SUPPLEMENTARY MATERIALS AND METHODS

Cell Cultures and Cell Stimulation – Primary cortical astrocytic cultures, purified microglial cultures from E21 embryonic rat pups, and N9 murine microglial cell line were obtained and maintained as previously described (Calegari et al, 1999; Bianco et al, 2005). Human glioblastoma cell line ADF was cultured in RPMI-1640 supplemented with 10% fetal calf serum, 100 IU/ml penicillin, 10mg/ml streptomycin, and 2mM L-glutamine at 37°C and 5% CO2. Induction of PM-derived MPs was done as follows: confluent glial cells (N9, or primary cortical astrocytes) were deeply washed with PBS at 37°C, and then stimulated with 100mM Benzoyl-ATP (BzATP) in Krebs-Ringer solution (KRH, 125mM NaCl, 5mM KCl, 1.2mM MgSO4, 1.2mM KH2PO, 2mM CaCl2, 6mM D-glucose, and 25mM HEPES/NaOH, pH 7.4) for 15-30 min at 37°C and 5% CO2. Alternatively, for the isolation of MPs released by cells under non-stimulating conditions, cells were cultured in their medium containing ultracentrifuged serum (110,000g for 1h at 4°C) at 37°C and 5% CO2 for 15-30 min or 24h (24h-basal MPs), respectively. When needed, cell cultures were treated with different chemicals and drugs before BzATP stimulation, as described.

Time-lapse Videomiscroscopy – Glial cells were incubated for 2 min at 37°C and 5% CO2 with FM1-43 (2mM), an amphipathic styryl dye which partitions reversibly in membranes, or with the fluorescent phosphocholine analog, NBD (3mM), which has the fluorophore attached to the non polar tail. After washing in KRH, cells were mounted on an inox cell chamber (Elettrofor, Italy), treated or not treated with 100mM BzATP, and observed at 37°C, 5% CO2, and 55–60% relative humidity using an inverted Zeiss Axiovert 200M microscope, equipped with a chamber Incubator S (Zeiss, Germany) controlled by a CTI Controller 3700 (Zeiss, Germany). Images were acquired with a 63X objective and a digital CCD camera system Micromax 512 BFT (Princeton Instruments, NJ, USA) and analysed using Metamorph Imaging Series 6.1 software.

Vesicle Isolation by Annexin V-Coated Beads – PS+ vesicles from PM blebbing were isolated as described previously (Bianco et al, 2005). Confluent astrocytes (40´103 cells/cm2) or confluent N9 microglia (2.7´105 cells/cm2) in 75cm2 flasks were stimulated with 100mM BzATP as described above in 1 ml of KRH solution. The supernatant containing shed MPs was collected and, after centrifugation (10 min 300g 4°C) in order to remove cells and debris, was incubated for 10 min at 4°C under gentle periodic rotation with streptavidin beads (Sigma-Aldrich, Italy) precoated with biotinylated annexin V (Sigma-Aldrich, Italy). Shed MPs bound to annexin V-coated beads were then separated from the supernatant by gravity sedimentation at 4°C, treated with a reducing sample buffer and analysed for western blotting.

Vesicle Isolation by Differential Centrifugation – Confluent astrocytes (40´103 cells/cm2) or confluent N9 microglia (2.7´105 cells/cm2) in 75cm2 flasks were stimulated with 100mM BzATP in KRH or maintained in ultracentrifuged medium as described above. The supernatant was then withdrawn and subjected to differential centrifugation at 4 C° as follows (all steps at 4°C): 5 min at 300g to discard cells and debris (P1 pellet); supernatant, 20 min at 1,200g to obtain P2 vesicle fraction ; supernatant, 30 min at 10,000g to obtain P3 vesicle population; supernatant, 1h at 110,000g to obtain P4 vesicles. The resulting pellets of vesicles (P2-P4) were either resuspended in SDS sample buffer for western blotting, or resuspended (and fixed when needed) for negative staining electron microscopy or fluorescence microscopy.

Quantification of PM-Derived MPs – Spectrophotometric quantification of FM1-43 labelled MPs was performed in KRH devoid of BSA. Cells were incubated with 500mM FM1-43, washed with KRH solution, treated with different chemicals and drugs as indicated, and then stimulated with 100mM BzATP. Supernatants were collected, centrifuged 10 min 300g 4°C to remove cells and debris, and then total fluorescence was assayed at 485/535 nm with a spectrophotometric system (1420 Multilabel Counter Victor 2- Wallac, Finland). Given that FM1-43 dye could be internalised into intracellular compartments during BzATP incubation, a set of experiments was preliminarly performed to distinguish the amount of FM1-43 fluorescence due to labelled PM-derived MPs from that due to exosome fraction (Heijnen et al, 1999). To this purpose supernatants from FM1-43 labelled, BzATP-stimulated cells were centrifuged at 10,000g, and the fluorescence of separated vesicle pellet was measured. Negligible contribution of labelled exosomes (10,000g pellet-depleted supernatant) to total BzATP-induced MPs fluorescence was observed (Suppl. Figure 1).

Western Blotting - MPs isolated from confluent astrocytes (40´103 cells/cm2) or confluent N9 microglia (2.7´105 cells/cm2) in 75cm2 flasks by the immunoaffinity or the differential centrifugation methods were resuspended in SDS sample buffer, run on a 12% polyacrylamide gel and blotted onto nitrocellulose filters (Millipore, Italy). Selected proteins were detected with specific Abs followed by HRP-conjugated secondary Abs (Jackson ImmunoResearch Laboratories, UK) and revealed using an ECL system (Supersignal from Pierce- Celbio Italy). Vesicle lysates were loaded in order to respect relative yields of different populations (P2, P3, P4) produced by cells under a given condition (BzATP; basal 30 min; basal 24h; see Cell Stimulation Section above). In particular, Comassie Blue staining experiments in triplicate (one representative example in Suppl. Figure 2C) allowed to assess an average yield of: ~ 2mg proteins for P2 fraction of MPs, ~ 2 mg proteins for P3, and ~ 8 mg proteins for P4, per 84´106 confluent N9 cells upon BzATP stimulation. Similar yields were obtained for each MPs fraction from confluent 45´106 astrocytes stimulated with BzATP. Based on these results, all subsequent western blotting analyses of BzATP-induced or basal MPs protein lysates were performed as follows: separate protein lysates of P2, P3 and P4 from (BzATP-stimulated) 84´106 confluent N9 cells or 45´106 confluent astrocytes were loaded each on a single lane and compared to 4mg total protein lysates from N9 or astrocytes, respectively, in the same running experiment (Suppl. Figure 2). Optical density of immunolabelled, ECL-exposed protein bands was measured with ImageJ software vers. 1.35i, and avarage values and SEM were calculated over three independent experiments.

Fluorescence Microscopy of vesicles – Resuspended vesicels were spotted on glass microscope slides, covered with glass coverslips, sealed and observed with an inverted Zeiss Axiovert 200M microscope; images were acquired with a 63X objective (Zeiss) and analysed using Metamorph Imaging Series 6.1 software. For NBD or FITC-Annexin V labelling, vesicles were resuspended in 20mL KRH solution (plus 1% BSA) containing 3mM NBD or FITC-Annexin V (Sigma, A2214, 1:100), respectively, and incubated for 10 min at RT. Staining for Na+/K+ ATPase (mouse monoclonal Ab, clone 6H, 1:250, a gift by Dr Pietrini G, University of Milan), CD63/LAMP3 (Santa Cruz Biotechnology, clone M-13, 1:100) and GLAST (polyclonal rabbit Ab against N-terminal epitope, 10mg/ml, a gift by Dr Pietrini G, University of Milan) were performed as follows: vesicles from differential centrifugation were fixed in 4% paraformaldehyde for 20 min at RT, and washed in PBS for 12h at 4°C. Primary antibodies were added in a 1:1 volume of PBS buffer containing 17% goat serum, 0.3% TritonX-100, and 0.45M NaCl, and incubations were allowed for 1h at RT. Primary Ab-conjugated vesicles were then washed with PBS and pelletted at their respective centrifugation speed, before incubation with fluorochrome-conjugated secondary Abs for 2h at RT and further washing in PBS. Re-pelletted labelled vesicles were then spotted on glass microscope slides and observed at fluorescence microscope (obj 63X).

Flow Cytometry Analysis – P2, P3 and P4 pellets were resuspended in KRH solution containing 1% BSA, and incubated with PE-Annexin V (1:100, Bender MedSystem, 1h at RT) or anti-A-SMase primary Ab (rabbit polyclonal Ab, 1:100, 12h at 4°C, Perrotta et al, 2007) or with NBD membrane dye. Pellets incubated with anti-A-Smase primary Ab were then washed with in KRH solution (1% BSA), re-pelletted, and then incubated with goat anti-rabbit Alexa Fluor 647 fluorochrome-conjugated secondary Ab (1:200, Molecular Probes Invitrogen, 1h at RT) with (double labelling analysis) or without (single labelling analysis) PE-Annexin V. All single (PE-Annexin V or anti-A-Smase Ab/Alexa 647 secondary Ab) and double (PE-Annexin V plus anti-A-Smase Ab/Alexa 647 secondary Ab) labelled MPs were further incubated with NBD membrane dye (3mM) 2 min at RT before FACS analysis for triple labelling experiments. Incubation of MPs with goat anti-rabbit Alexa647 secondary Ab was performed as control experiments. For each experiment, 10,000 positive events (positive for NBD membrane dye) were analysed. Analysis was performed using a log scale for forward scatter and side scatter parameters; beads of 2 mm were used as size marker. In order to detect surface A-SMase in intact cells by FACS, N9 cells were stimulated with 100mM BzATP, washed with ice cold PBS and blocked for 20 min with PBS (1% BSA) at different times upon BzATP stimulation, as indicated. Cells were then incubated with rabbit polyclonal anti A-SMase Ab for 45 min at RT in PBS containing 1% BSA (Perrotta et al, 2007). After washing in PBS, cells were labelled with Alexa Fluor 488-conjugated goat anti-rabbit secondary Ab (Molecular Probes) for 45 min at RT. An isotype-matched Ab was also used as control in separate experiments. Samples were then resuspended at 106 cells/ml in PBS and analysed by FACS.

A-SMase Cell Surface Labeling and Immunoprecipitation –A-SMase cell surface exposure upon BzATP administration was examined using the membrane-impermeable biotinylation reagent, NHS-SS-biotin (Pierce, Inc., IL, 0.5mg/ml in DMEM) in two successive incubations on ice. After washing in DMEM and in PBS containing 1mM CaCl2 and 1 mM MgCl2, cells were dissolved in lysis buffer (10mM Tris, 150mM NaCl, 1mM EDTA, 0.1% SDS, 1% Triton-X 100, protease inhibitor cocktail, pH 7.4), and then streptavidin-agarose beads (Pierce, Inc., IL) were added to the supernatant to isolate cell membrane proteins. The biotinylated proteins were eluted by boiling in SDS sample buffer and analyzed by SDS-PAGE followed by Western blotting. In order to evaluate the cell surface exposure of A-SMase, 25mg of total lysate for each sample were analysed.

Electron Microscopy of vesicles – P2, P3 and P4 pellets from BzATP-stimulated astrocytes or N9 cells were fixed with 4% paraformaldehyde in PBS for 12h at 4°C, and processed for negative staining electron microscopy. Fixed vesicles resuspensions (5mL) were adsorbed to 400-mesh formvar/carbon-coated grids for 5 min at RT; grids with adherent vesicles were then negative contrasted with 5mL of 1% uranyl acetate for 5 min at RT, and analysed with Philips CM10 transmission electron microscope in bright field.

Yo-pro uptake – Cells (either microglia or astrocytes) were plated on 96 multiwells at equal density (2000 cells/mL) and cultured either overnight (microglia) or for 4-5 days (astrocytes). Cells were then exposed to 1mM YO-PRO-1 iodide dye (Invitrogen) in different experimental conditions, rinsed with phosphate buffered solution (PBS) and dye uptake was quantified by spectrofluorimetric readouts at 485/535 nm with a spectrophotometric system (1420 Multilabel Counter Victor 2- Wallac, Finland). To favour P2X7 opening experiments were performed in the absence of extracellular calcium ions (Virginio et al, 1997).

Chemicals – ATP, BzATP, lipopolysaccharide (LPS), SB203580, biotinylated annexin V, FICT-annexin V, KN-62, U0126, geninstein, PP2, Brilliant Blue G (BBG), TNP-ATP, A-SMase, imipramine, apyrase, BAPTA-AM, cytocalasin D, PMA and ionomycin were from Sigma-Aldrich. YO-PRO-1, FM1–43, NBD C6-HPC, were from Molecular Probes (Invitrogen). Y-27632 dihydrochloride was purchased from Tocris. Manumycin A, GW4869 and PP2 were from Calbiochem. PE- annexin V was from Bender MedSystem.

4