Kidney cancer (mainly referring to nephroblastoma) and corresponding non-cancerous kidney tissues were collected from 10 patients at Shenzhen Children's Hospital (Shenzhen, China) between December 2019 and June 2020. The distance between the tumor and the matched normal adjacent tissue was >2 cm and it was histologically confirmed to be non-cancerous. The inclusion criteria were as follows: i) Diagnosed with kidney cancer; ii) had not received other adjuvant treatments including chemotherapy and radiotherapy prior to surgery; and iii) agreed to participate in the study. The exclusion criteria were as follows: i) Failed to cooperate with researchers; and ii) diagnosed with other diseases. These patients included 6 male and 4 female patients with an age range of 7–50 months (average age, 29.04±14.13 months). The age below 18 years was the most frequently used cutoff point for kidney cancer in China (24). The present study was approved by the Ethics Committee of Shenzhen Children's Hospital (Shenzhen, China). Written informed consent was obtained from the parents or legal guardians of the patients with kidney cancer prior to study enrollment. Animal experiments were performed in accordance with the guiding principles of the Institutional Animal Care and Use Committee of Shenzhen Children's Hospital (Shenzhen, China). Detailed information regarding the demographic and clinicopathological characteristics is provided in Table SI.

A total of three kidney cancer cell lines were used in the present study: The kidney Ewing sarcoma cell line SKNEP1, the kidney rhabdoid tumor cell line G401 and the kidney nephroblastoma cell line HANB. A cell line originally derived from human embryonic kidney cells, 293T, was also used. All cell lines were purchased from the Type Culture Collection of the Chinese Academy of Sciences and were cultured in RPMI-1640 and McCoy's 5A medium (HyClone; Cytiva) supplemented with 10% FBS (HyClone; Cytiva) at 37°C with 5% CO₂. The authenticity of all cell lines was verified by DNA fingerprinting (short tandem repeat profiling).

The full-length sequence of circKL (Geneseed Biotech Co., Ltd.) was cloned into a pLCDH vector (BioVector NTCC, Inc.), which was subsequently cotransfected with two assistant vectors pMD2.G (cat. no. 12259; BioVector NTCC, Inc.) and psPAX2 (cat. no. 12260; BioVector NTCC, Inc.) into 293T cells to produce a lentivirus. The control group was treated with the lentiviral vector. Each cell line was then transfected with the circKL overexpression lentivirus. Cells were subsequently selected following exposure to puromycin for 7 days, after which the results were validated by reverse transcription-quantitative (RT-q)PCR analysis.

According to the manufacturer's protocol, total RNA was extracted from kidney cancer tissues or cultured cell lines using the TRIzol^® reagent (Invitrogen; Thermo Fisher Scientific, Inc.), after which cDNA was synthesized using the PrimeScript^® RT Master Mix (Takara Bio, Inc.). NE-PER Nuclear and Cytoplasmic Extraction Reagents (cat. no. 78833; Thermo Scientific) were utilized to isolate the nuclear and cytoplasmic portions of cellular RNA. Subsequently, qPCR was performed using the SYBR^® Premix Ex Taq^™ II kit (Code: DRR081; Takara Bio, Inc.) and an ABI 7900 Sequence Detection system (Applied Biosystems; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. The thermocycling conditions were as follows: 95°C for 1 min, followed by 40 cycles of 95°C for 10 sec and 60°C for 30 sec according to the manufacturer's protocol. The primers used for qPCR were as follows: circKL forward, 5′-ATGGAATCGATGACGGGCTG-3′ and reverse, 5′-GCTTAGGGCAATGGACACCT-3′; linear KL forward, 5′-GTGCGTCCATCTGGGATACG-3′ and reverse, 5′-TGTCGCGGAAGACGTTGTT-3′; circ0056949 forward, 5′-GCATCTTTGACTGCCCCAATG-3′ and reverse, 5′-ATCCACACAGTCGTTGCGTT-3′; circ0056860 forward, 5′-ACGTGTTATTGATCTTCGCTGT-3′ and reverse, 5′-TTCAGCATCATTTGTCAATGGC-3′; circ0056861 forward, 5′-TCTTCGCTGTAACCCAAGAACA-3′ and reverse, 5′-GGAGACAGGGTTTTCGATGA-3′; circ0039504 forward, 5′-AGCTATGGCTGGAACTTCACC-3′ and reverse, 5′-ACATTCCGAAGAAGGTGCCAT-3′; miR-182-5p forward, 5′-ATCACTTTTGGCAATGGTAGAACT-3′ and reverse, 5′-TATGGTTTTGACGACTGTGTGAT-3′; GAPDH forward, 5′-ACAACTTTGGTATCGTGGAAGG-3′ and reverse, 5′-GCCATCACGCCACAGTTTC-3′. The 2^−ΔΔCq method was used to determine the fold change of expression (25).

Total protein was isolated from kidney tissues and cells using radioimmunoprecipitation (RIP) assay lysis buffer (Nanjing KeyGen Biotech Co., Ltd.), which was then added to PMSF to prevent degradation. Equal quantities of protein (20 µg) were resolved by 10% SDS-PAGE, separated and transferred to PVDF membranes (EMD Millipore) for 2 h at 300 mA. After blocking the membranes with 5% skimmed milk (Nestle) for 2 h at room temperature, the membrane was then incubated with the following antibodies overnight at 4°C: Anti-FBXW7 (1:1,000 dilution; cat. no. ab109617; Abcam) and anti-GAPDH (1:1,000 dilution; cat. no. ab8245; Abcam). Samples were then incubated with horseradish-peroxidase-conjugated secondary antibody (1:2,000 dilution; cat. no. ab288151; Abcam) at room temperature for 1 h. GAPDH was used as an internal control. Band densitometry analysis was performed using ImageJ software (version 1.8.0.112; National Institutes of Health).

After 3 µg of total RNA was extracted from SKNEP1 kidney cancer cells, samples were treated with Ribonuclease R (RNase R) (5 U/µg; cat. no. R0301; Geneseed, Inc.) or control solution for 20 min at 37°C. Purification was then performed using an RNeasy MinElute Cleanup Kit (cat. no. 74204; Qiagen GmbH) and the RNAs were quantified by RT-qPCR analysis.

SKNEP1 kidney cancer cells were exposed to 3 µg/ml actinomycin D (cat. no. SBR00013; MilliporeSigma) to degrade the linear mRNA transcript for 0, 8, 16 or 24 h. SKNEP1 cells were subsequently harvested, after which the stability of circKL and linear KL mRNA was analyzed by RT-qPCR.

G401 and SKNEP1 kidney cancer cells were digested and resuspended. Empty vector-transfected and circKL overexpression vector-transfected cancer cells (each, 5,000 cells/well) were seeded into a 96-well plate and incubated for 48 h at 37°C. Subsequently, 10 µl CCK-8 solution (cat. no. C0037; Beyotime Institute of Biotechnology) was added, followed by incubation at 37°C for 1 h prior to optical density measurement at 450 nm using a microtiter plate reader (Epoch 2; BioTek Instruments, Inc.).

To assess the colony formation ability of the cells 1×10³ cells were seeded in six-well plates and incubated at 37°C for 2 weeks. When macroscopic colonies (>50 cells) were evidently observed, the cells were fixed with 100% methanol for 15 min at room temperature and stained with 0.5% crystal violet for 10 min at room temperature. The colonies were counted using an inverted light microscope (magnification, ×100; Carl Zeiss AG). The numbers of colonies were then counted and measured using ImageJ software (version 1.8.0.112; National Institutes of Health). The colony formation efficiency was calculated as the number of colonies/plated cells ×100%.

A total of 3×10⁴ cells in serum-free RPMI-1640 medium (HyClone; Cytiva) were resuspended and added to the upper chamber of Transwell plates (8 µm pore size; Cell Biolabs, Inc.). The upper chamber of Transwell plates were not coated with Matrigel^® for migration assays. Furthermore, 600 µl medium containing 10% FBS was added to the lower chamber. After incubation for 24 h at 37°C, cells on the upper side of the filter were removed using a cotton swab. The cells which had migrated to the lower surface of the membrane were fixed using 4% paraformaldehyde and stained using 0.5% crystal violet at 37°C for 15 min. Cells were then counted in five different fields (magnification, ×200) under a light microscope (Carl Zeiss AG).

G401 and SKNEP1 kidney cancer cells (4×10⁵) were seeded in 6-well plates and transfected with vector or circKL. Subsequently, after a 70–80% confluent culture was reached, a linear wound was scratched with a sterile 200-µl pipette tip. Cells were washed 3 times with PBS and cultured with serum-free medium, after which cell wounds were imaged using an inverted microscope (Carl Zeiss AG) at 0 and 24 h. The degree of wound healing was expressed as the change in width between the two time-points. Representative images were obtained using a light microscope (magnification, ×100; Carl Zeiss AG) and analyzed using ImageJ software (version 1.8.0.112; National Institutes of Health).

G401 or SKNEP1 kidney cancer cells were seeded into 96-well-plates (5×10³ cells/well). miR-182 mimics (5′-UUUGGCAAUGGUAGAACUCACACU-3′) and negative control mimics (miR-NC) (5′-UCACAACCUCCUAGAAAGAGUAGA-3′) were obtained from Guangzhou RiboBio Co., Ltd. The complementary DNA fragment of G401 or SKNEP1 kidney cancer cells containing the wild-type (WT) or mutant type (MUT) sequence from the targeted 3′-untranslated region (UTR) of FBXW7 were subcloned downstream of the luciferase gene using the psiCHEK-2 vector (cat. no. C8021; Promega Corporation). Sequences of primers used to amplify the targeted 3′-UTR of FBXW7 mRNA were as follows: Forward, 5′-CCACTGACAGCTAGACACCTA-3′ and reverse, 5′-GAACCCAGGACAACTTGCCA-3′. The plasmid with the MUT sequence from the 3′-UTR of FBXW7 mRNA was generated using a Site-Directed Mutagenesis Kit (Shanghai Yeasen Biotechnology Co., Ltd.). With this kit, the binding sites predicted by TargetScan (http://www.targetscan.org) of miR-182-5p and circKL in the 3′-UTR of FBXW7 mRNA were mutated. For transfection, cells were seeded into 24-well plates and cultured overnight. miR-182 mimics (10 µl) or miR-NC (10 µl) and luciferase reporter plasmid containing the WT or MUT 3′-UTR of FBXW7 (5 µg) were transfected into cells using Lipofectamine^® 3000 (Thermo Fisher Scientific, Inc.) reagent 48 h prior to performing dual-luciferase reporter assays. Firefly and Renilla luciferase activities were examined by employing a Dual-Luciferase Reporter Assay system (Promega Corporation) in accordance with the manufacturer's protocol. Renilla luciferase activity was used as a normalization control.

RIP assays for argonaute RISC catalytic component 2 (AGO2) protein were performed using an anti-AGO2 antibody (EMD Millipore) according to the manufacturer's protocols. The relative expression levels of circKL, FBXW7 and miR-182-5p were assessed after RNA purification. For the MS2-based (Escherichia coli Bacteriophage MS2-based) immunoprecipitation assays, MS2 binding site Renilla luciferase (MS2bs-Rluc), MS2bs-circKL and MS2bs-circKL-mutant type (mut) plasmids were constructed using a pcDNA3.1 vector (cat. no. V79020; Invitrogen; Thermo Fisher Scientific, Inc.). Subsequently, 5 µg MS2bs-Rluc, MS2bs-circKL or MS2bs-circKL-mt was transfected into G401 and SKNEP1 cells using Lipofectamine^® 3000 reagent (Invitrogen; Thermo Fisher Scientific, Inc.) prior to performing immunoprecipitation. Normal Mouse immunoglobulin G (IgG) (dilution, 1:150; cat. no. 17–700; EMD Millipore) served as the control control. The degree of miR-182-5p enrichment was measured by RT-qPCR after purification of RNA complexes. For the in vitro assays, three replicates were performed for each experiment.

SKNEP1 kidney cancer cells (2×10⁷) that stably overexpressed circKL, or control vectors for the control group, were subcutaneously injected into randomly allocated BALB/c nude mice (five mice per circKL group and vector group; body weight, 20–25 g; age, 4 weeks). A total of 10 BALB/c nude mice were purchased from Shanghai Laboratory Animal Research Center. All mice were housed under specific pathogen-free conditions at 26°C and 20% humidity, with a 12-h light/dark cycle and ad libitum access to food and water. The tumors of the mice were measured with Vernier calipers every 4 days and their volumes calculated according to the following equation: 0.5 × width² × length. After 28 days, the nude mice were euthanized by intraperitoneal injection of 80–100 µl pentobarbital sodium (100 mg/kg). The tumors were then extracted and weighed.

A lung metastasis assay was also performed. SKNEP1 cells (5×10⁵) were injected through the tail veins of nude mice (four mice per circKL group and vector group). After 8 weeks, lung tissues were excised while mice were anesthetized with sodium pentobarbital (100 mg/kg). Tumors were subsequently paraffin-embedded and cut into 4-µm sections. The samples were dewaxed in xylene and rehydrated with an ethanol gradient. The sections were stained with hematoxylin (cat. no. ab245880; Abcam) for 5 min at room temperature and then with eosin (cat. no. ab245880; Abcam) for 2 min at room temperature. The number of macroscopically visible lung metastatic nodules was quantified and validated by a light microscope (magnification, ×40, ×100 and ×200; Carl Zeiss AG).

Tumor xenografts were fixed in 4% neutral formalin at room temperature for 24 h. Histology sections (4 µm-thick) were prepared, deparaffinized using xylene and hydrated using a graded series of alcohols. The slides were incubated with Ki-67 antibodies (1:300 dilution; cat. no. 9449; Cell Signaling Technology, Inc.) overnight at 4°C. Subsequently, horseradish-peroxidase-conjugated secondary antibodies (ready-to-use antibody 50–120 µl; cat. no. 8125; Cell Signaling Technology, Inc.) were applied and samples were incubated for 45 min at room temperature. The resultant signal was visualized using 3,3′-diaminobenzidine color reagent staining at room temperature for 3 min, after which the slides were counterstained with hematoxylin at room temperature for 5 min and dehydrated in ethanol and xylene. Finally, the staining was quantified and image acquisition was performed utilizing a light microscope (magnification, ×200; Carl Zeiss AG).

All statistical analyses were performed using SPSS 23.0 software (SPSS, Inc.). Values are expressed as the mean ± standard deviation of at least three independent experiments. Multigroup comparisons were performed using one-way ANOVA followed by Tukey's post hoc test. Comparisons between tumor and adjacent normal tissues were performed using a paired Student's t-test. A paired t-test was also used to compare the expression of circKL between two matched groups. P<0.05 was considered to indicate a statistically significant difference.

Following a previous analysis of high-throughput microarray sequencing data by our group (26), RT-qPCR was performed in the present study to verify the expression level of the top five downregulated circRNAs in 10 pairs of kidney cancer samples and adjacent normal kidney samples (Fig. 1A). The results confirmed that circKL was significantly downregulated in the tumor parts of all tumor-normal tissue pairs. Furthermore, the expression levels of circKL were downregulated in kidney cancer cell lines compared with those in normal kidney 293T cells (Fig. 1B). The circular structure and stability of circKL were further examined by performing RNase R and actinomycin D assays. The RNase R assay results revealed that, in contrast to linear KL mRNA, circKL was resistant to RNA exonuclease (P<0.01; Fig. 1C). Furthermore, actinomycin D assays confirmed that circKL had a significantly longer half-life than linear KL mRNA (P<0.01; Fig. 1D).

A vector that continuously expressed circKL to exogenously introduce circKL was constructed in the present study to explore the potential role of circKL in kidney cancer progression. G401 and SKNEP1 cells were transfected using a lentivirus that stably expressed circKL, after which the efficacy of the overexpression vector was verified (P<0.001; Fig. 2A). CCK-8 and colony formation assays revealed that overexpression of circKL significantly suppressed the proliferation and colony-formation ability of G401 and SKNEP1 cell lines in vitro (P<0.01; Fig. 2B-D). Further examination of the anti-tumor function of circKL was performed in mouse xenograft assays. Tumor volume curves revealed that overexpression of circKL inhibited tumor growth. Similarly, in subcutaneous tumors, the maximum tumor diameter and mean volume in the circKL group (0.984 cm and 650.467 mm³, respectively) were significantly smaller than those in the vector group (1.249 cm and 1061.87 mm³, respectively) after 28 days (P<0.01; Fig. 2E and F). In addition, Ki67 protein expression in the murine xenograft tumors of the two groups was analyzed by immunohistochemistry. The results demonstrated that Ki67 expression was markedly decreased in the tumor tissues of the circKL overexpression vector group (Fig. 2G).

Migration and invasion assays were performed to investigate whether circKL was able to suppress the metastatic capacity of kidney cancer cells. The results revealed that upregulation of circKL significantly inhibited the percentage of wound closure in G401 and SKNEP1 cells (P<0.01; Fig. 3A). The results of the Transwell assay demonstrated that circKL overexpression reduced the migration of G401 and SKNEP1 cells (P<0.01; Fig. 3B). In concordance with the in vitro experimental results, circKL overexpression inhibited the metastasis of SKNEP1 cells in murine xenograft models in vivo (P<0.01; Fig. 3C).

After different cellular proportions were detected by qPCR, it was revealed that circKL predominantly existed in the cytoplasm of cells (Fig. 4A). The Circular RNA Interactome database (freely accessible at http://circinteractome.nia.nih.gov) was therefore used to evaluate the potential interactions between circRNA and various miRNAs (27). Among the miRNA candidates, only miR-182-5p was predicted to bind to the circKL sequence at four possible interaction sites (Fig. 4B). In kidney cancer cell lines, RT-qPCR analysis revealed that miR-182-5p was significantly upregulated (Fig. 4C). Furthermore, the AGO2-related RIP assay confirmed the direct interaction between circKL and miR-182-5p. In addition, it was determined that miR-182-5p was predominantly enriched in the MS2bs-circKL overexpression vector group (P<0.01; Fig. 4D), indicating that circKL directly interacted with miR-182-5p and may act as a sponge for miR-182-5p.

TargetScan (http://www.targetscan.org) was used to predict the potential targeting genes of miR-182-5p (27). Among the possible genes, FBXW7 was identified as a putative downstream target gene of miR-182-5p (Fig. 5A). The results of the qPCR analysis revealed that FBXW7 was markedly downregulated in kidney cancer cells (Fig. 5B). Whether miR-182-5p was able to directly bind to the 3′-UTR of FBXW7 mRNA was subsequently examined. The relative luciferase activity of G401 and SKNEP1 kidney cancer cells was significantly decreased following the transfection of miR-182-5p and the wild-type 3′-UTR-FBXW7 plasmids. However, after co-transfection with the mutated luciferase reporter vector, no such effect was observed (P<0.01; Fig. 5C). The exogenous introduction of miR-182-5p contributed to the reduction of FBXW7 mRNA expression levels (P<0.01; Fig. 5D). In addition, AGO2-related RIP assays revealed that circKL, miR-182-5p and FBXW7 were all highly enriched in the anti-AGO2 G401 and SKNEP1 kidney cancer cell groups (P<0.01; Fig. 5E). Furthermore, the mRNA level of FBXW7 was markedly decreased following circKL overexpression (P<0.01; Fig. 5F). After transfection with miR-182-5p mimics, FBXW7 protein was decreased in both G401 and SKNEP1 kidney cancer cell lines (Fig. 5G). Western blot analysis also revealed that circKL overexpression significantly increased FBXW7 protein levels (Fig. 5H).