This article is mentioned in:

Introduction

Neuroblastoma (NB) is the most common extracranial solid tumor in children and is responsible for 10% of the mortality resulting from all pediatric tumors (1,2). The prognosis of many children's malignant tumors has improved by effective treatments. However, high-risk NB remains intractable, with merely 40% of patients achieving long-term survival despite the availability of multiple therapeutic methods (3,4). The tumorigenesis and progression of NB is reported to be complex, and is influenced by both the external environment and internal genetic factors (4,5). Early accurate diagnosis and timely intervention is important to improve prognosis. Therefore, further research investigating the biological targets and underlying molecular mechanisms is essential for improving the diagnosis and therapy of NB (6).

Long non-coding RNAs (lncRNAs) are generally defined as RNA transcripts that are >200 nt with no protein products (7–9). Non-coding RNAs, once considered as transcriptional waste, are actually a part of the regulatory network of transcriptional and post-transcriptional processes (1,10,11). They are ubiquitously expressed in mammalian genomes, and participate in the regulation of numerous biological processes, including DNA methylation, chromatin remodeling, transcription regulation and translation (10,12). It is reported that several lncRNAs are closely linked to NB initiation and progression, and involved in a variety of tumor-associated biological processes (13,14), such as cancer susceptibility 15 (CASC15) (15), MYCN upstream transcript (lncUsMycN) (16) and ETS1 antisense RNA 1 (pancEts-1) (17).

In our previous study, genome-wide lncRNA analysis was performed to detect NB-associated lncRNAs. Linc01105 was identified as differentially expressed between NB and normal adrenal gland tissues. It was also demonstrated that upregulation of linc01105 in NB was correlated with the International Neuroblastoma Staging System (INSS) stage (14). The present study found that hypoxia inducible factor-1α (HIF-1α) may bind at the linc01105 promoter to activate its transcription. In addition, linc01105 knockdown significantly suppressed proliferation and induced apoptosis via the p53/caspase signaling pathway. Furthermore, it is well established that, lncRNAs can act as microRNA (miRNA) sponges, and thus regulate their ability to target mRNAs (18,19). In the present study, it was demonstrated that linc01105 may inhibit tumor angiogenesis via the linc01105/miR-6769b-5p/vascular endothelial growth factor A (VEGFA) network.

Materials and methods

Human tissues and NB cell line

Human NB and adjacent normal adrenal gland tissues were collected from patients who underwent surgery between January 2011 and January 2017 in the Children's Hospital of Fudan University, Shanghai, China. Tumor tissues were diagnosed as NB by the pathology department and all patients were stage III/IV according to INSS. For the present study, 32 patients were enrolled,18 boys and 14 girls. The age of the patients ranged from two months to ten years. Informed consent was acquired from every patient's legal guardians. This study was approved by the Institute Research Ethics Committee at the Children's Hospital of Fudan University.

The NB cell line SH-SY5Y and SK-N-BE (2), the human umbilical vein endothelial cells (HUVECs) and 293T cells were obtained from the Cell Bank of Type Culture Collection of the Chinese Academy of Sciences in Shanghai. CHLA15, LA-N-5, and CHLA136 were gifts from professor Li Kai's lab of FuDan University. The SH-SY5Y and SK-N-BE (2) cells were cultured in DMEM/F12 (Biological Industries) with 10% fetal bovine serum (Biological Industries) in a humidified incubator with 5% CO2 at 37°C. HUVEC, 293T, and other NB cell lines(CHLA15, LA-N-5, and CHLA136 cells) were cultured in DMEM with 10% FBS (both Gibco; Thermo Fisher Scientific, Inc.). All cell culture dishes and culture plates were purchased from Hangzhou Xinyou Biotechnology Co., Ltd, China.

Nuclear and cytoplasmic extract preparation

Cells were grown to 90–100% confluency. Pre-cooled PBS was used to wash the cells twice. Cytoplasmic extraction buffer (0.5 ml; Invent Biotechnologies, Inc.) was then added to the cells, and set on ice for 5 min. The lysates were transferred to tubes and mixed rigorously for 15 sec. After centrifuging at 12,000 × g for 5 min at 4°C, TRIzol was added to the nuclear (pellet) and cytoplasmic fragments (supernatant) for RNA extraction.

Reverse transcription-quantitative PCR (RT-qPCR)

Total RNA was extracted using TRIzol reagent (Takara Bio, Inc.) and cDNA was synthesized using a cDNA Reverse Transcription kit (Takara Bio, Inc.). qPCR reactions were performed using SYBR Green PCR Master Mix (Takara Bio, Inc.) on the Step One Real-Time PCR system. The PCR program was: 94°C for 10 min, 40 cycles of 94°C for 60 sec and 60°C for 15 sec. GAPDH was used as internal control. Data was analyzed using GraphPad software (GraphPad Software, Inc.) and relative fold changes in expression were calculated using the formula 2−ΔΔCq (20). The following primer sequences were used in this study: Linc01105, forward 5′-TGTGCCATTCCATGTTATA-3′ and 5′-TTGAGGCTGAAGACCAAA-3; and GAPDH, forward 5′-GGAGCGAGATCCCTCCAAAAT-3′ and reverse 5′-GGCTGTTGTCATACTTCTCATGG-3′. U1 and actin were used as controls for the nuclear and the cytoplasmic extracts, respectively, and their primes were: U1, forward 5′-GACGGGAAAAGATTGAGCGG-3′ and reverse 5′-GCCACGAAGAGAGTCTTGAAGG-3′; actin, forward 5′-CATGTACGTTGCTATCCAGGC-3′ and reverse, 5′-CTCCTTAATGTCACGCACGAT-3′.

To quantitatively measure miRNAs, the All-in-OneTM miRNA qRT-PCR Detection kit (GeneCopoeia, Inc.) was used. The experimental procedure included two major steps. Firstly, single-step cDNA synthesis was performed. Poly-A polymerase was used to add poly-A tails to the 3′ end of miRNAs, and M-MLV RTase was used simultaneously, with a unique oligodT adaptor primer, to reverse transcribe the miRNA. Secondly, qPCR detection was performed. The All-in-OneTM miRNA qRT-PCR Mix, containing SYBR Green, specifically detects the reverse transcribed miRNA. The miR-6769b-5p-specific primer (cat. no. HmiRQP3816) and the U6 control primer (cat. no. HmiRQP9001) were purchased from GeneCopoeia, Inc.

Construction of shRNA and stable transfected cell lines

Linc01105 shRNA plasmids were designed and constructed by Genomeditech Co., using the pGMLV-SC5 RNAi vector. The control target sequence was 5′-TTCTCCGAACGTGTCACGT-3′. Three different shRNA target sequences were designed as follows: sh1, 5′-GCTCAGGAGAAAGAGCAAATG-3′; sh2, 5′-GCCTGCTGAGAAGGCTCATCT-3′; and sh3, 5′-GCAGCAACTCCTGTGCATGT-3′. Purified, endotoxin-free lentiviral vector and its auxiliary packaging vector plasmids were co-transfected into 293T cells (Cell Bank of Type Culture Collection of the Chinese Academy of Sciences) using the HG transgene reagent (Genomeditech Co.) for 10–12 h. Enhancing buffer (Genomeditech Co.) was added, and the medium was replaced with fresh medium after 8 h. After 48 h of incubation, the supernatant of the cell line, containing the lentiviral particles, was collected and concentrated to obtain a high titer of lentivirus concentrate, with a final virus titer of 5×108 transduction units/ml. SH-SY5Y cells were infected with different shRNA-expressing lentivirus, and stable infected cells were acquired following puromycin selection.

Cell proliferation and viability assay

Cell proliferation and viability were detected using the Cell Counting Kit-8 assay (Yeasen Biotechnology Co., Ltd.) and the cell viability kit (Beyotime Institute of Biotechnology), according to the suppliers' protocols. Cells were cultured in a 96-well plate at a concentration of 2,000 cells per well in 100 µl medium. The Cell-Light EdU DNA cell proliferation kit (Guangzhou RiboBio Co., Ltd.) was also used to assess cell proliferation. For the EdU assay, 10,000 cells were seeded in 24-well plates. Cells were cultured for 2 h with 50 µM EdU, then 4% formaldehyde was used to fix the cells for 20 min at room temperature. Cells were permeabilized with 0.5% Triton X-100 for 10 min, stained with 200 µl of Hoechst 33342 for 30 min, and visualized under a fluorescent microscope.

Cell apoptosis assay

After digestion with EDTA-free trypsin, cells were collected and washed twice with pre-cooled PBS. Cells (106) were transferred into 5 ml flow tubes. Then, 5 µl of Annexin V-APC and 10 µl of 7-AAD (Yeasen Biotechnology Co., Ltd.) were added on the cells in the dark at room temperature for 15 min. Apoptosis rates (percentages of Q2+Q3) were then analyzed by flow cytometry (BD Celesta; Beckman Coulter, Inc.) within 1 h.

Cell migration and invasion assay

Twelve-well culture plates with 8 µm micropore inserts were used for cell migration and invasion assays. Cells were serum-starved for 24 h prior to the assays. For the migration assay, 2×105 NB cells were placed into the upper wells in DMEM/F12 without FBS for 24 h. For the cell invasion assay, the upper side of the insert was coated with Matrigel (BD Biosciences), and 4×105 NB cells were placed into the upper wells without FBS for 24 h. The cells on the lower sides of the inserts were then fixed with 4% paraformaldehyde for 10 min and stained with 0.1% crystal violet for 30 min. Five fields were randomly selected and captured using a light microscope (Olympus Corporation), and the average count of the five fields was calculated.

Wound healing assay

Cells (1×105) were seeded in the two wells of culture inserts (Ibidi GmbH). After the cells were attached and the insert was removed, a standard gap was left between the cells. The gap closure was observed at ×100 magnification and photographs were captured using light microscopy at 0, 6, 12 and 24 h. Image J (National Institutes of Health) was used to measure the gap width.

Caspase-3 activity assay

Caspase-3 activity was detected using the caspase-3 activity kit (Beyotime Institute of Biotechnology). According to the manufacturer's instructions, cells were mixed with buffers containing caspase-3 substrate and incubated at 37°C for 2 h. Samples were the detected at 450 nm using a plate reader.

Western blotting

Cell proteins were extracted by lysis buffer (Beyotime Institute of Biotechnology), and western blotting was performed with standard procedures (21). Briefly, after blocking, the membranes were incubated with primary antibodies at 4°C overnight. The primary antibodies were as follows: Anti-BCL2 (cat. no. 60178; 1:2,000; ProteinTech Group, Inc.), anti-Bax (cat. no. 50599; 1:2,000; ProteinTech Group, Inc.), anti-caspase-3 (cat. no. 19677; 1:500; ProteinTech Group, Inc.), anti-caspase-9 (cat. No 10380; 1:500; ProteinTech Group, Inc.), anti-active caspase-3 (cat. no. F021507; 1:500; Abways Technology), anti-cleaved caspase-9 (cat. no. F016210; 1:500; Abways Technology), anti-p53 (cat. no. F024201; 1:500 Abways Technology), anti-poly (ADP-ribose) polymerase (PARP; cat. no. AP102-1; 1:1,000; Beyotime Institute of Biotechnology), anti-VEGFA (cat. no 19003; 1:1,000; ProteinTech Group, Inc.), and anti-tubulin (cat. no. ab210797; 1:1,000; Abcam). After washing with TBS/0.5% Tween 20 three times, the membranes were incubated with the relevant secondary antibody (goat anti-rabbit cat. no. CW0156, 1:2,000; or goat anti-mouse cat. no. CW0110, 1:5,000; both ComWin Biotech Co., Ltd.) for 1 h at room temperature. The protein signals were detected using an enhanced chemiluminescence substrate (EMD Millipore).

Transient transfections

Cells (104) were seeded in a 24-well plate one day in advance, and the cell confluence was ~30% at the time of transfection. Small interfering RNA (siRNA) or miRNA mimics or inhibitor (50 pmol) were added to OPTI-MEM (Gibco; Thermo Fisher Scientific, Inc.) to a final volume of 25 µl and mixed well. Then, 1 µl of Lipofectamine 2000 (Thermo Fisher Scientific, Inc.) was added to OPTI-MEM to a final volume of 25 µl. The two solutions were then thoroughly mixed at room temperature for 15 min. The 50 µl of transfection complex was added to cells in 0.45 ml media, and incubated for 48 h. The sequences were as follows: siVEGFA, 5′-GUGCUACUGUUUAUCCGUA-3′; negative control (NC) siNC, 5′-UUCUCCGAACGUGUCACGU-3′; miR-107 mimics, 5′-GCCUUCUGACUCCAAGUCCAGU-3′, miR-378a-3p mimics, 5′-UGAUAGCCCUGUACAAUGCUGCU-3′, miR-6769b-5p mimics, 5′-UCUACUCUUUCUAGGAGGUUGUGA-3′; mimics NC, 5′-GCACUUCUCCUCCCCACCCACCA-3′; miR-6769b-5p inhibitor, 5′-GCACUUCCUCCCCACCCACCA-3′; and inhibitor NC, 5′-UCACAACCUCCUAGAAAGAGUAGA-3′. HIF-1a and VEGFA plasmids (both in pCDH vector) were purchased from Shanghai Generay Biotech Co, Ltd, and transfected (4 µg) into cells with Lipofectamine 2000 (Thermo Fisher Scientific, Inc.). After 48 h of transfection, subsequent experiments were performed.

Luciferase assays

293T cells were seeded in 24-well plates. When cells reached 75–85% density, they were transfected with luciferase reporter plasmids containing the 3′ untranslated region (UTR) of VEGF or the promoter of linc01105 (both constructed by Generay Biotech Co., Ltd.) using Lipofectamine 2000, according to the manufacturer's instructions. After 48 h, luciferase activity (Dual Luciferase Reporter Gene Assay kit; Yeasen Biotechnology Co., Ltd) was measured using a microplate reader and Renilla luciferase activity was used for normalization.

Tube formation assay

HUVECs (5×103) were seeded into a 96-well plate that was pre-coated with Matrigel (BD Biosciences). Conditioned media (CM) supernatant from the stable-infected SH-SY5Y cells were added to the HUVEC cells. A VEGFA antibody (cat. no. 19003; 1:500; ProteinTech Group, Inc.) was used for 6 h pre-treatment in the tube formation assay, as indicated. After 6 h of incubation, tube formation was observed using phase-contrast light microscopy.

Database used for exploring miRNAs

To determine whether linc01105 functions as a miRNA sponge, first the DIANA-LncBase database was used to search for potential miRNA recognition elements on linc01105 (http://diana.imis.athena-innovation.gr). The search term was linc01105, and multiple miRNAs were predicted to bind. The CircNet database (http://circnet.mbc.nctu.edu.tw) was used to search for miRNAs that may bind with VEGFA. Finally, three miRNAs, miR-107, miR-6769b-5p, miR-378a-3p were identified as potential miRNAs that could bind with both linc01105 and VEGFA.

Chromatin immunoprecipitation (ChIP)

ChIP assay was performed as described by Morelli et al (22). The cells of one 10 cm dish were sonicated 4 times for 10 sec at pre-cooled conditions (Fisher Sonic Dismembrator; Thermo Fisher Scientific, Inc.) and then treated according to the standard protocol. Anti-HIF-1 (cat. no. 39665; Active Motif, Inc.) or control rabbit IgG (cat. no. 294670; Abmart, Inc.) antibodies were used to capture chromatin fragments from cell extracts. PCR was used to amplify the DNA fragment with the antibody and the input DNA was used as control.

Chromatin isolation by RNA purification (CHIRP)

Linc01105 antisense DNA (asDNA), β-galactosidase (lacZ) asDNA and linc01105 sense DNA probes were designed using an online probe designer (singlemoleculefish.com). Oligonucleotides were biotinylated at the 3′ end with an 18-carbon spacer arm. Cells were collected and subjected to CHIRP, using the method previously described by Chu et al (23). GO and KEGG analysis were performed using the DAVID Functional Annotation web-based tool (http://david.ncifcrf.gov).

Statistical analysis

All results are presented as mean ± standard deviation. Results from different groups were compared using the Student's test or one-way ANOVA followed by Turkey multiple comparisons test. Statistical analysis was performed using GraphPad Prism 5 (GraphPad Software, Inc.). P<0.05 was considered to indicate a statistically significant difference.

Results

Expression and localization of linc01105

Linc01105 was significantly upregulated in the 32 NB specimens compared with the matched normal adrenal gland tissues (Fig. 1A). The SH-SY5Y (non-MYCN amplification) cell line was selected as the in vitro model for NB in the present study (Fig. 1B), as our previous study had used SK-N-BE (2)C (MYCN amplification) and CHLA15 was too difficult to culture. Nuclear and cytosolic RNA of SH-SY5Y was extracted and detected by RT-qPCR. The differential enrichments of U1 small nuclear RNA and actin were used as controls for the nuclear and the cytoplasmic extracts, respectively. The results demonstrated that linc01105 expression levels were clearly increased in the cytosol relative to the nucleus (Fig. 1C), which suggested that linc01105 was mainly localized in the cytosol and thus may have a key role in post-transcriptional events. In order to investigate the function of linc01105, three individual shRNAs and a negative control shRNA were purchased from Genomditech. Following lentiviral infection, the RT-qPCR results confirmed that shRNA2 had the highest knockdown efficiency, and thus was selected for subsequent experiments (Fig. 1D).

Figure 1. Expression and cellular localization of linc01105 in NB. (A) Linc01105 expression levels in NB and normal tissues were measured by RT-qPCR. (B) Linc01105 expression levels in CHLA-15, CHLA-136, SH-SY5Y, L-A-N-5 and SK-N-BE(2) cells were measured by RT-qPCR. (C) Linc01105 expression in nucleal and cytoplasmic extracts of SH-SY5Y. Actin and U1 were used as cytoplasmic and nuclear internal control genes, respectively. (D) Linc01105 levels in SH-SY5Y cells infected with shRNA lentivirus against linc01105 or negative control were evaluated by RT-qPCR. **P<0.01. NB, neuroblastoma; RT-qPCR, reverse transcription-quantitative PCR; C, cytoplasm; N, nucleus; shRNA, short hairpin RNA.

Knockdown of linc01105 inhibits proliferation and viability

To further explore the biological function of linc01105 in NB cells, a shRNA2 stable-infected line was generated in SH-SY5Y cells. A CCK-8 assay was performed to investigate the effect of linc01105 on cell proliferation. Silencing of linc01105 significantly inhibited NB cell proliferation compared with negative control (Fig. 2A). In addition, cell viability was reduced following linc01105 silencing (Fig. 2B). Finally, similar results were observed with the Cell-Light EdU DNA cell proliferation assay (Fig. 2C). Taken together, these results revealed that downregulation of linc01105 may inhibit cell proliferation.

Figure 2. Linc01105 knockdown results in decreased cell proliferation. (A) CCK-8 assay was used to measure neuroblastoma cell proliferation. (B) Cell viability assay was performed. (C) Cell-Light EdU DNA assay was used to further assess cell proliferation. *P<0.05, **P<0.01 compared with control. shRNA, short hairpin RNA.

Knockdown of linc01105 contributes to increased apoptosis

Compared with the control group, knockdown of linc01105 induced cell apoptosis, as evidenced by flow cytometry analysis (Fig. 3A). To investigate this further, a caspase-3 activity assay was performed to evaluate the activation of caspase-3 (Fig. 3B). Knockdown of linc01105 resulted in increased caspase-3 activation compared with the control group. Western blot analysis for proteins associated with the Bcl-2 family and the p53/caspase pathway further confirmed that linc01105 knockdown resulted in an obvious apoptotic effect in NB cells (Fig. 3C). Bcl-2 family proteins and the p53/caspase pathway are important regulatory factors of apoptosis (24–26). As shown in Fig. 3C, the expression levels of the antiapoptotic protein Bcl-2 were decreased, while the expression levels of the pro-apoptotic protein Bax were increased following Linc01105 knockdown. Consistent with this, pro-caspase9/3 expression levels were downregulated, while cleaved caspase9/3 and PARP, a marker of caspase3 activation, were upregulated.

Figure 3. Linc01105 knockdown results in increased cell apoptosis. (A) Annexin V-APC/7-AAD double staining and flow cytometry were used to detect cell apoptosis. (B) Caspase-3 activity assay was performed to evaluate the activation of caspase-3. (C) Expression levels of apoptosis-associated proteins and VEGFA protein were evaluated by western blot analysis. *P<0.05, ***P<0.005 compared with control. APC, allophycocyanin; 7-AAD, 7-aminoactinomycin D; VEGFA, vascular endothelial growth factor A; shRNA, short hairpin RNA; PARP, poly (ADP-ribose) polymerase.

Knockdown of linc01105 inhibits migration and invasion

Transwell assays were performed to evaluate the effect of linc01105 knockdown on the migration and invasion of NB cells. The results demonstrated that silencing linc01105 significantly inhibited NB cell migration (Fig. 4A and B) and invasion (Fig. 4A and C), compared with the control group. Furthermore, analysis of the wound healing assay indicated that knockdown of linc01105 significantly suppressed NB cell migration (Fig. 4D and E).

Figure 4. Silencing of linc01105 inhibits migration and invasion in SH-SY5Y cells. (A) Representative images of Transwell migration and invasion assays. (B) Quantification of migration. (C) Quantification of invasion. (D) Representative images and (E) quantification of wound closure assay. **P<0.01, ***P<0.005 compared with control. shRNA, short hairpin RNA.

Linc01105 regulates the expression of miR-6769b-5p targeting VEGFA in NB cells

To explore the mechanism of linc01105 in NB progression, a potential link with VEGFA, a key angiogenic factor involved in NB angiogenesis, was explored. DIANA-LncBase and CirNet database analyses were used to search for potential miRNAs binding with both linc01105 and VEGFA (Fig. 5A and D). miR-107, miR-6769b-5p and miR-348a-3p were identified as potential miRNAs involved in this process. To examine this, the cDNA of linc01105 was cloned into a luciferase reporter plasmid (RLuc-Linc01105-WT). Subsequently, the luciferase reporter plasmid and different miRNA mimics were transfected into 293T cells. Luciferase activity was significantly decreased by miR-6769b-5p mimics transfection (Fig. 5B), but not by the other miRNA mimics tested. To avoid unspecific binding, the miR-6769b-5p binding site of linc01105 and VEGFA was mutated from ACCCACC to TGGGTGG. Transfection of miR-6769b-5p mimics significantly inhibited RLuc-Linc01105-WT activity, but had no effect on RLuc-Linc01105-Mut activity (Fig. 5C). Similar results were observed for the RLuc-VEGFA-WT and RLuc-VEGFA-MU activities (Fig. 5E). Next, the miRNA-6769b-5p expression levels were detected in NB tissues and adjacent normal tissues (Fig. 5F). To further confirm that VEGFA was a target gene of miRNA-6769b-5p, miRNA-6769b-5p mimics and inhibitor were transfected into SH-SY5Y cells (Fig. 5G). The results revealed that miRNA-6769b-5p mimics decreased the expression of VEGFA, while miRNA-6769b-5p inhibitor transfection increased VEGFA expression (Fig. 5H).

Figure 5. miR-6769b-5p binds to linc01105 and VEGFA in 293T cells. (A) The predicted binding site of miR-6769b-5p and linc01105. (B) The relative luciferase activities of linc01105 following transfection with miR-107, miR-378a-3p and miR-6769b-5p mimics. **P<0.01 compared with NC mimics. (C) The relative luciferase activities of the wild type or mutant linc01105 reporter plasmid following transfection with miR-6769b-5p mimics. **P<0.01 compared with NC mimics. (D) The predicted binding site of miR-6769b-5p and VEGFA. (E) The relative luciferase activities of the wild type or mutant VEGFA 3′UTR reporter plasmid following transfection with miR-6769b-5p mimics. **P<0.01 compared with NC mimics. (F) miR-6769b-5p expression levels in NB and adjacent normal tissues. U6 was used as the internal control gene. **P<0.01 compared with adjacent tissues. (G) Validation of miR-6769b-5p overexpression and knockdown by mimics or inhibitor transfection, respectively. **P<0.01 compared with NC. (H) VEGFA mRNA expression levels following transfection with either miR-6769b-5p mimics, inhibitor or their respective control. **P<0.01 compared with NC. VEGFA, vascular endothelial growth factor A; NC, negative control; UTR, untranslated region; WT, wild type; MU, mutant.

HIF-1α activates the expression of linc01105 by acting as a transcription factor

HIF-1α was predicted to bind with the linc01105 promoter region through analysis with the ALGGEN (http://alggen.lsi.upc.es) and JASPAR (http://jaspar.genereg.net) databases. Therefore, a HIF-1α-expressing plasmid was transfected in SH-SY5Y cells (Fig. 6A). The results demonstrated that overexpression of HIF-1α upregulated linc01105 levels (Fig. 6B). VEGFA expression levels were also upregulated following HIF-1α overexpression (Fig. 6C). Subsequently, the linc01105 promoter (2 kb upstream the transcript start site) was cloned into a luciferase gene reporter (RLuc-Linc01105-promoter-WT) and transfected into 293T cells. Luciferase activity was significantly increased following HIF-1α overexpression (Fig. 6D), confirming direct binding of HIF-1α with the promoter of linc01105. Finally, a CHIP-PCR assay also indicated that HIF-1α may bind directly with the promoter of linc01105 (Fig. 6E) and promote linc01105 transcription.

Figure 6. HIF-1α activates the transcription of linc01105. (A) Validation of HIF-1α overexpression in SH-SY5Y cells by transient transfection with a HIF-1α-expressing plasmid. (B) Linc01105 expression levels and (C) VEGFA mRNA expression levels were upregulated following HIF-1α overexpression in SH-SY5Y cells. (D) The relative luciferase activities following co-transfection with a linc01105 promoter reporter plasmid and the HIF1α overexpression plasmid. (E) Chromatin immunoprecipitation assay for HIF-1α and the promoter region of Linc01105. The promoter region fragments are shown as follows: A, 1–400 bp; b, 400–800 bp; c, 800–1,200 bp; d, 1,200-1,600 bp; e, 1,600-2,000 bp. *P<0.05, **P<0.01 and ****P<0.0001. HIF-1α, hypoxia inducible factor-1α; OE, overexpression; IP, immunoprecipitation; IgG, immunoglobulin G.

Linc01105 and miR-6769b-5p may participate in angiogenesis

HUVECs were used for tube-formation assays, in order to evaluate the angiogenesis potential. HUVECs were cultured with HUVEC-specific medium or the CM of SH-SY5Y cells. Silencing of linc01105 was demonstrated to inhibit endothelial cell tube formation (Fig. 7A and B). The tube formation ability of the CM was reduced by pretreatment of CM with VEGFA antibody, by VEGFA siRNA transfection in the SH-SY5Y cells, and by miR-6769b-5p overexpression by mimics transfection in the SH-SY5Y cells (Fig. 7C and D). The tube formation ability reduced by miR-6769b-5p overexpression was restored by VEGFA overexpression (Fig. 7C and D). Finally, silencing and overexpression of VEGFA were confirmed by western blotting (Fig. 7E).

Figure 7. Effects of linc01105, miR-6769b-5p and VEGFA on angiogenesis. (A) HUVECs were cultured in CM from SH-SY5Y cells that were either untreated (blank), infected with control virus (control) or infected with linc01105-shRNA virus (shRNA2). (B) Quantification of tube formation ability from panel A. **P<0.01 compared with control. (C) HUVECs were cultured as follows: With HUVEC-specific media or with SH-SY5Y CM as positive control; with CM pre-treated with VEGFA antibody; with CM of SH-SY5Y cells transfected with a VEGFA-targeting siRNA; with CM of SH-SY5Y cells transfected with miR-6769b-5p mimics; and with CM of SH-SY5Y cells transfected with miR-6769b-5p mimics and VEGFA-overexpressing plasmid (n=3). Tube formation was observed after 6 h of culture using phase-contrast light microscopy. (D) Quantification of tube formation abilities from panel C. *P<0.05, **P<0.01 and ***P<0.005, with comparisons indicated by lines. (E) Validation of VEGFA knockdown and overexpression by siRNA and plasmid transfection, respectively. VEGFA, vascular endothelial growth factor A; HUVECs, human umbilical vein endothelial cells; CM, conditioned media; siRNA, small interfering RNA; shRNA, short hairpin RNA; NC, negative control; OE, overexpression.

Linc01105 gene targets

To explore more functions of linc01105, a CHIRP assay was performed to identify gene locations that are directly bound by linc01105. A protein interaction network was constructed using STRING for the linc01105-specific binding gene targets (after exclusion of targets identified by the negative control), revealing significant interactions between 20 genes (Fig. S1 and Table SI); among these genes were numerous ribosomal proteins and heat shock protein 90, which have been previously reported to be associated to tumorigenesis (27,28). Indeed, GO analysis demonstrated that linc01105 may impact a variety of biological processes. However, no KEGG pathway was found to be enriched, most likely due to low number of the 20 genes used as input in the analysis.

Discussion

Advanced NB patients are typically associated with a poor prognosis and frequent relapses despite treatment with a variety of therapies (29,30). Therefore, it is extremely important to identify novel biomarkers in order to improve the prognostic outcome of pediatric patients with NB (31). With the development of genomics sequencing technologies, recent research has focused on lncRNAs (32,33). Increasing evidence suggests that lncRNAs participate in a number of biological processes and have important roles in human diseases, such as cancer (34,35).

Our previous study found that linc01105 was differentially expressed in NB and adjacent normal tissues through genome-wide lncRNA analysis (14). In addition, it was demonstrated that upregulation of linc01105 was correlated with NB INSS stage. In that previous study, the SK-N-BE(2)C cell line was used, which is a MYCN-amplification cell line; in the present study, the cell line SH-SY5Y was used, which is a non-MYCN amplification cell line, to further investigate the mechanism of linc01105 in NB. The present study focused on the molecular mechanism of linc01105. To explore the functions of linc01105, a nuclear-cytoplasmic fractionation was performed and the results revealed that linc01105 was mainly expressed in the cytoplasmic fraction. Subsequently, a loss-of-function assay was performed by linc01105 shRNA knockdown. Knockdown of linc01105 inhibited cell proliferation and promoted cell apoptosis. Notably, knockdown of linc01105 resulted in inhibition of migration and invasion. Furthermore, linc01105 silencing altered the expression of Bcl-2 family proteins and activated the p53/Caspase signaling pathway. These results indicated that linc01105 affected the apoptosis process via regulating Bcl-2 proteins and activating the p53/caspase signaling pathway.

One of the significant functions of lncRNAs, especially those located in the cell cytoplasm, is to bind with miRNAs, subsequently acting as a ‘sponge’ and inhibiting their expression (36–38). For example, in gastric cancer, linc01234 may have a role as a competing endogenous RNA to regulate core-binding factor subunit β expression by sponging miR-204-5p (36). Another study demonstrated that miR-29b-3p was directly inhibited by linc00511, which resulted in an increase in VEGFA expression in pancreatic ductal adenocarcinoma (39). Angiogenesis is crucial for tumor growth and is associated with tumor metastasis; notably, VEGFA is a major regulator of NB angiogenesis (40,41).

Bioinformatics analysis was performed to identify miRNAs that may bind with both linc01105 and VEGFA. The results revealed that miR-6769b-5p shared complementary binding sites with the VEGFA 3′UTR region and with linc01105 (924–948 nt region), which was confirmed by luciferase assay. RT-qPCR results indicated that miR-6769b-5p expression levels were lower in NB tissues compared with adjacent normal tissues. In addition, transfection with miR-6769b-5p mimics resulted in downregulation of VEGFA, while miR-6769b-5p inhibition resulted in upregulation of VEGFA expression. These findings suggested that linc01105, miR-6769b-5p and VEGFA mRNA constituted a competing endogenous RNA regulatory network. Silencing of linc01105 reduced the miR-6769b-5p competing adsorption and increased the levels of free miR-6769b-5p, thereby promoting the degradation of VEGFA by increasing binding to the 3′UTR region of VEGFA. Indeed, transfection with a miR-6769b-5p inhibitor suppressed VEGFA expression. By contrast, upregulation of linc01105 contributed to VEGFA upregulation via the competing endogenous RNA network. Of note, VEGFA knockdown had no influence on apoptosis but inhibited tumor migration (42,43). Therefore, the present study suggested that miR-6769b-5p may participate in NB tumorigenesis via interaction with VEGFA. The VEGF-mediated mechanism most likely has no effect on the p53/caspase signaling pathway, and the VEGFA network may only affect the migration phenotype of the linc01105 knockdown and not apoptosis.

The present study used the ALGGEN and JASPAR databases to predict transcription factors which may bind with the promoter of linc01105; among the predicted factors, HIF-1α was of particular interest, as high expression of HIF-1α has been widely reported to be associated with poor prognosis of NB (44,45). Over-expression of HIF-1α significantly promoted linc01105 expression levels. Furthermore, a luciferase assay confirmed that HIF-1α regulated the expression of linc01105 acting as a transcription factor and directly binding to its promoter. Additionally, the CHIRP assay identified 20 genes that directly bind with Linc01105, and which are potential candidates for future studies.

In conclusion, the present study demonstrated that linc01105 was significantly upregulated in NB tissues compared with normal tissues. Silencing of linc01105 resulted in the activation of the p53/caspase signaling pathway and the inhibition of NB cell proliferation, migration and invasion. In addition, silencing of linc01105 suppressed angiogenesis via miR-6769b-5p targeting of VEGFA. Taken together, these results suggested that the linc01105/p53/caspase pathway or the linc01105/miR-6769b-5p/VEGFA network may serve as candidate targets for future therapies for NB.

Supplementary Material

Supporting Data

Acknowledgements

The authors wish to thank Ms Xia Wenjun and Mr Huang Jianbo (Fudan University, Shanghai, China), who provided kind assistance with all of the experiments.

Funding

This study received financial support from Shanghai Key Disciplines (grant no. 2017ZZ02022), National Natural Science Foundation of China (grant nos. 81771633 and 81572324) and Science Foundation of Shanghai (grant nos. 17411960600 and 15ZR1404200).

Availability of data and materials

The datasets used during the present study are available from the corresponding author upon reasonable request.

Authors' contributions

DK and YM designed the study. YM and MJ collected the data and performed experiments. LB and LX analyzed and interpreted the data. MD and DK were involved in critical reviewing of the manuscript. All authors read and approved the final manuscript.

Ethics approval and consent to participate

This study was approved by the Institute Research Ethics Committee at the Children's Hospital of Fudan University. Informed consent was acquired from every patient's legal guardians.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

References