Cell culture
HeLa cells (ATCC) and MCF7 cells (a kind gift from Munira Kadhim, Oxford Brookes University) were grown in Dulbecco’s Modified Eagle Medium (DMEM)/ Ham’s F-12 Nutrient Mixture (HAM F12) medium plus 10% (v/v) Foetal Bovine Serum (FBS). MCF7 cells were transfected with an mEmerald-CD81 construct (mEmerald-CD81-10 Plasmid Addgene number 54031) and stably expressing colonies were selected using 600 µg/mL geneticin (G418) containing medium. The homogeneity of the mEmerald-CD81 positive MCF cell line was routinely checked by fluorescent microscopy (Primovert Cell and Tissue Culture Microscope; Zeiss) and flow cytometry (CytoFlex S flow cytometer; Beckman Coulter).
Preparation of conditioned media for EV extraction
Cells were seeded at ~ 2 × 106 MCF7 mEmerald-CD81-cells per T175 flask overnight. The cells were washed with PBS, medium was replaced with 25 mL DMEM F12 medium containing 5% (v/v) pre-cleared FBS (FBS previously centrifuged at 120,000×g for 16 h to pellet excess extracellular vesicles) and cells were cultured for 72 h. Conditioned medium was cleared by centrifugation at 300×g for 5 min followed by 16,500×g for 20 min at 4 °C. Approximately 50 mL of conditioned medium from two T175 flasks (approximately 90% confluent on collection), was concentrated using Vivaspin 20, 100 kDa (GE Healthcare) to 500 µL and used to isolate extracellular vesicles by size exclusion chromatography (SEC). Concentrated conditioned medium was either used immediately or frozen at − 80 °C until use.
Serum-free conditioned medium was made by seeding 4 × 106 MCF7 mEmerald-CD81-cells per T175 flask overnight. The cells were washed with PBS and medium replaced with 25 mL serum free DMEM F12/HAM medium for 48 h, to condition. The conditioned medium was then prepared as above.
Extracellular vesicle isolation by size exclusion chromatography (SEC)
Size exclusion columns were made using sepharose CL-2B (GE Healthcare). Briefly, 14 mL sepharose was added to an empty Econo-Pac chromatography Column (BioRad) and left to settle; a support bed was placed on top of column before washing twice with PBS. The 500 µL of concentrated medium sample was added and allowed to settle into the column, 10 mL of PBS was added, and 500 µL fractions were collected using PBS from the chromatography column. EV fractions (6 to 9) were kept, pooled and further concentrated using a Vivaspin 2, 5 kDa (GE Healthcare) to 200–400 µL for downstream experiments.
Nanoparticle tracking analysis (NTA)
Nanoparticle tracking analysis was performed using Zetaview (Particle Metrix). EV samples and stained EVs were diluted in PBS (usually 1 in 1000–10,000 dilution) and were analysed for size and concentration using NTA video capture for 0.5 s at 11 positions at room temperature. The acquisition conditions were set as follows: Sensitivity = 80, Shutter speed = 100, Frame rate = 30, Cycle number = 2, Minimum brightness = 25, Maximum size = 1000, Minimum size = 5, Trace length = 15. The videos were analysed using ZetaView software version 8.04.12.
Immunoblotting
EVs were concentrated in Vivaspin 2, 5 kDa, concentrator columns to 20 µL and lysed using 10X RIPA lysis buffer containing protease inhibitor cocktail III (1:100; Thermo Fisher). Lysates from cells or EVs (20 µg) were loaded into a 4–12% Bis–Tris NUPAGE gel (Invitrogen) either under reducing (with DTT; GM130) or non-reducing (Alix, TSG101, CD9 and CD81) conditions and run at 125 V for 90 min in NUPAGE MES SDS buffer. Gels were subject to a wet transfer onto a nitrocellulose membrane using NUPAGE transfer buffer (15% methanol) at 150 V for 2 h at 4 °C. Membranes were blocked in 5% (w/v) milk in 0.1% (v/v) PBST (PBS Tween-20) for 1 h. Membranes were probed with antibodies against: TSG101 (Abcam; ab83, mouse, 1:500) Alix (Abcam; Ab117600, mouse, 1:2500), CD9 (Cambridge Biosciences / System Biosciences; EXOAB-CD9A-1; rabbit 1:1000), CD81 (Abcam; ab79559, rabbit, 1:1000), and GM130 (Abcam; ab52649, rabbit, 1:1000). The membranes were then washed three times with PBST then probed in milk buffer with either anti-Rabbit conjugated horseradish peroxidase (HRP) (Promega; w4011, 1:5,000) or anti-mouse HRP conjugated secondary antibody (Promega; w402B, 1:20,000). The membranes were washed again three times with PBST, then developed using ECL reagent (Amersham; Pico) for 5 min (RT) and imaged using a Bio-Rad Gel Doc XR +.
MACsPlex exosome immunocapture assay
The MACsPlex exosome immunocapture assay was carried out using a modified microcentrifuge tube protocol47. Briefly, 1 × 109 EVs were incubated with 15 µL MACsPlex capture beads (Miltenyi Biotech) overnight rotating at 12 RPM. Beads were then washed in MACsPlex buffer and incubated with a cocktail of CD81-APC, CD63-APC and CD9-APC conjugated antibodies (Miltenyi Biotech) for 1 h. Beads were then washed twice with MACsPlex Buffer, resuspended in MACsPlex buffer, then fluorescence of the 39 bead populations were assessed by flow cytometry using a CytoFlex S flow cytometer (Beckman Coulter). Capture bead populations were resolved using channels B585 (PE) and B525 (FITC). Data was analysed by assessing the median fluorescence intensity (MFI) for APC (R660) for each gated bead population. Background bead MFI levels were subtracted from sample values and values were made relative to the average MFI for CD63/CD81/CD9 bead populations in MCF7 cells.
Transmission electron microscopy
EV samples (10 µL) were pipetted onto carbon 300 mesh copper grids (TAAB, C267), previously glow discharged for 20 s at 15 mA, and incubated for 2 min. Grids were then dabbed against filter paper and placed onto a 20 µL drop of 2% (w/v) uranyl acetate for 10 s, then left to air-dry before storage. Grids were visualised using Jeol JEM-1400 Flash transmission electron microscope (TEM) with a Gatan OneView 16 Megapixel camera at 100 kV.
EV dual-labelling with C5-maleimide-Alexa633
MCF7 mEmerald-CD81 EVs were stained with C5-maleimide-Alexa633 using a previously published protocol19. Briefly, this involved performing a BCA assay on un-lysed EVs and staining 60–100 µg EVs (total) in 50 µL PBS with 2.5 µL of 200 µg/mL C5—maleimide-Alexa633 (ThermoFisher) for 1 h, at room temperature, in the dark, with no agitation. The excess dye was removed using Exospin 3 kDa columns (Invitrogen) or by a making the sample up to 500 µL with PBS and performing an additional round of SEC.
EV dual-labelling with PKH26
MCF7 mEmerald-CD81 EVs were stained with 2 µM PKH26 (Sigma). 20 µL EVs in PBS were diluted with 80 µL diluent C; 1 µL of PKH26 (final concentration: 2 µM) was added and the EVs were incubated in the dark at room temperature for 10 min. The excess dye was removed using Exospin 3 kDa columns (Invitrogen) or by a making the sample up to 500 µL with PBS and performing an additional round of SEC.
Proteinase K and Triton X-100 treatment of EVs
Concentrated MCF7 CD81emGFP EVs were treated with 100 µg/mL Proteinase K (Invitrogen) in isolation or with 1% (v/v) Triton X-100 (Sigma) then incubated at 37 °C for 1 h, with vortexing every 15 min. After treatment(s), EVs were placed under coverslips for confocal imaging.
Exoview experiments
Exoview R100 (Nanoview BioSciences, UK) experiments were performed at Nanoview BioSciences UK facilities according to their protocols. Briefly, EVs were prepared and stained as above from conditioned media containing 10% (v/v) pre-cleared FBS. 20 µL of diluted EVs were incubated overnight on Exoview chips containing immunocapture spots for CD81, CD63, CD9 and MIgG, in triplicate. Exoview chips were washed three times with incubation solution then stained with antiCD81-AF555 antibody for 1 h. Exoview chips were washed three times, then imaged using the Exoview R100 and analysed for fluorescence and size using the ExoScan 2.5.5 acquisition software (NanoView Biosciences, UK).
Preparation of EVs in 1% agarose for confocal imaging
EVs were mixed with 2% (w/v) low melting agarose in TAE buffer (cooled to just above melting temperature) in a 1:1 ratio (10 µL:10 µL), then the mixture (20 µL) was pipetted onto a glass microscope slide with a 13 mm diameter cover slip placed on top. This was left to dry, then secured by clear nail polish.
Uptake assay
HeLa cells were seeded overnight in 8-well chamber slides with removable wells (Nunc Lab-TEK II Chamber slide system; Thermo Fisher) at a density of 20,000 cells/ well. Cells were treated with dual stained (mEmerald-CD81 EVs stained with either PKH26 or Maleimide) or unstained mEmerald-CD81 EVs for 3 h. The dose of EVs was equivalent to 50 × the growth area of the cells (e.g. EVs harvested from a T175 flask used to treat one well [growth area 1 cm2] would equal a 175 × dose). After EV treatment cells were optionally stained with cell mask orange or cell mask deep red (Thermo Fisher), washed with PBS and fixed in 4% (w/v) Paraformaldehyde for 15–20 min at room temperature. Fixed cells were then washed three times with ice cold PBS, incubated with 30 mM glycine for 5 min and washed before aspirating the buffer, removing the wells and mounting a coverslip on top with DAPI containing Prolong gold antifade mounting agent (Thermo Fisher).
Confocal imaging
EVs were imaged using a Zeiss LSM880 or Zeiss LSM800, as indicated on individual figure legends. Images of EVs in agarose were taken using a 100 × objective (LSM800; 100x = NA 1.4, LSM880; 100x = NA 1.46). Images of cells were captured using a 63 × objective (NA = 1.4). Images were acquired using 4 × line averaging. Z stacks were taken sectioning the height of the agarose/ cells.
Image analysis
Images were analysed using Zen black software (Zeiss) or ImageJ (FIJI)48. Colocalisation of dual labelled EVs was analysed through the JACOPx Plugin on ImageJ FIJI49. Pearson’s coefficient, Rank-weighted colocalisation (RWC)49, fluorescent particle counts and colocalisation based on centres of mass-particles coincidence were performed using the JACOPx Plugin settings with manually adjusted thresholds, matching the intensity of particles seen on the software with the original image captured. The percentage of dual stained EVs was calculated using the object based ‘colocalization based on mass-particle coincidence’ feature of the JACOPx plugin. This automatically quantified the number of particles and assessed overlap of pixels calculated between the channels. The values of coincidence were then used to calculate the percentage of dual stained EVs as a proportion of the individual EV particles counted.
Statistical analysis
Statistical analysis was performed on GraphPad Prism using 2-way ANOVAs and Student’s unpaired t-test, as stated in figure legends. All error bars are standard error mean (SEM) unless otherwise stated. All graphical figures were prepared using GraphPad Prism.
