Fig 1: BSA activates CFDA‐SE and results in false positives. (a) A solution of 0.05% BSA strongly activates CFDA‐SE, and this signal is not eliminated by detergent treatment. Plate reader data. (b) Flow cytometer data. 0.05% BSA activates CFDA‐SE resulting in fluorescent particles above threshold (right panel) which are indistinguishable from CFDA‐SE labelled EVs (middle panel). (c) Exospin SEC columns efficiently remove protein from EV samples. Graphs show analysis of 20 × 45 µL samples collected from the SEC column. In the top row 100 µL of 1% BSA was put through the column and the protein concentration of each measured by Nanodrop analysis. In the middle row the same experiment was repeated with 100 µL of rat plasma. In the lower row, 100 µL of rat plasma was purified using an SEC column and the first 180 µL elution collected. One‐hundred microlitres of this was put through a second SEC column, fractions were stained with CFDA‐SE and the proportion of particles above threshold measured. (d) The CD63 (B) antibody, which contains 1% BSA, activates CFDA‐SE in the absence of EVs. The isotype control IgG‐AF647 does not have a large effect. Plate reader data. (e) The antibody CD9, which does not contain BSA, does not activate CFDA‐SE in the absence of EVs. Plate reader data. BSA, bovine serum albumin; CFDA‐SE, carboxyfluorescein diacetate succinimidyl ester; EV, extracellular vesicle; SEC, size‐exclusion chromatography.