Characterization of EVs isolated from differently processed bovine milk

• Published in: Journal of extracellular vesicles Vol. 7; p. 83
• Main Authors: Kleinjan, Marije, Libregts, Sten F H M, Feitsma, Anouk, van Neerven, Joost, Wauben, Marca H M
• Format: Journal Article
• Language: English
• Published: Abingdon John Wiley & Sons, Inc 01-01-2018
• Subjects: CD63 antigen; CD9 antigen; Centrifugation; Cow's milk; Flow cytometry; Homogenization; Milk; Pasteurization; Pasteurized milk; Ribonucleic acid; RNA; Western blotting
• ISSN: 2001-3078

Background: We investigated how processing of bovine milk affected the EV quantity and composition by isolating EVs from homogenized, pasteurized or ultra-heat-treated (UHT) milk and comparing these EVs to raw bovine-milk-derived EVs. Methods: EVs from differently processed bovine milk were isolated using differential centrifugation followed by sucrose density gradient centrifugation. Density gradient fractions 4-6, 7-9 and 10-12 were pooled and analysed using high-resolution flow cytometry, cryo EM and western blot. Small RNA from EV containing fractions was isolated and concentrations small RNA were determined by Bioanalyzer. Results: The quantity of EVs as measured by high-resolution flow cytometry is not affected in pasteurized milk when compared to raw milk. However, homogenization and pasteurization resulted in a strong reduction of EVs in fraction 7-9. In UHT milk, the amount of EVs was drastically reduced. These results were confirmed by cryo EM. Western blotting showed that the general EV markers CD9 and CD63 were most prominent in fraction 7-9 of all kinds of milk, except for UHT-treated milk where no protein signals could be detected by western blotting. Remarkably, in raw milk, MHCI and MHCII were detected in fraction 7-9, whereas these markers were detected mostly in fraction 4-6 after pasteurization. This could indicate that MHCI/II-positive EV populations were lost or damaged during milk processing. After pasteurization, a clear loss of small RNA cargo was seen in fraction 7-9, but not in fraction 4-6. Furthermore, homogenization of milk clearly affected the distribution of MFG-E8 through the gradient. Summary/conclusion: Processing of milk affects the EV population. Depending on the type of processing, different effects on the total EV population or on EV subsets were observed. Although no clear effects on total EV numbers were observed after pasteurization, the total RNA yield was reduced and the EV integrity was probably affected (shift in buoyant density based on distribution of MHCI/II and miRNAs). Homogenization most likely affected mainly the MFG membranes in milk while UHT treatment had the most detrimental effect on EVs.