Comparison of Extracellular Vesicle Isolation Methods for miRNA Sequencing.
Author(s): Llorens-Revull M, Martínez-González B, Quer J, Esteban JI, Núñez-Moreno G, Mínguez P, Burgui I, Ramos-Ruíz R, Soria ME, Rico A, Riveiro-Barciela M, Sauleda S, Piron M, Corrales I, Borràs FE, Rodríguez-Frías F, Rando A, Ramírez-Serra C, Camós S, Domingo E, Bes M, Perales C, Costafreda MI
Publication: Int J Mol Sci, 2023, Vol. 24, Page
PubMed ID: 37569568 PubMed Review Paper? No
Purpose of Paper
This paper compared the size, morphology, and microRNA (miRNA, miR) profile of extracellular vesicles (EVs) isolated from a single plasma specimen using size exclusion chromatography (SEC), iodixanol gradient (GRAD), and a combination of SEC and GRAD. Additionally, miRNA profiles of a single plasma specimen that underwent next-generation sequencing (NGS) library preparation with three different kits were compared.
Conclusion of Paper
When technical replicates were pooled for each EV isolation method SEC method yielded more particles/mL than the GRAD or SEC+GRAD methods; whereas protein levels were comparable among the isolation methods. Both the morphology and the mean particle size of the isolated particles were consistent with EVs. Significantly more miRNAs were detected in EVs isolated by GRAD than SEC or SEC+GRAD and by SEC+GRAD than SEC. Reproducibility in miRNA detection among the three technical replicates was the highest when GRAD was used for EV isolation followed by SEC and SEC+GRAD.
In plasma, the number of reads corresponding to miRNAs was higher in the duplicate NEBNext Multiplex Small RNA Library Prep Set (NEB) and the NEXTFlex Small RNA-Seq Kit v3 (NEXT) libraries than the SMARTer smRNA-seq libraries (SMARTer). Significantly more miRNAs were detected in plasma using the NEB or NEXT libraries than the SMARTer libraries, and the reproducibility of miRNA detection in plasma between the two technical replicates was higher in NEB libraries than NEXT or SMARTer libraries.
Studies
-
Study Purpose
This study compared the size, morphology, and miRNA profile of extracellular vesicles (EVs) isolated from a single plasma specimen using size exclusion chromatography (SEC), iodixanol gradient (GRAD), and a combination of SEC and GRAD. Additionally, miRNA profiles of a single plasma specimen that underwent next-generation sequencing (NGS) library preparation with three different kits were compared. Plasma was separated from the blood of a single volunteer by centrifugation (centrifugation details not provided) and stored at -80°C. Plasma was centrifuged at 1200 g for 20 min at 4°C followed by 10,000 g for 30 min at 4°C before EV isolation. EVs were isolated from plasma aliquots by: 1) fractionation through qEV2/35 nm columns (SEC), 2) ultracentrifugation at 100,000 g for 3 h at 4°C followed by applying the resuspended pellet to a 40% iodixanol step gradient prepared with OptiPrep and centrifugation at 140,000 g for 18 h at 4°C (GRAD) and 3.) SEC followed by GRAD. Total and subsets of CD9+ EVs were enumerated using a FACSCalibur cytometer. Total proteins were quantified with the bicinchoninic acid (BCA) assay. Levels of total cholesterol were measured using an AU5800 instrument and levels of apolipoprotein A (apoA) and apolipoprotein B (apoB) were measured using an automated nephelometry system. Particle size and concentration were evaluated by nanoparticle tracking analysis and morphology was visualized by transmission electron microscopy. RNA was extracted from plasma and isolated EVs using the miRNeasy Mini Kit. RNA was quantified using a NanoDrop spectrophotometer and RNA integrity was evaluated using a Bioanalyzer 2100. RNA was spiked with 52 synthetic miRNAs and sequencing libraries were constructed from plasma RNA using the NEBNext Multiplex Small RNA Library Prep Set for Illumina, the NEXTFlex Small RNA-Seq Kit v3, and the SMARTer smRNA-seq Kit and from EV RNA using the NEBNext Multiplex Small RNA Library Prep Set. Libraries were sequenced using the NextSeq 500/550 High Output Kit v2.5 on a NextSeq 500 system.
Summary of Findings:
Among SEC fractions, levels of CD9 (the EV marker) were highest in fractions 8-12, protein levels were highest in fractions 18-19, total cholesterol was highest in fractions 14-16, and ApoA was highest in fraction 18. VLDL, IDL, LDL, and lipoprotein(a) were not detectable in any of the SEC fractions. Among the GRAD fractions, CD9 was highest in fractions 5-8, protein levels were highest in fractions 3, 4, and 6, and cholesterol ApoA and apoB were undetectable. Fractions from the combined SEC+GRAD method had the lowest protein levels of the EV isolation methods evaluated; for the SEC+GRAD method, CD9 levels were highest in fractions 4-7. When technical replicates were pooled for a given EV isolation method, the fractions with the highest level of CD9 was 8-12 for the SEC method and 4-8 for both the GRAD method, and SEC+GRAD methods. The technical replicate pools for the SEC EV isolation method had more particles/mL than the GRAD or the SEC+GRAD method (7.49 x1011±5.5x1010 versus 8.51x109 ±1.98x108 and 3.83 x109±5.53x108, respectively). Both the morphology and the mean particle size of the isolated particles were consistent with EVs (133.37±1.93 nm for SEC pools, 136.5±4 nm for the GRAD pools, and 170.5±4.57 nm for the SEC+GRAD pools) and. Protein levels were comparable between the pools of technical replicates for each EV isolation method and were considerably lower than levels in plasma (1.86 mg/mL for the SEC pool, 2.02 mg/mL for the GRAD pool, and 1.50 mg/mL for the SEC+GRAD pool versus 147 mg/ml in plasma).
Recovery of the spike-in control miRNA from plasma was most consistent when libraries were constructed with either the NEXT or NEB Kit, but specimens with the spike-in miRNA clustered by library preparation method following principal component analysis. The number of reads in plasma that corresponded to miRNAs was higher in the duplicate NEB (1,653,876 and 1,649,122) and NEXT (4,823,457 and 2,519,523) libraries than the SMARTer libraries (5,348 and 18,918). Significantly more miRNAs were detected in plasma using the NEB or NEXT libraries than the SMARTer libraries (397 or 385, respectively, versus 213, P<0.001). The reproducibility of miRNA detection in plasma between the two technical replicates was higher in NEB libraries than NEXT or SMARTer libraries (84.1% versus 58.7% and 61.5%, respectively).
Significantly more miRNAs were detected in EVs isolated by the GRAD EV isolation method than SEC or SEC+GRAD (336 versus 223 or 242, respectively, P<0.0001, both) and by SEC+GRAD than SEC (242 versus 223, P=0.037). The highest reproducibility in miRNA detection among the three technical replicates for EV isolation was observed when EVs were isolated using the GRAD method (72.3%) followed by SEC (49.8%) and SEC+GRAD (48.8%) methods.
Biospecimens
Preservative Types
- Frozen
Diagnoses:
- Normal
Platform:
Analyte Technology Platform Cell count/volume Flow cytometry Cell count/volume Light scattering RNA Next generation sequencing Steroid Clinical chemistry/auto analyzer Lipoprotein Clinical chemistry/auto analyzer Peptide Colorimetric assay Pre-analytical Factors:
Classification Pre-analytical Factor Value(s) Next generation sequencing Specific Technology platform NEBNext
NEXTFlex
SMARTer smRNA-seq
Analyte Extraction and Purification Analyte isolation method Size exclusion chromatography (SEC)
Iodixanol gradient (GRAD)
SEC and GRAD