Patients and samples. A drainage tube was inserted into the subarachnoid space of GBM patients using the lumbar puncture method before anesthesia. The drainage tube remained closed until cranial surgery. CSF was collected from the cistern of control patients during some meningioma surgeries, which employed CSF drainage for better tumor exposure. We collected 30–50 mL of CSF at 4°C for each patient. Tumor and brain tissues accessed from the Tumor and Tissue Bank at Nanjing Brain Hospital.
EVs isolation. EVs were isolated from GBM, CSF, and GSC culture medium using differential centrifugation at 4°C. Briefly, the CSF and GSC supernatant were initially centrifuged at 3,500 g for 10 min to remove suspended cells. The supernatant was then centrifuged two times at 4,500 g for 10 min each, followed by 10,000 g for 30 min to remove cellular debris. The pellet was re-suspended in phosphate-buffered saline (PBS) and further centrifuged at 100,000 g for 60 min. The 100,000 g pellet was centrifuged again after dilution with PBS. The final CSF and GSC-derived EV pellets were used immediately or re-suspended in PBS and stored at -80°C.
Transmission electron microscopy (TEM). CSF and GSC-derived EVs (10 µg) were re-suspended in PBS and loaded onto Formvar/Carbon-coated grids (Ted Pella Inc., Redding, CA, USA) fixed in 2.5% glutaraldehyde, stained with 2% uranyl acetate, and visualized with LEO 912AB Omega electron microscope (FEI Tecnai G2 Spirit Bio TWIN, USA).
Western blot analysis. Isolated EVs were digested with RIPA lysis buffer and centrifuged for the supernatant protein. The supernatants were mixed with Loading Buffer and heated. The samples loaded onto 10% SDS-PAGE gels. Proteins were transferred onto PVDF membranes, blocked with 5% non-fat milk for 2 h, and incubated overnight with primary antibody against CD63 (Abcam, ab59479), GAPDH (Abcam, ab181602), Hsp70 (Abcam) at 4°C. The membranes were washed three times with PBST and incubated with HRP-conjugated secondary antibody (Bioss, bs-40295G-HRP) for 2 h. The detection was imaged using a chemiluminescence instrument (Tanon 4200, Shanghai, China).
Nanoparticle tracking analysis (NTA). Size determination of CSF and GSC-derived EVs was analyzed using ZetaView (Particle Matrix GmbH, Microtrac, Meerbusch, Germany).
RNA sequencing (RNA-Seq). Using TRIzol, total RNA was extracted from CSF EVs and glioma tissues using the miRNeasy Micro Kit (Qiagen, Hilden, Germany). Small (i.e.,18–30 nt) RNAs were enriched by polyacrylamide gel electrophoresis (PAGE). Then the 3′ adapters added, and the 5′ adapters were ligated. The products were reverse transcribed using PCR amplification. Their 140–160 bp size products were enriched to generate a cDNA library and finally sequenced on an Illumina HiSeqTM2500 system (San Diego, CA, USA).
Bioinformatics analysis. To obtain high-quality reads, the raw data was further filtered according to the following rules: reads containing more than one low quality (Q-value ≤ 20) base or unknown nucleotides (N); reads without 3' adapters or having 5′ adapters; reads containing 3′ and 5′ adapters but no small RNA fragment between them; reads containing poly A in small RNA fragment; reads shorter than 18 nt. Then the tiny RNAs were aligned and identified. Significantly differentially expressed miRNAs were identified using miRNAs with a fold change ≥ 2 and P-value < 0.05 in comparison.
Real-time quantitative polymerase chain reaction (RT-qPCR). Total RNA was isolated from glioblastoma cells, exosomes and cells using Trizol reagent (Thermo Scientific, USA). The relative RNA concentration and quality were measured on a DU800 UV spectrophotometer (Beckman Counter, USA). cDNA was synthesized using a cDNA synthesis kit (Yeasen, Shanghai, China). qPCR was performed using qPCR SYBR Green Master Mix on a Lightcycler (QuantStudio 5, ABI, USA). The data were calculated using the comparative threshold cycle relative-quantification method.
Animals. All experimental animal procedures were reviewed and approved by the Ethics Committee for Animal Experimentation at Nanjing medical University (Nanjing, China) and performed in accordance with the Guide for the Care and Use of Laboratory Animals by the National Institutes of Health. Male mature BALB/c nude mice were purchased from the Animal Center of Nanjing medical university. The mice were housed at the animal feeding institution, where room temperature, humidity, and 12 h light/dark cycle were monitored and followed.
Isolation and characterization of glioma stem cells (GSCs). The detailed experimental protocol for GSC isolation has been described elsewhere (21). Briefly, GSCs were isolated and purified from GBM patient tissues and the extracted GSC primary spheroids. The GSCs were incubated in the DMEM/F12 (Gibco, Waltham, MA, USA) medium supplemented with B27(1:50, Invitrogen, Carlsbad, CA, USA), N2(1:100, Invitrogen), 10 ng/mL epidermal growth factor and 10 ng/mL fibroblast growth factor (FGF, Invitrogen).
The GSCs morphology were photographed and recorded in detail. The GSCs were labeled with different monoclonal antibodies (SOX2 and Nestin) for immunofluorescence analysis.
Glioblastoma cell culture. Two human glioblastoma cell lines (U87 and U251) were purchased from the Cell Bank of Chinese Academy of Sciences (Shanghai, China). The cells were cultured in DMEM medium supplemented with 10% FBS and 1% penicillin/streptomycin (Thermo Scientific, USA). The incubator was maintained at 37°C in a concentration of 95% humidity and 5% CO2.
Cell co-culture system. A total of 1 × 106 U87 or U251 cells were seeded on the lower chamber of a 6-well co-culture plate, while equal amount of 10 µg GSC-derived EVs were added to the upper chamber. The 6-well plate was cultured in a 37°C incubator for 24 h. Finally, glioblastoma cell lines were harvested for further analysis.
MiRNA transfection. The amount of 5 × 105 GSCs were seeded on the 6-well plate for 24 h. 5 µL Lipofectamine 6000 (Beyotime, Shanghai, China) was blended with 50 µL serum-free Opti-MEM medium for 5 min. Simultaneously, 100 pmol miRNA mimic or inhibitor were added into the 50 µL Opti-MEM medium. Then this medium was mixed and allowed to rest for 20 min. At last, the 100 µL mixture was taken to the MSCs culture medium and incubated for 24 h.
Intracranial xenograft model. 4-week-old male BALB/c nude mice were randomly divided into 2 groups (sh-NC and sh-miR-9). GSCs were transfected with stably expressed sh-NC and sh-miR-9 luciferase lentivirus. Intracranial tumor growth was measured by in vivo fluorescence imaging. Each mouse was anesthetized by initial 2% isoflurane in 100% O2 and maintained in 1% isoflurane. The mice were then given an intraperitoneal injection of D-luciferin (50 mg/mL). Xenografts were imaged by means of Caliper Lumina system (Caliper Life Science, Waltham, MA, USA). Mouse survival time was monitored until death.
EdU (5-ethynyl-2′-deoxyuridine) assay. The cell proliferation analysis was conducted by EdU assay. U87 and U251 cells were co-cultured with different groups of GSC-derived exosomes in a concentration of 10 µg/mL. After incubation for 24 h, 37°C, 20 µM EdU reagent was added and reacted for 2 h. Then the culture medium was discarded and the sample was washed with PBST. Cells were fixed with paraformaldehyde for 15 min. We added click addition solution and incubated for 30 min. Then 1× Hoechst 33342 was administered for 10 min in dark. We randomly selected six fields of view by fluorescence microscope (Carl Zeiss Meditec AG, Jena, Germany) with two observers blind to group assignments.
Cell viability assay. Cell Counting Kit-8 (CCK-8; Dojindo, Kumamoto, Japan) was employed here to analyze cell proliferation. Equal amount of about 1 × 103 cells was seeded in the 96-well plates and cultivated for several days. 10% CCK-8 dilutions were added to cell medium in every time point and reacted for 2 h. Subsequently, the plates were analyzed with Thermo Fisher microplate reader at the value of 450 nm wavelength (OD450). All the tests were performed at least 3 times.
Migration assay. Transwell assays were employed to evaluate the migration captivity. Briefly, 2×105 cells in 200 µL DMEM, supplemented with or without EVs, were seeded onto the upper chamber of 6.5 mm transwells (Corning Incorporated, Corning, NY, US). After incubation for 20 h, the interior U87 or U251 cells of the chamber, which did not penetrate the membrane, were removed. After that, the chambers were fixed with paraformaldehyde and then stained with crystal violet. Finally, the migrating cells were measured in five microscopic fields (200×) (Carl Zeiss, Germany).
Wound healing assay. Equal amount of glioblastoma cells were seeded onto the 6-well plates and incubated. The wound was made by tip scratching after cells reaching a density about 90% confluence and then cultured for 24 h to measure wound closure.
Dual luciferase reporter assay. The predicted binding sequence of DACT3 3′UTR region and the mutate sequence were amplified and cloned into luciferase reporter plasmid (PGL3-DACT3 WT and PGL3-DACT3 mut). The luciferase reporter plasmids were co-transfected with miR-9 mimics. And PGL3-empty vector was used as negative controls. After reaction for 48 h, the luciferase density was examined by the Dual Luciferase Reporter Assay Kit (Vazyme, Nanjing, China) according to the manufacturers’ protocol.
Statistical analysis. Statistical analyses were performed using GraphPad Prism 5 (GraphPad Software, CA, US) and R software version 3.2.1. Student’s t-test was used to determine significant differences between EV miRNAs derived from normal and glioma CSF. A two-sided P-value < 0.05 was considered statistically significant.