Comparison of methods for pre-processing, exosome isolation, and RNA extraction in unpasteurized bovine and human milk

Abstract

Milk is a highly complex, heterogeneous biological fluid that contains non-nutritive, bioactive extracellular vesicles called exosomes. Characterization of milk-derived exosomes (MDEs) is challenging due to the lack of standardized methods that are currently being used for milk pre-processing, storage, and exosome isolation. In this study, we tested: 1) three pre-processing methods to remove cream, fat, cellular debris, and casein proteins from bovine milk to determine whether pre-processing of whole milk prior to long-term storage improves MDE isolations, 2) the suitability of two standard exosome isolation methods for MDE fractionation, and 3) four extraction protocols for obtaining high quality RNA from bovine and human MDEs. MDEs were characterized via Transmission Electron Microscopy (TEM), Nanoparticle Tracking Analysis (NTA), and western immunoblotting for CD9, CD63, and Calnexin protein markers. We also present an optimized method of TEM sample preparation for MDEs. Our results indicate that: 1) Removal of cream and fat globules from unpasteurized bovine milk, prior to long-term storage, improves the MDE yield but not purity, 2) Differential ultracentrifugation (DUC) combined with serial filtration is better suited for bovine MDE isolation compared to ExoQuick (EQ) combined with serial filtration, however both methods were comparable for human milk, and 3) TRIzol LS is better suited for RNA extraction from bovine MDEs isolated by EQ and DUC methods. 4) TRIzol LS, TRIzol+RNA Clean and Concentrator, and TRIzol LS+RNA Clean and Concentrator methods can be used for RNA extractions from human MDEs isolated by EQ, yet the TRIzol LS method is better suited for human MDEs isolated by DUC. The QIAzol + miRNeasy Mini Kit produced the lowest RNA yield for bovine and human MDEs.

Fig 1. Pre-processing of bovine colostrum prior to long-term storage at -80°C.

Group (G)1: whole milk frozen immediately upon collection and processed post-thaw. G2: Whole milk processed to remove fat globules and cream prior to ultracold storage. G3: Whole milk processed to remove fat globules, cream, milk cells, and casein proteins prior to ultracold storage.

Fig 2. Bovine and human milk-derived exosome isolation via ExoQuick (EQ) precipitation and differential ultracentrifugation (DUC) methods.

Fig 3. Four RNA extraction protocols.

1) QIAzol + miRNeasy MiniKit (Q), 2) TRIzol LS (TLS), 3) TRIzol + RNA Clean and Concentrator Kit (Tri+RCC), and 4) TRIzol LS + RNA Clean and Concentrator Kit (TLS+RCC) used for the isolation of total RNA from bovine and human milk exosomes.

Fig 4. The effect of milk pre-processing prior to long-term storage for bovine milk-derived exosomes isolated via ExoQuick precipitation.

Group (G)1: whole milk frozen immediately upon collection and processed post-thaw. G2: Whole milk processed to remove fat globules and cream prior to ultracold storage. G3: Whole milk processed to remove fat globules, cream, milk cells, and casein proteins prior to ultracold storage. Size and distribution profiles of bovine milk-derived exosomes as determined by Nanoparticle Tracking Analysis (NTA) (A). Concentration [particles/mL] of bovine milk-derived exosomes (B). Relative protein abundance of two exosome-specific markers (CD9 and CD63) and a cellular marker (Calnexin) as determined by western immunoblotting (C). Total soluble protein isolated from human microglia cell culture (HMC3) is used to represent cellular protein profiles. Data are mean ± SEM with n = 2 independent trials/group. * Significant difference in exosome concentration between the pellets and supernatants (p ≤ 0.05).

Fig 5. The effect of milk pre-processing prior to long-term storage for bovine milk-derived exosomes isolated via differential ultracentrifugation.

Group (G)1: whole milk frozen immediately upon collection and processed post-thaw. G2: Whole milk processed to remove fat globules and cream prior to ultracold storage. G3: Whole milk processed to remove fat globules, cream, milk cells, and casein proteins prior to ultracold storage. Size and distribution profiles of bovine milk-derived exosomes as determined by Nanoparticle Tracking Analysis (NTA) (A). Concentration [particles/ml] of bovine milk-derived exosomes (pellet fraction) (B). Relative protein abundance of two exosome-specific markers (CD9 and CD63) and a cellular protein marker (Calnexin) as determined by western immunoblotting in the pellet fraction (C). Total soluble protein isolated from human microglia (HMC3) cells used to represent cellular protein profiles. Data are mean ± SEM with n = 2 independent trials/group. ^# Main effect of pre-processing (p ≤ 0.05). * Significant difference in exosome concentration between the pellets and supernatants (p ≤ 0.05).

Fig 6. Morphology of bovine milk-derived exosomes isolated via the ExoQuick (EQ) method and visualized by Transmission Electron Microscopy (TEM).

Group (G)1: whole milk frozen immediately upon collection and processed post-thaw. G2: Whole milk processed to remove fat globules and cream prior to ultracold storage. G3: Whole milk processed to remove fat globules, cream, milk cells, and casein proteins prior to ultracold storage. Scale bars: 200 nm—500 nm.

Fig 7. Morphology of bovine milk-derived exosomes isolated via differential ultracentrifugation (DUC) and visualized by Transmission Electron Microscopy (TEM).

Group (G)1: whole milk frozen immediately upon collection and processed post-thaw. G2: Whole milk processed to remove fat globules and cream prior to ultracold storage. G3: Whole milk processed to remove fat globules, cream, milk cells, and casein proteins prior to ultracold storage. Scale bars: 200 nm—1000 nm.

Fig 8. RNA yield [ng/μL], purity and quality of bovine milk-derived exosome pellets and supernatants isolated via ExoQuick (EQ) precipitation and differential ultracentrifugation methods (DUC).

RNA was extracted using four protocols, 1) QIAzol + miRNeasy MiniKit (Q), 2) TRIzol LS (TLS), 3) TRIzol + RNA Clean and Concentrator Kit (Tri+RCC), and 4) TRIzol LS + RNA Clean and Concentrator Kit (TLS+RCC). RNA concentration [ng/μL] (A), RNA purity—absorbance at 260 nm/280 nm (B), and absorbance at 260 nm/230nm (C), and 1% TAE agarose gel electrophoresis (D) of the RNA samples. Data are mean ± SEM with n = 6 independent trials/group. ^@ Main effect of RNA extraction protocol (p ≤ 0.05). * Significant difference in RNA concentration and purity between the pellets and supernatants (p ≤ 0.05).

Fig 9. Human milk-derived exosomes isolated via ExoQuick (EQ) precipitation and differential ultracentrifugation (DUC) method.

Size and distribution profiles of human milk-derived exosomes as determined by Nanoparticle Tracking Analysis (NTA) (A). Concentration [particles/mL] of human milk-derived exosomes (pellet fraction) (B). Relative protein abundance of two exosome-specific protein markers (CD9 and CD63) and a cellular protein marker (Calnexin) as determined by western immunoblotting in the pellet fraction (C). Total soluble protein isolated from human microglia cell culture (HMC3) is used to represent cellular total protein profiles. Data are mean ± SEM with n = 6 independent trials/group. * Significant difference in exosome concentration between the pellets and supernatants (p ≤ 0.05). S10–S12 Figs shows the complete immunoblot images of the protein targets.

Fig 10. Morphology of human milk-derived exosomes visualized by Transmission Electron Microscopy (TEM) with negative staining (uranyl acetate).

Human milk-derived exosomes were isolated via ExoQuick (EQ) precipitation and differential ultracentrifugation (DUC) methods. Scale bars: 200 nm.

Fig 11. RNA yield [ng/μL], purity and quality of human milk-derived exosome pellets and supernatants isolated via ExoQuick (EQ) precipitation and differential ultracentrifugation methods (DUC).

RNA was extracted using four protocols, 1) QIAzol + miRNeasy MiniKit (Q), 2) TRIzol LS (TLS), 3) TRIzol + RNA Clean and Concentrator Kit (Tri+RCC), and 4) TRIzol LS + RNA Clean and Concentrator Kit (TLS+RCC). RNA concentration [ng/μL] (A), RNA purity—absorbance at 260nm/280nm (B), and absorbance at 260nm/230nm (C), and 1% TAE agarose gel electrophoresis (D) of the RNA samples. Data are mean ± SEM with n = 3 independent trials/group. ^@ Main effect of RNA extraction protocol (p ≤ 0.05). * Significant difference in RNA concentration and purity between the pellets and supernatants (p ≤ 0.05).