Circular non‐coding RNA circ_0072088 serves as a ceRNA, targeting the miR‐1225‐5p/WT1 axis to regulate non‐small cell lung cancer cell malignant behavior

Abstract

Background

Circular RNA (circRNA) circ_0072088 has been reported to be associated with NSCLC cell growth, migration, and invasion. However, the role and mechanism of circ_0072088 on NSCLC development have not yet been determined.

Methods

Circ_0072088, microRNA‐1225 (miR‐1225‐5p), and Wilms' tumor (WT1) suppressor gene level was detected by reverse transcription‐quantitative polymerase chain reaction (RT‐qPCR). Migration, invasion, and apoptosis were detected using transwell and flow cytometry assays. Matrix metallopeptidase 9 (MMP9), hexokinase 2 (HK2), and WT1 were examined using western blot assay. The biological role of circ_0072088 on NSCLC tumor growth was examined by the xenograft tumor model in vivo. Circular RNA Interactome and TargetScan were used to predict the binding between miR‐1225‐5p and circ_0072088 or WT1, followed by confirmation using a dual‐luciferase reporter.

Results

Circ_0072088 and WT1 were highly expressed in NSCLC tissues and cells, and miR‐1225‐5p was decreased. Knockdown of circ_0072088 might repress migration, invasion, and glycolysis, and facilitate apoptosis of NSCLC cells in vitro. Circ_0072088 silencing also blocked NSCLC tumor growth in vivo. Mechanistically, circ_0072088 acted as a sponge of miR‐1225‐5p to regulate WT1 expression.

Conclusion

Circ_0072088 knockdown could inhibit cell growth, migration, invasion, and glycolysis partially by regulating the miR‐1225‐5p/WT1 axis, thus providing a promising therapeutic target for NSCLC treatment.

Circ_0072088 is highly expressed in NSCLC tissues and cell lines. Furthermore, circ_0072088 silencing might suppress NSCLC cell growth, migration, invasion, and glycolysis partially via regulating the miR‐1225‐5p/WT1 axis, which provides a promising therapeutic target for NSCLC treatment.

INTRODUCTION

As a malignant neoplasm of the respiratory system, lung cancer has been recognized to critically threaten human health worldwide. ¹ According to United States cancer statistics, there were about 235 760 new diagnoses and 131 880 deaths of lung cancer in 2021. ² Notably, non‐small cell lung cancer (NSCLC) is responsible for 85% of lung cancer cases, which is the most prevalent pathological subtype of lung cancer. ³ As NSCLC has untypical early symptoms, most patients are diagnosed at middle to late stages with local spread or distant metastasis. ⁴ Therefore, exploring effective biomarkers and understanding the molecular mechanism behind NSCLC progression is highly desirable for improving the prognosis of NSCLC patients.

Currently, some scholars have described that more than 90% of the mammalian genome is transcribed as noncoding RNAs. ⁵ Circular RNAs (circRNAs) are a distinct group of noncoding RNAs, which form a covalently closed circular structure by back‐splicing events. ⁶ , ⁷ The remarkable stability and high abundance of circRNAs endows a possibility of a potential attractive biomarker in various diseases, ⁸ including cancer. ⁹ In practice, it is becoming increasingly apparent that there is an association between circRNAs and the initiation of diverse human cancers. ¹⁰ , ¹¹ In terms of NSCLC, circ‐FOXM1 silencing has been reported to attenuate the malignant progression of tumor cells via repressing cell viability and migration. ¹² Analogously, circ_0000376 has also been found to serve as a competitive endogenous RNA (ceRNA) of miR‐1182 to accelerate migration, invasion, and glycolysis through upregulating NOVA2 in NSCLC. ¹³ Circ_0072088, a well‐known circRNA originating from zinc finger RNA binding protein (ZFR), has been verified to be a carcinogenic factor in cervical and esophageal squamous cell cancers. ¹⁴ , ¹⁵ It has also been reported that the upregulation of circ_0072088 might contribute to pancreatic ductal adenocarcinoma cell glycolysis in vitro. ¹⁶ Furthermore, recent evidence indicated that circ_0072088 could facilitate NSCLC development by increasing cell growth and metastasis. ¹⁷ However, the pathogenesis of circ_0072088 involved in NSCLC remains to be illustrated.

During the last decade, the circRNA‐microRNA (miRNA)‐mRNA regulatory mechanism network has attracted considerable attention in various cancers. ¹⁸ , ¹⁹ , ²⁰ In this study, there were some binding sites between circ_0072088 and miR‐1225‐5p. In addition, the inhibitory action of miR‐1225‐5p on tumor progression has also previously been seen in various tumors, such as osteosarcoma, glioblastoma, and thyroid cancer. ²¹ , ²² , ²³ Meanwhile, miR‐1225‐5p has previously been validated to take part in the regulation of proliferation and metastasis of NSCLC cells. ²⁴ Hence, this study was designed to determine whether circ_0072088 could mediate NSCLC cell malignant behavior by binding to miR‐1225‐5p.

METHODS

Clinical samples and cell culture

After signing their written informed consent, tumor tissues and matched adjacent normal tissues were obtained from 30 NSCLC patients who underwent surgery in Jingzhou Hospital Affiliated with Yangtze University. Collected tissues were stored at −80°C. The detailed clinical characteristics of patients are described in Table 1.

TABLE 1.

The clinicopathological features in non‐small cell lung cancer (NSCLC) patients.

Parameters	N = 30
Age, years
<60	12
≥60	18
Sex
Male	24
Female	6
Smoking history
Yes	22
No	8
Tumor location
Left lobe	16
Right lobe	14
Tumor size
<3 cm	8
≥3 cm	22
TNM stage
I + II	21
III	9
Lymph node metastasis
Yes	13
No	17

All cells were acquired from Procell and maintained in a 5% CO₂ humidified incubator at 37°C. Human bronchial epithelial cell line 16 HBE and NSCLC cell line H1299 were cultured in Roswell Park Memorial Institute‐1640 medium (RPMI‐1640; Invitrogen). NSCLC cell line A549 was allowed to grow in special medium (Invitrogen). Also, 10% fetal bovine serum (FBS: Gibco) and 1% penicillin–streptomycin (Gibco) were introduced into the medium.

Reverse transcription‐quantitative polymerase chain reaction (RT‐qPCR) assay

Using TRIzol reagent (Invitrogen), total RNAs were extracted and then analyzed using a NanoDrop ND‐1000 (Thermo Scientific). After transferring into cDNA using PrimeScript RT reagent kit, the RT‐qPCR reaction was carried out with specific paired primers and BeyoFast SYBR Green qPCR mix kit (Beyotime). Normalizing by glyceraldehyde‐3‐phosphate dehydrogenase (GAPDH) for circ_0072088, Wilms' tumor (WT1) suppressor gene, and U6 for miR‐1225‐5p, the data obtained were analyzed by the 2^−ΔΔCt method. The primers are listed in Table 2.

TABLE 2.

The sequences of primers for reverse transcription‐quantitative polymerase chain reaction (RT‐qPCR) used in this study.

Names	Sequences (5′‐3′)
Circ_0072088: Forward	TTTCCAAGCTGGCCCTTACG
Circ_0072088: Reverse	TCTGAACTGCCTGTAACTCCTC
miR‐1225‐5p: Forward	GCCGAGGTGGGTACGGCCCA
miR‐1225‐5p: Reverse	CTCAACTGG TGTCGTGGA
WT1: Forward	GCTATTCGCAATCAGGGTTACAG
WT1: Reverse	TGGGATCCTCATGCTTGAATG
U6: Forward	CTCGCTTCGGCAGCACA
U6: Reverse	AACGCTTCACGAATTTGCGT
GAPDH: Forward	GGAGCGAGATCCCTCCAAAAT
GAPDH: Reverse	GGCTGTTGTCATACTTCTCATGG

Subcellular fractionation assay

In this assay, the RNA from nuclear and cytoplasm fractions of NSCLC cells was isolated using a cytoplasmic and nuclear RNA purification kit (Norgen Biotek). Generally, 5 × 10⁶ cells were suspended in the lysis solution. After being centrifuged, the cytoplasmic supernatant was transferred into a clean tube. The nuclear pellet was collected to extract RNA in the cell disruption buffer. Circ_0072088 in different fractions was determined using RT‐qPCR assay, normalizing to GAPDH and U6.

Cell transfection

Small interference RNA circ_0072088 (si‐circ_0072088#1: 5′‐GATTTTCCAAGCTGGCCCTTA‐3′, si‐circ_0072088#2: 5′‐TTTCCAAGCTGGCCCTTACGT‐3′) and the scrambled siRNA control (si‐NC), and miR‐1225‐5p mimic/inhibitor (miR‐1225‐5p/in‐miR‐1225‐5p), and their negative controls (miR‐NC, in‐miR‐NC) were collected from RiboBio. Meanwhile, the WT1 overexpression vector (pc‐WT1, also known as WT1) was built using pcDNA empty vector (pc‐NC, Invitrogen). After being grown to 70%–80% confluence, NSCLC cell transfection was conducted using lipofectamine 3000 (Invitrogen).

Transwell assay

In brief, transfected cell suspension (serum‐free medium) was introduced into the upper chamber (invasion using chamber precoated with matrigel), while the bottom counterparts were filled with a medium mixed with 10% FBS. Then, 24 h later, 0.1% crystal violet was used to stain the migrated/invaded cells. The migrated or invaded cells were determined using a microscope.

Cell apoptosis assay

In short, the analysis of NSCLC cell apoptosis rate was implemented as per the supplier's direction of annexin V‐FITC/PI kit (Bender Med System). Generally, transfected cells floating in the supernatant were harvested. After being resuspended in binding buffer, the cells were incubated with 10 μL annexin V‐FITC and 5 μL PI (Bender Med System) at 4°C in the dark, followed by detection using a FACScan flow cytometer.

Glucose consumption and lactate production

In this experiment, the culture medium was collected after transfection for 48 h. Whereafter, the measurement of glucose consumption and lactate production was carried out in accordance with the instructions for the glucose detection and lactic acid detection kits (Biovision).

Western blot assay

In general, the lysates were prepared using radioimmunoprecipitation assay (RIPA) buffer (Beyotime). After fractionation using 10% SDS‐PAGE and electro‐transfer to PVDF membranes, the primary antibodies matrix metallopeptidase 9 (MMP9, 1:1000, ab119906; Abcam) hexokinase 2 (HK2, 1:1000; ab227198; Abcam), WT1 (1:500; ab201948; Abcam), and β‐actin (1:1000; ab253283; Abcam) were probed at 4°C. The next day, the membranes were incubated with the secondary antibody (1:2500, ab205719, Abcam) for 1 h. Detection of protein signals was performed using an ECL detection kit.

Tumor xenograft assay

BALB/C nude mice were used in the study, and A549 cells were infected with sh‐circ_0072088 or sh‐NC (Genechem). Subsequently, the mice (5‐week‐old males) from Vital River Laboratory were kept in a specific‐pathogen‐free environment, and randomly divided into two groups (n = 6 per group). They were subcutaneously injected using 2 × 10⁶ A549 cells stably expressing sh‐circ_0072088 or sh‐NC. Tumor volume was monitored every 7 days. Four weeks later, the excised tumors from the sacrificed mice were weighed, and subsequently stored at −80°C.

Dual‐luciferase reporter assay

The prediction of binding between miR‐1225‐5p and circ_0072088 or WT1 was, respectively, conducted using circular RNA Interactome (https://circinteractome.nia.nih.gov) and TargetScan (http://www.targetscan.org) software. In brief, wild‐type luciferase reporter constructs (wt‐circ_0072088 and wt‐WT1 3'UTR) and the site‐directed mutant constructs (mut‐circ_0072088 and mut‐WT1 3'UTR) were collected from GeneCopoeia. The cells were then cotransfected with the constructs and miR‐1225‐5p mimic/control, followed by analysis using a dual‐luciferase reporter kit (Promega).

Statistical analysis

In this study, statistical analysis was implemented according to GraphPad Prism7 (GraphPad Prism software). Statistical significance (p < 0.05) was determined by Student's t‐test (two groups) and one‐way analysis of variance (ANOVA) with Tukey's tests (multiple groups). All data are shown as the mean ± standard deviation (SD).

RESULTS

Expression patterns of circ_0072088 in NSCLC

As shown in Figure 1a, b, circ_0072088 was significantly upregulated in 30 NSCLC tumor tissues and NSCLC cell lines (A549 and H1299). Additionally, circ_0072088 was mainly distributed in the cytoplasm of NSCLC cells (Figure 1c, d), indicating its potential post‐transcriptional regulatory mechanism. In summary, dysregulated circ_0072088 might be involved in NSCLC progression.

FIGURE 1.

Circ_0072088 was elevated in non‐small cell lung cancer (NSCLC). (a) and (b) Circ_0072088 in tumor tissues, 16 HBE, A549, and H1299 cells was measured using reverse transcription‐quantitative polymerase chain reaction (RT‐qPCR) assay. (c) and (d) Subcellular fractionation analysis of circ_0072088 cellular localization. ***p < 0.001.

Effects of circ_0072088 knockdown on NSCLC cell malignant behavior

Considering the high expression of circ_0072088 in NSCLC, we knocked down circ_0072088 expression in A549 and H1299 cells. As shown in Figure 2a, the transfection efficiency of si‐circ_0072088#1 and si‐circ_0072088#2 was assessed and exhibited in these two cell lines. Subsequently, transwell assay showed that the capacities of migration and invasion were obviously constrained due to the downregulation of circ_0072088 in A549 and H1299 cells (Figure 2b, c). Apart from that, enhanced apoptosis of A549 and H1299 cells was observed by circ_0072088 deficiency relative to their control groups (Figure 2d). Similar to the flow cytometry assay results, the knockdown of circ_0072088 elicited an evident improvement in caspase‐3 activity in tumor cells relative to their controls (Figure 2e). It has been confirmed that glycolysis is the way in which tumor cells acquire energy, and that increase in glucose uptake and lactate production is a vital characteristic of glycolytic activation. ²⁵ , ²⁶ Thus, we further confirmed the effect of circ_0072088 silencing on glycolytic metabolism in NSCLC cell lines. As seen in Figure 2f, g, the reintroduction of si‐circ_0072088 could decrease glucose consumption and lactate production. Whereafter, we detected the expression of MMP9 (a related factor of migration/invasion) and HK2 (a glycolytic pathway‐related enzyme) in NSCLC cells. As presented in Figure 2h, in MMP9 and HK2 the protein levels were reduced after the transfection of si‐circ_0072088, verifying the repression role of circ_0072088 downregulation. Additionally, we further explored the effect of circ_0072088 on tumor growth in vivo by using a xenograft tumor mouse model. First, we detected the transfection efficiency of sh‐circ_0072088 in A549 cells (Figure 2i). Then our results showed that the tumor volume and weight decreased by circ_0072088 silencing in this xenograft formation assay (Figure 2j, k), indicating the inhibitory role of circ_0072088 knockdown on NSCLC cell growth in vivo. Overall, we determined that circ_0072088 deficiency could suppress NSCLC progression.

FIGURE 2.

The effects of circ_0072088 knockdown on non‐small cell lung cancer (NSCLC) cell malignant behavior. Tumor cells were transfected with si‐NC, si‐circ_0072088#1, and si‐circ_0072088#2. (a) The relative level of circ_0072088 was determined using reverse transcription‐quantitative polymerase chain reaction (RT‐qPCR assay). (b) and (c) Migration and invasion were detected using transwell assay. (d) and (e) Apoptosis rate and caspase‐3 activity analysis were evaluated using flow cytometry assay and special kit. (f) and (g) The corresponding kit analysis of glucose consumption and lactate production. (h) Protein levels of MMP9 and HK2 were determined using western blot assay (from different gels). (i) Circ_0072088 was examined in A549 cells with sh‐NC or sh‐circ_0072088 using RT‐qPCR assay, followed by inoculating subcutaneously into the nude mice. (j) and (k) Measurement of tumor volume and weight was carried out. ***p < 0.001.

Circ_0072088 directly interacted with miR‐1225‐5p

Then, we further explored the mechanism of action of circ_0072088 using online software. The results showed that miR‐1225‐5p has some binding sites with circ_0072088 (Figure 3a). Subsequently, in order to verify the interaction was mediated by the putative binding site, a dual‐luciferase reporter assay was carried out in NSCLC cells. As shown in Figure 3b, c, forced expression of miR‐1225‐5p resulted in a remarkable decline in the luciferase activity of the wt‐circ_0072088 reporter vector, but not that of the mut‐circ_0072088 reporter in tumor cells. Interestingly, downregulation of miR‐1225‐5p in 30 NSCLC tumor tissues was noticed versus 30 adjacent normal tissues (Figure 3d). Meanwhile, we further confirmed that the miR‐1225‐5p level was increased in NSCLC cell lines versus 16 HBE cells (Figure 3e). After that, we found that the miR‐1225‐5p level was strikingly decreased caused by circ_0072088 silencing relative to their respective control groups (Figure 3f, g). Collectively, circ_0072088 interacted with miR‐1225‐5p to block its expression.

FIGURE 3.

Circ_0072088 directly binds to miR‐1225‐5p. (a)–(c) Potential binding sites between circ_0072088 and miR‐1225‐5p were predicted, followed by verification using a dual‐luciferase reporter assay. (d) and (e) Reverse transcription‐quantitative polymerase chain reaction (RT‐qPCR assay) was used to assess miR‐1225‐5p in si‐NC, si‐circ_0072088#1, and si‐circ_0072088#2‐transfected cells. (f) and (g) MiR‐1225‐5p level was determined in tumor tissues, normal tissues, 16 HBE cells, and tumor cells. ***p < 0.001.

Circ_0072088 deficiency could suppress NSCLC cell malignant behavior by targeting miR‐1225‐5p

The association between circ_0072088 and miR‐1225‐5p on NSCLC progression was further explored. As shown in Figure 4a, miR‐1225‐5p was expressed at a low level in miR‐1225‐5p‐transfected tumor cells, indicating that the transfection efficiency was successful. Then, circ_0072088 silencing enhanced miR‐1225‐5p expression in tumor cells, and these influences were greatly counteracted in miR‐1225‐5p (Figure 4b). Functionally, the inhibitory effect of circ_0072088 downregulation on migration and invasion of A549 and H1299 cells was overturned by miR‐1225‐5p knockdown (Figure 4c, d), as evidenced by increased MMP9 level (Figure 4i). In addition, the reduced miR‐1225‐5p mitigated the facilitation role of si‐circ_0072088 on apoptosis rate in tumor cells (Figure 4e), accompanied by lowered caspase‐3 activity (Figure 4f). In terms of glycolysis, si‐circ_0072088‐mediated decrease in glucose consumption and lactate production level was significantly abrogated after the cotransfection with miR‐1225‐5p in A549 and H1299 cells (Figure 4g, h), as described by higher HK2 (Figure 4i). All in all, circ_0072088 knockdown repressed NSCLC progression by targeting miR‐1225‐5p.

FIGURE 4.

MiR‐1225‐5p downregulation could abolish circ_0072088 knockdown‐mediated non‐small cell lung cancer (NSCLC) cell malignant behavior. (a) The relative level of miR‐1225‐5p was assessed in miR‐NC or miR‐1225‐5p‐transfected tumor cells. (b)–(i) Tumor cells were transfected with si‐NC, si‐circ_0072088#1, si‐circ_0072088#1 + in‐miR‐NC, and si‐circ_0072088#1 + miR‐1225‐5p. (b) miR‐1225‐5p was assessed by reverse transcription‐quantitative polymerase chain reaction (RT‐qPCR) assay. (c) and (d) The measurement of migration and invasion was performed using transwell assay. (e) and (f) Analysis of apoptosis rate and caspase‐3 activity were conducted by flow cytometry assay and specific kits. (g) and (h) The determination of glucose consumption and lactate production was executed using commercial kits. (i) Assessment of MMP9 and HK2 protein levels was performed (from different gels). ***p < 0.001.

WT1 served as a target of miR‐1225‐5p in NSCLC cells

Next, the possible target mRNAs of miR‐1225‐5p were predicted using the online software TargetScan. As a result, there were some complementary sequences between miR‐1225‐5p and WT1 3'UTR (Figure 5a). Subsequently, the dual‐luciferase reporter assay exhibited that the luciferase activity was significantly dampened in A549 and H1299 cells cotransfected with wt‐WT1 3'UTR and miR‐1225‐5p compared with the miR‐NC group, while the luciferase activity revealed that there was no marked change in NSCLC cells cotransfected with mut‐WT1 3'UTR and miR‐1225‐5p (Figure 5b, c). In addition, the mRNA and protein levels of WT1 were expressed at a high level in 30 tumor tissues versus 30 adjacent normal tissues (Figure 5d, e). Synchronously, we also found a significant upregulation of WT1 in NSCLC cells (Figure 5f, g). In addition, miR‐1225‐5p upregulation hindered WT1 levels in tumor cells (Figure 5h, i). Together, miR‐1225‐5p modulated the abundance of WT1 by interacting with WT1.

FIGURE 5.

WT1 acted as a direct target of miR‐1225‐5p. (a)–(c) Putative binding sites between WT1 3'UTR and miR‐1225‐5p were predicted by TargetScan software and proven using dual‐luciferase reporter assay. (d) and (e) WT1 level was detected in non‐small cell lung cancer (NSCLC) tumor tissues and adjacent normal tissues (from different gels). (f) and (g) WT1 level was determined in 16 HBE cells, A549 cells, and H1299 cells by reverse transcription‐quantitative polymerase chain reaction (RT‐qPCR) assay and western blot assay (from different gels). (h) and (i) WT1 level in miR‐NC or miR‐1225‐5p‐transfected tumor cells was monitored (from different gels). ***p < 0.001.

WT1 overexpression weakened miR‐1225‐5p‐mediated NSCLC progression

We conducted rescue assays to further prove the link between miR‐1225‐5p and WT1. As illustrated in Figure 6a, the WT1 level was increased in pc‐WT1‐transfected tumor cells. The suppressive role of miR‐1225‐5p on WT1 expression was counteracted via the reintroduction of pc‐WT1 in tumor cells (Figure 6b, c). Moreover, the overexpression of miR‐1225‐5p impeded migration and invasion abilities in tumor cells, which was apparently mitigated via WT1 upregulation (Figure 6d, e). Simultaneously, our results also showed that the acceleration of apoptosis rate and caspase‐3 activity caused by miR‐1225‐5p was effectively abrogated after the cotransfection of pc‐WT1 in A549 and H1299 cells (Figure 6f, g). What is more, the miR‐1225‐5p‐triggered decrease in glycolysis was in part attenuated by WT1 upregulation in tumor cells, as described by higher glucose consumption and lactate production (Figure 6h, i). To further validate the effects of miR‐1225‐5p and WT1 on migration, invasion, and glycolysis, we tested the MMP9 and HK2 expression in tumor cells. Data exhibited that enhanced WT1 abolished miR‐1225‐5p‐mediated the decline of MMP9 and HK2 in tumor cells (Figure 6j). Collectively, the miR‐1225‐5p could restrain NSCLC progression by targeting WT1.

FIGURE 6.

WT1 overexpression could reverse the miR‐1225‐5p‐induced decrease in migration, invasion, glycolysis, and elevation in apoptosis. (a) WT1 was determined in pc‐DNA or WT1‐transfected tumor cells (from different gels). (b)‐(j) A549 and H1299 cells were transfected with miR‐NC, miR‐1225‐5p, miR‐1225‐5p + pc‐DNA, and miR‐1225‐5p + WT1. (b) and (c) WT1 level was assessed (from different gels). (d)–(g) Migration, invasion, apoptosis rate, and caspase‐3 activity were tested. (h) and (i) The detection of glucose consumption and lactate production. (j) MMP9 and HK2 protein level was examined (from different gels). ***p < 0.001.

Circ_0072088 affects WT1 expression by sponging miR‐1225‐5p

In order to confirm whether circ_0072088 exerts a carcinogenic role through the miR‐1225‐5p/WT1 axis, we carried out rescue experiments in tumor cells. Western blot assay suggested that si‐circ_0072088 retarded WT1 expression in tumor cells, whereas these impacts were relieved by in‐miR‐1225‐5p (Figure 7a, b). All in all, we concluded that circ_0072088 performed as a sponge of miR‐1225‐5p to regulate WT1 expression.

FIGURE 7.

WT1 was positively regulated by circ_0072088/miR‐1225‐5p. (a) and (b) WT1 level was monitored (from different gels) in tumor cells transfected with si‐NC, si‐circ_0072088#1, si‐circ_0072088#1 + in‐miR‐NC, and si‐circ_0072088#1 + miR‐1225‐5p. ***p < 0.001.

DISCUSSION

Nowadays, with the advent of next‐generation sequencing technology, an increasing number of circRNAs are being gradually identified in various tissues and species. ²⁷ , ²⁸ Furthermore, the unique circular structure of circRNAs confers a special molecular biological function, providing a novel perspective for the diagnosis and prognosis of multiple cancers, ²⁹ including NSCLC. ³⁰ Interestingly, a previous study reported that the altered expression of circRNAs has recently been documented to impact NSCLC carcinogenesis. ³¹ In the present study, our data identified that circ_0072088 exerts a significant increase in NSCLC, similar to the former research. ¹⁷ It has been verified that metastasis and aberrant cell proliferation are the overwhelming cause of mortality in patients with solid tumors. ³² , ³³ The current study indicated that reduced circ_0072088 constrained tumor cell malignant behavior in vitro. Also, in a mouse model experiment, circ_0072088 deficiency‐triggered tumor cell growth inhibition was also demonstrated. Moreover, altered glycolysis, a primary way by which cancer cells obtain energy, has been recognized as a defining hallmark of cancer. ³⁴ , ³⁵ In this study, circ_0072088 knockdown was confirmed to hinder the glycolysis of NSCLC cells, in accordance with a previous report of pancreatic ductal adenocarcinoma. ¹⁶ The above results suggest that circ_0072088 might exert an oncogene in NSCLC development, providing an underlying attractive biomarker.

To date, it has been widely accepted that circRNAs perform as essential regulators in tumor biology through sponging their target miRNAs. ³⁶ , ³⁷ In the study, the latent circ_0072088‐interacting miRNAs were searched, and miR‐1225‐5p was the candidate target miRNA of circ_0072088 for the first time. In addition, some studies suggested that miR‐1225‐5p, a well‐known tumor suppressor, has been verified to repress cell growth and metastasis in different human cancers, ³⁸ , ³⁹ containing NSCLC. ²⁴ In this study, our data highlighted that decreased circ_0072088 blocked tumor cell malignant behavior by upregulating miR‐1225‐5p. Intriguingly, circ_0072088 has also been proven to facilitate cell malignant behavior in colorectal cancer, esophageal squamous cell cancer, and renal cell carcinoma by interacting with multiple miRNAs. ¹⁵ , ⁴⁰ , ⁴¹ These reports suggest that circ_0072088 participates in complex regulatory networks, and endowed cell‐type‐specific management in varying tumors.

Canonically, miRNAs could induce mRNA degradation or translation inhibition via binding to target mRNA, and decreasing the corresponding protein expression. ⁴² In the present study our data first suggested the binding between miR‐1225‐5p and WT1. Further relevant studies have previously indicated that WT1 was originally identified in the homonymous renal neoplasm, representing a pro‐ and antioncogene role in different cancers. ⁴³ , ⁴⁴ , ⁴⁵ What is more, the promoting role of WT1 on NSCLC cell malignant behavior has been verified. ⁴⁶ , ⁴⁷ In agreement with these studies, the upregulation of WT1 was seen in NSCLC in the current study. Furthermore, functional analysis manifested that the elevated expression of WT1 could partly abolish miR‐1225‐5p‐mediated decline in NSCLC cell malignant behavior. More importantly, rescue assays suggested that miR‐1225‐5p inhibitor, at least in part, overturned circ_0072088 knockdown‐mediated WT1 decrease in NSCLC, supporting the regulatory role of the circ_0072088‐miR‐1225‐5p‐WT1 in NSCLC progression.

In conclusion, these findings revealed that circ_0072088 acts as a miR‐1225‐5p sponge and can increase WT1 expression, thereby boosting NSCLC development, which provides a crucial preclinical basis for NSCLC.

AUTHOR CONTRIBUTIONS

Xiaofang Zhu conducted the experiments, supervised the study and drafted the manuscript. Jing Wan and Xu You collected and analyzed the data. Wanli Yang and Lei Zhao contributed the methodology and edited the manuscript.

CONFLICT OF INTEREST STATEMENT

The authors report no conflicts of interest in this work.

Zhu X, Wan J, You X, Yang W, Zhao L. Circular non‐coding RNA circ_0072088 serves as a ceRNA, targeting the miR‐1225‐5p/WT1 axis to regulate non‐small cell lung cancer cell malignant behavior. Thorac Cancer. 2023;14(20):1969–1979. 10.1111/1759-7714.14943

DATA AVAILABILITY STATEMENT

The data sets used and/or analyzed during the current study are available from the corresponding author on reasonable request.