Skip to main content

Functional repertoire of EV-associated miRNA profiles after lipoprotein depletion via ultracentrifugation and size exclusion chromatography from autologous blood products.

Otahal A, Kuten-Pella O, Kramer K, Neubauer M, Lacza Z, Nehrer S, De Luna A

Publication: Sci Rep (2021) • Vol 11 • Page 5823PMID: 33712660View on PubMed

Purpose

This paper compared particle concentration, size, and protein and microRNA (miRNA, miR) expression in plasma, serum, and extracellular vesicle (EVs) isolated from plasma and serum using six different methods.

Conclusion

The particle concentrations were equivalent in the plasma/serum and in EVs isolated by size exclusion chromatography (SEC) alone or in combination with ultrafiltration, but were lower than in plasma or serum when EVs were isolated by ultracentrifugation (UC) alone or in combination with SEC, regardless of order or number of UC steps. While generally the particle concentration was comparable between serum and plasma EVs, the concentration was significantly lower in serum than plasma when EV isolation was with UC followed by SEC and another UC. The mode particle size was generally larger than in plasma or serum when EVs were isolated by UC, which the authors state reflects depletion of non-EV particles. As expected, protein concentration decreased with increasing number of EV isolation steps performed. Significantly lower protein concentrations were found in EVs isolated by SEC followed by UC or UC followed by SEC and another UC from serum than plasma. Western blot analysis showed less lipoprotein contamination in the UC pellet than in serum or plasma or in EVs isolated by SEC and ultrafiltration, but lipoprotein contamination was below the limit of detection in EVs isolated by UC followed by SEC and another round of UC. However, while the EV markers Alix and CD-9 were observed at high levels in the UC pellet, they were reduced when EVs were isolated by UC followed by SEC and the signal was mostly lost when EVs were isolated by UC followed by SEC and another round of UC. When a much higher amount of protein was loaded on the gel, CD9 and CD63 but not ALIX were detected in EVs isolated by UC followed by SEC and another round of UC from plasma but not serum. The majority of miRNAs detected in serum or plasma were found in both but the majority of miRNAs identified in EVs were only found in EVs isolated from plasma. Rank miRNA levels were strongly correlated between plasma and plasma EVs but only a weak correlation was observed for rank miRNA expression in serum and serum EVs. Importantly, the enriched miRNA in EVs differed between serum and plasma. It is important to note that miRNA levels were low with an average of one miRNA for 5.69% of particles from plasma and 0.26% of particles from serum. Nevertheless, the identified miRNAs in EVs from plasma and serum mapped to 2033 and 1042 targets, respectively, mapping the 640 and 597 reactome pathways, respectively. After the stringency was increased, enriched miRNA in serum EVS were found to cluster in the proliferation and TGFβ/SMAD signaling pathways.

Studies (1)

Purpose

This study compared particle concentration, size, and protein and miRNA expression in plasma, serum, and EVs isolated from plasma and serum using six different methods. Blood was collected from 4 female and 3 male healthy blood donors (25-45 years of age) into sodium citrate Vacutainer tubes and using the hypACT device. Plasma was obtained by centrifugation of the sodium citrate blood at 440 × g at room temperature and then at 1710 × g for 10 min at room temperature. The pellet was resuspended in plasma and stored along with the supernatant. Serum was obtained from hypACT blood by centrifugation at 1710 × g for 10 min at room temperature. Blood products (serum, plasma, plasma pellet) were centrifuged at 2500 x g for 10 min and EVs were isolated by UC (100,000 x g for 120 min at 4°C), SEC (qEVoriginal sepharose column), UC followed by SEC, SEC followed by UC, SEC followed by ultrafiltration (centrifugation at 10,000 × g in a Vivaspin PES 100 kD cut off filter), or UC followed by SEC and then UC again. Importantly, only a subset of EV isolation methods were used for specimens from each individual because of sample volume. Size and concentration of isolated EVs were investigated by nanoparticle tracking analysis. Protein levels were analyzed using the BioRad DC assay and levels of Alix, ApoB100/48, CD9, CD63, and ApoA1 were measured by Western blot. RNA was extracted from freeze-thawed plasma, serum, and EVs isolated via UC + SEC + UC using the miRNeasy Mini Kit. RNA was reverse transcribed using the miRCURY RT Kit and expression was profiled using the miRNome Panel I and normalized using the most stable miRNA as identified by NormFinder. A specimen was considered hemolyzed if the ratio of miR-23a-3p to miR-451a was >5.

Summary of Findings

The particle concentrations were much higher in serum and plasma than the respective pellets. The particle concentrations were equivalent in the plasma or serum and in EVs isolated by SEC alone or in combination with ultrafiltration but significantly lower than in the input when EVs were isolated by UC alone or in combination with SEC, regardless of order or number of UC steps (P< 0.0001). While generally the particle concentration was comparable between serum and plasma EVs, the concentration was significantly lower in serum than plasma when EV isolation was with UC followed by SEC and another UC (P< 0.0001). The mode particle size was generally larger than in plasma or serum when EVs were isolated by UC (P<0.05, most), but the difference was not significant in serum EVs isolated by SEC before UC or in plasma EVs isolated by UC followed by SEC and another round of UC. The authors state the increase in mode size observed after UC reflects depletion of non-EV particles. As expected, protein concentration decreased with increasing number of EV isolation steps performed with approximately 1% of the protein in plasma or serum found in EVs isolated by UC and further reductions when subjected to SEC before UC and even more when EVs were isolated by UC followed by SEC and another round of UC.  Importantly, significantly lower protein concentrations were found in EVs isolated by SEC followed by UC or UC followed by SEC and another UC from serum than plasma (P<0.05 and P<0.01, respectively). Western blot analysis showed less lipoprotein contamination in the UC pellet than in serum or plasma or in EVs isolated by SEC and ultrafiltration, but lipoprotein contamination was below the limit of detection in EVs isolated by UC followed by SEC and another round of UC. However, while the EV markers Alix and CD-9 were observed at high levels in the UC pellet, they were reduced when isolated by UC followed by SEC and the signal was mostly lost when EVs were isolated UC followed by SEC and another round of UC. When a much higher amount of protein was loaded on the gel, CD9 and CD63 but not ALIX were detected in EVs isolated by UC followed by SEC and another round of UC from plasma but not serum. The majority of miRNAs detected in serum or plasma were found in both (248 of 278) but only 48 miRNAs were detected in EVs purified by UC followed by SEC and another round of UC from serum and plasma with an additional 108 and 1 miRNA detected only in EVs isolated from plasma and serum, respectively. Interestingly, rank miRNA levels were strongly correlated between plasma and plasma EVs isolated by UC followed by SEC and another round of UC (ρ=0.9592, P<0.001) but only a weak correlation was observed for serum and serum EVs (ρ=0.3879, P=0.006). Importantly, the enriched miRNA in EVs differed between serum and plasma. While serum was found by ratio of miR-23a-3p to miR-451a >5 to be hemolyzed, hemolysis was not detected in serum EVs, plasma, or plasma EVs. It is important to note that miRNA levels were low with an average of one miRNA for 5.69% of particles from plasma and 0.26% of particles from serum.  Nevertheless, the identified miRNAs in EVs from plasma and serum mapped to 2033 and 1042 targets, respectively, mapping the 640 and 597 reactome pathways, respectively. After the stringency was increased, enriched miRNAs in serum EVs were found to cluster in the proliferation and TGFβ/SMAD signaling pathways.

  • Biospecimens

    FluidBlood
    FluidPlasma
    FluidSerum
  • Preservation Types

    FrozenNone (Fresh)
  • Diagnoses

    Normal
  • Platforms

    RNAReal-time qRT-PCR
    ProteinColorimetric assay
    ProteinWestern blot
    Cell count/volumeLight scattering
  • Pre-analytical Factors

    ClassificationPre-Analytical FactorValue(s)
    Biospecimen Aliquots and ComponentsBlood and blood products
    Plasma
    Plasma vesicles
    Serum
    Serum vesicles
    Analyte Extraction and PurificationAnalyte isolation method
    SEC
    SEC and UC
    SEC and ultrafiltration
    UC
    UC and SEC
    UC followed by SEC and UC
    Western blot SpecificTargeted peptide/protein
    Alix
    ApoA1
    ApoB100/48
    CD63
    CD9