ABSTRACT: Milk-derived extracellular vesicles (mEVs) have been proved to play a critical role in intercellular communication, mainly through the microRNAs (miRNAs) that they carry, to regulate biological functions of the target cells. Given miRNAs are evolutionarily conserved, EVs present in commercial milk may play a role in the physiology and health consumers. It is therefore essential to know the effects of technological treatments such as skimming and spray drying on the EV content of milk powders and on the cargo of bioactive molecules, in particular miRNAs, that they convey. Since goat’s milk or goat milk based formulas are considered as a healthy alternative for infants with cow’s milk sensitivities, including allergy, we undertook to analyze the EV content of skimmed and unskimmed goat's milk powders and to characterize their RNA content, in particular their miRNomes. mEVs were isolated using an optimized protocol based on Size Exclusion Chromatography (SEC) and compared regarding morphology, number and size by Transmission Electron Microscopy (TEM) and Nanoparticle Tracking Analysis (NTA). Their RNA and protein content were determined and their miRNomes established, using RNA sequencing. In this study we demonstrated that goat milk powders, skimmed or not upstream the spray drying treatment, contained many mEVs, ranging from 5.4 1011 to 2.5 1012 particles per mL of reconstituted milk, with an average size between 136.8 and 160.6 nm. We also demonstrated that mEVs carried significant amounts of RNA, including miRNAs. Using RT-qPCR, mRNAs encoding five of the major milk proteins were detected, suggesting that mEVs originated from mammary epithelial cells. We established the goat milk powder miRNome by identifying 351 miRNAs of which 233 are common to the 262 miRNAs previously profiled in raw goat milk. The 20 most abundant miRNAs (TOP 20) account for 80% of the total reads and the hierarchy of this TOP 20 miRNAs is somewhat overturned when comparing goat milk powder and raw goat milk. Surprisingly, whereas the comparison of raw from cow and goat milk confirmed the prevalence of miR-148a, miR-21-5p and miR-26a/miR-30a-5p, let-7a-5p and let-7f, which occupied ranks 1 and 2, respectively, in powders, were relegated to ranks 6 and 10 and 5 and 11 in raw goat and cow milk, respectively. Conversely to what was previously reported, we provide evidence that: i) EVs of typical morphology are present in goat milk powders; ii) mEVs survived the technological processes used to produce the powders; iii) their miRNA cargo is protected from degradation even though their miRNomes are not an exact mirror of miRNomes of EVs derived from fluid raw milk.
