Hsa_circ_0010235 functions as an oncogenic drive in non-small cell lung cancer by modulating miR-433-3p/TIPRL axis

Background Non-small cell lung cancer (NSCLC) is a threat to human health. Circular RNAs (circRNAs) have been proved to function in NSCLC development. In this study, the role of circRNA hsa_circ_0010235 in NSCLC progression and the possible molecular mechanism were explored. Methods Expression of hsa_circ_0010235, miRNA (miR)-433-3p and TOR signaling pathway regulator-like (TIPRL) was examined by quantitative real-time PCR (qRT-PCR). Cell viability and clonogenicity were detected by cell counting kit-8 (CCK-8) assay and colony formation assay, respectively. Flow cytometry was performed to monitor cell apoptosis and cell cycle distribution. Western blot assay was employed to evaluate the protein levels of TIPRL, light chain 3 (LC3)-II/I and p62. Cell metastasis was assessed by Transwell and wound healing assays. The targeted relationship between miR-433-3p and hsa_circ_0010235 or TIPRL was confirmed by dual-luciferase reporter and RNA immunoprecipitation (RIP) assays. Furthermore, the role of hsa_circ_0010235 in vivo was investigated by xenograft assay. Results Hsa_circ_0010235 and TIPRL were highly expressed in NSCLC tissues and cells, while miR-433-3p was downregulated. Depletion of hsa_circ_0010235 or gain of miR-433-3p repressed proliferation and autophagy but promoted apoptosis in NSCLC cells. Hsa_circ_0010235 sponged miR-433-3p to upregulate TIPRL expression, so as to affect NSCLC development. Hsa_circ_0010235 knockdown also blocked tumor growth in vivo. Conclusion Hsa_circ_0010235 knockdown suppressed NSCLC progression by regulating miR-433-3p/TIPRL axis, affording a novel mechanism of NSCLC progression.

circRNA hsa_circ_0010235 (Position: chr1: ) was detected to be upregulated in NSCLC tissues [11], while its explicit role in NSCLC development remains to be studied.
MicroRNAs (miRNAs) are also non-coding RNAs, serving as important regulators in tumor development by post-transcriptionally downregulating expression of their target genes, whose length was only about 22 nucleotides [12,13]. MiRNAs could act as potential diagnosis and therapy biomarkers for NSCLC patients [14]. In addition, miRNAs was implicated in tumorigenesis and progression of NSCLC [15]. MiR-433 was substantiated to be a tumor suppressor in hepatocellular carcinoma (HCC), breast cancer and retinoblastoma [16][17][18], as well as in NSCLC [19]. In this study, miR-433-3p is predicted to be a target of hsa_circ_0010235 using online tools, while whether the miR-433-3p was involved in hsa_circ_0010235-mediated NSCLC progression needs to be investigated.
Mammalian target of rapamycin (mTOR) signaling was verified to play pivotal roles in tumor cell migration and invasion [20]. TOR signaling pathway regulator-like (TIPRL, the mammalian ortholog of the yeast protein TIP41) was initially recognized in yeast to interact with TAP42 and inhibit mTOR signaling [21]. TIPRL was upregulated in hepatocellular carcinoma (HCC), and it could facilitate TRAIL (a potential anti-cancer agent) resistance of HCC cells [22]. In NSCLC, enforced expression of TIPRL facilitated cell autophagy, and it could serve as an ideal therapeutic target [23]. Here, TIPRL was found to have binding position with miR-433-3p, and the interaction of TIPRL with hsa_circ_0010235-mediated NSCLC progression was also explored.
In this study, the significant upregulation of hsa_ circ_0010235 was detected in NSCLC tissues and cells. Functionally, the impact of hsa_circ_0010235 on NSCLC cell proliferation, autophagy, migration, invasion and apoptosis in vitro, as well as on tumor growth in vivo was investigated. The molecular mechanism by which hsa_circ_0010235 affected NSCLC development was also explored.

Clinical tissues and cell lines
Fifty two pairs of NSCLC tissues and corresponding adjacent normal tissues were resected from NSCLC patients enrolled at the First Affiliated Hospital of Zhengzhou University from 2013-2015 and preserved at -80 ℃. All patients were chosen based on the guidelines supplied by World Health Organization (WHO) and the International Association for the Study of Lung Cancer (IASLC) [24]. Follow-up of these 52 patients was implemented from date of surgery until end of this study or death. All participators signed informed consent.

RNase R and actinomycin D treatment
To confirm the stability of hsa_circ_0010235, total RNA extracted from H1299 and A549 cells was digested with RNase R (Geneseed, Guangzhou, China) or not (Mock). 0.5 h later, resulting RNA was subjected to qRT-PCR assay to examine the relative expression of hsa_ circ_0010235 and ALDH4A1 mRNA.
Actinomycin D treatment was also applied to validate the stability of hsa_circ_0010235. H1299 and A549 cells were treated with Actinomycin D (an inhibitor of transcription) (Sigma-Aldrich, St. Louis, MO, USA) or dimethyl sulfoxide (DMSO) solution for 0 h, 4 h, 8 h, 12 h or 24 h, followed by qRT-PCR assay to determine the relative expression of hsa_circ_0010235 and ALDH4A1 mRNA.

Cell counting kit-8 (CCK-8) assay
The current assay was performed to detect cell viability. In brief, H1299 and A549 cells were placed into 96-well plates at 0 h, 24 h, 48 h or 72 h post transfection, then 10 μL CCK-8 reagent was dropped into each well, and incubated for 2 h. Later, optical density (OD) value at 450 nm of each well was recorded using a microplate reader (Bio-Rad Laboratories, Inc., Hercules, CA, USA).

Flow cytometry
Flow cytometry was carried out to monitor cell apoptosis and cell cycle distribution. For cell apoptosis detection, an Annexin V-fluorescein isothiocyanate (FITC)/propidium iodide (PI) Apoptosis Detection Kit (BD Biosciences, Franklin Lakes, NJ, USA) was used. After transfection, H1299 and A549 cells were harvested and re-suspended in binding buffer, then double-stained with 5 μL Annexin V-FITC and 10 μL PI reagent in the dark for 15 min. Subsequently, apoptotic cells (Annexin V-FITC +) were monitored by a flow cytometer (BD Biosciences).
For cell cycle examination, transfected H1299 and A549 cells were immobilized by 75% ethanol and digested with RNase A, followed by the addition of PI solution (Sigma-Aldrich). Later, cells in G0/G1, S and G2/M phases were detected utilizing a flow cytometer.

Transwell assay
For migration determination, transfected H1299 and A549 cells (1 × 10 5 ) suspended in medium without serum were placed onto the top chambers. Complete medium was added to the bottom chambers. After incubation for 48 h, cells migrated through the polycarbonic membrane were fixed using 4% paraformaldehyde and stained with crystal violet, then observed and counted under a microscope (magnification: × 100).
The invasion ability of NSCLC cells was assessed with 5 × 10 5 cells plated on the top chambers pre-coated with Matrigel (BD Biosciences). Other procedures were same to the migration assay.

Wound healing assay
After transfection, H1299 and A549 cells were seeded into 24-well plates and maintained until 80-90% confluence. Then a sterile pipette tip (10 μL) was used to scratch the single cell layer across each well to make a scratch wound. The detached cells were washed away using phosphate buffer saline (PBS). Later, serum-free medium was added, followed by capture of phase-contrast images at 0 h or 24 h post incubation exploiting an inverted microscope.
RIP assay was also conducted to testify the relationship between miR-433-3p and hsa_circ_0010235 or TIPRL using the EZ-Magna RIP Kit (Millipore, Billerica, MA, USA) referring to the user's manual. Briefly, cell lysate of H1299 and A549 cells in lysis buffer was mixed with RIP binding buffer containing magnetic beads conjugated with human Ago2 antibody (1:50; ab32381; Abcam) or IgG antibody (1:50; ab109761; Abcam). Subsequently, RNA was extracted and qRT-PCR assay was implemented to examine the enrichment of hsa_circ_0010235, miR-433-3p and TIPRL.

In vivo tumorigenesis assay
BALB/c nude mice (male, 5 weeks old) were purchased from Beijing Laboratory Animal Center (Beijing, China). H1299 cells (2 × 10 6 ) stably transfected with lentiviral small hairpin RNA (shRNA) targeting hsa_ circ_0010235 (sh-hsa_circ_0010235) or sh-NC were subcutaneously inoculated into right flank of nude mice (n = 5). Then, the volume of formed tumors was monitored every 5 days and calculated using the formula: volume (mm 3 ) = 0.5 × length × width 2 . 30 days later, all mice were euthanatized, then tumor tissues were removed for weigh, qRT-PCR and Western blot assays. Fig. 1 Hsa_circ_0010235 was highly expressed in NSCLC tissues and cells. a QRT-PCR assay for the relative expression of hsa_circ_0010235 in NSCLC tissues (n = 52) and normal tissues (n = 52). b QRT-PCR assay for the relative expression of hsa_circ_0010235 in 16HBE, H1299, A549, H1581 and H23 cells. c Overall survival rate of NSCLC patients with high or low hsa_circ_0010235 expression. d, e QRT-PCR assay for the relative expression of hsa_circ_0010235 and ALDH4A1 mRNA in RNA isolated from H1299 and A549 cells digested with RNase R or not (Mock). f, g QRT-PCR assay for the half-life of hsa_circ_0010235 and ALDH4A1 in H1299 and A549 cells treated with Actinomycin D or DMSO. *P < 0.05

Statistical analysis
All data were acquired from 3 independent experiments. Obtained data were analyzed utilizing SPSS 20.0 software (SPSS, Inc., Chicago, IL, USA) and exhibited as mean ± standard deviation. The overall survival rate of NSCLC patients was determined via the Kaplan-Meier method with log-rank test. Correlation among the relative expression levels of hsa_circ_0010235, miR-433-3p and TIPRL in 52 cases of NSCLC tissues was determined by Spearman's correlation coefficient. Difference significance was calculated by Student's t-test or one-way analysis of variance followed by Tukey's test, and P < 0.05 was recognized to be statistically significant.

Hsa_circ_0010235 was highly expressed in NSCLC tissues and cells
At first, the expression level of hsa_circ_0010235 in NSCLC tissues was evaluated by qRT-PCR assay.
The data showed that hsa_circ_0010235 expression in NSCLC tissues (n = 52) was higher than that in normal tissues (n = 52) (Fig. 1a). Additionally, hsa_circ_0010235 expression was also increased in H1299, A549, H1581 and H23 cells relative to 16HBE cells (Fig. 1b). Among the four, H1299 and A549 cells were selected for further experiments due to their higher hsa_circ_0010235 expression than H1581 and H23 cells, which have certain representativeness. Follow-up results of these 52 patients revealed that NSCLC patients with higher hsa_circ_0010235 expression had lower overall survival rate. The median expression value of hsa_circ_0010235 was used as the cutoff (P = 0.0376) (Fig. 1c). Moreover, clinicopathological characteristics of these 52 NSCLC patients were shown in Table 1, and hsa_circ_0010235 expression was correlated with smoking history, tumor size, TNM stage, lymph node metastasis and recurrence. After digestion with RNase R, ALDH4A1 mRNA expression, rather than hsa_circ_0010235 expression, was obviously reduced in H1299 and A549 cells, suggesting that hsa_circ_0010235 was more stable than ALDH4A1 (Fig. 1d, e). After Actinomycin D disposition, the half-life of hsa_circ_0010235 was more than 24 h, further longer than that of ALDH4A1 mRNA (Fig. 1f, g). Collectively, hsa_circ_0010235 was highly enriched in NSCLC tissues and cells, and high expression of hsa_circ_0010235 could predict low survival rate of NSCLC patients.

Hsa_circ_0010235 knockdown inhibited proliferation and autophagy but facilitated apoptosis in NSCLC cells
The dysregulation of hsa_circ_0010235 in NSCLC prompted us to investigate the role of hsa_circ_0010235 in NSCLC cells. Si-hsa_circ_0010235#1 and si-hsa_ circ_0010235#2 were introduced into H1299 and A549 cells, with si-NC as control. The knockdown efficiency of the two was determined qRT-PCR assay and exhibited in Fig. 2a. Additionally, si-hsa_circ_0010235#1 was chosen for later assays since it induced better knockdown efficiency. And, hsa_circ_0010235 was successfully overexpressed in NSCLC cells via transfection with its overexpression vector, with pCD-ciR as a control (Fig. 2b). CCK-8 assay uncovered that hsa_circ_0010235 knockdown reduced in the cell viability of NSCLC cells; On the contrary, overexpression of hsa_circ_0010235 elevated the cell viability of H1299 and A549 cells (Fig. 2c,  d). Moreover, hsa_circ_0010235 knockdown inhibited clonogenicity in NSCLC cells, but overexpression of hsa_circ_0010235 triggered reverse results (Fig. 2e). Data from flow cytometry showed that depletion of hsa_circ_0010235 promoted cell apoptosis and blocked cell cycle at G0/G1 phase, while hsa_circ_0010235 overexpression resulted in opposite results ( Fig. 2f-h). Hsa_circ_0010235 knockdown-induced the downregulation of LC3-II and upregulation of p62, as well as the hsa_circ_0010235 overexpression-induced upregulation of LC3-II and downregulation of p62 manifested that hsa_circ_0010235 positively affected autophagy in NSCLC cells (Fig. 2i, j). Above results revealed that hsa_circ_0010235 knockdown inhibited proliferation and autophagy but facilitated apoptosis in NSCLC cells.

Depletion of hsa_circ_0010235 repressed metastasis of NSCLC cells
The functional effects of hsa_circ_0010235 on migration and invasion of NSCLC cells were also studied. As shown in Fig. 3a, b, hsa_circ_0010235 deficiency efficiently reduced the number of migrated and invaded H1299 and A549 cells; reversely, hsa_circ_0010235 overexpression increased the number of migrated and invaded H1299 and A549 cells. The results of the wound healing assay suggested that silencing of hsa_circ_0010235 apparently inhibited cell motility of H1299 and A549 cells, but upregulation of hsa_circ_0010235 elevated cell motility (Fig. 3c, d). Taken together, depletion of hsa_ circ_0010235 repressed metastasis of NSCLC cells.

Depletion of hsa_circ_0010235 blocked tumor growth in vivo
H1299 cells stably expressing sh-hsa_circ_0010235 or sh-NC were subcutaneously inoculated into nude mice (n = 5) to construct xenograft model in vivo, divided into sh-hsa_circ_0010235 group and sh-NC group. Results showed that both the tumor size (Fig. 8a) and the tumor weight (Fig. 8b) were less in sh-hsa_circ_0010235 group than those of sh-NC group. Moreover, expression of hsa_ circ_0010235 (Fig. 8c) and TIPRL (Fig. 8e, f ) was downregulated, while miR-433-3p (Fig. 8d) was upregulated in tumor tissues of sh-hsa_circ_0010235 group relative to those of sh-NC group. Taken together, depletion of hsa_ circ_0010235 inhibited tumor growth in vivo.

Discussion
Circular RNAs (circRNAs) with dysregulated expression were associated with growth and metastasis of NSCLC, which could serve as biomarkers of lung cancer diagnosis, prognosis and therapy response [28]. In this project, the functional effects of hsa_circ_0010235 on NSCLC progression were corroborated for the first time. Depletion of hsa_circ_0010235 could repress NSCLC cell proliferation, autophagy, mobility and tumorigenesis, highlighting its carcinogenic role in NSCLC. Increasing evidence showed that circRNAs were involved in the pathophysiological processes of NSCLC, affecting most cellular behaviors of tumor cells, including proliferation, migration, invasion, cell cycle epithelialmesenchymal transition and drug resistance [29]. With the development of sequencing technology and computational algorithms, increasing circRNAs were discovered to be meritorious in human cancers [30]. Through human circular RNA microarray and qRT-PCR assay, hsa_circ_0010235 was previously reported to be aberrantly upregulated in NSCLC tissues relative to normal tissues [11]. In this study, we also found the upregulation of hsa_circ_0010235 in NSCLC tissues and cells. Also, our data initially showed the hsa_circ_0010235 knockdown-induced inhibitory effects on cell proliferation, autophagy, metastasis and tumorigenesis of NSCLC cells, suggesting that hsa_circ_0010235 functioned as an oncogene in NSCLC.
With diverse functions, circRNAs are involved in cancer development in different ways, such as acting as sponges or decoys of miRNAs or proteins, scaffolds or transporters [31]. Classically, circRNAs function by sponging miRNAs. In our study, three online tools Fig. 7 Silencing of miR-433-3p or overexpression of TIPRL could attenuate hsa_circ_0010235 knockdown-induced inhibitory effects on NSCLC progression. H1299 and A549 cells were transfected with si-NC, si-hsa_circ_0010235#1, si-hsa_circ_0010235#1 + anti-miR-433-3p or hsa_circ_0010235#1 + TIPRL. a QRT-PCR assay for the relative mRNA expression of TIPRL in transfected cells. b, c Western blot assay for the protein level of TIPRL in transfected cells. d, e CCK-8 assay for the cell viability of transfected cells. f Colony formation assay for the colony formation ability of transfected cells. g Flow cytometry for the apoptotic rate of transfected cells. h, i Flow cytometry for the cell cycle distribution in G0/G1, S and G2/M phases of transfected cells. j, k Western blot assay for the protein levels of LC3-I, LC3-II and p62 in transfected cells. l, m Transwell assay for the migration and invasion of transfected cells. n Wound healing assay for the migration capacity of transfected cells. o Spearman's correlation analysis for the expression levels of hsa_circ_0010235 and TIPRL mRNA in 52 NSCLC tissues. *P < 0.05 were utilized to forecast the target miRNAs of hsa_ circ_0010235, miR-433-3p was identified as candidates by DLRA and RIP assay. Shi et al. alleged that miR-433-3p could decelerate esophageal squamous cell carcinoma (ESCC) proliferation and metastasis by downregulating growth factor receptor-bound protein 2 (GRB2) [32]. MiR-433-3p also inhibited cell growth, invasion and migration in human glioma by targeting cyclic adenosine monophosphate (AMP) response element-binding protein (CREB) [33]. Other works also highlighted the tumor-suppressor role of miR-433-3p in hepatocellular carcinoma (HCC), breast cancer, retinoblastoma and NSCLC [16][17][18][19]. As reported by Li et al. miR-433-3p was downregulated in NSCLC tissues, and its overexpression triggered anti-proliferative and anti-metastatic effects on NSCLC cells [19]. Consistently, we also detected the downregulation of miR-433-3p in NSCLC tissues and cells. Functionally, miR-433-3p repressed NSCLC cell proliferation, autophagy and metastasis. Furthermore, inhibition of miR-433-3p weakened hsa_circ_0010235 knockdowninduced repressed impact on NSCLC progression.
Generally, miRNAs could bind to the 3′UTRs of their target mRNAs to block the translation, so as to affect their functions [34]. In this study, TargetScan predicted that 3′UTR of TIPRL had complementary sites with miR-433-3p, and the target relationship was confirmed by DLRA and RIP assay. Former literature testified that TIPRL deficiency promoted the apoptosis of lung cancer H1299 cells disposed by cisplatin, suggesting its important role in lung cancer [35]. TIPRL expression was increased in NSCLC tissues in comparison to normal tissues, and TIPRL could contribute to autophagy so as to facilitate NSCLC development through the eukaryotic initiation factor 2α (eIF2α)-Activating Transcription Factor 4 (ATF4) pathway [23]. In addition, TIPRL was found to correlated with lower overall survival rate of NSCLC patients; TIPRL knockdown suppressed NSCLC metastasis, indicating that TIPRL acted as an oncogene in NSCLC [36]. Similarly, our data also showed that TIPRL was upregulated in NSCLC. Moreover, exogenous introduction of TIPRL also attenuated hsa_circ_0010235 knockdown-induced NSCLC progression inhibition. Above results revealed the significant role of hsa_circ_0010235/miR-433-3p/ TIPRL axis in NSCLC progression.

Conclusion
In conclusion, hsa_circ_0010235 was upregulated in NSCLC. The tumor suppressor role of hsa_circ_0010235 knockdown in NSCLC progression in vitro and in vivo was testified. Our findings also corroborated that hsa_ circ_0010235 affected NSCLC development by regulating Fig. 8 Depletion of hsa_circ_0010235 blocked tumor growth in vivo. Nude mice were injected with H1299 cells stably expressing sh-hsa_ circ_0010235 or sh-NC (n = 5). a Volume of formed tumors measured every 5 days. b Weight of formed tumors measured after 30 days. c, d QRT-PCR assay for the relative expression of hsa_circ_0010235 c and miR-433-3p d in formed tumors. e, f QRT-PCR and Western blot assays for the mRNA e and protein f levels of TIPRL in formed tumors. *P < 0.05