Limits...
Isolation and characterization of platelet-derived extracellular vesicles.

Aatonen MT, Ohman T, Nyman TA, Laitinen S, Grönholm M, Siljander PR - J Extracell Vesicles (2014)

Bottom Line: Therefore, we optimized an EV isolation protocol and compared the quantity and protein content of EVs induced by different agonists.Ca(2+) ionophore generated a large population of protein-poor and unselectively packed EVs.These activation-dependent variations render the use of protein content in sample normalization invalid.

View Article: PubMed Central - PubMed

Affiliation: Division of Biochemistry and Biotechnology, Department of Biosciences, University of Helsinki, Helsinki, Finland.

ABSTRACT

Background: Platelet-derived extracellular vesicles (EVs) participate, for example, in haemostasis, immunity and development. Most studies of platelet EVs have targeted microparticles, whereas exosomes and EV characterization under various conditions have been less analyzed. Studies have been hampered by the difficulty in obtaining EVs free from contaminating cells and platelet remnants. Therefore, we optimized an EV isolation protocol and compared the quantity and protein content of EVs induced by different agonists.

Methods: Platelets isolated with iodixanol gradient were activated by thrombin and collagen, lipopolysaccharide (LPS) or Ca(2+) ionophore. Microparticles and exosomes were isolated by differential centrifugations. EVs were quantitated by nanoparticle tracking analysis (NTA) and total protein. Size distributions were determined by NTA and electron microscopy. Proteomics was used to characterize the differentially induced EVs.

Results: The main EV populations were 100-250 nm and over 90% were <500 nm irrespective of the activation. However, activation pathways differentially regulated the quantity and the quality of EVs, which also formed constitutively. Thrombogenic activation was the most potent physiological EV-generator. LPS was a weak inducer of EVs, which had a selective protein content from the thrombogenic EVs. Ca(2+) ionophore generated a large population of protein-poor and unselectively packed EVs. By proteomic analysis, EVs were highly heterogeneous after the different activations and between the vesicle subpopulations.

Conclusions: Although platelets constitutively release EVs, vesiculation can be increased, and the activation pathway determines the number and the cargo of the formed EVs. These activation-dependent variations render the use of protein content in sample normalization invalid. Since most platelet EVs are 100-250 nm, only a fraction has been analyzed by previously used methods, for example, flow cytometry. As the EV subpopulations could not be distinguished and large vesicle populations may be lost by differential centrifugation, novel methods are required for the isolation and the differentiation of all EVs.

No MeSH data available.


Related in: MedlinePlus

Size distributions of the total EV populations by NTA and TEM. Vesicle size distributions are shown as percentages of the total EV populations analyzed by NTA (A, 9 independent experiments) and TEM (B, 4 independent experiments). Diameters of vesicles in TEM micrographs were measured manually from 53–81 images/activation and proportioned to a scale bar. At least 400 vesicles were calculated for each condition. Representative images of NTA (insert in A) and the uranyl acetate–stained total EVs induced by Ca2+ ionophore (insert in B, original magnification 4800×). Table comparing the conditions showing statistical significances from A (C). Statistical significances were determined by t-test (paired two-sample for means, two-way) assuming unequal variances. P-values of less than 0.05 (*), less than 0.01 (**) and less than 0.001 (***) were considered significant.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4125723&req=5

Figure 0004: Size distributions of the total EV populations by NTA and TEM. Vesicle size distributions are shown as percentages of the total EV populations analyzed by NTA (A, 9 independent experiments) and TEM (B, 4 independent experiments). Diameters of vesicles in TEM micrographs were measured manually from 53–81 images/activation and proportioned to a scale bar. At least 400 vesicles were calculated for each condition. Representative images of NTA (insert in A) and the uranyl acetate–stained total EVs induced by Ca2+ ionophore (insert in B, original magnification 4800×). Table comparing the conditions showing statistical significances from A (C). Statistical significances were determined by t-test (paired two-sample for means, two-way) assuming unequal variances. P-values of less than 0.05 (*), less than 0.01 (**) and less than 0.001 (***) were considered significant.

Mentions: Next, possible agonist-dependent differences in the size distribution of the EVs were analyzed by NTA, which showed that 94–99% of EVs were <500 nm and 65–82% were <250 nm. When the main peaks of the size distribution profiles from the differentially induced EVs were compared, the EVs from the agonist-treated platelets, excluding Ca2+ ionophore, tended to be slightly bigger than the unstimulated (Fig. 4A), while for the Ca2+ ionophore–induced EVs, the predominant size was <100 nm. The activation-dependent differences diminished at the size range of 100–250 nm. At the 250–500 nm range, most of the EVs were induced by the TC co-stimulus. Although agonist-dependent size differences were observed, they were not statistically significant. Strikingly, the most vesicles were in the 100–250 nm size range from both the total EV pool, 59–65% (Fig. 4A) and the EXO pool, 64–77% (data not shown). The centrifugation step (20,000×g) to obtain the EXO supernatant, free of MPs, reduced the number of vesicles within the 250–500 nm size range and increased the percentage of vesicles <100 nm, but the most of the EVs were still 100–250 nm. This result was independent of the activator used (Supplementary Fig. 2), and shows that a large population of EVs can be dismissed if the commonly used differential centrifugations of MPs (10–20,000×g) and EXOs (100,000×g of a pre-filtrate) are used to harvest the EVs.


Isolation and characterization of platelet-derived extracellular vesicles.

Aatonen MT, Ohman T, Nyman TA, Laitinen S, Grönholm M, Siljander PR - J Extracell Vesicles (2014)

Size distributions of the total EV populations by NTA and TEM. Vesicle size distributions are shown as percentages of the total EV populations analyzed by NTA (A, 9 independent experiments) and TEM (B, 4 independent experiments). Diameters of vesicles in TEM micrographs were measured manually from 53–81 images/activation and proportioned to a scale bar. At least 400 vesicles were calculated for each condition. Representative images of NTA (insert in A) and the uranyl acetate–stained total EVs induced by Ca2+ ionophore (insert in B, original magnification 4800×). Table comparing the conditions showing statistical significances from A (C). Statistical significances were determined by t-test (paired two-sample for means, two-way) assuming unequal variances. P-values of less than 0.05 (*), less than 0.01 (**) and less than 0.001 (***) were considered significant.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4125723&req=5

Figure 0004: Size distributions of the total EV populations by NTA and TEM. Vesicle size distributions are shown as percentages of the total EV populations analyzed by NTA (A, 9 independent experiments) and TEM (B, 4 independent experiments). Diameters of vesicles in TEM micrographs were measured manually from 53–81 images/activation and proportioned to a scale bar. At least 400 vesicles were calculated for each condition. Representative images of NTA (insert in A) and the uranyl acetate–stained total EVs induced by Ca2+ ionophore (insert in B, original magnification 4800×). Table comparing the conditions showing statistical significances from A (C). Statistical significances were determined by t-test (paired two-sample for means, two-way) assuming unequal variances. P-values of less than 0.05 (*), less than 0.01 (**) and less than 0.001 (***) were considered significant.
Mentions: Next, possible agonist-dependent differences in the size distribution of the EVs were analyzed by NTA, which showed that 94–99% of EVs were <500 nm and 65–82% were <250 nm. When the main peaks of the size distribution profiles from the differentially induced EVs were compared, the EVs from the agonist-treated platelets, excluding Ca2+ ionophore, tended to be slightly bigger than the unstimulated (Fig. 4A), while for the Ca2+ ionophore–induced EVs, the predominant size was <100 nm. The activation-dependent differences diminished at the size range of 100–250 nm. At the 250–500 nm range, most of the EVs were induced by the TC co-stimulus. Although agonist-dependent size differences were observed, they were not statistically significant. Strikingly, the most vesicles were in the 100–250 nm size range from both the total EV pool, 59–65% (Fig. 4A) and the EXO pool, 64–77% (data not shown). The centrifugation step (20,000×g) to obtain the EXO supernatant, free of MPs, reduced the number of vesicles within the 250–500 nm size range and increased the percentage of vesicles <100 nm, but the most of the EVs were still 100–250 nm. This result was independent of the activator used (Supplementary Fig. 2), and shows that a large population of EVs can be dismissed if the commonly used differential centrifugations of MPs (10–20,000×g) and EXOs (100,000×g of a pre-filtrate) are used to harvest the EVs.

Bottom Line: Therefore, we optimized an EV isolation protocol and compared the quantity and protein content of EVs induced by different agonists.Ca(2+) ionophore generated a large population of protein-poor and unselectively packed EVs.These activation-dependent variations render the use of protein content in sample normalization invalid.

View Article: PubMed Central - PubMed

Affiliation: Division of Biochemistry and Biotechnology, Department of Biosciences, University of Helsinki, Helsinki, Finland.

ABSTRACT

Background: Platelet-derived extracellular vesicles (EVs) participate, for example, in haemostasis, immunity and development. Most studies of platelet EVs have targeted microparticles, whereas exosomes and EV characterization under various conditions have been less analyzed. Studies have been hampered by the difficulty in obtaining EVs free from contaminating cells and platelet remnants. Therefore, we optimized an EV isolation protocol and compared the quantity and protein content of EVs induced by different agonists.

Methods: Platelets isolated with iodixanol gradient were activated by thrombin and collagen, lipopolysaccharide (LPS) or Ca(2+) ionophore. Microparticles and exosomes were isolated by differential centrifugations. EVs were quantitated by nanoparticle tracking analysis (NTA) and total protein. Size distributions were determined by NTA and electron microscopy. Proteomics was used to characterize the differentially induced EVs.

Results: The main EV populations were 100-250 nm and over 90% were <500 nm irrespective of the activation. However, activation pathways differentially regulated the quantity and the quality of EVs, which also formed constitutively. Thrombogenic activation was the most potent physiological EV-generator. LPS was a weak inducer of EVs, which had a selective protein content from the thrombogenic EVs. Ca(2+) ionophore generated a large population of protein-poor and unselectively packed EVs. By proteomic analysis, EVs were highly heterogeneous after the different activations and between the vesicle subpopulations.

Conclusions: Although platelets constitutively release EVs, vesiculation can be increased, and the activation pathway determines the number and the cargo of the formed EVs. These activation-dependent variations render the use of protein content in sample normalization invalid. Since most platelet EVs are 100-250 nm, only a fraction has been analyzed by previously used methods, for example, flow cytometry. As the EV subpopulations could not be distinguished and large vesicle populations may be lost by differential centrifugation, novel methods are required for the isolation and the differentiation of all EVs.

No MeSH data available.


Related in: MedlinePlus