Background
Malignant glioma is the most common primary tumor in the human central nervous system. Despite many advances in surgical techniques, chemotherapeutics and radiation therapy, the prognosis of patients with malignant gliomas remains obstinately poor. Glioma progression depends on tumor blood vessel growth. Glioma cells secrete vascular endothelial growth factor and other pro-angiogenic factors to promote the growth of vascular endothelial cells. Moreover, glial vascular endothelial cells also secrete a variety of factors that promote tumor growth. The interactions of these secreted factors can promote glioma growth [
1]. Therefore, anti-angiogenesis therapeutics are considered as important treatments for malignant glioma, and an in-depth study of angiogenesis in glioma is in need.
The
FUS (Fused in sarcoma) gene is located at chromosome 16p11.2 and consists of 15 exons encoding a protein of 526 amino acids belonging to the FET (FUS/EWS/TAF15) protein family. As a DNA/RNA-binding protein with a gene regulation function, it is involved in regulating intracellular RNA transport, mRNA synthesis, alternative splicing, and polyadenylation site selection [
2]. It has been found that
FUS mRNA or protein expression is up-regulated in liposarcoma [
3], breast cancer [
4], cervical cancer [
5], and other cells.
FUS can promote the malignant progression of non-small cell lung cancer [
6]. Silencing of the
FUS expression inhibits the proliferation and migration of neuroblastoma cells and increased their chemosensitivity to cisplatin [
7]. A recent study has confirmed that
FUS regulates the expression of 19 circRNAs, including
circ_3279 and
circ_5306, via binding to introns flanking the splicing junction [
8]. But the function of
FUS in vascular endothelial cells has not yet been reported.
CircRNA is a non-coding RNA with a covalent loop structure, which can perform biological functions via various modes of regulation. For example, circRNAs can affect gene expression or transcription by regulating transcription and alternative splicing [
9]. CircRNAs may also act as molecular sponges of microRNAs (miRNAs) or competitive endogenous RNAs to regulate translation of the target genes [
10]. Previous studies have shown that circRNAs play regulatory roles in the malignant biological behavior of glioma cells. For example,
circ-TTBK2 and
circ-HIPK3 promote malignant progression of glioma cells [
11,
12].
Hsa_circ_0000177 is significantly up-regulated in human glioma cells, promoting cell proliferation, invasion in vitro, and growth of glioma in vivo [
13]. Human
circ_002136 (
hsa_circ_0000005,
hsa_circCDK11A_001) formed through the looping of linear
CDK11A-VT1 (
Homo sapiens cyclin-dependent kinase 11A transcript variant 1; GenBank: NM_024011) is located at chromosome 1 and is 49,639 bps in length. To date, the function and mechanism of
circ_002136 have not been clarified.
MiRNAs regulate the expression of target genes at the post-transcriptional level via binding to the 3′-untranslated region (3’-UTR) of the target genes. Studies have shown that over-expression of
miR-138-5p suppresses tumor cell proliferation, invasiveness and induces apoptosis in pancreatic cancer [
14] and bladder cancer [
15]. The
SOX13 gene is located at chromosome 1q31.3–32.1 and consists of 14 exons belonging to the SOX gene family (Sex-related region Y, Sry-related high-mobility group box). It has been found that
SOX13 is highly expressed in oligodendroglioma [
16] and can regulate angiogenesis [
17]. However, the roles of
miR-138-5p and
SOX13 in glial vascular endothelial cells and their potential molecular mechanisms remain poorly defined.
SPON2 (
Spondin 2,
Mindin) is a member of the Mindin F-Spondin superfamily, which encodes secreted proteins and extracellular matrix proteins.
SPON2 is highly expressed in ovarian cancer [
18] and breast cancer [
19]. In patients with hepatocellular carcinoma, the expression of
SPON2 is positively correlated with prognosis [
20]. These suggest that
SPON2 plays an essential role in the development of various tumors. However, the function of
SPON2 in vascular endothelial cells of glioma is still unknown.
In our study, we first identified the endogenous expressions of FUS, circ_002136, miR-138-5p, SOX13, and SPON2 in GECs, and explored possible regulatory interactions among the factors mentioned above and their impacts on the GEC angiogenesis. Our results provide a new insight into the mechanism of angiogenesis in glioma as well as new strategies for anti-angiogenesis treatment of glioma.
Methods
Cell culture and the preparation of glioblastoma (GBM) cell-conditioned medium
The immortalized human cerebral microvascular endothelial cell (EC) line hCMEC/D3 was provided by Dr. Couraud from the Institut Cochin, Paris, France. Cells were cultured in endothelial basal medium (EBM-2) (Lonza, Walkersville, MD, USA), supplemented with 5% fetal bovine serum (FBS) “Gold” (PAA Laboratories, Pasching, Austria), 1% penicillin-streptomycin (Life Technologies, Paisley, UK), 1% chemically defined lipid concentrate (Life Technologies), 1 ng/mL human basic fibroblast growth factor (bFGF) (Sigma-Aldrich, Beijing, China), 1.4 μM hydrocortisone, 5 μg/mL ascorbic acid (Sigma-Aldrich), and 10 mM N-2-hydroxyethylpiperazine-N-ethane-sulphonic acid (HEPES) (PAA Laboratories). ECs were limited with the passage below 30. The human glioblastoma cell line (U87MG) and human embryonic kidney 293 T (HEK293T) cell line were purchased from the Shanghai Institutes for Biological Sciences Cell Resource Center (Shanghai, China) and were cultured in high-glucose Dulbecco’s modified Eagle medium (DMEM, GIBCO, Carlsbad, CA, USA), supplemented with 10% FBS. All cells were maintained in a humidified incubator at 37 °C with 5% CO2.
GBM cell-conditioned medium was collected from the human glioblastoma cell line U87 plated in 100-mm-diameter Petri dishes. Cells that had grown to near confluency were washed twice with serum-free medium and incubated in serum-free EBM-2 medium for 24 h. The supernatant was harvested, centrifuged at 2000×g at 4 °C for 10 min and supplemented with 5% FBS, 1% penicillin-streptomycin, 1% chemically defined lipid concentrate, 1 ng/ml bFGF, 1.4 μM hydrocortisone, 5 μg/ml ascorbic acid, 10 mM HEPES, and stored at 4 °C. The GBM cell-conditioned medium was used to culture human cerebral microvascular endothelial cells (hCMEC/D3) for 24 h to produce GECs.
Quantitative real-time PCR
Total RNAs were extracted from ECs and GECs with Trizol reagent (Life Technologies, Carlsbad, CA, USA) according to the manufacturer’s description. RNA concentration and quality were determined for each sample with a Nanodrop Spectrophotometer (ND-100; Thermo Fisher Scientific, Waltham, MA, USA) using the 260/280 nm ratio. One Step SYBR PrimeScript RT-PCR Kit (Takara Biomedical Technology, Dalian, China) was used to qualify the expression levels of
FUS (NM_004960.3),
circ_002136(NM_024011),
CDK11A(NM_024011.3),
SOX13(NM_005686.2), and
SPON2(NM_012445.3). In addition, RNase R was used to confirm the existence of
circ_002136 and eliminate the influence of linear RNAs. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as an endogenous control. The expression levels of
miR-138-5p (NR_029700.1) were detected using the TaqMan MicroRNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA) and TaqMan Universal Master Mix II (Life Technologies). The U6 housekeeping gene was included as an endogenous control. All qRT-PCR reactions were performed using the 7500 Fast RT-PCR System (Applied Biosystems). The relative quantification 2-
ΔΔCt method was applied to calculate the gene expression values. Primers and probes used in this study are shown in Table
1.
Table 1
Primers and probes used for RT-qPCR
Primer | FUS | F: GCCAAGATCAATCCTCCATGAGTAGTG |
R: TCCACGGTCCTGCTGTCCATAG |
circ_002136 | F: CTTTCCGAGACATTTGCTGG |
R: CATGGAGATCACAATAAGGAACTC |
P:FAM + TCTTCTTCTCCTCTGTCTTCC+MGB |
GAPDH | F: GGACCTGACCTGCCGTCTAG |
R: TAGCCCAGGATGCCCTTGAG |
P:FAM + CCTCCGACGCCTGCTTCACCACCT+Eclipse |
CDK11A | F: AGAGGAAGAGGAGGAGGAGGAGAC |
R: CGAACCGTGACTCTGGAACAACC |
SOX13 | F: CTGGACTTCAACCGAAATTTGA |
R: GTTCCTTCCTAGAAACCTCTCC |
SPON2 | F: GATTGTAGACAGCGCCTCAGTTCC |
R: GACGCACTCAGCCTCTTCTTCG |
Probe | MiR-138-5p U6 | 002284(Applied Biosystems) |
001973(Applied Biosystems) |
Western blot assay
Total protein was extracted from cells using ice-cold radioimmunoprecipitation assay (RIPA) buffer (Beyotime Institute of Biotechnology, Jiangsu, China) supplemented with protease inhibitors (10 mg/mL aprotinin, 10 mg/mL phenyl-methylsulfonyl fluoride [PMSF], and 50 mM sodium orthovanadate). The BCA protein assay kit (Beyotime Institute of Biotechnology) was used to determine the protein concentration of the supernatant. Equal amounts of protein samples (50 μg) were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and electrically transferred onto polyvinylidene difluoride (PVDF) membrane (Millipore, Shanghai, China). Non-specific binding was blocked by incubation with 5% fat-free milk in Tris-buffered saline containing 0.1% Tween-20 (TBST) at room temperature for two hours.
The membranes were subsequently incubated with primary antibodies as follows: FUS (1:2000; Proteintech, Chicago, IL, USA), SOX13 (1:500; Proteintech), SPON2 (1:1000; Affinity, Cincinnati, OH, USA), and GAPDH (1:10,000; Proteintech) at 4 °C overnight. The membranes were washed and incubated with HRP-conjugated secondary antibodies (Santa Cruz Biotechnology, Santa Cruz, CA, USA), diluted at 1:5000 at room temperature for two hours. Immunoblots were visualized using an enhanced chemiluminescence kit (ECL; Santa Cruz Biotechnology) and detected by ECL Detection Systems (Thermo Scientific, Beijing, China) and then scanned using Chemi Imager 5500 V2.03 software. The relative integrated density values (IDVs) were calculated by FluorChem 2.0 software and then normalized to that of GAPDH.
Plasmid construction and cell transfection
Short-hairpin RNAs (shRNAs) directed against human FUS, circ_002136, CDK11A, SOX13, and SPON2 gene were constructed in pGPU6/GFP/Neo vector (GenePharma, Shanghai, China) to generate silencing plasmids. The human SOX13 gene coding sequence was ligated into the pIRES2-EGFP vector (GenScript, Piscataway, NJ, USA) to construct the SOX13-over-expression plasmid. pGPU6/GFP/Neo and pIRES2-EGFP empty vectors without targeting sequences were used as negative controls (NCs). GECs were seeded into 24-well plates and transfected with the plasmids via Opti-MEM I and Lipofectamine LTX and Plus Reagents (Life Technologies) when they reached approximately 80% confluence. The stable transfected cell lines were created by selection on culture medium containing 0.4 mg/mL geneticin (G418) (Sigma-Aldrich, St. Louis, MO, USA). G418-resistant cell clones were established after approximately four weeks of use. For co-transfection of FUS (−) and circ_002136 (−), cells with stably knocked down FUS were transfected with pGPU6/circ_002136 (−)/Blasticidin. G418 and Blasticidin dual-resistant clone were selected. qRT-PCR was then performed to measure the transfected efficiencies.
Furthermore,
miR-138-5p agomir [
miR-138-5p (+); GenePharma],
miR-138-5p antagomir [
miR-138-5p (−)], and their respective NCs were transiently transfected into GECs using Lipofectamine 3000 Reagents (Life Technologies). For co-transfection of
circ_002136 (−) and
miR-138-5p agomir/ antagomir or
SOX13 (+) and
miR-138-5p agomir, cells with stably knocked down
circ_002136 or overexpressed
SOX13, or their respective NCs, were transiently transfected with
miR-138-5p agomir/antagomir or their NCs. All transient-transfected cells were harvested after 48 h. The sequences of all shRNA templates are shown in Table
2. The transfection efficiencies are shown in Additional file
1: Figure S1.
Table 2
Sequences of shRNA template
FUS | Sense | CACCGCCCTACGGACAGCAGAGTTTCAAGAGAACTCTGCTGTCCGTAGGGTTTTTTG |
Antisense | CGGGATGCCTGTCGTCTCAAAGTTCTCTTGAGACGACAGGCATCCCAAAAAACCTAG |
Circ_002136 | Sense | CACCGCTATGGAAGACAGAGGAGAAGTTCAAGAGACTTCTCCTCTGTCTTCCATAGTTTTTTG |
Antisense | CGATACCTTCTGTCTCCTCTTCAAGTTCTCTGAAGAGGAGACAGAAGGTATCAAAAAACCTAG |
CDK11A | Sense | CACCAGAUCUACAUCGUGAUGAATTTTCAAGAGAUUCAUCACGAUGUAGAUCUTGTTTTTTG |
Antisense | GATCCAAAAAAAGAUCUACAUCGUGAUGAATTTCTCTTGAAUUCAUCACGAUGUAGAUCUTG |
SOX13 | Sense | CACCGGAAGATCCTGCAAGCCTTCCTTCAAGAGAGGAAGGCTTGCAGGATCTTCCTTTTTTG |
Antisense | GATCCAAAAAAGGAAGATCCTGCAAGCCTTCCTCTCTTGAAGGAAGGCTTGCAGGATCTTCC |
SPON2 | Sense | CACCGGGCGCTGATGAAGGAGATCGTTCAAGAGACGATCTCCTTCATCAGCGCCCTTTTTTG |
Antisense | CCCGCGACTACTTCCTCTAGCAAGTTCTCTGCTAGAGGAAGTAGTCGCGGAAAAAACCTAG |
NC | Sense | CACCGTTCTCCGAACGTGTCACGTCAAGAGATTACGTGACACGTTCGGAGAATTTTTTG |
Antisense | GATCCAAAAAATTCTCCGAACGTGTCACGTAATCTCTTGACGTGACACGTTCGGAGAAC |
Cell viability assay
Cell Counting Kit-8 (CCK-8, Beyotime Institute of Biotechnology) assay was conducted to determine the viability of GECs. Cells were plated in 96-well plates at a density of 2000 cells/well and incubated in GBM cell-conditioned medium for 24 h. Each well was incubated with 10 μL CCK-8 solution at 37 °C for 2 h. Optical density values were evaluated at 450 nm using the SpectraMax M5 microplate reader (Molecular Devices, San Jose, CA, USA).
Cell migration assay
The migration ability of GECs was assessed using a 6.5 mm Transwell with 8.0 μm Pore Polycarbonate Membrane Insert (#3422, Corning, NY, USA). The upper chamber was used to incubate cells resuspended in 200 μL serum-free medium at a density of 5 × 105 cells/mL. The lower chamber was filled with 600 μL GBM cell-conditioned medium.
After incubation at 37 °C for 48 h, non-migrated cells on the top surface of the membrane were carefully removed. Migrated cells on the lower surface of the membrane were fixed with methanol and glacial acetic acid at the ratio of 3:1 and stained with 10% Giemsa solution (Dinguo, Beijing, China). Then, images of the stained cells were taken with an inverted microscope, and the cell numbers in five randomly selected fields were counted for statistical analysis in each well.
Matrigel assay was performed to evaluate the tube formation of GECs. Pre-chilled 96-well plates were coated with 100 μL Matrigel (BD Biosciences, Bedford, MA, USA) per well and incubated to polymerize at 37 °C for 30 min. ECs were resuspended in 100 μL GBM cell-conditioned medium and seeded onto the surface of the polymerized Matrigel at a density of 4 × 105 cells/mL, followed by incubation at 37 °C for 6 h. Olympus DP71 microscopy (Olympus, Tokyo, Japan) was used to acquire three or more images at random from each culture, and ImageJ software was used to measure the total tubule length and the number of branches.
Reporter vector construction and luciferase reporter assays
The putative binding sequences and mutant sequences of miR-138-5p in circ_002136 and SOX13 3′-UTR were amplified by PCR and cloned downstream of the pmirGLO Dual-luciferase miRNA Target Expression Vector (Promega, Madison, WI, USA) to construct a dual-luciferase reporter vector (circ_002136Wt/circ_002136Mut and SOX13–3’-UTR-Wt/SOX13–3’-UTR-Mut, GenePharma). Circ_002136Wt/circ_002136Mut or SOX13–3’-UTR-Wt/SOX13–3’-UTR-Mut dual-luciferase vectors and miR-138-5p agomir (or agomir NC) plasmids were co-transfected into HEK293T cells using Lipofectamine 3000. The dual-luciferase activity was measured 48 h after transfection. The Dual-Luciferase Reporter Assay System (Promega) was used to analyse luciferase activity. Relative luciferase activity was expressed as the ratio of firefly luciferase activity to Renilla luciferase activity.
For the SOX13-SPON2 reporter constructs, the SPON2 promoter region (− 1, 000 to + 200 bp) was amplified from human genomic DNA by PCR and then subcloned into pGL3-Basic-Luciferase vector (Promega) containing a firefly luciferase reporter gene, yielding the wide-type plasmid (SPON2-Wt). In addition, to test the binding specificity, corresponding mutants of putative SOX13 binding sites were created to form the reporter vector SPON2-mutated-types (SPON2-Mut1, SPON2-Mut2, SPON2-Mut3 and SPON2-Mut4) (GenePharma). The human full-length SOX13 gene was constructed into pEX3 (pGCMV/MCS/Neo) plasmid vector (GenePharma). HEK293T cells were co-transfected with pGL3 vector (either with wide-type promoter regions or mutated promoter regions) and pEX3-SOX13 (or pEX3 empty vector) using Lipofectamine 3000. The promoter activity of constructed plasmid was normalized with the co-transfected reference vector (pRL-TK) and expressed as relative to the activity of pEX3 empty vector, which the activity set to 1.
RNA pull-down assay
The interaction between RBP FUS and circ_002136 introns was detected using Pierce Magnetic RNA-Protein Pull-Down Kit (ThermoFisher) according to the manufacturer’s protocols. In brief, circ_002136 transcripts were transcribed using T7 RNA polymerase (Ambion Life, Taoyuan, Taiwan). Biotin RNA Labeling Mix (Ambion Life) was used to biotin-label the purified RNAs, and then the positive control (Input), negative control (Antisense RNA), and biotinylated RNAs were mixed and co-incubated with proteins extracted from ECs at room temperature. Magnetic beads were added to prepare a probe–magnetic bead complex. Then the bead complex was washed with Handee spin columns and boiled in SDS buffer. Finally, retrieved proteins were detected by western blot, including GAPDH as a control.
RNA-binding protein immunoprecipitation assay
GECs lysates from different groups were incubated with RIP buffer containing magnetic beads conjugated with anti-human argonaute 2 (Ago2) antibodies (Millipore, Billerica, MA, USA). Normal mouse IgG (Millipore) was used as NCs. The samples were incubated with Proteinase K, and then immunoprecipitated RNAs were isolated. The RNA concentration was measured using a spectrophotometer (NanoDrop, ThermoScientific), and the RNA quality was assessed using a bioanalyzer (Agilent, Santa Clara, CA, USA). Purified RNAs were extracted and applied in qPCR for reverse transcription analysis.
ChIP assay
ChIP assays were performed using the Simple ChIP Enzymatic Chromatin IP Kit (Cell Signaling Technology, Danvers, MA, USA) according to the manufacturer’s protocol. Briefly, GECs were crosslinked with EBM-2 containing 1% formaldehyde and collected in a lysis buffer containing 1% PMSF. Chromatin was digested by micrococcal nuclease, and 2% aliquots of lysate were used as an input control. Lysates were incubated with 3 μg anti-
SOX13 antibody (Proteintech) or normal rabbit IgG, followed by immunoprecipitation with protein G agarose beads and incubation at 4 °C overnight with gentle shaking. DNA crosslink was reversed by the addition of 5 mol/L NaCl and Proteinase K at 65 °C for 2 h, and finally, DNA was purified. Immunoprecipitated DNA was amplified by PCR using specific primers. Primers used for ChIP PCR are shown in Table
3.
Table 3
Primers used for ChIP experiments
SPON2 | PCR1 | F:TTTACCGAGTGCTAGAGCCG | 165 | 59.6 |
R: AGGCTGCTGTGGCTGTTT | | |
PCR2 | F: CTTACGACGCAGGGTCTGG | 237 | 59.8 |
R:CGGCTCTAGCACTCGGTAAA | | |
PCR3 | F:ACCCAAGAAAATCAGCCAAAGC | 211 | 60.2 |
R:TCACTGTGGAATCGCGTGAG | | |
PCR4 | F:GTGCCCAGCATCTATTCTGGT | 207 | 59.9 |
R: CTACAGCGTCCCACAGACC | | |
FUS | PCR1 | F:AGTGTTTTGCAGTTACAAGACCTG | 97 | 59.8 |
R:GGAAAGTGAGACTCAGAGACCC | | |
PCR2 | F: CGTCTTGGCTCACTGCAACT | 167 | 60.5 |
R:GTCAGGAGTTCGAGACCAGC | | |
In vivo Matrigel plug assay
Four-week-old male BALB/C athymic nude mice were purchased from the Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China). The animals were fed with autoclaved food and water during the experiment. All animal procedures were performed in strict accordance with the protocol approved by the Administrative Panel on Laboratory Animal Care of China Medical University (Shenyang, China). In brief, 3 × 106 GECs resuspended in 400 μL solution containing 80% Matrigel were subcutaneously injected. Plugs were harvested after 4 days and then weighed, photographed, and dispersed in 400 μL PBS (with overnight incubation at 4 °C) to collect the hemoglobin. Hemoglobin content was measured using Drabkin’s reagent solution (Sigma-Aldrich) according to the manufacturer’s instructions.
Statistical analysis
Quantitative data were presented as mean ± standard deviation (SD). GraphPad Prism v5.01 (GraphPad, La Jolla, CA, USA) software was used for statistical analysis. Student’s t-test (two-tailed) or one-way ANOVA, followed by Bonferroni’s post-hoc test, was employed to evaluate all statistical analyses. Differences were considered statistically significant when P < 0.05.
Discussion
This study demonstrated for the first time that FUS, circ_002136, SOX13 and SPON2 were highly expressed, while miR-138-5p was under-expressed in GECs. FUS bound to circ_002136, while circ_002136 acted as a molecular sponge for miR-138-5p, which had a negative regulatory effect on SOX13 and regulated glioma angiogenesis. SOX13 promoted the expression of SPON2 and increased the angiogenic capacity of GECs. SOX13 bound to the FUS promoter region to up-regulate the expression of FUS, forming a positive feedback loop that promoted glioma angiogenesis.
This study found that
FUS was highly expressed in GECs and that
FUS silencing inhibited the viability, migration, and tube formation of GECs. It has been reported that
FUS plays an important role in regulating various biological processes. For example,
FUS and DNA damage-inducible transcript 3 are able to form a fusion oncogene
FUS-CHOP, which increases the invasive capability of human mucus-like and round cell liposarcoma cells [
21].
FUS protein is also a pathological hallmark of neurodegenerative diseases such as amyotrophic lateral sclerosis and frontotemporal lobe degeneration [
22,
23]. This study further found that
circ_002136 was highly expressed in GECs and that
circ_002136 silencing was reduced in GEC angiogenesis, suggesting that
circ_002136 functioned as an oncogene in GECs.
The regulation of glioma development by circRNAs has become a major research focus.
Circ_0001649 regulates glioma cell growth, colony formation, and apoptosis [
24].
Circ-FBXW7 encodes FBXW7-185aa, which can shorten the half-life of c-Myc and slow the proliferation of glioma cells [
25]. CircRNAs also play an important role in regulating EC function.
Circ_0003575 is up-regulated in oxLDL-induced human umbilical vein endothelial cells (HUVECs) and promotes HUVEC proliferation and angiogenic capacity [
26].
CircHECTD1 accelerates silica-induced transformation from pulmonic ECs to mesenchymal cells [
27]. As previously reported,
CDK11A is abnormally expressed in penile squamous cell carcinoma [
28] and hepatocellular carcinoma [
29]. In this study, we observed that the expression of linear
CDK11A was not up-regulated in GECs and that the degradation of
CDK11A by RNase R did not affect the expression of
circ_002136, suggesting that
circ_002136 and linear
CDK11A are two mutually independent RNAs that might perform different functions, a finding consistent with the function of
circ-SHKBP1 [
30].
The present study revealed that
FUS contained a binding site for
circ_002136, which was proven through RNA-IP and RNA pull-down assays. It was further confirmed that
FUS silencing significantly reduced the expression of
circ_002136, while the expression of
CDK11A was not affected. Previous studies have shown that RBPs can interact with circRNAs to perform diverse biological functions. In doxorubicin-induced heart failure, the RBP
Qki5 interacts with circRNAs
Ttn,
Fhod3, and
Strn3 to improve cardiac function [
31]. In lung cancer cells, the RBP
TNRC6A binds to and regulates the formation of
circ_0006916 accelerating lung cancer cell growth [
32].
Our findings demonstrated that
miR-138-5p was down-regulated in GECs and that
miR-138-5p over-expression significantly reduced the angiogenic capacity of GECs, and vice versa, suggesting that
miR-138-5p acted as a tumor suppressor in GECs. These findings followed those of other research groups.
MiR-138-5p inhibits proliferation and enhances radiation-induced DNA damage and autophagy in nasopharyngeal carcinoma [
33].
MiR-138-5p contributed to the TNF-α-induced insulin resistance through inducing autophagy in HepG2 cells by regulating SIRT1 [
34]. Our study also verified that a binding site existed between
circ_002136 and
miR-138-5p, indicating that
circ_002136 might act as a
miR-138-5p sponge to modulate its functions in GECs. Further studies manifested that knockdown of
circ_002136 significantly up-regulated the expression of
miR-138-5p.
MiR-138-5p silencing increased the expression of
circ_002136 and vice versa. These findings, as well as those from the RNA-IP assays, suggested reciprocal repression between
circ_002136 and
miR-138-5p, which might operate in a RISC manner. Furthermore, the co-effect of
circ_002136 and
miR-138-5p on the angiogenesis of GECs was verified. The results showed that
miR-138-5p reversed
circ_002136 knockdown-mediated inhibition of viability, migration, and tube formation of GECs. Accumulated evidence confirms that circRNAs serve as miRNA sponges by targeting miRNAs. For example, circRNA
ZNF609 functions as a competitive endogenous RNA to regulate
FOXP4 expression by sponging
miR-138-5p in renal carcinoma [
35].CircRNA
MYLK binds to
miR-29a and promotes epithelial-mesenchymal transition and xenograft growth, angiogenesis, and metastasis of bladder cancer [
36].
This study confirmed high expression of
SOX13 in GECs.
SOX13-silencing attenuated the angiogenic capacity of GECs and vice versa, indicating that
SOX13 might act as an oncogene in GECs.
SOX13 acts as an effective regulator of embryonic development, stem cell maintenance, tissue homeostasis, and multiple cancer development [
37]. Previous studies have confirmed that
SOX13 is up-regulated in tumors such as renal clear cell carcinoma [
38] and colorectal cancer [
39].
SOX13 regulates T lymphocyte differentiation by promoting the development of γδ T cells and inhibiting the differentiation of αβ T cells [
40].
One of the most common modes of miRNAs action is the inhibition of gene expression at the transcriptional and post-transcriptional levels by binding to the 3′-UTR of the target mRNA [
41]. Our results revealed that
SOX13 was a target gene for
miR-138-5p and that
miR-138-5p binds to the ACCAGC sequence in the 3’-UTR of
SOX13. In GECs,
circ_002136 silencing significantly reduced the expression of
SOX13, while over-expression and silencing of
miR-138-5p significantly reduced or increased, respectively, the expression of
SOX13. Further studies revealed that the reduction in
SOX13 expression mediated by
circ_002136 knockdown was reversed by
miR-138-5p silencing.
MiR-138-5p performs many biological functions by regulating target genes. For example, in non-small cell lung cancer,
miR-138-5p reduces the expression of
GPR124 and suppresses Gefitinib resistance [
42].
Our present study demonstrated that
SPON2 performed oncogene functions in GECs as a result of
SPON2 up-regulating angiogenesis. This was confirmed by the inhibition of angiogenesis in
SPON2-knockdown GECs.
SPON2 is an innate immune modulator of host cells that recruits inflammatory cells and regulates neuronal development. Previous studies have shown that
SPON2 plays an important role in Egr-1-mediated inhibition of proliferation, migration, and tube formation of vascular ECs in colorectal cancer [
43]. In addition, luciferase reporter assays and ChIP assays certified that
SOX13 was directly associated with and activated the
SPON2 promoter, thereby up-regulating
SPON2 expression. Further findings demonstrated that over-expression of
SOX13 significantly increased the expression of
SPON2 and that
SOX13 promoted the angiogenesis of GECs by transcriptionally up-regulating
SPON2. Moreover, our data indicated that
miR-138-5p down-regulated the expression of
SPON2 and inhibited the viability, migration, and tube formation of GECs by negatively regulating
SOX13. Based on the above results, knockdown of
circ_002136 negatively regulated the expression of
SOX13 by targeting
miR-138-5p and further down-regulated
SPON2 expression to inhibit the glioma angiogenesis.
Based on our prediction, ChIP assays were performed to confirm that
SOX13 bound to the
FUS promoter and promoted
FUS transcription. Over-expression of
SOX13 up-regulated
FUS expression, whereas inhibition of
SOX13 diminished
FUS expression, validating our hypothesis that a positive feedback loop was formed between
FUS/
circ_002136/
miR-138-5p/
SOX13 to regulate the angiogenesis of GECs. Similar positive feedback loops have been reported in previous studies. In the
TDP43/SNHG12/miR-195/SOX5 pathway of glioma cells,
SOX5 promotes the transcription of
SNHG12 and forms a positive feedback loop to regulate the biological behavior of glioma cells [
44].
In the GAS5/miR-196a-5p/FOXO1/PID1 (MIIP) pathway of glioma stem cells,
FOXO1 promotes
GAS5 transcription and forms a positive feedback loop that regulates the biological behavior of glioma stem cells [
45]. Regulation of the positive feedback loop and its effect on the function of glioma cells and angiogenesis is now an important research focal point in biomedical science.