Summit v6

Cells were trypsinized, washed in PBS and fixed in 70% ethanol for 4 hr. Cells were then permeabilized with PBS containing 0.1% Triton X100 and 200 µg/ml RNase A for 2 hr at 37°C and stained with 167 µg/ml propidium iodide for 30 min. DNA content was measured by flow cytometry on a CytoFLEX flow cytometer (Beckman Coulter) and analyzed Summit v6.2 software (Beckman Coulter).

KU812 cells and MEFs were washed three times with PBS and then cultured for 36 h before exosome isolation. Conditioned media were harvested (120 ml) and centrifuged (480 × g for 5 min, 2000 × g for 10 min, 10 000 × g for 30 min, all steps at 4°C) as described previously (Théry et al, 2006; Ji et al, 2013). Crude exosomes were isolated by repeated ultracentrifugation (100 000 × g for 60 min at 4°C, twice) as described previously (Théry et al, 2006).
For flow cytometry, exosomes were labelled with anti‐CD63‐fluorescein isothicyanate (FITC) (561924; BD Biosciences, San Jose, CA, USA) and anti‐CD81‐allophycocyanin (APC) (561958; BD Biosciences) and analysed on a MoFlo Astrios device using Summit v6.2 software (Beckman Coulter, Brea, CA, USA). Before labelling, antibodies were centrifuged at 45 000 × g for 5 min to exclude antibody aggregates. For assessment of size calibration, silica beads (100, 500, and 1000 nm; Kisker Biotech, Steinfurt, Germany) with a refractory index near to biological material were recorded. 100 nm silica beads clearly discriminated from laser noise. For exosome analysis, events overlaying laser noise were excluded. An effect of swarm/coincidence was excluded as pooled, single‐stained exosomes did not show any signs of double positivity.

Flow cytometry FACS of Cy5-labeled bacterial cells was done with an ultra-high-speed cell sorter MoFlo Astrios EQ (Beckman Coulter, California, USA) using the software Summit v6.2 (Beckman Coulter), as represented in Additional file 7: Figure S2. To standardize measurements and assess bacterial size, silica calibration beads (100, 500, and 1000 nm, Kisker Biotech, Germany) with refractive indexes close to that of biological material were recorded. The sorting of Cy5-labeled bacteria was performed as followed: background noise of the machine was first detected using the parameters forward scatter and side scatter. 488-nm FSC1-Height-Log vs 488-nm SSC2-Height-Log was then used to show the different sizes of silica beads in the first measurement and the scattering of the bacteria in subsequent measurements. Bacteria were pre-gated and displayed on a third scatter plot with 488 nm SSC-Area-Log vs 640 nm 671/30-Area-Log axes. Cy5-positive bacteria were then sorted out into tubes with a maximum event rate of 50,000 events/s. Reanalysis of the samples showed a purity of more than 99%.