Learn how EV researchers characterize EV preparations using particle abundance, size distribution, marker analysis, and purity assessment in line with MISEV2023.

An Introduction to EV Characterization

Extracellular vesicles (EVs) have gained significant attention as important mediators of intercellular communication and as promising candidates for biomarker discovery and therapeutic delivery. However, translating this potential into robust research and clinical applications remains challenging. Current isolation methodologies are not perfectly selective and frequently co-isolate non-vesicular extracellular particles (NVEPs), making downstream characterization essential for assessing EV identity and sample composition. Compounding this challenge, EVs are highly heterogeneous and exist at the sub-micron scale, making their characterization technically demanding and difficult to standardize.

To help address these challenges, the International Society for Extracellular Vesicles (ISEV) has developed guidelines to support best practices in EV research. MISEV2023 recommends that EV characterization aim to establish EV presence, estimate particle abundance, assess size distribution, and evaluate the extent of non-vesicular material remaining in the preparation. Importantly, MISEV is not a rigid checklist; the appropriate depth of characterization depends on the source material, isolation strategy, and the biological claims being made.

Key takeaways 

• No one-size-fits-all approach: The level of characterization required depends on the source material, isolation method, and downstream application.
• One assay is rarely sufficient: EV identity is generally supported by multiple complementary measurements rather than a single technique.
• Four key domains: Robust EV characterization should describe the source material, particle size distribution and abundance, EV-associated markers, and sample purity.

What EV characterization data should you report? 

No universal workflow exists for EV characterization. The required depth of analysis depends on the starting material, isolation method, and intended downstream application. However, a robust analytical package typically includes data spanning four key domains:

1. Source and Starting Material

Characterization begins with documenting the starting material to provide context for downstream analyses. This includes:

• Origin: Detailed information about the cell type, biofluid, or tissue source, including relevant culture conditions or donor characteristics.
• Quantity: The volume of biofluid, number of cells, or mass of tissue used as input material.

2. Particle Size Distribution and Abundance

This involves estimating the size distribution and concentration of particles present in the final preparation.

• Size Distribution: EV populations are heterogeneous and span a broad range of sizes. Characterizing the size distribution provides important context for interpreting biological origin, isolation performance, and sample quality. Unexpected changes in the distribution may indicate co-isolated contaminants.
• Particle Abundance: Particle abundance is commonly reported as concentration (particles/mL) and, where appropriate, as total particle yield to facilitate comparisons between samples and isolation methods.

3. Presence of EV-Associated Markers

To support the presence of membrane-bound vesicles, MISEV recommends evaluating proteins from multiple categories.

• Transmembrane or GPI-anchored proteins: Surface-associated markers such as CD9, CD63, and CD81 provide evidence consistent with membrane-bound structures.
• Cytosolic proteins: Internal proteins such as TSG101, ALIX, and Syntenin support a cellular origin and the presence of luminal content.

Combining markers from multiple categories provides stronger evidence for EV identity than relying on a single protein.

4. Assessment of Purity (Non-EV Components)

Because no isolation method is perfectly selective, researchers should evaluate the presence of co-isolated materials that may influence downstream analyses.

• Co-isolated material: Assessing markers associated with common contaminants, such as albumin, ApoA1, or ApoB in plasma-derived samples, helps determine the extent of non-EV material in the preparation.
• Purity metrics: Ratios such as particles-to-protein or particles-to-lipid can provide additional information regarding sample composition, although these measurements should be interpreted cautiously and in conjunction with other characterization data.

By documenting these criteria, researchers can establish a transparent baseline description of their EV preparations. In practice, however, this typically requires the integration of multiple complementary methods.

Why are orthogonal methods so important? 

No single technique can independently satisfy all EV characterization requirements. Because EV preparations are heterogeneous and isolation methods remain imperfect, researchers often rely on orthogonal approaches that interrogate distinct physical and biochemical properties to strengthen confidence in EV identity.

Robust characterization therefore benefits from the convergence of multiple, independent lines of evidence:

• Physical characterization: Techniques such as NTA and TRPS measure particle size distributions and abundance.
• Biochemical profiling: Methods such as Western blotting, ELISA, and flow cytometry evaluate EV-associated proteins and assess the presence of co-isolated contaminants.
• High-resolution imaging: Techniques including TEM and Cryo-EM provide direct visualization of particle morphology and can reveal membrane-bound vesicular structures alongside non-vesicular material.

Ultimately, EV characterization is not based on a single “perfect” assay. Rather, robust evidence emerges from combining complementary methods that together provide a coherent picture of particle identity, composition, and purity.

Expert perspective:

“Reliable EV characterization remains technically challenging. Current isolation approaches often co-isolate contaminants such as lipoproteins, protein aggregates, and cell debris, resulting in labour-intensive characterization workflows.

At EXIT071, we use the Leprechaun platform for our internal EV characterization. By integrating size measurements, particle abundance, and phenotypic analysis within a semi-automated workflow, Leprechaun enables multi-dimensional characterization while supporting streamlined and reproducible EV research. This integrated approach also facilitates the identification and analysis of EV subpopulations.”

FAQ 

1. What purity controls matter most?

MISEV2023 recommends assessing EV preparations using markers from multiple categories rather than relying on a single measurement. A robust purity assessment typically includes:

• EV-associated membrane proteins to support the presence of membrane-bound vesicles.
• Cytosolic EV-associated proteins to provide evidence of intracellular content and cellular origin.
• Markers of non-EV co-isolated structures to evaluate the extent of contaminants such as lipoproteins, protein aggregates, or abundant serum proteins.

Together, these measurements provide a more complete picture of EV identity and sample composition. 

2. What positive EV markers should I use?

No universal EV marker exists. While markers such as CD9, CD63, and CD81 are widely used, not all EV populations express these proteins equally.

Marker selection should therefore be guided by the biological source, EV subtype of interest, and the specific scientific question being addressed. Using complementary markers from multiple categories generally provides stronger evidence than relying on a single protein.

3. Why are purity controls so important? 

Current EV isolation methods are not perfectly selective and often co-isolate non-vesicular extracellular particles (NVEPs), including lipoproteins, protein aggregates, and cell debris.

These contaminants can influence measurements of particle abundance, molecular cargo, and biological activity. Assessing purity therefore helps researchers interpret their results more confidently and reduces the risk of drawing conclusions from mixed particle populations.

Conclusion 

Characterizing EVs requires integrating multiple complementary measurements—from particle abundance and size distribution to marker profiling and purity assessment. At EXIT071, we address these challenges using the Leprechaun platform, our semi-automated EV characterization system that combines immunoaffinity capture, fluorescence detection, and SP-IRIS within a unified workflow.

Interested in learning more about the current EV characterization landscape or how Leprechaun enables multi-dimensional EV analysis? Explore our related articles and discover how EXIT071 supports EV research and development.