Identify differentially expressed genes (DEGs) from the gene expression omnibus database
Gene expression omnibus (GEO) (https://www.ncbi.nlm.nih.gov/geo/) was adopted to select GSE128004 and GSE121513 datasets. The GSE128004 dataset is the exosomal data of neuroblastoma cancer patients, including 15 patients and 3 normal controls. In the GSE121513 dataset, the number of embryonal stem cell lines, normal fetal adrenal cortex samples, normal fetal adrenal neuroblast samples, and neuroblastoma samples were 8, 7, 7, and 95, respectively. The R language package (Deseq2 package) was used to screen the DEGs in tissue samples, with adjusted P<0.01 and |log2FC|>2 as the threshold.
Cell lineage and culture
Cell lines of NB in human, including IMR-32, SK-N-SH, SK-N-BE and SH-SY5Y, were obtained from the National Collection of Authenticated Cell Cultures (Shanghai, China) Dulbecco’s modified Eagle’s medium (DMEM; HyClone, Life Sciences, Shanghai, China) was utilized to culture these cell lines, supplemented with 10% FBS (BI, Kibbutz, Israel), penicillin G (100 units; Beyotime, Beijing, China), and streptomycin (100 units, HyClone). The medium was placed in a 37 °C 5% CO2 incubator.
Extraction of RNA and real-time quantitative polymerase chain reaction (qRT-PCR) analysis
We extracted total RNA from cells and tissues using TRIzol reagent (Invitrogen, Carlsbad, California, USA) based on the manufacturer’s recommendations. Reverse Transcription Kit (Fermentas, MA, USA) was adopted to synthesize complementary DNA (cDNA) from each 2 μg RNA sample. SYBR premix PCR kit (Thermo) was utilized to conduct qRT-PCR on a qTOWER real-time PCR instrument (Jena, Germany). The primer sequences are listed below. The miRNA expression level was measured by the comparative threshold cycle value, and the relative change was computed by the 2-ΔΔCt method with the normalization of the target gene to that of U6 expression. miR-32-5p forward primer: CGCGCGTATTGCACATTACTAA and reverse primer: AGTGCAGGGTCCGAGGTATT. The forward and reverse primers of U6 are as follows: CTCGCTTCGGCAGCACA and AACGCTTCACGAATTTGCGT, respectively.
Western blotting
Western blotting was carried out in tissues and cultured cells. Cells were harvested after drug treatment, and phosphate-buffered saline (PBS) was used for wash. RIPA lysis buffer containing protein inhibitor (Roche, CA, USA) was used to isolate total proteins, and protein samples were quantified by BCA kit (ThermoFisher) via spectrophotometer at 562 nm. Twenty microgram protein was separated on SDS-PAGE gel. After half-an-hour blocking of 5% BSA at ambient temperature, primary antibodies were adopted to incubate blots: VPS4B (Ab224736, Abcam, 1:1000), EGFR (#4267, CST, 1:1000), Snail (#3879, CST, 1:1000), MMP2 (Ab97779, Abcam, 1:1000), MMP9 (Ab228402, Abcam, 1:1000), E-cadherin (#14472, CST, 1:1000), and GAPDH (#5174, CST, 1:1000). After overnight incubation at 4 degrees Celsius, we rinsed the membrane in TBST thrice to avoid the unbound primary antibodies. Then, a secondary antibody of goat anti-rabbit IgG H&L (HRP) was utilized to incubate the membrane at ambient temperature for 2 h, which was then washed by TBST. The protein bands were detected using the imaging system (Protein Simple, USA).
Plasmid construction and transfection
miR-32-5p mimic/inhibitor as well as the negative control (NC) were provided by GenePharma company (Shanghai, China). A 6-well plate was utilized to subculture cells a day before transfection. Utilizing Lipofectamine® 2000 reagent (Invitrogen; Thermo Fisher Scientific Inc.), the cells of SK-N-SH were mock transfected (mimic NC, 100 pmol) or subjected to transfection with miR-32-5p mimic (100 pmol), and IMR-32 cells were mock transfected (inhibitor NC, 100 pmol) or transfected with miR-32-5p inhibitor (100 pmol) at 70–80% confluency. For VPS4B expression vector construction, the full-length cDNA encoding VPS4B was PCR amplified with primers (F: 5’-CCCAAGCTTATGTCATCCACTTCGCCCAAC-3’ R: 5’-CGGAATTCTTAGCCTTCTTGACCAAAATCTTC-3’). VPS4B was ligated to the appropriate sites of the pcDNA3.1 vector to generate the recombinant expression vector pcDNA3.1-VPS4B, which was propagated in E. coli strain DH5α. DNA sequence analysis was used to confirm the VPS4B cDNA sequence. After that, the identified pcDNA3.1-VPS4B plasmid was utilized to transfect SK-N-SH cells.
Wound healing analysis
Cells (1×105/500 μL per well) were placed in a plate containing 12 wells and cultured into a confluent monolayer. A line was scrapped on the adherent cells with a sterile pipette tip of 10 μL, washed, and incubated with 2% FBS DMEM. The cells were traced and photographed after 0, 12 and 24 h, respectively. Software of Image J (http://rsbweb.nih.gov/ij/) was used to compute the scratch width.
2.7 Transwell assay
Transwell chambers (Costar, MA, USA) were adopted to study the cell migration as well as its invasion with or without Matrigel (Corning, Kennebunk, ME, USA). We placed 700 mL medium which contained 10% FBS in the lower chamber; 3×104 cells suspended in 300 mL medium which contained 1% FBS were seeded into the upper chamber for 24 and 48 h, respectively. We fixed the migrated or invaded cells across the membrane for 1/3 h and stained them for 1/4 h. Finally, the cells were counted and images were collected by an Olympus BX51 microscope.
Luciferase assay
TargetScanHuman (release 7.1) and miRDB were used to predict miR-32-5p’s complementary sequence at the 3’-UTR of VPS4B. Utilizing Dual luciferase 3’-UTR reporter gene detection to verify VPS4B as miR-32-5p’s direct target. To construct the luciferase plasmid, the 3’-UTR sequence fragment with or without putative binding sites was inserted into pGL3 vectors (Promega, Madison, WI, USA), defined as wild-type (wt) and mutant regions. The wt plasmid of VPS4B 3’-UTR and VPS4B 3’-UTR mutant were constructed. The wt sequence of VPS4B was 5’-ACTGATACCTTTCACTGTGCAATC-3’ and mut sequence was 5’-CTGATACCTTTCACTACATGCACTC-3’. SK-N-SH cells (5×105/well) were planted in plates of 6 wells and cultured for 24 h and co-transfected with Wt-VPS4B 3’-UTR or Mut-VPS4B 3’-UTR and miRNA-32-5p mimic/mimic NC using Lipofectamine 2000 (Thermo Fisher Scientific, Inc.). Two days after post-transfection, we harvested and lysed the cells. The activities of the Firefly and Renilla luciferases were measured to detect the luciferase reporter gene from cell lysates (Promega). Finally, the relative activity of Firefly luciferase to that of Renilla luciferase was normalized, and the fold-changes of the reporters were computed below: RLU Firefly luciferase/RLU Renilla luciferase×100%.
Metastasis in nude mice
Nude mice at the age of four to six weeks, obtained from Beijing Vital River Laboratory Animal Technology Co., Ltd., were kept under conditions of normal room temperature and humidity. Mice were given free access to conventional food as well as water. All of the animal procedures were carried out under the manuals for the care and use of experimental animals and under the approval of the IACUC of the China Academy of Chinese Medical Science. NB metastases were produced via tail vein injection of IMR-32 cells (100 mL, 2×106) into the nude mice. 1 week after injection, we divided the mice into the normal saline control group (n=5) and two treatment groups: DHA group (100 mg/kg, n=5), DHA plus mimic group (DHA 100 mg/kg+mimic 0.5 OD/day, n=5). Mice were injected DHA and miR-32-5p mimic intraperitoneally daily for 3 weeks. Subsequently, all mice were killed after anesthesia by chloral hydrate, and the lung nodules were counted and photographed. After euthanasia, we recovered the tumor, whose wet weight for each was weighed. The xenograft tumors were surgically cut and fixed for hematoxylin and eosin /immunohistochemistry (IHC) staining. In addition, the EMT protein expression was detected using the tumor tissues by Western blotting.
IHC analysis
Tumor tissue samples were fixed and embedded in paraffin. To retrieve antigens, citrate buffer (pH 6.0) was used in the heat process of tissue slides. Then, the slides were stained using 3,3’-diaminobenzidine (DAB) (Phoenix Biotechnologies), and Meyer’s hematoxylin was utilized for counterstaining. Afterwards, we dehydrated them in ethanol. The slides were observed using a fluorescent microscope.
Statistical methods
Mean±standard deviation (SD) was utilized to express all data. We analyzed the comparison between groups of two using student’s t-test, and explored the comparison among various groups using one-way analysis of variance. Values of P<0.05 (*P<0.05, **P<0.01, ***P<0.001) were markers of statistical significance.