ZRANB2/SNHG20/FOXK1 Axis regulates Vasculogenic mimicry formation in glioma
- 11 Downloads
Glioma is the most common intracranial neoplasm with vasculogenic mimicry formation as one form of blood supply. Many RNA-binding proteins and long non-coding RNAs are involved in tumorigenesis of glioma.
The expression of ZRANB2, SNHG20 and FOXK1 in glioma were detected by real-time PCR or western blot. The function of ZRANB2/SNHG20/FOXK1 axis in glioma associated with vasculogenic mimicry formation was analyzed.
ZRANB2 is up-regulated in glioma tissues and glioma cells. ZRANB2 knockdown inhibits the proliferation, migration, invasion and vasculogenic mimicry formation of glioma cells. ZRANB2 binds to SNHG20 and increases its stability. Knockdown of SNHG20 reduces the degradation of FOXK1 mRNA by SMD pathway. FOXK1 inhibits transcription by binding to the promoters of MMP1, MMP9 and VE-Cadherin and inhibits vasculogenic mimicry formation of glioma cells.
ZRANB2/SNHG20/FOXK1 axis plays an important role in regulating vasculogenic mimicry formation of glioma, which might provide new targets of glioma therapy.
KeywordsZRANB2 SNHG20 FOXK1 Glioma Vasculogenic mimicry formation
Cell Counting Kit-8
Forkhead box K1
Long non-coding RNAs
Normal brain tissues
Normal human astrocyte
Quantitative real-time PCR
STAU1 binding site
STAU1-mediated mRNA decay
Vascular endothelial cadherin
Zinc-finger RAN-binding domain containing protein 2
Glioma is globally recognized as the most common primary intracranial neoplasm [1, 2]. Despite the existence of various treatment methods including surgery, radiation and chemotherapy, the median survival time of patients suffering glioma is no more than 15 months [3, 4]. Although glioma tissue is characterized by angiogenesis and vasculogenesis , tumor treatment effects of anti-angiogenic drugs including bevacizumab are far from satisfaction [6, 7]. Vasculogenic mimicry (VM) formation was first discovered in 1999 and regarded as a new form of blood supply independent of blood vessels . The study of VM formation may bring light to the treatment of glioma.
RNA-binding protein (RBPs) complexes are one class of proteins binding specifically to certain RNAs to form RNA-binding proteins (RNPs), which can regulate transcription, editing, alternative splicing, polyadenylation, translocation, etc. Considering these variable functions, RBPs are expected as important targets for cancer treatment . ZRANB2 (zinc-finger RAN-binding domain containing protein 2) is one kind of RNA-binding proteins originally identified in rat juxtaglomerular cells . ZRANB2 could inhibit the BMP (bone morphogenetic proteins) signaling pathway by binding to Smad protein in HEK293T cells . ZRANB2 was also reported highly expressed in ovarian serous papillary carcinoma . However, no report of ZRANB2 expression in glioma tissues and cells and involvement in the regulation of VM formation has been reported.
Long non-coding RNAs (LncRNAs) are non-coding RNA molecules with a total length of more than 200 nucleotides. Recent studies have shown that lncRNAs regulate gene expression in epigenetic regulation, transcriptional regulation, post-transcriptional regulation and translational regulation , which have potential value in diagnosis and treatment of glioma. SNHG20 was originally identified in hepatocellular carcinoma, localized to 17q25.2, and highly expressed in hepatocellular carcinoma, promoting hepatocellular carcinoma proliferation and migration, and was negatively correlated with patient prognosis . It also played a cancer-promoting role in colorectal cancer, non-small cell lung cancer, cervical cancer, and breast cancer [14, 15, 16, 17]. There are no reports of SNHG20 in regulating glioma VM.
The Staufen1 (STAU1)-mediated mRNA decay (SMD) pathway is one of the ways in which lncRNAs degrade mRNAs in mammalian cells. The Alu element of lncRNAs can form the STAU1 binding site (SBS) by specifically binding to the Alu element in the 3’UTR of the target gene. The target gene mRNA is prone to recruit the RNA helicase and ATPase frameshift increase protein 1 (UPF1), forming the complex STAU1-UPF1 which allows the degradation of target gene mRNA [18, 19].
The transcription factor FOXK1 (Forkhead box K1, FOXK1) is an important member of the forkhead family of proteins. Studies have shown that FOXK1 has different levels of expression in different tumors and plays different roles. FOXK1 was highly expressed in colorectal cancer, and FOXK1 and FOXK2 transfered DVL (Dishevelled)-related proteins into the nucleus, which positively regulated Wnt/β-catenin signaling pathway . However, FOXK1 works as a tumor suppressor in breast cancer, and its expression was positively correlated with the prognosis of breast cancer . At present, the regulation of VM in glioma by FOXK1 is unknown.
Matrix metalloproteinase 1 (MMP1) and matrix metalloproteinase 9 (MMP9) belong to the family of metalloproteinases that degrade extracellular matrix. It has been demonstrated that MMP1 and MMP9 were positively associated with VM formation in clear cell renal carcinoma and ovarian cancer, respectively [22, 23]. Vascular endothelial cadherin (VE-Cadherin) had been shown to be highly expressed in gliomas and was associated with tumor malignancy and promotes vasculogenic mimicry formation in glioma . There has been no report whether FOXK1 regulates transcription of MMP1, MMP9 and VE-Cadherin and regulates the VM formation in glioma.
This study first examined the endogenous expression of ZRANB2, SNHG20 and FOXK1 in glioma tissues and U87 and U251 glioma cells, and studied the effects of ZRANB2, SNHG20 and FOXK1 on VM formation in glioma. Then we explored the probable molecular mechanism of ZRANB2, SNHG20 and FOXK1 in regulating VM formation. This study aims to provide new theoretical basis for the study of the mechanism of glioma development and provide new targets for the treatment of glioma.
Human tissue samples
Human glioma tissues and normal brain tissues (NBTs) were acquired from patients diagnosed with glioma undergoing neurosurgeries at Shengjing Hospital of China Medical University. All patients voluntarily signed informed consent before surgery and the consent was permitted by the Ethics Committee of Shengjing Hospital of China Medical University. Surgical resection specimens were placed in liquid nitrogen immediately after isolation before use. The pathological grades were determined by the pathologists according to WHO classification and were divided into low-grade glioma (WHO I-II: n = 15) and high-grade glioma (WHO III-IV: n = 15). Normal brain tissues acquired from patients undergoing traumatic surgeries without any clear brain diseases before (n = 15) were used as negative controls.
The human glioma cell lines (U87 and U251) and human embryonic kidney cell line (HEK293T) were purchased from Shanghai Institutes for Biological Sciences Cell Resource Center. Dulbecco’s modified Eagle medium (DMEM)/high glucose with 10% FBS was used to culture U87 and HEK293T cells while DMEM/F12 medium with 10% FBS was used to culture U251 cells. Normal human astrocyte (NHA) cells (Catalog #1800, 1 million cells per vial) were purchased from ScienCell Research Laboratories (Carlsbad, CA, USA) and were cultured in RPMI-1640 medium. All the cells were maintained in a humidified incubator at 37 °C with 5% CO2.
Quantitative real-time PCR (qRT-PCR)
Total RNA was extracted from human tissue samples and HA, U87 and U251 cells using Trizol reagent (Life Technologies Corporation, Carlsbad, CA). RNA quality and concentration were measured via 260/280 nm absorbance with Nanodrop Spectrophotometer (ND-100, Thermo, Waltham, MA). SNHG20, FOXK1 mRNA, GAPDH mRNA concentration were detected using One-Step SYBR PrimeScript RT-PCR Kit (TakaraBio, Inc., Japan) in 7500 Fast RT-PCR System (Applied Biosystems, USA). The relative quantification (2-△△Ct) was calculated after the expression levels were normalized using GAPDH as endogenous control.
Total proteins were gained after cells were lysed in RIPA buffer (Beyotime Institute of Biotechnology) for 30 min and were then centrifuged at 17,000 g at 4 °C for 45 min. Protein concentrations were determined by BCA protein assay kit (Beyotime Institute of Biotechnology, China). Protein samples were subjected to SDS-PAGE and electrophoretically transferred to PVDF membranes. After incubated in Tween-Tris-buffered saline (TTBS) containing 5% nonfat milk at room temperature for 2 h, membranes were then incubated with primary antibodies as follows: ZRANB2 (1:500, Proteintech, Rosemont, IL), FOXK1 (1:200, Santa Cruz, USA), MMP1 (1:500, Proteintech, Rosemont, IL), MMP9 (1:500, Proteintech, Rosemont, IL), VE-Cadherin (1:500, Affinity Biosciences, USA), GAPDH (1:10000, Proteintech, Rosemont, IL). After washed three times with TTBS, membranes were incubated with horseradish peroxidase conjugated secondary antibodies at room temperature for 2 h. Immunoblots were visualized with ECL (enhanced chemiluminescence) Kit (Beyotime Institute of Biotechnology, China) and scanned using ChemImager 5500 V2.03 software (Alpha Innotech, San Leandro, CA). Integrated density values (IDVs) were calculated using GAPDH as an internal control.
Short-hairpin RNA against ZRANB2 (ZRANB2(−)), short-hairpin RNA against FOXK1 (FOXK1(−)), short-hairpin RNA against STAU1 (STAU1(−)), short-hairpin RNA against UPF1 (UPF1(−)), ZRANB2 full length plasmid (ZRANB2(+)) and their respective non-targeting sequence (negative control, NC) were synthesized (GenePharma, Shanghai, China). Short-hairpin RNA against SNHG20 (SNHG20(−)) and its respective non-targeting sequence (negative control, NC) were synthesized (GeneChem, Shanghai, China). SNHG20 full length plasmid (SNHG20(+)) and FOXK1 full length plasmid (FOXK1(+)) and their respective non-targeting sequence (negative control, NC) were synthesized (GeneScript, Piscataway, Nj, USA). FOXK1 sequence with assumptive SNHG20 binding site was amplified by PCR and cloned into pmirGLO Dual-Luciferase Vector (GenePharama, China) as wild type (FOXK1–3’UTR-Wt). The corresponding FOXK1 sequence without assumptive SNHG20 binding site (GenePharama, China) was cloned into pmirGLO Dual-Luciferase Vector as mutant type (FOXK1–3’UTR-Mut). Cells were seeded in 24-well plates (Corning, NJ, USA) and transfected using Lipofectamine 3000 reagent (Life Technologies Corporation, Carlsbad, CA) according to its instructions. The stable transfected cells were selected using G418 (0.4 mg/ml), puromycin (1 μg/ml) or hygromycin (500 μg/ml). The silencing and over-expressions efficiency were measured by qRT-PCR. To investigate the effect of ZRANB2 on glioma, cells were divided into five groups: control, ZRANB2(−)-NC, ZRANB2(−), ZRANB2(+)-NC and ZRANB2(+) groups. To investigate the effect of SNHG20 on glioma, cells were divided into five groups: control, SNHG20(−)-NC, SNHG20(−), SNHG20(+)-NC and SNHG20(+) groups. To explore whether SNHG20 mediated the regulation of ZRANB2 affecting the behaviors of glioma, cells were divided into five groups: control, ZRANB2(−) + SNHG20(−), ZRANB2(−) + SNHG20(+), ZRANB2(+) + SNHG20(−), ZRANB2(+) + SNHG20(+) groups. To investigate the effect of FOXK1 on glioma, cells were divided into five groups: control, FOXK1(−)-NC, FOXK1(−), FOXK1(+)-NC and FOXK1(+) groups. To explore whether FOXK1 mediated the regulation of SNHG20 affecting the behaviors of glioma, cells were divided into five groups: control, SNHG20(−) + FOXK1(−), SNHG20(−) + FOXK1(+), SNHG20(+) + FOXK1(−) and SNHG20(+) + FOXK1(+) groups.
Cell proliferation assay
Cell Counting Kit-8 (CCK-8, Beyotime Institute of Biotechnology, China) assay was used to evaluate the proliferation of cells. Cells were seeded in 96-well plates at the density of 2000 cells per well. After 72 h, 10 μl CCK-8 solution was added and cells were incubated at 37 °C for 2 h. The absorbance was measured at 450 nm on the SpectraMax M5 microplate reader (Molecular Devices, USA).
Cell migration and invasion assay
Cells were resuspended in 100 μl serum-free medium at the density of 2 × 105 cells per milliliter and seeded in the upper chamber for migration (or precoated with 500 ng/ml Matrigel solution (BD, Franklin Lakes, NJ) for invasion) with 8 μm pore size polycarbonate membrane (Corning, NY, USA) while 600 μl medium containing 10% FBS was added to the lower chamber. 24 h later, cells migrated or invaded to the lower side of membrane surface were fixed and stained 20% Giemsa. Three random fields under a microscope were chosen to count cell numbers.
In vitro VM tube formation assay
24-well plates were coated with 300 μl Matrigel solution per well and placed at 37 °C for 30 min. Cells were resuspended and seeded with serum-free medium at a density of 3 × 105 cells per well in the Matrigel-coated wells and incubated at 37 °C for more than 24 h. Photos of cells were taken under the inverted microscope (Olympus, Tokyo, Japan). Numbers of VM tube structures in three random fields were counted.
CD34-periodic acid-schiff (PAS) dual-staining
5 μm formalin-fixed and paraffin-embedded tissue specimens underwent dewaxed in xylene, hydrated in gradient ethanol and boiled in EDTA antigen-unmasking solution. After cooled gradually at room temperature, the tissue specimens were incubated with peroxide, blocked with goat serum and incubated with CD34 primary monoclonal antibody (1:50, Proteintech, Rosemont, IL) at 4 °C for 16 h. After incubation with secondary antibody at 37 °C for 10 min, the tissue specimens were stained with DAB kit (MaiXin Biotech, China). Periodic acid solution, schiff solution and hematoxylin were used for the next PAS staining. VM density was counted under microscope in five random fields.
Chromatin immunoprecipitation assay
Chromatin immunoprecipitation (CHIP) kit (Beyotime Institute of Biotechnology, China) was used according to the manufacturer’s protocol. 1 × 107 cells were cross-linked with 1% formaldehyde for 10 min and then treated with glycine at room temperature for 5 min. Cells were lysed in buffer solution containing PMSF and resuspended. Chromatin was digested by micrococcal nuclease. Immunoprecipitates were incubated with 3 μg anti-FOXK1 antibody (Santa Cruz, USA) or normal rabbit IgG and incubated with Protein G agarose beads at 4 °C overnight with gentle shaking while 2% lysates were used as input reference. DNA crosslink was reversed with 5 M NaCl and proteinase K at 65 °C for 2 h. Then the DNA was purified and amplified by PCR using the following primers: one of putative binding site of FOXK1 in MMP1 promoter using the primers 5′- TGAGTAAGATATCAGTCTTGACGCA - 3′ and 5′- GTGTCTCCCACCTTTCCCAC - 3′, generating a 101 bp product; the other putative binding site of FOXK1 in MMP1 promoter using the primers 5′- CCTAGCACCAAGGAGCGAAG -3′ and 5′- CCGGATGATGAAAAGGCTGG - 3′, generating a 107 bp product; control group of putative binding site FOXK1 in MMP1 promoter using the primers 5′- TGCTTGTCATAAGGGGTAAAGGA -3′ and 5′- ACCAAATCCTAGCAATGCCT -3′, generating a 177 bp product; the putative binding site of FOXK1 in MMP9 promoter using the primers 5′- AGATCACGCCACTGCACTC -3′ and 5′- CGGGCAGGGTCTATATTCACC - 3′, generating a 147 bp product; control group of putative binding site FOXK1 in MMP9 promoter using the primers 5′- GACAGCCCCAAGTGCCAATA -3′ and 5′- CCCCCACTTGCCATCAATG -3′, generating a 101 bp product; the putative binding site of FOXK1 in VE-Cadherin promoter using the primers 5′- GGATGACACAACTGGCCAGA -3′ and 5′- GACTCCAGCTCTAAGGTGCC - 3′, generating a 119 bp product; control group of putative binding site FOXK1 in VE-Cadherin promoter using the primers 5′- AGGAGTCCCAAGGAGAGCTT -3′ and 5′- TGATGGGGTCAGAATGGCTG -3′, generating a 141 bp product.
Reporter vector constructs and luciferase reporter assay
HEK293T cells were seed in 96-well plates and co-transfected with FOXK1–3’UTR-Wt (or FOXK1–3’UTR-Mut) reporter plasmid and SNHG20(+)-NC or SNHG20(+), respectively. The relative luciferase activities were measured 48 h after co-transfection using Dual-Luciferase reporter assay kit (Promega).
EZ-Magna RNA-binding protein immunoprecipitation kit (Millipore, USA) was used according to the manufacture’s protocol. 3 × 107 cell lysate was incubated with RIP buffer with magnetic beads conjugated with 5 μg human anti-Ago2 or normal mouse IgG (as negative control). Samples were incubated with Proteinase K and immunoprecipitated RNA was then isolated. The concentration of RNA was measured by spectrophotometer (NanoDrop, USA). The RNA samples were purified and analyzed by qRT-PCR.
Nascent RNA capture
Click-iT® Nascent RNA Capture Kit (Thermo Fisher Scientific, USA) was used to detect nascent according to the manufacture’s protocol. Briefly, nascent RNAs were marked with 0.2 mM 5-ethymyl uridine (EU) and the EU-nascent RNA was captured on magnetic beads for subsequent qRT-PCR.
mRNA stability assay
Cells culture medium were added 5 μg/ml actinomycin D (ActD, NobleRyder, China) to inhibit the de novo RNA synthesis. Total RNA was extracted at 0, 3, 6, 9, 12, 15 h and its concentrations were measured by qRT-PCR. The half-life of RNA was determined by its level at certain point of time compared with time zero.
Xenograft mouse model in vivo
For the vivo study, 4-week-old athymic nude mice (BALB/c) were purchased from the Cancer Institute of the China Academy of Medical Science and randomly divided into five groups: control, ZRANB2(−)-NC + SNHG20(−)-NC, ZRANB2(−), SNHG20(−) and ZRANB2(−) + SNHG20(−) groups. All the experiments with mice were conducted strictly in accordance with the Animal Welfare Act approved by the Ethics Committee of China Medical University.
For subcutaneous xenografts, 3 × 105 glioma cells were injected into the right flank area of nude mice. Tumor volumes were measured every 5 days until ultimate 45 days according to the estimate formula: volume (mm3) = length × width2/2. VM of nude mice xenograft model was detected by CD34-PAS dual-staining later. For survival analysis, 3 × 105 glioma cells were stereotactically infected to the right striatum of nude mice. Kaplan-Meier survival curve was applied for the survival analysis.
All descriptive data are expressed as mean ± SD. SPSS 19.0 software was used for statistical analysis with Student’s t-test or one-way ANOVA. Differences were considered statistically significant if P < 0.05.
ZRANB2 was up-regulated in glioma tissues and cells, and knockdown of ZRANB2 inhibited VM formation
SNHG20 was up-regulated in glioma tissues and cells, and knockdown of SNHG20 inhibited VM formation
ZRANB2 promoted VM formation in glioma cells by increasing the stability of SNHG20
FOXK1 was down-regulated in glioma tissues and cells, and overexpression of FOXK1 inhibited VM formation
SNHG20 promoted degradation of FOXK1 through SMD pathway and enhanced VM formation in glioma cells
FOXK1 bound to the promoters of MMP1, MMP9 and VE-cadherin and inhibited transcription
ZRANB2 knockdown combined with SNHG20 knockdown suppressed tumor growth, generated the longest survival time and yielded the lowest VM formation in vivo
RBP exerts a variety of biological regulatory functions in tumors, and binding to lncRNA or mRNA and enhancing its stability had been reported in many studies . This study found that ZRANB2 specifically bound to SNHG20. More importantly, ZRANB2 did not affect the nascent SNHG20 but increased the stability of SNHG20 by prolonging its half-life. Consequently, ZRANB2 promoted VM formation in glioma cells at least by increasing the stability of SNHG20. The similar in vitro and in vivo approaches can be found in previous research . Similar to the results of our study, RBP serine and arginine rich splicing factor 1 (SRSF1) increases the stability of the DNA repair gene DNA ligase 1 (LIG1) mRNA in non-small cell lung cancer and enhances its expression in a mTOR-dependent manner, thereby promoting the proliferation of non-small cell lung cancer cells and attenuated the apoptosis . Poly(rC) binding protein 1 (PCBP1) protein binds to p27 mRNA and enhanced its stability in ovarian cancer cells, thereby enhancing p27 mRNA translation, inhibiting cell cycle and promoting apoptosis . ELAV like RNA binding protein 1 (ELAVL1, HuR) is highly expressed in osteosarcoma, and by binding to high mobility group AT-hook 1 (HMGA1) 3’UTR increases the stability and expression of HMGA1 mRNA, promoting the progression of osteosarcoma ; HuR also binds and increases the expression of lncRNA SRY (sex determining region Y)-box 5 (Sox5) and promotes the migration and invasion of human tongue squamous cell carcinoma . Lin-28 homolog B (LIN28B) protein binds and allows translation of AKT serine/threonine kinase 2 (AKT2) mRNA in ovarian cancer to inhibit apoptosis of ovarian cancer cells via AKT2/FOXO3A/BIM axis ; LIN28B also promotes proliferation and migration of high-grade serous ovarian cancer by binding to NEAT1 and enhancing its stability .
Early studies have shown that lncRNA regulates the progression of a variety of tumors including epigenetic regulation, transcriptional regulation, post-transcriptional regulation, translational regulation [34, 35, 36]. For example, HOX transcript antisense RNA (HOTAIR) inhibits cyclin dependent kinase inhibitor 2A (CDKN2A) promoter activity by DNA methylation in osteosarcoma cells . Moreover, metastasis associated lung adenocarcinoma transcript 1 (Malat1) regulates SP1 downstream protein expression in lung cancer by enhancing the stability and transcriptional activity of SP1 . In gastric cancer, knockdown of EGFR-AS1 reduces EGFR expression by decreasing the stability of EGFR mRNA . AFAP1-AS1 regulates the expression of miR-423-5p in nasopharyngeal carcinoma through competitive endogenous RNA pathway . H19 acts as a precursor of miR-673 in breast cancer to enhance breast cancer cell proliferation and invasion . Similarly, our study finds that SNHG20 is highly expressed in glioma tissues and cells, and knockdown of SNHG20 reduces VM of glioma by reducing the expression of VM-related molecular MMP1, MMP9, VE-Cadherin. The results suggest that SNHG20 plays as oncogenic role in gliomas. Earlier studies also prove SNHG20 as tumor-suppressive gene in other tumors consistent with our findings. For example, knockdown of SNHG20 inhibits the expression of β-catenin, inhibits the Wnt/β-catenin signaling pathway, and increased the apoptosis of ovarian cancer cells . SNHG20 is up-regulated in osteosarcoma tissues and cells and knockdown of SNHG20 up-regulates the expression of apoptosis-related protein Bax, decreases the expression of Bcl-2, inhibits the activity of caspase-3 and caspase-9, and promotes the apoptosis of osteosarcoma cells . SNHG20 is also highly expressed in non-small cell lung cancer, down-regulates the expression of the cell cycle-associated protein kinase inhibitor P21, which has been shown to directly bind to and affect G1/S transformation related kinases such as CyclinD/CDK4, CyclinD/CDK6, and CyclinE/ CDK2, and promotes tumor cell proliferation, induces apoptosis, and leads to cell cycle arrest . In addition, in colorectal cancer, high expression of SNHG20 is an important independent risk factor for colorectal cancer patients and silencing SNHG20 increases p21 mRNA and protein expression levels and reduced CyclinA1 expression . SNHG20 is highly expressed in cervical cancer and promotes proliferation and invasion of cervical cancer cells through the miR-140-5p/ADAM10/MEK-ERK axis . In addition, studies find that SNHG20 regulates breast cancer cell proliferation, migration and invasion in breast cancer through the miR-495 / HER2 axis . In bladder cancer, SNHG20 promotes bladder cancer cell proliferation and inhibits apoptosis through the Wnt/β-catenin pathway . Moreover, SNHG20 is up-regulated in hepatocellular carcinoma and can be used as an independent prognostic factor .
Staufen1 (STAU1)-mediated mRNA decay is one way of mRNA degradation. For instance, lncRNA-TINCR can form TINCR-STAU1 complex to degrade cyclin dependent kinase inhibitor 2B (CDKN2B) mRNA through the SMD pathway in gastric cancer . During somatic differentiation, the differentiation-related protein KRT80 mRNA binds to TINCR and STAU1 via its Alu elements and finally is degraded by UPF1 . Besides, the paired box 3 (PAX3) mRNA whose product is a protein essential for myoblast differentiation is also degraded by the SMD pathway . In addition, the SMD pathway mediates the degradation of growth associated protein 43 (GAP43) mRNA in mouse C2C12 myoblasts . In our study, dual luciferase assay and RNA-IP assay demonstrate that SNHG20 and FOXK1 mRNA bind to STAU1 through the SBS site. Knockdown of STAU1 or UPF1 prolongs the half-life of FOXK1 mRNA and enhanced its stability expression. This study demonstrates for the first time that SNHG20 promotes the degradation of FOXK1 mRNA through the SMD pathway and regulates the VM formation in gliomas.
FOXK1 belongs to transcription factor of forkhead family and plays various biological functions in different tumors. For example, in colorectal cancer, FOXK1 acts as a DVL-interacting protein, and transfers DVL-related proteins into the nucleus with FOXK2, which positively regulates the Wnt/β-catenin signaling pathway . In lung cancer, FOXK1 acts as a downstream target of circMAN2B2/miR-1275 and plays a role as an oncogene . It is reported that overexpression of FOXK1 inhibits apoptosis of esophageal cancer cells and promoted cell proliferation and migration . However, the expression level of FOXK1 in breast cancer is significantly lower than that in adjacent lung cancer tissues; FOXK1 knockdown decreases the expression of E-cadherin and increases the expression of N-cadherin, which suggests that FOXK1 inhibits the invasion of breast cancer cells by inhibiting epithelial-mesenchymal transition (EMT) . Our study finds that FOXK1 is down-regulated in glioma tissues and cells and is negatively correlated with pathological grade of glioma. Overexpression of FOXK1 significantly inhibits the proliferation, migration and invasion of glioma cells. FOXK1 significantly reduces VM-related molecules MMP1, MMP9 and VE-Cadherin and reduced VM in glioma cells. In consideration of the different reports of FOXK1 as an oncogene or tumor suppressor in different tumors, we refer to the TCGA database for FOXK1 expression in glioma, as well as survival curves (Fig. 4c-d). Compared with tissues, FOXK1 expresses lower in glioma tissues, but the number of cases in NBTs is only 5 and the results shows no significant difference. However, survival curve analysis shows that glioma cases of FOXK1 high-expression have longer survival than that with FOXK1 low-expression (P < 0.0001). Statistical analysis of the TCGA database partly caters to the results of this study.
Moreover, further researches are needed to check for the results on tumor cells issued from fresh tumors and maintained in culture. Moreover, whether knockdown and overexpression of these molecules in normal cells could induce similar VM-related effects needs further studies.
In summary, our study found for the first time that ZRANB2 and SNHG20 are up-regulated in glioma tissues and glioma cells. ZRANB2 bound to SNHG20, and SNHG20 acted on FOXK1 via SMD pathway. Knockdown ZRANB2 reduced the stability of SNHG20, and reduced the degradation of FOXK1 mRNA by SMD pathway subsequently. FOXK1 inhibited transcription by binding to the promoters of MMP1, MMP9 and VE-Cadherin and inhibited the proliferation and migration, invasion and VM formation of glioma cells. This study demonstrated that the ZRANB2/SNHG20/FOXK1 axis played an important role in regulating the expression of VM-related molecules MMP1, MMP9 and VE-Cadherin, and regulated the VM formation of glioma. Our results reveal a new mechanism that identifies novel possible targets which might be useful for future therapeutic interventions.
This work is supported by grants from the Natural Science Foundation of China (81872073, 81872503, 81672511 and 81573010), Liaoning Science and Technology Plan Project (No.2017225020, 2015225007), Project of Key Laboratory of Neuro-oncology in Liaoning Province (112-2400017005), special developmental project guided by central government of Liaoning Province (No. 2017011553-301) and outstanding scientific fund of Shengjing hospital (No. 201304).
This work is supported by grants from the Natural Science Foundation of China (81872073, 81872503, 81672511 and 81573010), Liaoning Science and Technology Plan Project (No.2017225020, 2015225007), Project of Key Laboratory of Neuro-oncology in Liaoning Province (112–2400017005), special developmental project guided by central government of Liaoning Province (No. 2017011553–301) and outstanding scientific fund of Shengjing hospital (No. 201304).
Availability of data and materials
The datasets during this study are available from the corresponding author on reasonable request.
YL contributed to the experiment design, manuscript draft, and data analysis. XL contributed to the experiment implementation, manuscript draft and data analysis. YX designed the experiments. XL, SS, CY and JC Performed the experiments. ZL analyzed the data. LL, JM and TM conceived or designed the experiments, performed the experiments, and wrote the manuscript. All authors read and approved the final manuscript.
Ethics approval and consent to participate
Human glioma tissues and normal brain tissues were acquired from patients diagnosed with glioma undergoing neurosurgeries at Shengjing Hospital of China Medical University. All patients voluntarily signed informed consent before surgery and the consent was permitted by the Ethics Committee of Shengjing Hospital of China Medical University.
For the vivo study, 4-week-old athymic nude mice were purchased from the Cancer Institute of the China Academy of Medical Scienc. All the experiments with mice were conducted strictly in accordance with the Animal Welfare Act approved by the Ethics Committee of China Medical University.
Consent for publication
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
- 15.Guan YX, Zhang MZ, Chen XZ, Zhang Q, Liu SZ, Zhang YL. Lnc RNA SNHG20 participated in proliferation, invasion, and migration of breast cancer cells via miR-495. J Cell Biochem. 2018;119:7971-81.Google Scholar
- 28.Martinez-Terroba E, Ezponda T, Bertolo C, Sainz C, Remirez A, Agorreta J, et al. The oncogenic RNA-binding protein SRSF1 regulates LIG1 in non-small cell lung cancer. Lab Investig. 2018.Google Scholar
- 42.He S, Zhao Y, Wang X, Deng Y, Wan Z, Yao S, et al. Up-regulation of long non-coding RNA SNHG20 promotes ovarian cancer progression via Wnt/beta-catenin signaling. Biosci Rep. 2018. https://doi.org/10.1042/BSR20170681.
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.