Hsa_circ_0070661 inhibits cancer progression through miR-556-5p/TEK axis in lung adenocarcinoma

Abstract

BACKGROUND:

Lung adenocarcinoma (LUAD) has a high incidence and poor prognosis, and multiple circRNAs (circRNAs) have been found to regulate LUAD.

OBJECTIVE:

This study focuses on the effect and mechanism of hsa_circ_0070661 in LUAD.

METHODS:

LUAD tissues and para-cancerous tissues were collected from 38 patients diagnosed with LUAD in our hospital. Hsa_circ_0070661, miR-556-5p and TEK Receptor Tyrosine Kinase (TEK) levels were evaluated using western blotting and RT-qPCR, and the targeting relationship was detected by luciferase reporter and RIP assays. Cell migration, viability, apoptosis-related proteins, (Bcl-2 and Bax) and tumor growth in vivo were assessed by Transwell, CCK-8, western blotting and xenograft assays, respectively.

RESULTS:

Results indicated downregulation of hsa_circ_0070661 and TEK in LUAD cell lines and tissues, whereas upregulation of miR-556-5p. Hsa_circ_0070661 upregulation restrained the viability, migration and tumor growth of LUAD cells, and promoted apoptosis. Hsa_circ_0070661 could directly target miR-556-5p to upregulate TEK expression in LUAD. MiR-556-5p upregulation promoted the malignant phenotypes of LUAD cells and reversed the anti-cancer effect of hsa_circ_0070661 overexpression, while TEK upregulation inhibited LUAD progression and somewhat eradicated the cancer-promoting effect of miR-556-5p upregulation.

CONCLUSIONS:

Hsa_circ_0070661 sponges miR-556-5p to inhibit LUAD development via regulating TEK, providing a promising molecular target for LUAD clinical therapy.

1. Introduction

Lung cancer is the highly widespread cancer, accounting for more than 18% of global cancer-related deaths [1, 2]. Lung adenocarcinoma (LUAD) is the most common histological type of non-small cell lung cancer (NSCLC), a major category of lung cancer that accounts for 85% of all lung cancers, accounting for approximately 40% of all lung cancers [3, 4]. A complete diagnostic system consisting of mediastinoscopy, sputum cytology, radionuclides, bronchoscopy, and X-rays can identify patients with LUAD to a large extent [5]. However, approximately 70% of patients are already in stage III-IV at the time of diagnosis of LUAD, resulting in a 5-year survival rate of less than 20% [6, 7]. In addition, traditional chemoradiotherapy has a tendency to fail in the treatment of LUAD due to the development of drug resistance [2]. To better understand the pathophysiology of LUAD and enhance its treatment efficacy, it may be useful to identify efficient molecular targets.

A unique endogenous RNA called circular RNAs (circRNAs) is created by a non-classical reverse splicing process [8]. Its lengthy half-life and resistance to ribonuclease digestion are caused by the ability of its downstream splice donor sites to covalently bind to the upstream splice acceptor sites [8, 9]. CircRNA has been shown to be a rich and conserved RNA and is generally expressed in a cell type, complex tissue or stage-specific manner [10, 11]. Current research has supported circRNA’s crucial biological roles, particularly as a predictive biomarker for human cancer [12, 13, 14]. For example, hsa_circ_0000792 is highly expressed in LUAD and has significant diagnostic accuracy for LUAD [15]. High circXPO1 expression was related with poor survival of LUAD [16]. Hsa_circ_0070661 is located on chr4 :110580166-110608872, and its related gene symbol is CCDC109B. There is no research to prove its effect on human diseases. Therefore, the hsa_circ_0070661 function and molecular mechanism in LUAD need to be preliminary investigated.

MicroRNAs (miRNAs) are short non-coding RNAs of about 18–25 nucleotide lengths that play a wide and important role in biological process regulation together with apoptosis, migration, proliferation, and differentiation [17]. MiRNAs have been found to be sponges of circRNAs and interact with seed sequences on the 3’-untranslated region (3’ UTR) of the target mRNAs [18]. Notably, a growing number of molecular mechanisms based on circRNA-miRNA-mRNA have been shown to play an significant regulatory role in LUAD. For instance, hsa_circ_0008274 overexpression inhibits LUAD cells invasion and proliferation by down-regulating miR-578 and activating HMGA2 [19]. CircFBXW7 repress the migration and proliferation of LUAD cell lines by controlling miR-942-5p/BARX2 axis [20]. MiR-556-5p has been widely reported to participate in the development and incidence of human malignant tumors, inhibits breast cancer (BC) [21], colorectal cancer [22] and meningioma [23], while promotes prostate cancer [24]. Importantly, Feng et al. [25] suggested that miR-556-5p was heavily expressed in NSCLC tissues, and its knockdown reduced cancer tumorigenesis and cell viability, and persuaded apoptosis. In this paper, the binding of hsa_circ_0070661 to miR-556-5p was found through the target prediction website. Therefore, we examined the association among miR-556-5p and hsa_circ_0070661 in LUAD.

TEK receptor Tyrosine kinase (TEK) also known as TIE2 is an endothelial tyrosine kinase receptor of vascular growth factor responsible for angiogenesis [26]. According to previous studies, TEK could regulate endothelial cell proliferation, migration, survival and inflammation by binding to Angiopoietin [27, 28]. Due to the malignant proliferation and migration characteristics of tumor cells, the role of TEK in cancer attracts many researchers’ attention. For instance, Yang et al. identified that TEK knockdown interacted with SIRT7 could reduce adriamycin-dependent breast cancer cell malignancy [29]. TEK expression was found to be downregulated in papillary thyroid carcinoma, and its knockdown also could decrease the proliferation and migration of papillary thyroid carcinoma cells [30]. However, TEK was identified as an anticancer in LUAD, which could be regulated by miR-486-5p [31] or miR-19a-3p [32]. The previous study on TEK in LUAD suggested that TEK could be regulated by different miRNAs to play the anticancer role in LUAD. This study found that TEK was a target gene of miR-556-5p. So far, the relationship between TEK and miR-556-5p in LUAD is still not be explored.

Here, we inspected the expression of hsa_circ_ 0070661 in LUAD as well as its effect LUAD cells biological process of in vitro as well as in vivo. Additionally, we focused on the specific regulatory mechanism of hsa_circ_0070661 acting on miR-556-5p/TEK Receptor Tyrosine Kinase (TEK) axis. The findings of this study should offer useful tools for LUAD diagnosis and therapy.

2. Materials and methods

2.1. Sample Tissues

This study included 38 cases of LUAD patients treated in our hospital between January 2021 and January 2022 LUAD tissue and corresponding para-cancerous tissue were collected during operation and stored in liquid nitrogen. The inclusion criteria were as follows: 1) All patients were diagnosed with LUAD by three pathologists; 2) All patients did not diagnose with other diseases; 3) All patients or their families signed informed consent forms. The exclusion criteria were as follows: 1) The patients had LUAD history; 2) The patients underwent radiation, chemotherapy or other treatments; 3) The characteristics of some patients were lost. Table 1 listed the characteristics of all patients. All participants and their families signed informed consent forms after being informed of the study. Additionally, this study adhered to the Declaration of Helsinki and was carried out with endorsement from our hospital’s Ethics Committee.

Table 1.

The characteristics of all patients with LUAD

Characteristics	All patients ($n=$ 38)
Age
56	18
$⩾$ 56	20
Gender
Male	24
Female	14
Smoking status
Current	7
Former	16
Never	5
Tumor stage
I	13
II	11
III	8
IV	6
Tumor differentiation
Well	17
Moderate	14
Poor	7

2.2. Cell culture

H1650 and A549 two cell lines for LUAD and BEAS-2B (the human bronchial epithelial cell line) acquired from ATCC (USA), and one LUAD cell line (PC9) acquired from Sigma-Aldrich (USA) were incubated in DMEM (Thermo Fisher Scientific, USA) containing 10% FBS (HyClone, USA) and 1% penicillin/streptomycin (Gibco, USA). Cells were cultured at 37${}^{\circ }$C in a 5% CO${}_2$ incubator. Cell passaging was carried out when cell fusion reached 80%. The logarithmic stage cells of more than 3 generations were selected for subsequent experiments.

2.3. Cell transfection

H1650 and A549 cells were selected to establish stable transfection cell lines. Overexpression vector loading hsa_circ_0070661 and TEK (circ-OE and TEK-OE; 2 $\mu$g/mL; GenePharma, China) or 2 $\mu$g/mL empty vector as negative control (NC); miR-556-5p mimic or mimic-NC (75 nM; SwitchGear Genomics, USA) was transfected into PC9 as well as H1975 cells via Lipofectamine™2000 transfection reagent (Invitrogen, USA). Finally, the transfection efficiency was detected by RT-qPCR after transfection for 48 hours at 37${}^{\circ }$C in a 5% CO${}_2$ incubator.

2.4. RT-qPCR assay

Total RNA isolated by RNAiso Plus (TAKARA, Japan) and miRNA extracted by miRNEasy EFPE Kit (BioTeke, China) were treated with ultraviolet spectrophotometer and agaLUADose gel electrophoresis to determine their concentration, purity and integrity. The transcription was then performed by means of the TaqMan reverse transcription kit (ABI, USA) and the TaqMan miRNA reverse transcription Kit (ABI). RT-qPCR was shown by means of Hieff qPCR SYBR Green Master Mix kit (Yeasen, China) on ABI 7500 Rapid RT-PCR system (ABI). The RT-qPCR conditions were as follows: 95${}^{\circ }$C for 5 min, as well as 40 cycles of 95${}^{\circ }$C for 10 s and 60${}^{\circ }$C for 30 s. The data were analyzed by 2${}^{-\mathrm{\Delta }\mathrm{\Delta }\text{𝐶𝑡}}$[33], and normalized against either U6 for miR-556-5p and GAPDH for hsa_circ_0070661 and TEK. Table 2 lists all of the primers that were utilized in this research.

Table 2.

Real-time PCR primer synthesis list

Gene		Sequences
hsa_circ_0070661	Forward	5’-CCCATGGTTCATTTTGATTG-3’
	Reverse	5’-AGGTGGCACCACGGTACTAT-3’
miR-556-5p	Forward	5’-AT AGT AAT AAG AAA GAT GAG-3’
	Reverse	5’-TGT TGA AGG TAG TAA TAA AAA-3’
TEK	Forward	5’-CCAGCCCTGCTGATACCAAA-3’
	Reverse	5’-ATGTGCATGAGGTCCCAAGG-3’
U6	Forward	5’-CTCGCTTCGGCAGCACA-3’
	Reverse	5’-AACGCTTCACGAATTTGCGT-3’
GADPH	Forward	5’-AGAAAAACCTGCCAAATATGATGAC-3’
	Reverse	5’-TGGGTGTCGCTGTTGAAGTC-3’

2.5. Subcellular localization and RNase R tolerance test

Hsa_circ_0070661 levels in subcellular site of H1650 and A549 cells was located via a PARI kit (Norgenbiotek, Canada) and quantified by RT-qPCR. Briefly, subcellular localization was performed in 1$\times$10${}^4$ cells. Firstly, the collected cells (1$\times$10${}^4)$ were resuspended in cell separation buffer, followed by placed for 10 min on ice. Formerly, the samples were centrifuged, the nuclear precipitate and supernatant were obtained and preserved in cell disruption buffer. For the normalization, U6 as well as GAPDH were employed as the nuclear and cytoplasmic control.

Total RNA in H1650 and A549 cells acquired as described in RT-qPCR assay was utilized for RNase R tolerance test. In brief, qualified total RNA was divided into RNase R group and control group, treated with or without RNase R (3 U/$\mu$g; Guangzhou Geneseed Biotechnology, China). The two groups were incubated for 30 mins at 37${}^{\circ }$C, and hsa_circ_0070661 and its linear transcript CCDC109B levels via a RT-qPCR assay were determined.

2.6. CCK-8 assay

H1650 and A549 cells after transfection for 48 h were transferred to a 96-well plate (4$\times$10${}^3$ cells per well). After that, 10 $\mu$L of CCK-8 solution (TransGen, China) was pipetted to each well at 0, 24, 48, and 72 h, followed by another two hours of incubation at 37${}^{\circ }$C in a 5% CO${}_2$ incubator. The absorbance values were read at a 450 nm wavelength with the help of automatic microplate reader (Bio-Rad, USA).

2.7. Cell migration assay

H1650 and A549 cells after transfection for 48 h were suspended in serum-free medium (5$\times$10${}^5$ cells/mL), and seeded into the upper compartments (200 $\mu$L) of each Transwell insert (Corning, USA). Besides, 600 $\mu$l of complete medium was filled into the lower chamber. After 24 h incubation at 37${}^{\circ }$C in a 5% CO${}_2$ incubator, the migratory cells were immobilized and visualized with 4% paraformaldehyde and crystal violet (0.5%) for 15 min. Afterward, the cells images were taken and added up with an optical microscope (Olympus, Japan).

2.8. Western blotting

Total cell protein was lysed with RIPA buffer (Thermo Scientific, USA), and protein concentration were quantified by BCA kit (Thermo Scientific). The proteins were then separated in equal quantities using 10% SDS-PAGE and transferred to a PVDF membrane (Millipore, USA), which was sealed in 5% skim milk at 37${}^{\circ }$C for 1 h. Subsequently, Abcam (USA) primary antibodies [TEK (1:1000, ab86002), Bax (1:1000, ab32503), Bcl-2 (1:1000, ab32124) and GAPDH (1: 1000, ab8245)] were incubated with PVDF membranes for 12 h at 4${}^{\circ }$C. Then membranes were incubated with HRP labeled secondary antibody (1:50,000, ab6721, Abcam) at 37${}^{\circ }$C for 1 h. Finally, protein signals were visualized with the help of ECL substrates (Millipore). Finally, the image was collected with acquired using a LAS-4000UVmini image analyzer (Fujifilm, Japan) and the gray value was analyzed in Quantity One software (Bio-Rad, USA).

2.9. Xenograft tumor model

Ten BALB/c nude mice, aged 5 weeks and weighing 18–22 g, were purchased from Hunan SJA Laboratory Animal (China). A549 cells in circ-OE or empty vector groups were prepared into cell suspension, and each nude mouse was subcutaneously injected with 200 $\mu$L cell suspension (1$\times$10${}^6$ cells) to induce xenograft tumor. Tumor volume was recorded weekly to generate tumor growth curves. Six weeks after injection, the mice were euthanized by intravenously-administered pentobarbital and the tumors were taken out and weighed. The Animal Care Committee of our hospital has approved all procedures for this animal study.

2.10. Luciferase reporter assay

CircInteractome (https://circinteractome.nia.nih.gov/ index.html) [34] was used to predict the binding sites between hsa_circ_0070661 and miR-556-5p, whereas TargetScan (http://www.targetscan.org/vert_71/) [35] was used to predict the binding sites between TEK and miR-556-5p. Full length sequence of Hsa_circ_0070661 and TEK 3’-UTR with miR-556-5p complementary sequence were cloned into pmirGLO vector (Promega, USA) to generate hsa_circ_0070661 and TEK wild type (WT). On the other hand, mutant luciferase reporter vectors (MUT) were constructed by mutating the miR-556-5p binding sequence with the help of QuickMutation Site-directed Mutagenesis Kit (Beyotime, China). Subsequently, hsa_circ_0070661/TEK-WT/MUT were transfected into H1650 and A549 cells together with miR-556-5p mimic or miR-NC. Lipofectamine 2000 transfection reagent (Invitrogen) was used for transfection. After 48 h, we assessed luciferase activity with the mean of dual-luciferase reporting (DLR) kit form Promega.

2.11. RIP assay

This experiment was conducted using a Magna RIP (RNA immunoprecipitation) RNA-binding protein immunoprecipitation kit (Millipore) under strict kit instructions. Magnetic beads were incubated with anti-Ago2 or IgG antibody (Millipore) in a rotator at 25${}^{\circ }$C for 30 min. Protein samples from H1650 and A549 cells lysed by RIPA lysis buffer were exposed to pretreated magnetic beads, and then maintained overnight at 4${}^{\circ }$C. Finally, the beads were washed, and then miR-556-5p and hsa_circ_0070661 levels were evaluated by RT-qPCR.

2.12. Statistical analysis

SPSS (version 20.0, IBM, USA) was utilized for the statistical analysis of the data. Continuous data from all experiments performed in triplicates were in the format of mean $\pm$ standard deviation (SD). Student’s $t$-test was employed in testing contrasting the two groups’ values. For multiple groups evaluation, a one-way ANOVA with Tukey’s test was applied. Pearson test was used for correlation analysis among miR-556-5p, hsa_circ_0070661 and TEK. 0.05 implies a remarkable difference.

3. Results

3.1. Hsa_circ_0070661 was lowly expressed in LUAD

To evaluate the abnormal hsa_circ_0070661 expression in LUAD, RT-qPCR was performed. The results showed that the expression level of hsa_circ_0070661 in LUAD tissues was 60% lower than that in adjacent tissues (Fig. 1A), and the expression level of hsa_circ_0070661 was reduced with the increase of tumor stage (Fig. 1B). Similarly, compared with BEAS-2B cells, hsa_circ_0070661 was uniformly down-regulated in all three kinds of LUAD cell lines (H1650, PC9 and A549), and H1650 and A549 showed lower hsa_circ_0070661 expression (Fig. 1C). In addition, to determine the subcellular localization of hsa_circ_0070661, we also examined its relative abundance in the nucleus and cytoplasm. It was found that the proportion of hsa_circ_0070661 in the cytoplasm of H1650 and A549 cells was greater in the nucleus (Fig. 1D). This suggests that hsa_circ_0070661 is mainly located in the cytoplasm. In addition, an RNase R tolerance test was performed, and observed that hsa_circ_0070661 remained after RNase R treatment (Fig. 1E).

Figure 1.

Hsa_circ_0070661 was lowly expressed in LUAD. (A) Relative levels of hsa_circ_0070661 in LUAD tumors and adjacent normal tissue were assessed via RT-qPCR. (B) Relative levels of hsa_circ_0070661 in different tumor stages were assessed via RT-qPCR. (C) Relative hsa_circ_0070661 expression in a human bronchial epithelial cell line (BEAS-2B) and LUAD cell lines (PC9, H1650 and A549). 0.001 vs. BEAS-2B. (D) Subcellular fractionation assay was used to detect transcript abundance of hsa_circ_0070661 in cytoplasm and nucleus of H1650 and A549 cells. (E) Relative hsa_circ_0070661 and its linear transcript (CCDC109B) expression in the RNase R-treated group was ascertained via RT-qPCR analysis. 0.001 vs. Control. (F) Relative levels of hsa_circ_0070661 in H1650 and A549 cells transfected with either Empty vector and circ-OE, as assessed via RT-qPCR. 0.001 vs. Empty vector.

In addition, overexpression vector of hsa_circ_0070661 was transfected into LUAD cells. As shown in Fig. 1F, the expression level of hsa_circ_0070661 in LUAD cells in the circ-OE group was up-regulated by more than 4-fold compared with the empty vector group. These results suggest that hsa_circ_0070661 is down-regulated in LUAD and may play a role in the cytoplasm through the circular structure.

3.2. Hsa_circ_0070661 up-regulation suppressed the migration, proliferation and tumor growth of LUAD cells while facilitated apoptosis

Next, we investigate the role of hsa_circ_0070661 in LUAD cells, the changes of LUAD cells viability and migration were analyzed by CCK-8 and Transwell assay. We observed that the viability of H1650 and A549 cells in the circ-OE group was about 35% and 45% of that in the empty vector group, respectively (Fig. 2A). Furthermore, hsa_circ_0070661 overexpression reduced the migration level of H1650 and A549 cells by 80% and 55%, respectively (Fig. 2B). Moreover, western blotting showed an obvious up-regulation of Bax and down-regulation of Bcl-2 protein levels in the circ-OE group than that in the empty vector group (Fig. 2C). Additionally, we investigated the impact of hsa_circ_0070661 on the progress of LUAD cells in vivo through xenograft tumor in nude mice. Six weeks after the model was established, the tumor size in the circ-OE group was significantly smaller than in the empty vector group, and tumor growth curve and tumor weighing showed that the weight and volume of tumor in the circ-OE group were lessen as compare to the empty vector group (Fig. 2D). These results suggest that up-regulation of hsa_circ_0070661 could destroy the malignant phenotype of LUAD cells in vitro and in vivo.

Figure 2.

Hsa_circ_0070661 up-regulation suppressed the migration, proliferation, and tumor growth of LUAD cells while facilitated apoptosis. (A) Effect of Empty vector and circ-OE transfection on the viability of H1650 and A549 cells, as determined by CCK-8 assay. 0.001 vs. Empty vector. (B) Effect of Empty vector and circ-OE transfection on H1650 and A549 migration, as evaluated by Transwell assay. 0.001 vs. Empty vector. (C) Empty vector and circ-OE transfection’s effect on apoptosis-related protein levels (Bax and Bcl-2), as ascertained via western blotting analysis. 0.05, 0.001 vs. Empty vector. (D) Representative photos of xenograft tumors from each group, 6 weeks after subcutaneous administration of A549 cells and injection with Empty vector and circ-OE ($n=$ 5). Effect of hsa_circ_0070661 overexpression on the volume and weight of tumors in BALB/c nude mice. vs. Empty vector.

3.3. Hsa_circ_0070661 sponged miR-556-5p

Based on the LUAD inhibitory properties of hsa_ circ_0070661 revealed above, we continued to explore the downstream miRNA of hsa_circ_0070661. As shown in Fig. 3A, the binding sites for hsa_circ _0070661 and miR-556-5p were predicted by circInteractome. To verify their binding, we executed a luciferase reporter assay. The luciferase activity of wild type hsa_circ_0070661 (WT) vectors could be inhibited by miR-556-5p mimic, but the luciferase activity of MUT vectors showed no change (Fig. 3B). Next, we used RIP to verify whether hsa_circ_0070661 and miR-556-5p could bind Ago2 protein. According to the data, Ago2 antibodies enriched hsa_circ_0070661 and miR-556-5p levels (Fig. 3C). Based on these data, we hypothesize that hsa_circ_0070661 directly bound to miR-556-5p. Then, we performed RT-qPCR and Pearson correlation analysis in clinical sample of LUAD patients. The outcomes presented that miR-556-5p expression in LUAD was up-regulated by 6.5-fold and negatively associated with the expression of hsa_circ_0070661 (Fig. 3D and E). In addition, miR-556-5p levels in H1650 and A549 increased 5.0 and 7.0 times, respectively, compared to BEAS-2B (Fig. 3F). Subsequently, we examined the regulation impact of miR-556-5p by hsa_circ_0070661 at the cellular level. The level of miR-556-5p in hsa_circ_0070661 over-expressed LUAD cells was 20% of the empty vector group, and the up-regulation of hsa_circ_0070661 reversed miR-556-5p mimic induced up-regulation of miR-556-5p (Fig. 3G). These outcomes specified that miR-556-5p was over-expressed in LUAD and its up-regulation was caused by hsa_circ_0070661.

Figure 3.

Hsa_circ_0070661 sponged miR-556-5p. (A) Speculated binding site of miR-556-5p within hsa_circ_0070661 was predicted by circInteractome database (https://circinteractome.nia.nih.gov/index.html). (B) Luciferase activity was estimated in H1650 and A549 cells carrying miR-NC/miR-556-5p mimic and hsa_circ_0070661 WT/MUT. 0.001 vs. miR-NC. (C) Data of RIP assay employed to ascertain the association among miR-556-5p and hsa_circ_0070661. 0.001 vs. Anti-IgG. (D) Relative expressions of miR-556-5p in 38 sets of LUAD tumors and adjacent normal tissue were assessed using RT-qPCR analysis. (E) Results of Pearson correlation among miR-556-5p and hsa_circ_0070661 expressions in LUAD tissue. (F) Relative miR-556-5p expression in one human normal BEAS-2B and LUAD cell lines (H1650 and A549). 0.001 vs. BEAS-2B. (G) MiR-556-5p levels in H1650 and A549 cells transfected with Empty vector, circ-OE, mimic-NC, mimic and OE$+$mimic were detected using RT-qPCR analysis. 0.001 vs. Empty vector; 0.001 vs. mimic-NC; 0.001 vs. OE$+$mimic.

3.4. Hsa_circ_0070661 blocked the malignant behavior of LUAD cells through targeting miR-556-5p

Next, biological function experiments was performed to verify the role of miR-556-5p in LUAD and its interaction with hsa_circ_0070661 in vitro. Further analysis by Transwell assay and CCK-8 assay exhibited that miR-556-5p up-regulation increased viability and migration of LUAD cells by about 1.5 and 1.6-fold, respectively, while hsa_circ_0070661 overexpression attenuated this effect (Fig. 4A and B). In addition, western blotting displayed that miR-556-5p overexpression of LUAD cells decreased Bax protein and increased Bcl-2 protein, but this effect could also be reversed by transfection of circ-OE (Fig. 4C). In conclusion, miR-556-5p accelerates the malignant behavior of LUAD cells and mediates the anticancer impact of hsa_circ_0070661.

Figure 4.

Hsa_circ_0070661 blocked the malignant behavior of LUAD cells through targeting miR-556-5p. H1650 and A549 cells were transfected with Empty vector, circ-OE, mimic-NC, mimic and OE$+$mimic. (A) Cell proliferation in these above-mentioned LUAD cells was assessed through CCK-8 assay. (B) Cell migration in these above-mentioned LUAD cells was assessed through Transwell assays. (C) Cell apoptosis-related protein levels (Bax and Bcl-2) in these above-mentioned LUAD cells was assessed through western blotting. 0.05, 0.001 vs. OE$+$mimic; 0.05, 0.001 vs. mimic-NC; 0.05, 0.001 vs. Empty vector.

3.5. MiR-556-5p targeted TEK

We additionally refined the downstream target of miR-556-5p. Based on the TargetScan database prediction, we found that TEK sequences have miR-556-5p binding site (Fig. 5A). Luciferase analysis showed that the luciferase activity decreased by more than 55% in LUAD cells containing TEK-WT vector and miR-556-5p mimic, while in LUAD cells containing TEK-MUT vector and miR-556-5p mimic, there was no significant change in luciferase activity (Fig. 5B). Moreover, the mean level of TEK mRNA in LUAD tumors was almost 50% lower in normal tissues (Fig. 5C). In H1650 and A549 cells, TEK mRNA levels were 50% and 40% of those in BEAS-2B cells, respectively (Fig. 5D). Moreover, the expression of TEK in LUAD tumors was negatively associated with miR-556-5p levels (Fig. 5E). To confirm the association among miR-556-5p and TEK in LUAD, cells were transfected with either TEK-OE or miR-556-5p mimic. According to the quantitative result of TEK protein in cells, the level of TEK protein increased more than 1.5 times when TEK was overexpressed, however, these changes might be reversed by mean of introducing miR-556-5p mimic (Fig. 5F). In conclusion, miR-556-5p could target and negatively regulate TEK.

Figure 5.

MiR-556-5p targeted TEK. (A) Binding site of miR-556-5p within TEK, predicted using TargetScan database (http://www.targetscan. org/vert_71/). (B) Luciferase activities were determined among H1650 and A549 cells transfected with miR-NC/miR-556-5p mimic and TEK WT/MUT. vs. miR-NC. (C) Relative TEK mRNA levels in 38 sets of LUAD tumors and adjacent normal tissue were assessed via RT-qPCR. (D) Relative TEK mRNA expression in one human normal BEAS-2B and LUAD cell lines (H1650 and A549). vs. BEAS-2B. (E) Analysis via Pearson correlation among miR-556-5p and TEK expressions in LUAD tissue. (F) TEK protein expressions in H1650 and A549 cells transfected with Empty vector, TEK-OE, mimic-NC, mimic and TEK-OE$+$mimic, as measured via western blotting analysis. 0.001 vs. Empty vector; 0.001 vs. mimic-NC; 0.001 vs. TEK-OE$+$mimic.

3.6. High TEK expression inhibited the malignant behaviors of LUAD cells and reversed the tumorigenic effects of miR-556-5p upregulation

We continued to study the impacts of miR-556-5p and TEK on LUAD cell behavior in vitro. The results of CCK-8 assay and Transwell assays uncovered that transfection of TEK-OE slowed LUAD cell proliferation and migration, but miR-556-5p upregulation reversed this effect (Fig. 6A and B). In addition, high levels of TEK could increase Bax protein levels and decrease Bcl-2 protein levels in LUAD cells, while this could also be reversed by miR-556-5p mimic (Fig. 6C). These results suggest that miR-556-5p modulates TEK to achieve its promoting-cancer function in LUAD cells.

Figure 6.

High TEK expression inhibited the malignant behaviors of LUAD cells and reversed the tumorigenic effects of miR-556-5p upregulation. H1650 and A549 cells were transfected with Empty vector, TEK-OE, mimic-NC, mimic and TEK-OE$+$mimic. (A) Cell proliferation in these above-mentioned LUAD cells was assessed through CCK-8 assay. (B) Cell migration in these above-mentioned LUAD cells was assessed through Transwell assays. (C) Cell apoptosis-related protein levels (Bax and Bcl-2) in these above-mentioned LUAD cells was assessed through western blotting. 0.05, 0.001 vs. TEK-OE$+$mimic; 0.05, vs. mimic-NC; 0.05, 0.001 vs. Empty vector.

4. Discussion

This paper investigated the impact of hsa_circ _0070661 on the generation and development of LUAD. Our analysis presented that hsa_circ_0070661 was under-expressed in LUAD, and its overexpression could inhibit the migration and survival of LUAD cells, encourage cell apoptosis, and obstruct the formation of tumors in vivo. Moreover, we exhibited that hsa_circ_0070661 sponged miR-556-5p, and miR-556-5p targeted TEK.

Accumulated literatures have indicated that circRNA possesses the ability to regulate the malignant actions of LUAD cells. For example, high circXPO1 expression helps the survival, invasion and migration of LUAD cells in vivo as well as in vitro, and inhibits apoptosis [16]. Silencing circ-ENO1 inhibits the progression of LUAD, which is specifically manifested as delaying glycolysis of LUAD cells, inhibiting proliferation, metastasis and epithelial-mesenchymal transition (EMT), and inducing apoptosis [36]. Hsa_circ_0005576 promotes LUAD cell proliferation and osimertinib resistance [37]. There have been no reports on hsa_circ_0070661. Here, we first time study the mechanism and role of hsa_circ_0070661 in LUAD. Similar to various circRNA involved in LUAD development, we exhibited that hsa_circ_0070661 expression was down-regulated in LUAD. Over-expression of hsa_circ_0070661could suppress the LUAD cell growth and migration and induce cell apoptosis. These results suggest that hsa_circ_0070661 plays a role of tumor suppressor in LUAD.

Since circRNA has the effect of sponge miRNA, we continue to explore the downstream of miRNA of hsa_circ_0070661. In this study, the target miRNA of hsa_circ_0070661, miR-556-5p, was screened out by bioinformatics tools. MiR-556-5p plays different roles in different tumors. Overexpression of miR-556-5p promotes the growth of cancer cells in prostate cancer [24]. In breast cancer, miR-556-5p knockdown promotes cancer cell invasion and migration and EMT [21]. Our suspection was that this is due to the heterogeneity of the tumor. Furthermore, previous one report found that miR-556-5p plays a carcinogenic role in NSCLC, and its downregulation can effectively improve the sensitivity of NSCLC to cisplatin [25]. Consistent with the results reported by Feng et al. [25], our study observed miR-556-5p overexpression in LUAD. Furthermore, up-regulation of miR-556-5p promoted LUAD migration and viability, as well as inhibited apoptosis. The mechanism by which miR-556-5p is regulated by circRNA has also been discovered. MiR-556-5p is absorbed by circ-LAMP1 sponge to inhibit the growth and metastasis of cholangiocarcinoma [38], and is suppressed by circSNHG1 sponge to inhibit meningioma tumorigenesis [23]. By targeting analysis, we ascertained that miR-556-5p was the downstream miRNA of hsa_circ_0070661, and was negatively regulated by hsa_circ_0070661. In addition, miR-556-5p overexpression partially eliminates the inhibitory effect of hsa_circ_0070661 upregulation on the malignant actions of LUAD. miR-556-5p was discovered to promote LUAD progression and mediates the antitumor effect of hsa_circ_0070661.

TEK receptor Tyrosine kinase (TEK) is an endothelial tyrosine kinase receptor of vascular growth factor responsible for angiogenesis and angiogenesis [26]. Studies have shown that TEK mutates in disease, resulting in abnormal endothelial cell growth, elongated, overlapping cells, and reduced extracellular fibronectin [39]. Reports indicate that TEK is participated in the development of cancer. For instance, TEK enhances invasion and metastasis of breast cancer cells as an oncogenic factor [29]. As a tumor suppressor in clear cell renal cell carcinoma, TEK downregulation predicts the poor prognosis of cancer patients [40]. In LUAD, TEK was identified as crucial for an anticancer, which significantly inhibiting the proliferation, migration as well as adhesion of cancer cells [41]. Consistent with the above analysis, this study found that TEK was down-regulated in LUAD and that its overexpression inhibited the malignant behavior of cancer cells. Furthermore, TEK was found to be the target of miR-556-5p, reversing the promoting-cancer effects of miR-556-5p on LUAD cell growth. In short, TEK suppressed LUAD cell progression via targeting miR-556-5p.

In this study, we have systematically explained the mechanism of hsa_circ_0070661/miR-556-5p/TEK axis in LUAD progression, however, there are still some limitations in this study. Firstly, this study showed that the hsa_circ_0070661 expression was different in A549, PC9 and H1650 cells. As we known, A549 is K-Ras mutant positive cell line, whereas PC9 and H1650 are BRAF mutant positive cell line. Therefore, whether the effects of hsa_circ_0070661 on different LUAD cells induced by different mutant are not same is worthy to explore. Due to the clinical samples collected from recently years, the prognosis is well so that the prognostic correlation of hsa_circ_0070661 needs to be analyzed by following up with these patients. In addition, the downstream signaling pathways that could be regulated by TEK remain to be further studied.

5. Conclusion

The effects of hsa_circ_0070661 on LUAD progression and its interaction with miR-556-5p/TEK were discussed for the first time. We found that hsa_circ_0070661 sponged miR-556-5p and released TEK to prevent the migration, tumor growth as well as proliferation of LUAD, and encourage cell apoptosis. To our knowledge, this research deepens the knowledge of LUAD pathogenesis, offers valued biomarkers for the LUAD diagnosis, and provides a favorable molecular target for LUAD clinical treatment.

Funding

Not applicable.

Ethics approval

The study was approved by the Wuhan No. 1 Hospital Ethics Committee (China). The processing of clinical tissue samples strictly conforms to the ethical standards of the Helsinki Declaration. A documented informed consent has been signed by each patient.

The Animal Care Committee of Wuhan No. 1 Hospital approved this animal experiment, which was conducted in agreement with the ARRIVE criteria.

Approval of participation

All patients have written informed consent.

Publication consent

Participants approved the publishing with their permission.

Availability of data and material

This article contains all of the data that were created or examined during this investigation.

Author contributions

Conception: YPC and YYW.

Interpretation or analysis of data: YPC and YYW.

Preparation of the manuscript: YYW.

Revision for important intellectual content: YPC.

Supervision: YPC.

Conflict of interest

The author states that there are no conflicts of interest.