Effective Visualization and Easy Tracking of Extracellular Vesicles in Glioma Cells.

Cell Culture The KW10 cell-line used in this study was developed by us and we have extensively characterized it as a patient-derived glioblastoma cell-line [ 23 ]. As the fetal bovine serum (FBS) used for culturing most of the cell-lines is often contaminated with bovine EVs, we have performed ultra-centrifugation of the medium at 100,000 g for 18 h to get rid of them and thereby ensured that the medium was free of extraneous EVs. We cultured the GB cell-line KW10 in Dulbecco’s Modified Eagle Medium/Nutrient Mixture F-12 (DMEM/F12) with 10% FBS (EV depleted) and collected the Conditioned Medium (CM) after growth of cells in this medium for 48–72 h. We have also cultured human umbilical vein endothelial cells (HUVECs) on fibronectin (Invitrogen # 33016015) coated plates with endothelial cell media (Sigma # 211–500) for their use in uptake and tube formation assays.

Isolation and Characterization of EVs

The detailed steps are as follows: A.

Conditioned media processing

The collected conditioned medium was centrifuged at 4 °C for 10 mins at 800 x g in a swinging bucket rotor (Eppendorf 5804R; rotor ID A-4-44) to remove cellular debris. Later, the conditioned media was filtered through a 0.45 μm syringe filter with supor membrane (Pall # 4654) and stored at − 80 °C for later use (maximum 1–2 months) or at 4 °C for immediate use. B. EV isolation 20% PEG10000 (Sigma cat. - 92,897) solution was prepared with deionized water and passed through a 0.22 μ filter (Millipore # SLGP033RS) [ 24 ]. 5 ml conditioned medium was taken in a 15 ml centrifuge tube and another 5 ml of 20% PEG10000 solution was added to it. The above solution was mixed by inverting the tube 10 times and incubated on ice for 1 h. The mixture was centrifuged at 3000 x g for 30 min at 4 °C in a swinging bucket rotor (Eppendorf 5804R; rotor ID A-4-44). An off-white pellet was visible. The supernatant was removed carefully by decanting without disturbing the EV pellet. Another round of centrifugation was performed at 3000 x g for 5 mins to remove the residual liquid in the tube by pipetting. Collectively, the entire pellet was dissolved in 200 μl of PBS. The dissolved pellet was collected in a fresh 1.5 ml micro-centrifuge tube. [Note: 20 ml of conditioned media was used for protein labeling. Avoid keeping CM at − 20 °C because it may result in a meager yield during EV isolation. The use of freshly prepared conditioned medium is recommended to get higher yield during EV isolation. If the pellet is not visible, the volume of the conditioned media can be increased to obtain a larger pellet.] C.

EV characterization by Western Blotting

EVs and cells were lysed using 1X RIPA buffer and 20 μg of protein was loaded into each well for western blotting. Next, we characterized the EVs by western blotting using antibodies to several EV specific markers like TSG101 (1:3000 dilution; Pierce - MA123296 ), CD63 (1:1000 dilution; Abcam - ab59479) and HSP70 (1:3000 dilution Cloud clone- MAA873Hu21). Cellular markers like - calnexin (1:2000 dilution; CST - 2679S) which are not expressed by EVs served as negative control. The secondary antibodies used were IRDye 680RD Goat anti-Mouse IgG (1:8000 Dilution; Licor 926–68,070), IRDye 800CW Goat anti-Rabbit IgG (1:8000 dilution; Licor 926–32,211) and Donkey anti-Goat IgG HRP (1:7000 dilution; Invitrogen - A15999) (Fig. 2 a). Protocol for EV Immune-Labeling The flow-chart for EV immune-labeling is shown in Fig. 1 . All labeling experiments were performed at least 7 times by various people in the lab and we achieved similar results. The steps followed for EV immune-labeling procedure are as follows. Fig. 1 Strategy for labeling of Extracellular Vesicles (EV) derived from glioma cells. Flow chart detailing the steps involved in labeling of EVs in cell-free system The Isolated EVs Were Permeabilized Using 0.001% (Final Concentration) Triton X-100 for 5 Min. [Note: We tried different concentrations like- 0.01, 0.05 and 0.001% of Triton X-100; amongst them, we found that only 0.001% concentration of Triton X-100 was optimum for maintaining the integrity of EVs.] PEG10000 was added to a final concentration of 10% as mentioned above (no incubation required) and the suspension was spun at 3000 x g at room temperature (RT) for 5 mins. The supernatant was carefully removed and the pellet was dissolved in 100 μl of PBS. Primary antibodies to TSG101, CD63 and calnexin (final concentration of 0.01 μg/μl) were individually added to the dissolved EV suspension of KW10 cells. Likewise, 1 μg of each of these antibodies were individually added to 100 μl of EV preparation in a microfuge tube. The whole mixture was incubated for 90 mins with gentle shaking at RT. Next, 100 μl of 20% PEG10000 solution was added to the EV suspension in step 3, and the whole mixture was centrifuged at 3000 x g for 5 mins. The obtained pellet was re-suspended in 100 μl of 1x PBS. The step of PEG precipitation of EVs was repeated twice to remove excess antibodies. The EV pellet was suspended in 100 μl of PBS. Alexa Fluor labeled species specific secondary antibodies (Invitrogen Catalog no. A11005 and A11012) diluted 1:100 were added to the EV suspension and the complex was incubated for 1 h with gentle shaking in the dark. [Note: Primary and secondary antibody concentration may vary depending on the size of the pellet observed during EV isolation.] The unbound secondary antibodies were washed of using PEG10000 at speed of 3000 x g thrice in a centrifuge as mentioned in step-4. The pellet was dissolved in 100 μl of PBS. PEG can precipitate macromolecules. Hence, Sephadex G-25 column (G2580 Sigma) was used to remove the unbound antibodies and primary-secondary antibody conjugates which may give false positive signal. The solution was passed through the column to remove background noise. Then PEG solution was added to flow through for further precipitation of EVs. The mixture was centrifuged at 3000 x g for 5 mins and the supernatant was removed with a pipette. The pellet was dissolved in 20 μl of PBS. 3–4 μl of dissolved pellet of each labeled EV-antibody complex was placed on pre-cleaned microscope slides. A thin smear was drawn using coverslip and kept for drying in dark for 5–10 min. Further the slides were analyzed using confocal and STED microscopy (Leica TCS SP8 STED 3X). We also performed dSTORM imaging on Nanoimager S (ONI Oxford). [Note: We used anti-mouse secondary antibodies Alexa 488 (Invitrogen A11001) and Alexa 555 (Invitrogen A21422) for dSTORM imaging.] [Note: We tried different columns to remove background noise. Sephadex G-25 showed 80–90% efficiency in eliminating background noise. Sephacryl (range 20–8000 KDa) worked best for removing unbound primary-secondary antibody conjugates but gave a very low yield. Data provided here was generated using Sephadex G-25 column. Column packing must be done carefully and air bubbles must be strictly avoided inside the column]. To prove that the method of detecting EVs by immunofluorescence is as sensitive to the methods that are based on detection of EVs by PKH67 labeling, we performed a dual labeling experiment, wherein upon antibody labeling of EVs, we also labeled the same EVs using PKH67 dyes as well. Dual Labeling with PKH67 Dye After completing the secondary antibody staining at step-5 (in above), unbound secondary antibody from each reaction was washed using PEG10000 as mentioned above. After the 3rd wash, the pellet was dissolved in 50 μl of PKH diluent and 0.2 μl of PKH67 (PKH67GL Sigma) was added to it. The mixture was incubated at RT for 30 mins with gentle shaking. 50 μl of 20% PEG10000 solution was added to the mixture for the EV precipitation. Then the mixture was centrifuged at 3000 x g for 5 mins. The supernatant was removed very carefully and the EVs were suspended in 50 μl of PBS. [Note: You may not get a visible pellet here if you begin with less than 20 ml of CM]. The mixture was passed through the column (Sephadex G25). Flow through was collected and 100 μl of 20% PEG10000 was added to it. The mixture was centrifuged at 3000 x g for 10 mins (visible pellet may not be seen). The supernatant was removed very carefully and 10 μl of PBS was added to it. [Note: Sephadex G25 column is not efficient to remove unbound PKH dye]. Finally, 3–4 μl of dissolved pellet was placed on pre-cleaned slide. A thin smear was drawn as mentioned above (step 8) and mounted for visualization by confocal microscopy (Leica TCS SP5 II). Next, we checked the integrity of the EVs by transmission electron microscopy (TEM) using gold labeled secondary antibodies.

Sample Preparation for TEM

To perform gold labeling of EVs, we replaced Alexa Fluor labeled secondary antibody with gold nanoparticle tagged secondary antibodies (anti-mouse gold IgG, sigma G7527 and anti-rabbit gold IgG, Sigma G7402) and followed a similar immune-staining protocol as described above. To observe the immuno-gold labeled EVs under TEM, we processed our samples using the uranyl acetate staining protocol as described below. Reagents Required 2% paraformaldehyde in PBS, 4% uranyl acetate (pH 4) and 2% methylcellulose in filtered and deionized water. Working Solution 1 volume of 4% uranyl acetate was mixed with 9 volumes of 2% methylcellulose and the solution was kept in the dark. 3 to 5 μl of labeled and unlabeled EVs were applied on copper grids (Ted Pella Prod# 01810) respectively and incubated for 30 mins to settle at RT. 2% paraformaldehyde (PFA) solution was applied on parafilm. The grids were placed inversely on the drop of 2% PFA solution for fixation of the sample and incubated further for 20 mins in RT. Similarly, few drops of deionized water was applied on parafilm. Then, grids were placed on water drop to remove extra paraformaldehyde without any incubation (Note: PFA solution was prepared in PBS. PBS can precipitate uranyl acetate, hence washing step is necessary to obtain proper resolution of the samples). A 90 mm petri-dish was taken and parafilm was kept inside the petri-dish. Later, few drops of freshly prepared working solution was added on the top of parafilm. The whole preparation was kept on ice (in dark). Finally, grids were placed on the drop of working solution and incubated for 10 mins (on ice and in dark condition). Grids were removed by using clean fine forceps and extra solution was soaked using blotting paper. Finally, grids were kept in special grid holder. This was followed by imaging using TEM (FEI Tecnai T20). Images were taken at 100 kV.

Tracking of EVs in Recipient Cells

At a final step of EV labeling, the labeled EV pellet was dissolved either in 20 μl of endothelial medium or in sterile PBS. The EVs were incubated for 6 h on HUVECs (Invitrogen cat no. C0035C) which were pre-seeded on the coverslips. The coverslips were washed with PBS thrice and fixed with 4% PFA solution for 5 mins. Further, the coverslips were washed 3 times using PBS and incubated with 1 μg/ml DAPI (Invitrogen #D1306) for 10 mins in dark. Next, the coverslips were washed 3 times using PBS and finally mounted on slides for confocal imaging (Leica TCS SP5 II). Images were acquired using 63X oil immersion lens. Angiogenesis Assay for Checking Functionality of EVs Angiogenesis assay was performed according to the method of Guo et al. [ 25 ]. 1 × 10 4 number of HUVECs were plated on the growth factor reduced Matrigel (Corning product # 354230) in μ-slide (from ibidi cat no. 81506) in three different conditions that included – negative control, unlabeled EVs and labeled EVs. Then the slides were incubated at 37 °C in CO 2 incubator for 6 h. Later, the cells were analyzed for their tube forming potential using AngioTool software of National Institute of Health (NIH) [ 26 ]. We have performed analysis of every image under similar configuration of AngioTool64 software version 0.6a (02.18.14).

Statistical Analysis

Statistical analysis was conducted using GraphPad Prism 5 software. Two-tailed t-test (Mann-Whitney U) was performed for analyses of size distribution of EVs. Bars in all figures represent mean ± SEM.

Protocol for EV Immune-Labeling The flow-chart for EV immune-labeling is shown in Fig. 1 . All labeling experiments were performed at least 7 times by various people in the lab and we achieved similar results. The steps followed for EV immune-labeling procedure are as follows. Fig. 1 Strategy for labeling of Extracellular Vesicles (EV) derived from glioma cells. Flow chart detailing the steps involved in labeling of EVs in cell-free system The Isolated EVs Were Permeabilized Using 0.001% (Final Concentration) Triton X-100 for 5 Min. [Note: We tried different concentrations like- 0.01, 0.05 and 0.001% of Triton X-100; amongst them, we found that only 0.001% concentration of Triton X-100 was optimum for maintaining the integrity of EVs.] PEG10000 was added to a final concentration of 10% as mentioned above (no incubation required) and the suspension was spun at 3000 x g at room temperature (RT) for 5 mins. The supernatant was carefully removed and the pellet was dissolved in 100 μl of PBS. Primary antibodies to TSG101, CD63 and calnexin (final concentration of 0.01 μg/μl) were individually added to the dissolved EV suspension of KW10 cells. Likewise, 1 μg of each of these antibodies were individually added to 100 μl of EV preparation in a microfuge tube. The whole mixture was incubated for 90 mins with gentle shaking at RT. Next, 100 μl of 20% PEG10000 solution was added to the EV suspension in step 3, and the whole mixture was centrifuged at 3000 x g for 5 mins. The obtained pellet was re-suspended in 100 μl of 1x PBS. The step of PEG precipitation of EVs was repeated twice to remove excess antibodies. The EV pellet was suspended in 100 μl of PBS. Alexa Fluor labeled species specific secondary antibodies (Invitrogen Catalog no. A11005 and A11012) diluted 1:100 were added to the EV suspension and the complex was incubated for 1 h with gentle shaking in the dark. [Note: Primary and secondary antibody concentration may vary depending on the size of the pellet observed during EV isolation.] The unbound secondary antibodies were washed of using PEG10000 at speed of 3000 x g thrice in a centrifuge as mentioned in step-4. The pellet was dissolved in 100 μl of PBS. PEG can precipitate macromolecules. Hence, Sephadex G-25 column (G2580 Sigma) was used to remove the unbound antibodies and primary-secondary antibody conjugates which may give false positive signal. The solution was passed through the column to remove background noise. Then PEG solution was added to flow through for further precipitation of EVs. The mixture was centrifuged at 3000 x g for 5 mins and the supernatant was removed with a pipette. The pellet was dissolved in 20 μl of PBS. 3–4 μl of dissolved pellet of each labeled EV-antibody complex was placed on pre-cleaned microscope slides. A thin smear was drawn using coverslip and kept for drying in dark for 5–10 min. Further the slides were analyzed using confocal and STED microscopy (Leica TCS SP8 STED 3X). We also performed dSTORM imaging on Nanoimager S (ONI Oxford). [Note: We used anti-mouse secondary antibodies Alexa 488 (Invitrogen A11001) and Alexa 555 (Invitrogen A21422) for dSTORM imaging.] [Note: We tried different columns to remove background noise. Sephadex G-25 showed 80–90% efficiency in eliminating background noise. Sephacryl (range 20–8000 KDa) worked best for removing unbound primary-secondary antibody conjugates but gave a very low yield. Data provided here was generated using Sephadex G-25 column. Column packing must be done carefully and air bubbles must be strictly avoided inside the column]. To prove that the method of detecting EVs by immunofluorescence is as sensitive to the methods that are based on detection of EVs by PKH67 labeling, we performed a dual labeling experiment, wherein upon antibody labeling of EVs, we also labeled the same EVs using PKH67 dyes as well.

Availability of Data and Materials Datasets and materials are available by the corresponding author.