Abstract
The release of extracellular vesicles (EVs) in cell cultures as well as their molecular cargo can be influenced by cell culture conditions such as the presence of foetal bovine serum (FBS). Although several studies have evaluated the effect of removing FBS-derived EVs by ultracentrifugation (UC), less is known about the influence of FBS heat inactivation (HI) on the cell-derived EVs. To assess this, three protocols based on different combinations of EV depletion by UC and HI were evaluated, including FBS ultracentrifuged but not heat inactivated (no-HI FBS), FBS heat inactivated before EV depletion (HI-before EV-depl FBS), and FBS heat inactivated after EV depletion (HI-after EV-depl FBS). We isolated large (L-EVs) and small EVs (S-EVs) from FBS treated in the three different ways, and we found that the S-EV pellet from HI-after EV-depl FBS was larger than the S-EV pellet from no-HI FBS and HI-before EV-depl FBS. Transmission electron microscopy, protein quantification, and particle number evaluation showed that HI-after EV-depl significantly increased the protein amount of S-EVs but had no significant effect on L-EVs. Consequently, the protein quantity of S-EVs isolated from three cell lines cultured in media supplemented with HI-after EV-depl FBS was significantly increased. Quantitative mass spectrometry analysis of FBS-derived S-EVs showed that the EV protein content was different when FBS was HI after EV depletion compared to EVs isolated from no-HI FBS and HI-before EV-depl FBS. Moreover, we show that several quantified proteins could be ascribed to human origin, thus demonstrating that FBS bovine proteins can mistakenly be attributed to human cell-derived EVs. We conclude that HI of FBS performed after EV depletion results in changes in the proteome, with molecules that co-isolate with EVs and can contaminate EVs when used in subsequent cell cultures. Our recommendation is, therefore, to always perform HI of FBS prior to EV depletion.