Fig 2: Western Blot of lipoprotein (ApoA1, ApoB100/48) and EV markers (Alix, CD9, CD63). (A) 10 µg protein was loaded per lane (except for PL, 20 µg) and membrane Ponceau S stained after transfer is shown. (B) Detection of CD63 and CD9 (80 µg per lane, except for 20 µg for PL) in UC + SEC + UC pellets. (C) UC + SEC + UC pellets obtained after removing supernatant. PL: platelet lysate, S2: plasma or serum supernatant after 2500 × g centrifugation, S100: supernatant after 100,000 × g centrifugation, P100:.pellet after 100,000 × g centrifugation, fat: buoyant layer after ultracentrifugation, a: marker (Invitrogen, #LC5800), b: marker (Invitrogen, #LC5699). Full-length blots/gels are presented in Supplementary Data 2.

Fig 3: Evaluation of exosomes sheathed on PSiNPs in E-PSiNPs. a Hydrodynamic diameter of PSiNPs and E-PSiNPs by DLS analysis. b TEM images of PSiNPs and E-PSiNPs. Scale bar: 200 nm. c Colocalization of DiO (green) and PSiNPs (red) in E-PSiNPs by confocal microscopy. Scale bar: 20 µm. d Colocalization of CD63 (green) and PSiNPs (red) in E-PSiNPs by confocal microscopy. Scale bar: 20 µm. e Immunoblotting analysis of exosome markers (TSG101 and CD63) and ER marker (calnexin) expressed in E-PSiNPs exocytosed from Bel7402 cells. f Yield of E-PSiNPs when Bel7402 cells were pretreated with 200 µg mL-1 PSiNPs for 6 h and then incubated in fresh medium containing 15 nM DMA or 10 µM ionomycin for 16 h by ICP-OES. Data were represented as mean ± SD (n = 3). **P < 0.01, ***P < 0.001 (one-way ANOVA with Fisher’s LSD test). Source data are provided as a Source Data file

Fig 4: Immunofluorescence: UW228-2 cells and large extracellular vesicles (EVs) are CD63+. Medulloblastoma cell lines produce large extracellular vesicles (EVs) which can be harvested from serum-free conditioned media after 24 h. (A) Medulloblastoma cells produce EVs in vitro which contain polymerised actin and CD63, but no nucleus. The medulloblastoma cell line UW228-2 was cultured on glass bottom plates for 24 h; fixed and probed for the tetraspanin and EV marker CD63 (FITC—green). Polymerised actin was labelled using Alexa Fluor 555 phalloidin (f-ACTIN—red). Nuclei were counterstained with DAPI (DAPI—blue). Images of large EVs (yellow arrows) and cells were captured using the Perkin Elmer Operetta system at ×40 magnification. The individual images were captured in black and white using the appropriate emission filters and a coloured composite image created using Columbus software. Scale bar represents 20 µM. (B) Transmission electron microscopy: UW228-2 cell and adjacent large EV with limiting membrane. Large EVs have a limiting membrane, contain organelles but no nucleus and are < 6 µm in size. Cells and EVs were applied to Poly-D-lysine coated ACLAR film and fixed with glutaraldehyde before processing for Transmission Electron Microscopy. Serial sections were taken to assess sample purity, EV membrane integrity, size, and content comparative to the parent cell. Images were captured using a Biotwin Philips TECNAI G2 microscope at ×1900 magnification. Scale bar represents 5 µM. Left panel shows a cell and a large EV. Right panel: serial sections taken through the entire EV (4 images shown). The wide-field image shows an EV of 5.5 µM diameter at its central plane of focus, alongside a parent medulloblastoma cell of 18 µM. (C) Flow cytometry: Cell viability Annexin V/PI assessment. Cell apoptosis was assessed by flow cytometry. Unstained cells (i) and cells treated with 200 µM Etoposide (ii; Annexin V positive gate) or heated (iii; PI positive gate) provided positive controls which were used to assign accurate quadrant gates to bivariate scatter plots showing PI versus Annexin V APC. Live cells appear in Q4, Annexin only in Q2 (Early apoptosis), dual labelled in Q3 (Late apoptosis) and PI only (Dead cells). UW228-2 cells were cultured for 24 h in complete media (iv; DMEM with 10% FBS) or serum-free media (v; DMEM only). Representative plots of triplicate experiments shown. (D) Imagestream (ISX) analysis: Annexin V/PI assessment in cells. Imaging flow cytometry allows visual distinction between early and late stages of apoptosis. Cells in standard culture conditions were labelled using an Annexin FITC/PI kit and examined using the Imagestream (ISX) Mark II. Images were captured at x60 magnification and representative gallery images demonstrate Annexin V FITC only membrane labelling, indicating early apoptosis (upper panels) and dual Annexin V FITC/PI labelling, indicating late apoptosis (lower panels). (E) Cell viability prior to EV isolation. UW228-2 cells can be cultured in serum-free conditions for 24 h with no loss in viability. No difference in viability could be seen between culture conditions: Percentage unlabelled cells (live), Annexin V positive only (Early apoptosis), Annexin V and PI positive (Late Apoptosis) or PI positive only (DEAD) after 24 h in complete (grey bars) or serum-free medium (open bars). (F) Large EV isolation and characterisation. EV isolation protocol from cultured cells was optimised to harvest large EVs with minimal processing. Cells were seeded into large flasks (225 cm3) and allowed to adhere. Media was changed after 24 h for 50 mL serum-free DMEM and cells cultured at 37 °C for a further 24 h. Trypsinised cells were collected by 2 successive centrifugation steps at 300× g 5 min keeping the supernatant each time. EV containing supernatant was filtered using a double layered 5 µm pore nylon membrane by gravity. Filtered, cell-free supernatant was centrifuged at 2000× g for 30 min using a bench top centrifuge with swing out buckets. The resulting cell and EV pellets were processed appropriately for the downstream technique: For flow cytometry cells were processed and analysed separately, whilst cells were spiked into wells for direct comparison by microscopy.

Fig 5: The ISX can accurately assign diameter to large EVs and facilitate individual event analyses to phenotype heterogeneous EV populations. (A) The median diameter of CMO + EVs from UW228-2 cells was comparable across 10 samples. The erode mask was applied to the bright field images of those EVs which were included in the CMO+ >MESF 50 gate. The diameter of individual events within this gate was exported from IDEAS into PRISM for analysis. This figure shows a median diameter (represented by the black bar) of 922–1129 nm across the 10 samples and visualises the broad size range within each sample. (**** p > 0.0001 Kruskal-Wallis test). (B) The fluorescence intensity of the CMO was proportional to EV diameter. Bivariate plots showing CMO intensity against diameter consistently demonstrates a proportional relationship between EV size and intensity of the membrane label across samples run in triplicate experiments. 2 representative samples shown. (Regression analysis p < 0.0001). (C) Comparison of EV phenotypes across triplicate experiments. Individual histograms displaying EV diameter for each phenotype demonstrated differential size ranges according to EV marker(s) expression (i a and i b). In each case, those labelled with CMO only and not expressing either CD63, CD9 or LAMP1 have a smaller median diameter compared with EVs expressing CD63 and/or CD9 or LAMP1 (ii a and ii b). In both cases, those EVs co-expressing both CD63 with CD9 (iii a) or CD63 with LAMP1 (iii b) are significantly larger than CMO only. (** p < 0.005, *** p < 0.0004, **** p < 0.0001).