differences (Figure 4(e)), suggesting that hsa_circ_0003176 interacts directly with miR-182-5p in NSCLC. In addition, the expression of miR-182-5p affected by hsa_circ_0003176 both in vitro and in vivo was detected. The results indicated that the expression of miR-182-5p was dramatically decreased in A549 and H1299 cells (Figure 4(f)) and in subcutaneous xenograft (Figure 4(g)) with hsa_circ_ 0003176 overexpression, indicating that hsa_circ_0003176 could negatively regulate miR-182-5p expression. Taken together, the above results showed that hsa_circ_0003176 interacted directly with miR-182-5p to inhibit its expression.
To explore further whether hsa_circ_0003176 suppressed the proliferation, migration, and invasion of NSCLC cells via regulating miR-182-5p, the NSCLC cell lines (A549 and H1299) were transfected with hsa_circ_0003176 overexpression plasmid alone or cotransfected with miR-182-5p mimics. CCK-8 and colony formation assay indicated that overexpressing miR-182-5p could reduce the proliferation inhibition to both A549 and H1299 cells caused by overexpression of hsa_ circ_0003176 (Figures 5(a) and 5(b)). Wound-healing assay showed that the migration inhibited by overexpressing hsa_ circ_0003176 in both A549 and H1299 cells was also alleviated    Figure 5(c)). Transwell assay showed that the reduced invasive cell number of both A549 and H1299 cells in the hsa_circ_0003176 group was partially increased when transfection with miR-182-5p mimics ( Figure 5(d)). These data suggested that hsa_circ_ 0003176 suppressed the proliferation, migration, and invasion of NSCLC cells by downregulating miR-182-5p expression.

hsa_circ_000 3176 Enhanced RBM5 Expression via
Regulating miR-182-5p. circRNAs can regulate the expression of the target genes of miRNAs by acting as a ceRNA to sponge microRNA [19]. We predicted 132 potential common target genes of miR-182-5p by RITA, miRmap, miRanda, and Tar-getScan databases, including RBM5 ( Figure S1a). RBM5 (RNA-binding protein 5) has been confirmed as a tumor suppressor in different cancers, including lung cancer [20]. Therefore, RBM5 was identified as potential targets of miR-182-5p for further study. We further verified the expression level of RBM5 in NSCLC and its relationship with miR-182-5p based on bioinformatics and RT-qPCR. The data from    Figure 5: hsa_circ_0003176 suppressed the proliferation, migration, and invasion of NSCLC cells by downregulating miR-182-5p expression. (a) The viability of NSCLC cell lines (A549 and H1299) transfected with hsa_circ_0003176 overexpression plasmid alone or cotransfected with miR-182-5p mimics was detected using CCK-8 assay. (b) The colony formation ability of NSCLC cell lines (A549 and H1299) transfected with hsa_circ_0003176 overexpression plasmid alone or cotransfected with miR-182-5p mimics was detected using colony formation assay. (c) The migration ability of NSCLC cell lines (A549 and H1299) transfected with hsa_circ_0003176 overexpression plasmid alone or cotransfected with miR-182-5p mimics was detected using wound-healing assay. (d) The invasion ability of NSCLC cell lines (A549 and H1299) transfected with hsa_circ_0003176 overexpression plasmid alone or cotransfected with miR-182-5p mimics was detected using transwell invasion assay. Data was shown as mean ± SD. N = 3~5, * * p < 0:01, and * * * p < 0:001.

10
Disease Markers GEPIA database indicated the expression of RBM5 was downregulated in LUAD tumor tissues compared with control tissues ( Figure S1b). And highly expression of RBM5 showed better overall survival ( Figure S1c). Furthermore, the expression of RBM5 was significantly decreased in both NSCLC cell lines and tumor tissues compared with control (Figures 6(a) and 6(b)). The above results suggested that the RBM5 was significantly downregulated in NSCLC and might be a target gene of miR-182-5p.
3.6. hsa_circ_0003176 Suppressed the Proliferation, Migration, and Invasion of NSCLC Cells by Upregulating RBM5 Expression. In order to know whether hsa_circ_ 0003176 participated in the regulation of NSCLC cell proliferation, migration, and invasion by regulating RBM5 expression, the designed RBM5 shRNA (shRNA#1, shRNA#2, and shRNA#3) was used to silence the expression of RBM5 in both A549 and H1299 cells. The results indicated that all of the shRNAs for RBM5 significantly downregulated the expression of RBM5, and both shRNA#1 and shRNA#2 showed good interference effect ( Figure S3). Thus, one of them (shRNA#1) was selected for further study. Subsequently, the NSCLC cell lines (A549 and H1299) were transfected with hsa_circ_ 0003176 overexpression plasmid alone or cotransfected with shRBM5 plasmid, and RT-qPCR and western blot assay showed that transfection of sh-RBM5 plasmid can reduce the mRNA and protein expression of RBM5 in A549 and H1299 cells overexpressing hsa_circ_0003176 (Figures 7(a) and 7(b)). CCK-8 and colony formation assay indicated that silencing RBM5 could reduce the proliferation inhibition to both A549 and H1299 cells caused by overexpression of hsa_circ_ 0003176 (Figures 7(c) and 7(d)). Wound healing and transwell invasion assay showed that the migration ability and invasion ability inhibited by overexpressing hsa_circ_0003176 in both A549 and H1299 cells could be partially alleviated by silencing RBM5 (Figures 7(e) and 7(f)). Taken together, these results indicated that hsa_circ_0003176 suppressed the proliferation, migration, and invasion of NSCLC cells by upregulating RBM5 expression.

Discussion
Lung cancer is one of the common malignant tumors, among which NSCLC accounts for about 80% of the incidence of lung cancer [21]. Most of the patients are found in the advanced stage, and the prognosis is very poor due to the difficulty of the screening and diagnosis [22]. At present, the main treatments are surgery, radiotherapy, chemotherapy, and targeted therapy, but the overall cure rate is low, and there are many adverse reactions and defects that seriously affect the quality of life [23]. Numerous studies have shown that circRNAs play important roles in the NSCLC [24]. However, the biological functions and regulatory mechanisms of most circRNAs in NSCLC remain still overexpression plasmid alone or cotransfected with shRBM5 plasmid was detected using wound-healing assay. (f) The invasion ability of NSCLC cell lines (A549 and H1299) transfected with hsa_circ_0003176 overexpression plasmid alone or cotransfected with shRBM5 plasmid was detected using transwell invasion assay. Mock indicated negative control corresponds to hsa_circ_0003176 overexpression plasmid, and shNC indicated negative control corresponds to shRBM5 plasmid. Data was shown as mean ± SD. N = 3~5, * * p < 0:01, and * * * p < 0:001.
13 Disease Markers unknown. Therefore, exploring the roles and mechanisms of circRNAs in NSCLC is helpful for early diagnosis, targeted therapy, and improvement of prognosis of NSCLC. In the present study, we found that hsa_circ_0003176 suppressed the NSCLC progression via regulating miR-182-5p/RBM5 axis, and these findings indicated that hsa_circ_0003176 might be a novel molecular target for NSCLC treatment.
Circular RNAs (circRNAs) are a class of endogenously conserved noncoding RNAs that have been shown to be involved in transcriptional and posttranscriptional gene regulation, thereby regulating the occurrence and development of various diseases, including NSCLC [8,25,26]. For example, Zhang et al. demonstrated that the highly expression of circRNA_0001946 was positively correlated with lower TNM stage, less lymph node metastasis [27]. Yang et al. indicated that circRNA_001846 showed 78.2% sensitivity and 81.1% specificity values with an AUC value of 0.872, suggesting that circRNA_001846 might be a potential diagnostic biomarker for NSCLC [28]. Ishola et al. demonstrated that overexpression of circRNA C190 promoted NSCLC by modulating the EGFR/ERK pathway [29]. In the present study, we identified hsa_circ_0003176 showed typical characteristics of circRNAs, which was downregulated in both NSCLC tissues and cell lines. Functionally, we demonstrated that overexpression of hsa_circ_0003176 suppressed the proliferation, migration, and invasion of NSCLC cells in vitro and inhibited NSCLC tumor growth and metastasis in vivo. These results suggested that hsa_circ_0003176 acted as a tumor suppressor gene in NSCLC progression.
It has been reported in the literature that circRNAs can act as miRNA sponges to regulate the expression levels of miRNAs and downstream target genes [30,31]. In NSCLC, several studies indicated that circRNA could act as sponge of miRNA that regulated the downstream gene expression. For instance, circ-MYBL2 inhibited cell proliferation and apoptosis by sponging oncogenic miR-28 in NSCLC [32]. hsa_circ_0005909 was proved to promote NSCLC growth, metastasis, and drug resistance by regulating SOX4 expression via sponging miR-338-3p [33]. In the present study, we proved that hsa_circ_0003176 regulated RBM5 expression via sponging miR-182-5p. Zhang et al. showed that miR-182-5p was significantly increased in NSCLC tumor tissues and was correlated with tumor size and cell proliferation, suggesting that miR-182-5p played an oncogenic role in NSCLC [34]. Yang et al. found that miR-182-5p was upregulated in the NSCLC tissues and promoted the NSCLC metastasis and epithelial-mesenchymal transition (EMT) by regulating EPAS1 [18]. In consistent with these results, we demonstrated that hsa_circ_0003176 suppressed the proliferation, migration, and invasion of NSCLC cells by downregulating miR-182-5p expression.
In addition, it has been demonstrated that RBM5 acts as tumor suppressor in different cancers. Compared with normal tissues, RBM5 was significantly lower in medulloblastoma and RBM5 overexpression repressed cell proliferation and migration in vitro [35]. Zilun et al. demonstrated that long noncoding RNA LINP1 (LncRNA LINP1) inhibited apoptosis and promoted proliferation by repressing RBM5 in gastric cancer cells, suggesting that RBM5 plays a tumor suppressor role in gastric cancer [36]. Jamsai et al. and Sutherland indicated that act as a tumor suppressor in the lung by decreasing adenocarcinoma nodule numbers and tumor size [20,37]. Su et al. showed that RBM5 overexpression improves the antitumor effect by inducing autophagy in human lung adenocarcinoma cells [38]. In consistent with these findings, we demonstrated that hsa_ circ_0003176 suppressed the proliferation, migration, and invasion of NSCLC cells by upregulating RBM5 expression. Taken together, we demonstrated that hsa_circ_0003176 suppresses the progression of NSCLC via regulating miR-182-5p/RBM5 axis.
In this study, our results demonstrated that hsa_circ_ 0003176 inhibited the growth and metastasis of NSCLC in vitro and in vivo and elucidated that hsa_circ_0003176 regulated the malignant biological functions of NSCLC cells by targeting the expression of RBM5 through sponge adsorption of miR-182-5p in vitro, but hsa_circ_0003176 inhibiting NSCLC progression by regulating miR-182-5p/ RBM5 axis has not been verified in animals. In the next step, we will further validate this result in animals, expand clinical samples to verify the clinical significance of hsa_circ_ 0003176/miR-182-5p/RBM5 signaling axis, and explore the underlying molecular mechanism of hsa_circ_0003176 dysfunction in NSCLC.

Conclusion
Our results indicated that hsa_circ_0003176 suppressed the progression of NSCLC via regulating RBM5 expression by sponging miR-182-5p. These findings provided potential therapeutic targets to develop new strategy for NSCLC treatment.

Conflicts of Interest
The authors declare that there is no conflict of interest regarding the publication of this paper.

Authors' Contributions
Fangfang Yang was responsible for the idea and project design and was involved in the manuscript writing, editing, approval, and decision to publish; YanLi Pei performed the clinical sample detection and data collection; Wei Xu performed the experimental data analysis and literature search; Lei Rong provided fund collection. All authors read and approved the final manuscript.