Hsa_Circ_0001947/MiR-661/DOK7 Axis Restrains Non-Small Cell Lung Cancer Development

Hsa_circ_0001947 is associated with multiple cancers, but its function in non-small cell lung cancer (NSCLC) is ambiguous and needs further research. The targeting relationship among circ_0001947, miR-661, and downstream of tyrosine kinase 7 (DOK7) was predicted by database and further verified by dual-luciferase reporter assay, while their expressions in cancer tissues and cells were detected by quantitative real-time polymerase chain reaction (qRT-PCR). After transfection, cell biological behaviors and expressions of miRNAs, miR-661 and DOK7 were determined by cell function experiments and qRT-PCR, respectively. Circ_0001947 was low-expressed in NSCLC tissues and cells. Circ_0001947 knockdown intensified cell viability and proliferation, induced cell cycle arrest at S phase, suppressed apoptosis and evidently enhanced miR-510, miR-587, miR-661 and miR-942 levels, while circ_0001947 overexpression did the opposite. MiR-661 was a target gene of circ_0001947 that participated in the regulation of circ_0001947 on cell biological behaviors. Furthermore, DOK7, the target gene of miR-661, partly participated in the regulation of miR-661 on cell viability. Hsa_circ_0001947 acts as a sponge of miR-661 to repress NSCLC development by elevating the expression of DOK7.

Based on the above evidence, we aimed to figure out the influence of the circ_0001947/miR-661 axis on the biological behaviors of NSCLC cells and to explore whether circ_0001947 and miR-661 have complementary binding sites to determine potential therapeutic targets for NSCLC.

Ethics Statement
We obtained NSCLC tissues and adjacent non-tumor tissues from 40 patients who had been diagnosed and operated in the Sanmen People's Hospital between March 2018 and July 2019. All participants signed written informed consent. The present research was ratified by the Ethics Committee of the Sanmen People's Hospital (2018020039).
In this research, tissue specimens were acquired via surgical resection and snap-frozen in a -80 o C refrigerator (TSE600V, Thermo Scientific, USA). We used the quantitative real-time polymerase chain reaction (qRT-PCR) to examine the circ_0001947, miR-661 and DOK7 expressions.

Cell Processing
First, to fathom out the role of circ_0001947 on cells, the cells were assigned into the control group, NC group, circ_0001947 group, si-NC group, and si-circ_0001947#2 group. In the last four groups, cells were transfected with NC, overexpressed circ_0001947, si-NC and si-circ_0001947#2, respectively. Second, to ascertain the functions of si-circ_0001947#2 and miR-661 I on cells, the cells were separated into the control group, si-NC + IC group (co-transfection with si-NC and miR-661 IC), si-circ_0001947#2 + IC group (co-transfection with si-circ_0001947#2 and miR-661 IC), si-circ_0001947#2 + I group (co-transfection with si-circ_0001947#2 and miR-661 I), and si-NC + I group (co-transfection with si-NC and miR-661 I). Third, to investigate the functions of overexpressed circ_0001947 and miR-661 M on cells, cells were divided into the control group, NC + MC group (co-transfection with NC and miR-661 MC), circ_0001947 + MC group (co-transfection with overexpressed circ_0001947 and miR-661 MC), circ_0001947 + M group (co-transfection with overexpressed circ_0001947 and miR-661 M), and NC + M group (co-transfection with NC and miR-661 M). Fourth, to comprehend the roles of miR-661 I and siDOK7 on cells, cells were classified into the IC + siNC group (co-transfection with miR-661 IC and siNC), I + siNC group (co-transfection with miR-661 I and siNC), I + siDOK7 group (co-transfection with miR-661 I and siDOK7), and IC + siDOK7 group (co-transfection with miR-661 IC and siDOK7). Fifth, to investigate the functions of miR-661 M and overexpressed DOK7 on cells, cells were divided into the MC + NC group (co-transfection with miR-661 MC and NC), M + NC group (co-transfection with miR-661 M and NC), M + DOK7 group (co-transfection with miR-661 M and DOK7), and MC + DOK7 group (co-transfection with miR-661 MC and overexpressed DOK7).

qRT-PCR
RNA from lung tissues and cells was acquired using the RNA Extraction Kit (N065, China). Next, to quantify the mRNA expression, cDNA synthesis was conducted in line with instructions of the cDNA first chain synthesis kit (N118, China) and miRNA cDNA first-strand synthesis kit (KR211, China), respectively. Then, cDNA was amplified and quantitatively analyzed by the FastFire qPCR PreMix (FP207-02, Tiangen) in a C1000 Touch PCR system (E1138, Bio-Rad, USA). GAPDH and U6 were employed for normalization controls. The results were normalized by the 2 -ΔΔCt method [20]. The primers were listed in Table 1.

Cell Counting Kit-8 (CCK-8)
We used the CCK-8 kit (abs50003, absin, China) to assess cell viability. Based on the above groups, cells (1 × 10 4 per well) were transfected and then maintained in an incubator for a specified time (24 or 48 h). Next, 10 μl CCK-8 solution was adopted to treat cells for 4 h. Then, the optical density at a wavelength of 450 nm was determined under a microplate reader (Z742711, Sigma, USA).

Apoptosis Analysis
The measurement of cell apoptosis was performed using the Annexin V-FITC/propidium iodide (PI) kit (C1062S, Beyotime, China). Cells (5 × 10 4 per well) were treated according to the above grouping conditions. After treatment, 195 μl binding buffer was added to the cell suspensions. Thereafter, 5 μl Annexin V-FITC and 10 μl PI were added to the cell suspensions and mixed well. The mixture was placed in an incubator at 37 o C for 15 min (min). Finally, we used a flow cytometer (DxFLEX, Beckman, USA) to examine the FITC and PI signals.

Colony Formation Assay
In this research, the cells in good growth condition were made into a single-cell suspension. Cell suspension (1 × 10 4 cells/ml) was seeded into a 6-well plate and then transfected based on the above groups. After incubation for 14 days, we used the 4% paraformaldehyde (P0099, Beyotime) and Giemsa dye solution (G8220, Solarbio, China) to fix and stain cells, respectively. Finally, we used a microscope (Z723975-1EA, Sigma) to count the clone numbers of the cells.

Cell Cycle Assay
Cell Cycle and Apoptosis Analysis Kit (C1052, Beyotime) was used in this research. After treatment, cells (1 × 10 5 cells/ml) were collected and centrifuged. After being washed, the cells were immobilized by 70% cold ethanol solution (E111991, Aladdin, China) at 4 o C overnight. Then, the cells were washed and reacted with prepared 0.5 ml PI stain solution at 37 o C in the dark for 30 min. Finally, the percentage of NSCLC cells at different stages of cell cycle was examined using flow cytometer.

Statistical Analysis
Data were gathered and analyzed by GraphPad Prism 8.0 (GraphPad Software Inc., USA). The measurement data were expressed as mean ± standard deviation (SD). The levels of circ_0001947, miR-661, and DOK7 in 40 NSCLC tissues and adjacent non-tumor tissues were analyzed by paired-samples t test. An independent sample t test was employed to compare the difference of two groups. The multiple group comparisons were completed through a one-way analysis of variance (ANOVA) followed by Tukey's post hoc test. p < 0.05 was considered as statistically significant.

Circ_0001947 Was Low-Expressed in NSCLC Tissues and Cells and Si-Circ_0001947#2 Intensified the NSCLC Cell Viability
To explore the level of circ_0001947 in NSCLC tissues or cell lines, we checked the data from qRT-PCR and clarified that circ_0001947 expression was lower in NSCLC tissues than that in adjacent non-tumor tissues (Fig. 1A), while the same trend was discovered in NSCLC cells when compared with that in 16HBE cells (Fig. 1B). Since circ_0001947 expression was notably down-regulated in H1299 cells and slightly restrained in A549 cells relative to other NSCLC cells, H1299 and A549 cells were singled out for subsequent experiments. Next, we transfected si-circ_0001947#1, si-circ_0001947#2, circ_0001947 overexpression plasmids, or their NCs into Table 1. Gene sequence primers.

Si-circ_0001947#2 Intensified Cell Proliferation, Induced Cell Cycle Arrest at S Phase and Restrained Apoptosis, while Overexpressed circ_0001947 did the Opposite
The flow cytometer and colony formation assay exhibited that si-circ_0001947#2 markedly restrained apoptosis and enhanced cell proliferation, while overexpressed circ_0001947 generated the inverse effects (Figs. 2A-2D). Moreover, si-circ_0001947#2 induced cell cycle arrest at S phase whereas circ_0001947 overexpression facilitated the entry of G0/G1 phase but restrained the entry of G2/M and S phases (Figs. 2E and 2F).

MiR-661 Was Overexpressed in NSCLC Tissues and Cells, and Participated in the Regulation of Circ_0001947 on Cell Viability
Our study revealed that miR-661 was overexpressed in NSCLC tissues and cells (Figs. 4A and 4B). Furthermore, we explored the role of miR-661 in the regulation of circ_0001947 on the viability of H1299 and A549 cells. Interestingly, the miR-661 level was intensified by si-circ_0001947#2 but was weakened by miR-661 I, while the above effects were overturned by co-transfection of si-circ_0001947#2 and miR-661 I (Fig. 4C). Co-transfection of overexpressed circ_0001947 and miR-661 M reversed the regulation of miR-661 M / overexpressed circ_0001947 on the miR-661 level (Fig. 4D). Moreover, cell viability was elevated by si-circ_0001947#2 but was inhibited by miR-661 I, the two tendencies were reversed by co-transfection of si-circ_0001947#2 and miR-661 I (Fig. 4E). However, co-transfection of overexpressed circ_0001947 and miR-661 M partially offset the regulation of miR-661 M / overexpressed circ_0001947 on cell viability (Fig. 4F).

MiR-661 Was Involved in the Regulation of circ_0001947 on Apoptosis, Proliferation, and Cell Cycle of NSCLC Cells
As delineated in Fig. 5A, the results unveiled that apoptosis was largely elevated by miR-661 I but was suppressed by si-circ_0001947#2, which was mitigated by co-transfection of si-circ_0001947#2 and I. Meanwhile, an increase in apoptosis induced by overexpressed circ_0001947 or a decrease in apoptosis triggered by miR-661 M was reversed by co-transfection of overexpressed circ_0001947 and miR-661 M (Fig. 5B). Then, we also proved that co-transfection of si-circ_0001947#2 and miR-661 I neutralized the inhibition of miR-661 I and the promotion of si-circ_0001947#2 on cell proliferation (Fig. 5C). Moreover, co-treatment of overexpressed circ_0001947 and miR-661 M partially offset the regulation of miR-661 M / overexpressed circ_0001947 on cell proliferation (Fig. 5D). The detection of cell cycle confirmed that si-circ_0001947#2 and miR-661 M induced cell cycle arrest at S phase, while miR-661 I and overexpressed circ_0001947 facilitated the entry of G0/G1 phase (Figs. 5E and 5F). Besides, the effect of si-circ_0001947#2 or miR-661 I on cell cycle was partially reversed by cotransfection of si-circ_0001947#2 and I , while the effect of overexpressed circ_0001947 or miR-661 M was partially reversed by co-transfection of overexpressed circ_0001947 and miR-661 M (Figs. 5E and 5F).

MiR-661 Targeted DOK7
We used TargetScan v7.2 analysis to discover that miR-661 targeted DOK7 (Fig. 6A). Additionally, the relative luciferase activity was lower in the miR-661 M group than that in the miR-661 MC group when the sequence of DOK7 was wild type (Figs. 6B and 6C), while there was no obvious difference between miR-661 M and MC groups when the sequence of DOK7 was mutant.

DOK7 Was Partly Involved in the Regulation of miR-661 on NSCLC Cell Viability
To figure out the role of DOK7 in the regulation of miR-661 on H1299 and A549 cells, we measured the alteration of DOK7 expression. Co-transfection of miR-661I and siDOK7 partially reversed the regulation of miR-661 I or siDOK7 on DOK7 expression (Fig. 8A). In the meantime, the effect of miR-661 M on restraining DOK7 expression was discovered to be mitigated by overexpressed DOK7 (Fig. 8B). Besides, the promotion of cell viability induced by siDOK7 and the inhibition of that induced by miR-661 I were reversed by co-transfection of miR-661 I and siDOK7 (Fig. 8C). The enhancement of miR-661 M on cell viability was offset by overexpressed DOK7 (Fig. 8D).

Discussion
In our study, we tested the expression of circ_0001947 in NSCLC tissues, adjacent non-tumor tissues, and multiple NSCLC cell lines. Subsequently, we observed changes in cell viability, proliferation, apoptosis, and miRNAs expressions after circ_0001947 overexpression or silencing in H1299 and A549 cells. On this basis, we further verified the interactions between circ_0001947 and miR-661 as well as between miR-661 and downstream target genes, suggesting that miR-661 and downstream target genes might be the mediators of circ_0001947 in regulating the biological behaviors of NSCLC cells. Our findings revealed the role of circ_0001947 in modulating certain vital features of NSCLC.
First of all, we have clarified that circ_0001947 was low-expressed in NSCLC tissues and cell lines. Moreover, early research has demonstrated that circ_0001947 facilitates gastric cancer progression via modulating the miR-6894-5p/ANTXR1 axis [21]. Han et al. has put forward that circ_0001947 restrains acute myeloid leukemia development by modulating the miR-329-5p/CREBRF axis [12]. Our study complemented the research deficiency of circ_0001947 in NSCLC, which expounded for the first time that circ_0001947 overexpression inhibited cell viability and proliferation, induced cell cycle arrest at G0/G1 phase, and promoted apoptosis, while circ_0001947 knockdown exerted the opposite effects, manifesting the involvement of circ_0001947 in the progression of NSCLC.
Additionally, by searching the online database CircInteractome, we detected miR-510, miR-587, miR-661, and  playing crucial roles as "miRNA sponges" in the "circRNA-miRNA-mRNA" axis [22]. In our research, to probe into potential downstream pathways that may be implicated in the function of circ_0001947 in NSCLC, we performed database analysis and finally discovered that miR-661 may be a potential target of circ_0001947 which was further supported by the dual-luciferase reporter assay. MiR-661 has been confirmed to be abnormally expressed in a variety of malignant tumors and closely associated with the initiation and progression of tumors [23,24]. Wang, et al. has disclosed that miR-661 is overexpressed in NSCLC tissues and cells and it accelerates NSCLC development via targeting RUNX3, which is consistent with findings of our research [25]. Liu et al. has proposed that miR-661 facilitates the invasion and metastasis of NSCLC cells via restraining RB1 [26]. Then, we provided the first demonstration that miR-661 inhibitor suppressed cell viability and proliferation, induced cell cycle arrest at G0/G1 phase, and elevated apoptosis, while the above effects were overturned by co-transfection of circ_0001947 knockdown and miR-661 inhibitor. It was clarified that miR-661 might serve as a regulator in the regulation of circ_0001947 on the cell biological behaviors of NSCLC cells.
Furthermore, in order to explore the target of miR-661 in NSCLC, we used database prediction software to predict the target gene of miR-661 and confirmed that DOK7 might be a potential target of miR-661. Luciferase assay corroborated that miR-661 could directly bind to 3'-UTR of DOK7. The DOK7 gene, a member of the DOK family, is located at the end of the short arm of chromosome 4 [27]. Recent studies have revealed that the DOK7 gene and its family are closely related to the initiation and development of many malignant tumors, such as lung cancer, breast cancer, and leukemia [28][29][30], but the detailed mechanism has not been fully understood. At present, there are few studies on the specific role of DOK7 in NSCLC. Only one study confirmed that DOK7 was greatly under-expressed in lung cancer, and DOK7 overexpression suppressed the proliferation and migration of lung cancer cells [27]. Consistent with the previous study, our data proved that DOK7 was down-expressed in NSCLC tissues and cells, the expression of which was discovered to be elevated by circ_0001947 overexpression through targeting miR-661. Meanwhile, DOK7 was partly involved in the regulation of miR-661 on NSCLC cell viability. It suggested that circ_0001947-modulated miR-661 could further modulate the upregulation of DOK7 in NSCLC cells.
However, despite the evidence that DOK7 is the target gene of miR-661, the upstream and downstream pathways of DOK7 are not tested, which may be a defect of this study. In addition, this study lacks in vivo experiments, which will be listed in our future experimental projects, so as to further improve the accuracy of this research.
In conclusions, altogether, this research emphasized that circ_0001947 acted as a sponge of miR-661 to restrain NSCLC progression via modulating DOK7, revealing that circ_0001947 might be a potential new biomarker for the treatment of NSCLC.