Advances in extracellular vesicle research: tools and techniques

Abstract

Extracellular vesicles (EVs) are lipid bilayer vesicles produced endogenously by all cell types and they contain cytosolic, extracellular, and membrane signaling molecules. Due to the physiological and pathological significance of EVs, they have emerged as potential biomarkers of diseases and as vectors for delivering therapeutic molecules. For accurate functional and clinical analysis of EVs, reliable methods for isolation, labelling, imaging and characterization are required. In this review, we discuss the progress in research on methods for isolating and analyzing EVs over the past few years. While different reviews have focused on different aspects of EV research, a detailed, comparative report spanning isolation, imaging, labelling and characterization is lacking. Our review will provide a comprehensive analysis of currently available techniques for EV research by discussing traditional and emerging methods of isolation including ultracentrifugation, ultrafiltration, chromatographic-based methods, precipitation-based methods and microfluidics approaches with their advantages and disadvantages. Furthermore, for tracking and visualization of EVs, various labeling strategies, including fluorescent labeling, nanoparticle labeling, radiolabeling and genetic labeling are summarized. In addition to imaging and characterization of EVs, various state-of-the-art technologies, including electron microscopy, fluorescence microscopy, flow cytometry, mass spectrometry and other molecular profiling techniques have been discussed. This review aims to help researchers select appropriate methods for isolating and analyzing EVs, in both fundamental and clinical research, by comparing the performance, scalability and reproducibility of methods through application-specific suitability. Future research should focus on improving detection methods, developing novel analytical platforms and integrating artificial intelligence to improve the specificity of EV characterization methods.

---

1. Introduction

Extracellular vesicles (EVs) are diverse nanosized particles, produced by all kinds of cells.¹ EVs were formerly considered cellular trash and underappreciated by the scientific community. However, investigations have shown that they facilitate cellular communication by transporting DNA, RNA, lipids and proteins to nearby and distant cells.^2,3 EVs have been extensively studied for their therapeutic significance. Recently, EVs have gained recognition as potential circulating disease biomarkers, often called “liquid biopsies”. They have been shown to facilitate long-range intercellular communication within the body.^4,5 Biofluids may contain substantial EVs that transport diverse molecules from parent cells to recipient cells, including proteins, mRNA, miRNA and DNA, functioning as cellular surrogates. The word “EVs” is employed in this review to refer to several subtypes of extracellular vesicles, regardless of their size, biogenesis, and morphological structures.^6,7

EVs are promising biomarkers for disease diagnosis.⁸ For diagnostic purposes, high yield is more important than purity. However, for therapeutic purposes, including cellular reprogramming, immunotherapy, and drug delivery, a pure, well-defined, and intact EV subpopulation is required to guarantee consistency, safety, and regulatory compliance.⁹ Since EVs participate in cellular communication by transporting their molecular cargo, their role in intercellular communication needs further exploration.¹⁰ These issues require current isolation and separation methods to determine how cell origins or states affect EV composition. Recent discoveries of new EV subpopulations, including exosomes, have highlighted the need to improve isolation and characterization methods. While ultracentrifugation (UC), ultrafiltration (UF), immunoaffinity (IA) capture, size exclusion chromatography (SEC) and precipitation are the most general approaches, new techniques, such as multistep combination methods and microfluidics, are being explored. Moreover, a combination of different methods is the most appropriate strategy for isolating EVs due to their diverse origins and heterogeneity. Additionally, conventional approaches face difficulties in large-scale production due to significant variability between inter- and intra-batch samples. Most existing methods are time-consuming, requiring tedious manual work, and are susceptible to contamination by proteins and lipoproteins. EV isolation and separation methods for therapeutic applications must be scalable, fully automated, and highly selective for specific EV subpopulations, as well as efficient in handling complex biological samples. This review provides a comprehensive overview of different methods for EV isolation, separation, and fractionation from diverse types of cells, emphasizing the more advanced and recent techniques.

2. Traditional methods for isolation of EVs

2.1. Ultracentrifugation (UC)

2.1.1. Differential ultracentrifugation (DUF). DUF is a pioneering technique initially developed for the isolation of EVs, and remains a widely regarded technique (Fig. 1). As with all centrifugation techniques, it is based on the density, size, and shape of particles, with larger and denser particles settling quickly.¹¹ As one of the earliest and most commonly employed methods for EV isolation, primarily due to its simplicity and the absence of extra requirements for markers, it is preferable for high-concentration samples. However, the heterogeneous nature of extracellular fluid can lead to the co-formation of different vesicles and protein aggregates at 100 000 × g; the method is not considered ideal and requires further purification. Moreover, excessive centrifugal force may compromise the integrity of particles, which may influence functional and downstream evaluations.^12,13 Regarded as the “gold standard” for EV isolation, DUF uses different centrifugal forces and speeds to separate molecules based on differences in size and density. In the initial steps, debris and large EVs are effectively, removed after the resuspension of the pellet in phosphate-buffered saline (PBS). This technique is quite simple and affordable, as no specialized materials and expertise are required that guarantee compatibility and consistency with more volume sizes.¹⁴ Tian et al. compared the different methods for EVs derived from plasma and characterized them with flow cytometry. The results showed that the UC method had the highest isolation purity in comparison to precipitation and size-based methods.¹⁵ This method, therefore, can be a preferred option while isolating EVs from biological samples. Nevertheless, the obtained results are difficult to standardize and susceptible to variations in biological material, the specific rotor type, and the timing of centrifugation. Particularly, when this kind of information is missing, the inter-study comparison cannot be reliable.¹⁶

Fig. 1 Methods for isolation of EVs. The figure summarizes the different techniques developed for separation of EVs from different types of biological fluids based on their distinct biochemical and physical properties. The EVs isolated through these methods were also validated for the presence of surface biomarkers, such as Alix, TSG101, Hsp70, CD81, CD9 and CD63.

2.1.2. Density gradient ultracentrifugation (DGUC). DGUC is an excellent approach for separation and purification of nano-sized biological molecules, from blood, plasma, cell homogenate and complex matrices.¹⁷ The use of DGUC as a research method has significantly improved the investigation of intracellular structures unraveling cellular mechanisms, i.e. compartmentalization of membrane and lipoprotein metabolism. Recently, DGUC has captured the interest of researchers as it has been proven to be a robust method for purification of EVs. This method is designed for optimal EV preparation that must be suitable for structural and functional analysis in comparison to alternative methods.¹⁸ Based on DUC, DGUC is an advanced method in which samples are mixed with an inert medium composed of either caesium chloride or sucrose using a density gradient. It works by exploiting the differences in the density of the medium and the particles. When centrifugal force is applied in the density gradient medium, the particles tend to accumulate in specific regions, which creates different regions and enables collection of particles with increased purity. This method is suitable for isolation of particles with a minimal difference in their sedimentation coefficients.¹⁹ Tengler et al. compared the density gradient ultracentrifugation and differential ultracentrifugation for isolation of EVs from saliva of head and neck cancer (HNC) patients. In order to obtain EVs with high purity and particle-to protein yield, the authors designed the optimized protocol (UCopt) by coupling the sequential centrifugation along with density gradient purification. The EVs isolated through this optimized protocol showed increased level of surface markers, directly related to the HPV (human papillomavirus) profile and tumor stage.²⁰
2.1.2.1. Rate-zonal centrifugation (RZC). RZC uses sedimentation rate properties of particles to separate them from original samples. In RZC, a shallow density gradient is created and samples that contain EVs are layered on the top of the gradient. During centrifugation, particles pass through the density gradient and are separated into different regions depending on their density, with the denser particles tending to be settled at the bottom of the centrifugation tubes. RZC is different from ultracentrifugation as it makes use of particle diameters and allows them to distinguish particles of similar density having different diameters.²¹ Rikkert and his colleagues used this approach for EVs isolation from plasma samples. With the method linear optiprep, the criteria used for separation of EVs was diameter instead of density. The EVs (100–300 nm) derived from platelets were separated from larger size particles through centrifugation at 20 °C, 2772g for 100 minutes. The results confirmed the separation of EVs from platelets, with abundance of EVs present in initial fractions and platelets in later fractions. The results were validated with different analytical techniques. In the RZC centrifuge tube, the density of the medium should exhibit a linear increase from top to bottom of the tube, with values lower than those of the experimental samples.²²
2.1.2.2. Isopycnic ultracentrifugation. Isopycnic centrifugation involves the movement of particles to a region of steep density gradient, where the gradient's density matches the particles buoyant density, called the isopycnic position. Once particles are at this position, they remain stationary even after prolonged centrifugation. In contrast to RZC, where particles settle due to less steep gradient, exosomes in isopycnic centrifugation remain suspended in their isopycnic position throughout the centrifugation process.²³ Isopycnic centrifugation effectively resolves the concern regarding co-precipitation due to shared physical characteristics during isolation of EVs. A sucrose-based density gradient, showing a gradual increase from top to bottom, allows EVs to settle into layers of corresponding isopycnic under centrifugal force, leading to the removal of contaminants.

Although it offers both improved isolation efficiency and purity over UC, isopycnic centrifugation still has notable drawbacks, including complicate handling procedures and expensive equipment.²⁴

2.1.2.3. Cushioned–density gradient ultracentrifugation (C–DGUC). Another method was to use the one-step sucrose cushion centrifugation (SCC) technique in which the cell supernatant or body fluids are directly added to the sucrose cushion, which does not require preconcentration step. After collecting and diluting the sucrose layer with PBS, it can be centrifuged to obtain EVs. By improving both the integrity and yield of EVs, this method shows significant improvements. The samples obtained through this method are highly pure and suitable for structural and functional analysis. The method also prevents the aggregation of particles and even at a high-speed particles remain intact.²⁵

2.2. Ultrafiltration (UF)

UF is also known as membrane filtration and has a molecular weight cut-off (MWCO) value of 10–100 kDa. These filtration membranes are typically composed of polyethersulfone, cellulose, or hydrogenated salts, with cellulose-based film being most commonly used. This technique concentrates EVs from a large volume of raw material into a small volume, streamlining the downstream process. Due to this efficiency, UF is one of the alternative methods to UC when a faster process is required. Two microfilters are also easy to operate, have a high recovery rate, and are connected sequentially for continuous flow filtration.²⁶

2.2.1. Tandem configuration filtration (TCF). TCF uses two filters with different filtration pore sizes in series, a 200 nm membrane on top and a 20 nm membrane below, to separate large particles (apoptotic bodies) from small particles (proteins) and leave the desired ingredient (EVs) in the middle layer. A time-consuming, continuous filtration procedure lead to the creation of the “ExoMir™ Exosome Isolation” kit. Co-separating non-EVs is still possible. Vesicles traveling through the membrane might block the pores of the filtration membrane, causing substantial damage and increased costs. The EVs are also damaged by shear stresses.²⁷

2.2.2. Tangential-flow filtration (TFF). TFF for EV isolation involves concentrating and filtering the conditioned cell culture supernatant simultaneously using a peristaltic system with a high molecular weight cut-off membrane filter. Researchers invented TFF, which avoids clogging by having particles flow parallel to the membrane surface rather than perpendicular to it.²⁸ The results showed an 18-fold increase in TFF-EVs production compared to UC-EVs and an improvement in its anti-apoptotic effect.²⁹ An ultrafiltration-based EVs isolation chip (ExoTIC) is straightforward to use and yields good results.³⁰ Lou et al. presented ultrafiltration-TiO2 series EV isolation using phospholipid affinity. Phospholipid affinity-based separation employed the metal–phosphate interaction on the lipid bilayer. Alternative urine sample processing methods include this rapid hybrid approach, which can process several urine samples and yield high-purity EVs.³⁰

2.3. Precipitation-based methods (polymer precipitation)

Another popular method for EV separation is polymer-based precipitation, which reduces EV solubility by forming a hydrophobic microenvironment through interaction between the water molecules surrounding the EVs and the very hydrophilic polymers. The most popular polymer precipitation methods are listed below.

2.3.1. Polyethylene glycol (PEG) precipitation. EVs are usually precipitated by adding a water-excluding polymer, such as PEG, to the samples. After binding to water molecules, PEG promotes the precipitation of exosomes and other particles from the solution. Afterwards, the vesicles can be precipitated by centrifugation and used for different analyses.³¹ This isolation approach is fast, easy, and cost-effective, requiring minimal technical skills. Moreover, it's also suitable for different starting volumes in laboratory and clinical applications; however, the major limitation of this method is its lack of selectivity. Also, in addition to the exosomes, the PEG precipitates protein aggregates, other proteins and EVs. Therefore, filtration and ultracentrifugation are necessary for sample pretreatment to reduce contamination.^32,33

2.3.2. Lectin-induced agglutination. Lectin precipitation can be used as a substitute for PEG precipitation. Lectins represent a family of proteins that interact with carbohydrate moieties present on the exosome surface and aid in isolation. After binding, lectin changes exosome solubility, leading to their precipitation from solution.³⁴ Commonly, ultracentrifugation is performed to remove other contaminants, including carbohydrates, from the sample as a pretreatment step. The sample is then incubated overnight with phytohemagglutinin at 1 mg L⁻¹ or concanavalin A, followed by precipitation of exosomes after centrifugation. Like PEG precipitation, lectin precipitation is also an easy and time-efficient approach with no expertise requirement; however, co-precipitation of other contaminants is limited, except that they are extremely glycosylated.³⁵ In comparison to the ultracentrifugation methods, precipitation methods are affordable and cost-effective, allowing vesicle recovery at a lower speed. Furthermore, these methods yield pure vesicles, which are suitable for downstream analysis, such as proteomic profiling. To reduce the contamination risk, there are specific measures with the kits to minimize contamination. However, the data are still inconclusive regarding the effect of these contaminants on EV function, with some reports suggesting possible cytotoxic effects.¹²

2.4. Chromatography-based isolation

2.4.1. Size exclusion chromatography (SEC). SEC separates molecules on the basis of their size. It uses gel-based polymers as stationary phase, and passing of mobile phase allows elution of sample component from the column (Fig. 1). The porous stationary phase allows separation of molecules by Stokes radii, allowing larger particles to come out first and small molecules takes the long path and therefore elute later.³⁶ The polymeric bed is composed of different materials such as sephacryl, sepharose, sephadex or biogel P. This technique is widely used due to high purity, significant yield and small volume requirement. SEC also keeps exosomes and their molecular cargo intact, and further downstream processing can be performed.³⁷ According to Takov et al., SEC yields are higher than those of UC; however, purity is compromised. To address this issue, a dual SEC column system has been introduced to separate exosomes of different sizes from human urine samples.³⁸ Similarly, Guo et al. made a significant breakthrough by introducing a straightforward dichotomic SEC using a CL-6B column, optimizing the bed volume from 10 mL to 20 mL, and replacing multiple elution steps with a streamlined two-step process. This approach is well-suited for isolating EVs and proteins from human serum, FBS, and FBS-free cell culture supernatants. Experimental results showed that this technique can improve reproducibility in clinical applications while achieving high-quality EVs with an excellent recovery rate.³⁹ All these advancements in SEC through improving resin efficiency and expanding bed volume led to a simpler two-step workflow and advanced FPLC approaches, which represent a significant step towards more standardized and clinically more relevant isolation protocols for EVs.

2.4.2. Affinity chromatography. This immunoaffinity capture technology depends on marker proteins on the surface of EV membranes, like CD81, CD63, CD82, CD9, Rab5, epithelial cell adhesion molecules, annexins, and programmed cell death 6 interacting protein. Numerous approaches based on immunoaffinity capture, have used affinity columns, microtiter plates, and magnetic beads. Immunoaffinity capture gives a lower yield but extraordinary purity. This is because it isolates subtypes of EV more precisely, relying on specific markers instead of isolating all EV subtypes all together.⁴⁰ A column-based CD9-HPLC-IAC (CD9 antibody-immobilized HPLC-immunoaffinity chromatography) technique was established by Zhu et al.⁴¹ This technique is efficient, and could be monitored in real-time. This results in extraction of EVs from trace serum within 30 minutes. This provides substantial improvements in EVs purity along with lower contamination by proteins and apolipoproteins compared to the SEC and UC procedures.⁴² Moreover, the immunoaffinity procedure is sophisticated in all aspects, safeguarding the integrity of EVs.^41,43

The purification of EVs remains a significant challenge in advancing fundamental research and commercializing EV-based products. Barnes et al. used heparin affinity chromatography (HAC) as a novel approach for EV fractionation and purification. The method was found to be effective in purification due to its reliability, scalability and compatibility with automation.⁴⁴ Likewise, Zimmerman et al. used numerous advanced strategies for purification of EVs and eliminated molecular contaminants from plasma using multimodal chromatographic techniques. The comparison of ultracentrifugation and multimode chromatography-based methods discovered 1054 and 716 distinctive groups of protein in different EV isolates from plasma. The established approaches produced comparable results in terms of purity of isolated EVs, facilitating more straightforward application, scalability and throughput, as well as suitability for diverse downstream analyses and future clinical implementations (Table 1).⁴⁵

Table 1 Comparison of advantages and disadvantages of EVs' isolation techniques

Technique	Advantages	Disadvantages	Ref.
Ultracentrifugation	• Straightforward, Simple operation, extensively used and a standard method	• Time consuming	46 and 47
	• No requirement for extensive chemicals	• High-cost equipment
	• Cost-effective	• Structural and functional integrity may compromise
	• High yield	• Risk of contamination
		• Increased sample volume is required
Ultrafiltration	• Fast and versatile	• Low recovery due to membrane adhesion	30 and 46
	• Good integrity of EVs	• Membrane clogging
	• High yield	• Structural integrity may compromise due to applied pressure and force
	• Scalability
Chromatography (SEC)	• Efficient and budget friendly	• Intricate procedure	48 and 49
	• Nondestructive	• Time consuming
	• Maintaining structural and functional properties	• Co-isolating contamination of same size
	• High scalability
Immuno-affinity capture	• Increased specificity and purity	• Limited scalability, reduced yield	40 and 44
	• Easily accessible	• Requirement of specific markers of EVs
	• No requirement for expensive instruments,	• Loss of functional proteins on EV surface
	• Useful for small volume of samples	• No single marker for separation of EVs
		• Requirement of specific antibodies
Polymer-based precipitation	• Simple and economical	• Unstable quality of kits	32 and 46
	• Availability of standard kits	• Decreased purity
	• Scalability	• Co-precipitation of other contaminants
	• High yield	• Structural and functional integrity may be compromise
	• No need for special equipments	• The precipitation agent may interfere with downstream processes
Microfluidics	• Continuous and fast approach	• Minimal sample can be loaded	50 and 51
	• High purity	• Complex and costly procedure
	• Offering isolation and characterization at the same time	• Reequipment of skilled staff

---

3. Emerging methods for isolation of EVs

3.1. Immuno-captured based isolation

3.1.1. Magnetic-based Immunocapture. Immunoaffinity assays employed for isolation of EVs can use a magnetic bead of a size of sub-micron and coated with antibody. This gives improved yield, specificity and sensitivity. The improvement is due to the bigger surface area, almost constant capture mode, and no restrictions on the volume of the sample. This procedure of isolation comprises of antibody-coated immunomagnetic beads which attach to the surface molecules related to EVs. The beads are then mixed with samples to form complexes with EVs. Finally, a magnetic field separates them from the sample by prompting their movement. In magnetic immunocapture, EVs are separated by conjugation to specific antibodies on magnetic beads.⁵² Chen et al. developed two-step magnetic beads-based (2MBB) method for EVs isolation. The method involves use of magnetic beads that are tagged with capture molecules. These molecules can identify biomarkers on the surface of EVs, including CD63. The trapped EVs can be eluted from magnetic beads or hydrolyzed directly for subsequent analysis. Likewise, Valle-Tamayo et al. designated a two-step method of separation, for astrocyte-derived EVs (ADEVs) from plasma samples. Firstly, EVs are separated from plasma free of fibrin using polymer-based precipitation, and then, they are concentrated with the help of ACSA-1-targeted magnetic beads immunocapture.⁵³ In another study, Chou et al. introduced LipoMin, an optimized glycosaminoglycan (GAG)-functionalized magnetic bead-based reagent designed for the rapid removal of LP (lipoprotein particles). The efficacy of LipoMin reagent was checked against EVs obtained from UC pellets and SEC fractions. The method is fairly simple as it involves 10 minutes of incubation with LipoMin, followed by the exclusion of the supernatant containing LP.⁴⁹

3.1.2. DNA-directed immobilization (DDI). Other than old methods for separation of EVs, DDI is a tender and meticulous procedure that separates EVs by attaching antibodies specific to EV surface markers on a solid surface through a DNA linker. This facilitates effective capture of EVs and their subsequent discharge using a DNAse enzyme. To avoid the difficulties linked with immunoaffinity capture, Brambilla et al. presented a different approach for capturing sEVs and releasing them through DNA-guided immobilization of anti-CD63 antibody. The method uses a flexible DNA linker for enhancement of release and binding of EVs, allowing for smooth release by the endonuclease activity of DNase I.⁵⁴ Similarly, Lu et al. created multi-input binding system that performs logic computations and produces dual-output tandem chips for the separation of EV's subpopulation. By means of very specific dual-aptamer binding and sensitive tandem microchips to isolate tumor PD-L1 and non-tumor PD-L1 EVs successively. The system differentiates between healthy people and cancer patients, presenting unique indicators of immunological heterogeneity.⁵⁵

In a different study, Pham et al. established a simple methodology that combined a fluorescence polarization-based homogeneous assay and DNA-associated aptamer. This is done to circumvent the requirement of separating free ligands from the bound ones for detection of EV. Enhanced specificity comes from the immobilization of EVs with antibodies. This is then followed by detecting it with a DNA aptamer that targets distinctive EV biomarkers. This twin tactic guarantees the exclusion of non-EV substances from the sample, prior to quantification of biomarker-positive EVs.⁵⁶ Combined together, these DDI-based advanced technologies serve as a game-changer toolkit for the separation of diverse subpopulations of EVs, their characterization, and the identification of various biomarkers. The combination of DNA-based aptamer technologies, modular microchip systems, and innovation in microarray signifies improvement in diagnostics based on EVs in a clinical setting.

3.2. Microfluidic devices

The microfluidic devices for EVs isolation are small in size, utilizing a lab-on-a-chip system and accurate fluid handling, ensuring efficient separation of EVs from biological samples. These chips offer several advantages, like rapid processing, high throughput, and low volume requirements in comparison to traditional techniques, like ultracentrifugation, which can isolate EVs based on their surface markers, or other physical properties through techniques, like immunoaffinity capture, affinity-based capture, deterministic lateral displacement (DLD), and hydrodynamic focusing.⁵¹ Hua et al. developed a double TFF-based microfluidic technology to isolate exosomes from human blood and cell supernatants. The microfluid device included two modules. Each module separates exosomes, large vesicles, and free biomolecules using two polymethylmethacrylate (PMMA) plates with uniform serpentine channels and a nanoporous membrane. Compared to UC microfluidic chip-based separation has reduced instrumental and consumable costs, reduced duration, increased purity, and a higher recovery rate.²⁹ Cong et al. developed an in vivo method for detection of PD-L1⁺ EVs that can allow separation of EVs based on microfluidics device and their quantification analysis of PDL1+ through hybridization chain reaction and aptamer recognition. The constant monitoring and quantification of PD-L1⁺ EVs that show their direct correlation with different tumor stages. Based on PD-L1⁺ EVs concentration, healthy and diseased persons can be distinguished efficiently.⁵⁰ Similarly, Guo et al. introduced a one chip isolation approach with EXoCPR, a magnetic nanoparticles microfluidic system for isolation and purification of EVs from different samples in a short time. The ExoCPR chip is an integration of two systems; bubbles driven micromixers and IFAST (immiscible filtration assisted by surface tension) technology. The micromixer helps remove the unnecessary steps of off-chip oscillatory incubation and extra pipetting, therefore improving the formulation of EVs with immunomagnetic beads. The pure EVs were collected after going through non-miscible interface, where hydrophobic and hydrophilic impurities were removed, binding nonspecifically with SIMI.⁵⁷

4. Methods for labeling of EVs

4.1. Fluorescent labeling

Fluorescent labelling of EVs offers a new paradigm to examine their composition and functionality.⁵⁸ A range of methods can be employed for fluorescent labeling of EVs, including but not limited to membrane staining with hydrophobic/lipophilic dyes, water-soluble dyes within vesicles, membrane permeable dyes for internal protein, tagging and immunostaining of surface proteins (Fig. 2). Unfortunately, protein-dependent labeling of vesicles may be influenced by protein abundance bias and can affect subsequent functional analyses that are based on targeted proteins. Due to the membrane barrier, water-soluble dyes act separately from vesicles and cannot be taken up by preformed vesicles. Although carboxyfluorescein diacetate succinimidyl ester (CFDA-SE), a nonfluorescent membrane-permeant molecules can enter vesicles through passive diffusion and produce fluorescence, they only function if vesicles have active esterases. Therefore, their activity can be influenced by vesicle content. Membrane staining with lipophilic markers offers bright and impartial labeling, as different cyanine-derived dyes with single-particle sensitivity have been produced across visible light range.⁵⁹

Fig. 2 Methods for Labeling of EVs. The figure summarizes different methods, including fluorescent labeling, nanoparticle labeling, genetic labeling and radiolabeling etc. developed for labeling of EVs for detailed investigation of their biodistribution, mechanism of action and function.

4.1.1. Non-covalent labeling. Carbonaceous dyes are a group of hydrophobic fluorescent dyes which label the plasma membrane and other hydrophobic biological components, including membrane of the vesicle. When they bind to the lipid membrane, their fluorescence intensity increases significantly. These dyes exhibit strong quenching efficiency and extended fluorescence lifetimes. These dyes diffuse across the entire membrane and can be used at optimal concentrations to visualize the entire structure of EVs. Common carbocyanine dyes include DilC18⁷ (DiR), DiOC18³ (DiO), DilC18³ (DiI), and DilC18⁵ (DiD). For in vitro tracking of EVs, DiO and DiL are commonly used because they exhibit minimal fluorescence before passing through the membrane and their strong signal after membrane entry. Chen et al. examined four different lipophilic dyes for labeling of EVs, including di-8-ANEPPS, DiI, PKH26 and PKH67. The results showed 100% labeling efficiency with reduced damage to membrane of EVs by both Di-8-ANEPPS and DiI. The aggregation of DiI is found to be less in comparison to PKH dyes, suggesting its suitability for membrane labeling. Further, results from nan-flow cytometry also suggest that lipophilic dyes, such as DiI, do not affect the EV immunophenotying.⁶⁰ Similarly, a salt-change method was developed to improve the labeling efficiency of DiI.⁶¹ The DiI aggregates were removed through adjustment of NaCl concentration without ultracentrifugation. The proposed method obtained higher incorporation of dye into EVs in comparison to other traditional methods by a factor of 290 and guaranteed consistent staining of EVs by reducing non-specific labeling.⁶¹

4.1.2. Covalent labeling. A method based on protein ligase catalysis was developed by Pham et al. where they covalently immobilized the target peptides’ epidermal growth factor receptor (EGFR)-binding peptide or anti-EGFR nanobodies with EVs. The authors achieved maximum of 10⁴ ligands in each EV (higher copy number labeling) without reducing their biocompatibility and integrity. The results showed improved delivery of paclitaxel and increased accumulation of EGFR-directed EVs in lung cancer xenografts with higher expression of EGFR in a mouse model.⁶² Similarly, Jiang et al. incorporated phosphatidylcholine into EVs by using phospholipase D (PLD) mediated alkyne integration, followed by fluorescence labeling with Cy5-azide by click chemistry. Their method allows real-time monitoring of EVs uptake in live cells through clathrin-mediated endocytosis and macropinocytosis. The labeling method was reported to be biocompatible, and EVs retained their functionality of cargo transfer, including miRNA transfer.⁶³ Bhatta et al. presents a metabolic glycoengineering method for tagging azido group on the surface of EVs in parent cells through incorporation of unnatural sugar. The tagging of azido groups on EV surface allowed click chemistry conjugation with maleimide-functionalized Toll-like receptor 9 agonists, therefore improving the activation of dendritic cells and antitumor response from T cells. The study showed potential of EVs in immunotherapy after surface covalent modification.⁶⁴ Likewise, Jiang et al. covalently labeled thiol moieties in EVs with increased expression of HER2 using maleimide fluorescent dyes (C5-makeimide-Alexa 633). Visualization with super-resolution microscopy showed changes in tetraspanin (CD9/CD63) expression in trastuzumab-resistant cells, associated with HER2 density. This approach allowed quantification of covalently labeled EVs and non-disruptive monitoring of treatment response in plasma samples.⁶⁵

4.2. Nanoparticles labeling

4.2.1. Metal nanoparticles labeling. Metal NPs are most commonly used for labeling of EVs. For visualization of EVs through photoacoustic imaging (PAI), magnetic resonance imaging (MRI) and computed-tomography (CT), gold is a good contrast metal. Zhang et al. labeled exosomes derived from engineer macrophages with extremely small iron oxide nanoparticles (ESIONPs). The small iron oxide nanoparticle (ESIONPs) incorporated exosomes (ESIONPs@EXO) were incubated with macrophages to release exosomes containing NPs. Exosomes cause ferroptosis in the endothelial cells and act as contrast agents for MRI during angiogenesis imaging with stable integrity and activity. The comparison of ESIONPs@EXO with free NPs showed improved suppression of tumor angiogenesis and reduction in tumor volume by 60% in a murine model.⁶⁶ Buttiens et al. investigated the effect of different metal nanoparticles (Iron oxide, silicon dioxide, silver and gold) on secretion and composition of EVs derived from human breast cancer cells and murine melanoma. The results showed that increased concentration of EVs was observed in both gold and silver nanoparticles, while iron oxide nanoparticles caused 40% reduction in EVs secretion. Further, an increased expression of miR-21/miR-126 was found in EVs exposed to NPs, increasing the immunosuppressive and pro-angiogenic activity of EVs.⁶⁷ Similarly, Toomajian et al. labeled metastatic breast cancer cell-derived EVs with superparamagnetic iron oxide (SPIO) NPs which showed increased concentration in metastatic tissues.⁶⁸

4.2.2. Quantum dots (QDs) labeling. A recent breakthrough in EV research is labeling of EVs by coupling with QDs, photoluminescent nanosized semiconductors. Different methods have been developed for their conjugation using click chemistry-based approach. The conjugation of EVs with QDs could be customized by changing the ratio of QD to EV or by introducing a catalyst. This conjugation chemistry approach is found to be stable in the living system and can also be stored for one week.⁶⁹ The separation of uncoupled QDs from coupled QD-EV can be performed through SEC, allowing detection of EV specific signals. It was found that QDs-coupled EVs were persistent and can withstand fixation and visualization during high resolution in tissues and cells. These conjugates were found to be better in terms of their photostability in comparison to the DiI EV dye.⁷⁰ Zhang et al. labeled EVs from semen (sEVs) and excised tissues of rat brain (bEVs). They successfully tracked QD-sEVs from mucosal lining of vagina and also monitored the interaction of QD-bEVs with BV-2 microglial cells in real time.⁷¹ Vinduska et al. developed a method to detect surface protein marker of exosomes by coupling with QDs and their enrichment through immunomagnetic beads. The exosomes were isolated through magnetic beads based on expression of CD81 protein. The protein of interest was identified through primary antibody and simultaneous detection with QDs labeled secondary antibody with fluorescence spectrometry. The findings were confirmed through detection of exosomal surface proteins in cancer cells and normal cells.⁷² All together, these advances will help study localization of EVs, their movement inside the cells as well as their functional characteristics, including drug delivery, therapeutic agents and role as disease biomarkers.

4.3. Genetic labeling

Researchers can track and analyze the EVs and its cargo simultaneously by engineering cells to produce EVs with two distinct reporter molecules, providing a more thorough understanding of EV biology and function than they could with a single reporter. This technique is known as a dual-reporter system for EVs.⁷³ Dual-reporter systems are a novel platform in EV biology research, enabling simultaneous use of bioluminescent and fluorescent signals to monitor EV secretion, distribution, and uptake with higher sensitivity and spatial resolution. A dual-reporter system provides a robust method to track EV life cycle in mammalian cells from biogenesis and excretion to cellular uptake. Olson et al. developed a new dual-reporter platform that utilized Gaussia luciferase (gLuc) and green fluorescent protein (GFP) or red fluorescent protein (RFP) to enable pathway-specific EV tracking. They employed a variety of signal peptides to selectively direct reporters through either the ER–Golgi pathway or the exosome-mediated pathway.⁷⁴ Similarly, Sánchez et al. engineered EVs with tricistronic plasmid using dual fluorescence luminal proteins for long term expression in HEK293SF cells, that release bifluorescently labeled EVs. The EVs were characterized for their normal morphology and biogenesis and quantified through fluorescence cytometry.⁷⁵ These results illustrate that dual-reporter systems not only allow real-time and pathway-specific EV tracking, but also provide tremendous tools for investigating the molecular regulation of EV biogenesis and release under a range of cellular conditions.

4.4. Radiolabeling

The most popular technique for radiolabeling EVs is surface or membrane radiolabeling. This entails either directly incorporating the radionuclide into the lipid membrane or attaching it to the membrane proteins through the establishment of covalent chemical bonds.⁷⁶ Radiolabeling of EVs is an effective approach for in vivo analysis. The positron emission tomography (PET) and single-photon emission computed tomography (SPECT) are two common nuclear imaging approaches for this purpose. Two general EV labeling approaches are membrane radiolabeling and intraluminal labeling. Membrane radiolabeling, also known as surface radiolabeling, involves covalent modification of radionuclides to the membrane proteins of EVs through dual functional chelators, including NOTA, DOTA and DTPA, whereas in intraluminal labeling, lipophilic complexes, including ^99mTc-HMPAO and ^111In-oxine, entrap the radionuclides inside the lumen of EVs.⁷⁷ Faruqu et al. radiolabeled melanoma cell-derived exosomes through intraluminal and surface labeling approaches by entrapment of 111indium into EV lumen with tropolone delivery method and by chelating 111indium with covalent labeling of DPTA-anydride, a dual functional chelator.⁷⁸ Patel et al. used 89zirconium desferrioxamine to label the membrane of EVs. Different in vivo and ex vivo methods were combined to track labeled EVs in animals. Autoradiography, immunohistochemistry and PET were used for cellular and anatomical distribution of EVs.⁷⁹ Several radionuclides which are being used in clinical research are licensed, and these radiolabeling approaches have promising potential for conducting EV research from clinical diagnosis to therapeutics.

5. Methods for imaging of EVs

5.1. Electron microscopy (EM)

EM is regarded as a standard imaging method for ultrastructure analysis of nanosized structures. With resolution around 0.5 nm, EM may offer detailed structural information of vesicles as this limit is smaller than exosomes. Before analyzing samples with EM, they need to be fixed and processed because EM cannot capture EVs in their native confirmation. However, staining and dehydration may change native membrane structure, preventing direct interpretation of mechanical properties and hydration state. Among EM, scanning electron microscope (SEM) and transmission electron microscope (TEM) are two main types that are commonly used to analyze the detail structure of EVs.⁸⁰ The morphological and surface features analysis of EVs by SEM is used to measure the size. It can offer valuable information about three-dimensional structure of EVs, including size and particle aggregation. Currently, TEM is the most popular technique used for analysis of EVs.⁸¹ Both TEM and SEM can be used for structural analysis of EVs, but sample preparation is a major limitation. Analysis of EVs through TEM reveals structural damage with standard isolation methods, including ultracentrifugation, as this technique is reported to alter their structure and cause aggregation of particles that results in membrane deformation.⁸² However, milder separation techniques, including polymer-based precipitation, often preserve the integrity of EVs. TEM is also employed to distinguish EVs from other non-vesicular particles. For instance, Priglinger et al. used TEM to visually differentiate between original vesicles, with characteristic lipid bilayers and contaminants, which are devoid of clear membrane structures. Moreover, the use of immunogold labeling is performed during TEM to identify CD9, CD63 and flotillin-1, which are surface markers. The verification of the origin and the subtypes of EVs is easy with this method.⁸³ Likewise, Leng et al. used TEM to study exosomes of arthritic patients. An excellent dispersion of exosomes was observed with quasi-circular or cup-shaped structures having dual membrane.⁸⁴

In comparison to TEM, SEM is used to study samples sequentially through fine point beam of electrons instead of broad electron beam. The focus of SEM analysis is surface of samples to give detailed three-dimensional images of EVs in contrast to TEM that mostly provides two-dimensional images.⁸⁵ Elsner et al. analyzed morphogenetic changes during formation and release of EVs from aortic endothelial (AoEnd) and myocardial endothelial (MyEnd) cell lines using different techniques, including SEM and TEM, under inflammatory and starvation conditions. The outcomes offer novel insights into ultrastructural and morphological changes during biogenesis and release of exosomes and MVs by endothelial cells.⁸⁶ Together, these studies highlight the importance of TEM and SEM in the research and visualization of EVs, however, the deficiency of standardized protocols remains a challenge.⁸⁷

5.2. Cryo-EM

Cryo-EM is a brilliant method of identifying and characterizing various EV types found in body fluids. Cryo-EM is a TEM variation that enables samples to maintain their normal aqueous state in contrast to that of air-dried samples. The method has some advantages and enables analyzing single EV. Cryo-TEM is the best technique to study the structure of EVs, because it allows analyzing and characterizing morphologically peripheral coronal layer which plays significant role in EV functions.⁸⁸ Cryo-TEM is the technique of choice to image biomolecules present in biological fluid without artifacts generated by drying. Prevention of the destruction of electron beam and apt to capture membrane structure and lumen of vesicles are also employed.⁸⁹ Emelyanov et al. characterized EVs derived from cerebrospinal fluid (CSF) of Parkinson's disease patients with Cryo-EM. Cryo-EM enables visualization of large spectrum of EVs with different size and morphological characteristics of lipid membrane and internal structures of vesicles. The authors explored diversity in structures of CSF-derived vesicles, highlighting their different functions.⁹⁰ The native conformation of EVs from human CSF in post-traumatic brain injury was analyzed by Kurtjak et al. Vesicles were obtained through SEC and analyzed by AFM and cryo-TEM. Different structural variations of vesicles were found with a size range of 60–90 nm, and multi-membrane EVs were also reported, in addition to single and double membranes.⁹¹ Morandi et al. explained the formation of fusion intermediates between endosome-like liposomes and EVs. The authors captured those intermediates and explained the fusion mechanism which is pH- and protein-dependent, with a reversible nature, and is distinct from viral fusion.⁹² In addition, Parra et al. used cryo-TEM to observe the EVs from seminal plasma and observed their size and morphological diversity based on their origin.⁹³ Cryo-EM will retain EVs in a near-native hydrated condition, enabling visualization of the thickness, shape, and the inner structure of the membrane. These characteristics give qualitative data on membrane integrity and stability, but quantification of membrane mechanics is difficult.

5.3. Fluorescent microscopy

Fluorescent microscopy is mostly used for imaging and visualization of EVs. The interaction of EVs with other cells can also be studied with fluorescent microscopy, as EVs are involved in transportation of molecules between different cells.⁹⁴ Bos et al. studied the secretion and absorption of EVs in E. coli treated with membrane-active antimicrobial peptide, polymyxin B. The authors revealed the dual role of EVs in antibiotic sequestration and membrane repair. The fluorescently labeled EVs were tracked in real time under a high-resolution fluorescent microscope. The real-time monitoring of EVs showed their role in restoration of cell viability and reducing antibiotic stress.⁹⁵ Wallucks and his colleagues reported a label-free, high throughput platform by integrated iSCAT (label-free interferometric scattering) and fluorescent microscopy for rapid analysis of EVs. With dual illumination scheme, the sizes of EVs were found to be from 35 nm to 200 nm. The method provides concurrent analysis of physical and biochemical characteristics of EVs.⁹⁶ Label-free approaches, such as two-photon FLIM (Fluorescence lifetime imaging microscopy), have also appeared promising. Sorrells et al. found that EV NAD(P)H FLIM indicated their metabolic status and were modulable by parent cell conditions, offering a functional, dye-free readout.⁹⁷ These studies contribute to EVs research and their clinical applications by combining fluorescence microscopy with single-EV analysis and machine learning.

5.3.1. Bioluminescent imaging (BLI). The specific reaction between luciferase chemical and its substrate produces bioluminescence. Unlike fluorescence imaging, BLI is not impacted by the auto-luminescence of mammalian tissues because it does not require external stimulation. Due to its extremely high signal-to-noise ratio, bioluminescence imaging can be employed for both real-time monitoring and in vivo measurement.⁹⁸ In order to generate more stable signals and brighter image than traditional luciferases, Gupta et al. developed NanoLuc-CD63 and ThermoLuc-CD63 constructs which allowed efficient biodistribution of EVs in liver and lungs.⁹⁹ For monitoring EVs secreted from a single cell, Liao et al. constructed pH-sensitive luciferase probe with reduced toxicity. Live cell imaging showed trafficking of acidic EVs with pH ∼ 6.5, through microtubules filaments. The quantification of EVs showed increased secretion of EVs under hypoxic conditions and BLI intensity was found to be directly related to CD81+ exosome secretion.¹⁰⁰ Similarly, to track EVs of different sizes in the mice, Perez et al. developed PalmReNL, a Bioluminescence Resonance Energy Transfer (BRET) EV reporter. After intravenous injection of EVs, BRET-based imaging showed that small EVs were confined to the lungs and those with larger sizes were distributed to the liver.¹⁰¹ Collectively, these studies support the integration of BLI with single EV analysis and put forward EV research to non-invasive diagnosis and targeted therapies.

5.3.2. Confocal microscopy. A specialized form of fluorescence microscopy is confocal microscopy. Confocal microscopy can generate sharp images of the section plane of focus lacking baseline fluorescence when compared to other microscopy options. In confocal microscopy, laser light is directed on thin layer of specimen in one optical plane. The image scanning is performed by a point-to-point flow over the specimen to create a perfect image. The emitted fluorescence goes back through the mirror and lens where it is detected through photomultiplier tube. The unwanted and out of focus lights that enter in the pinhole are not picked and rejected by the detector.¹⁰² For imaging of fluorescent samples, confocal microscopy has become an important technique. The major benefit of confocal microscopy is its ability to construct 3D structure of specimen that can be reconstructed with the help of deconvolution software by scanning several planes. The recent technological advancements led to different types of confocal microscopes, with confocal laser scanning microscopy (CLSM) and laser confocal scanning microscopy (LCSM) being the most common. Uptake of EVs is important in the cells as EVs play a role in intercellular communication. EVs can be tracked inside the cell by labeling through fluorescent molecules.¹⁰³ Kim et al. proposed a unique assay by using three-dimensional florescence confocal microscope. EVs were captured by 3D confocal microscopy after their processing through microfluid-based nanofiltration device. Sake and his colleagues used confocal microscopy to study the role of a small molecule 634, that causes induction of EVs release having immunostimulatory potency via induction of Ca²⁺ influx. This is an innovative approach for EV-based vaccine development and immune studies.¹⁰⁴ Confocal microscopy is popular to observe labeled EVs and assess their localization and uptake by cells. Although the probes are mostly qualitative, the fluidity of the membrane and their interactions with the surface can be reported indirectly by use of environment-sensitive probes.

5.3.3. Super resolution microscopy (SRM). SRM is also an excellent approach for analysis of EVs. Different types of SRM were established, including stimulated emission depletion (STED), DNA-PAINT, photoactivated localization microscopy (PALM) and direct stochastic optical reconstruction microscopy (dSTORM). McNamara et al. used dSTORM for three-dimensional visualization of EVs and tracking a cluster of molecules, including tetraspanins CD9 and CD81 in vesicle surface of single EVs. The study validates the presence of membrane microdomain on EV surface and offers insights into EV heterogeneity, complexity and structure.¹⁰⁵ Similarly, Dechantsreiter et al. used SRM to compare EVs derived from human monocyte-derived macrophages (MDMs). MDMs were induced to become different cells (M0-, M- and M2- like cells) releasing EVs at the same intensity. SRM of EVs secreted from a single cell type showed the presence of HLA-DR and class II MHC proteins on 40% of EVs derived from M1-like MDMs, that was 2-fold than EVs derived from M0 and M2 like MDMs. Interestingly, the EVs derived from macrophages of lung cancer patients also showed expression of HLA-DR that is double than the amount of EVs secreted by M1. The quantification and profiling of single cell-derived EVs in all four types of macrophages showed distinct secretion types, with five of them being constant in multiple samples. A M1-like sub-population of MDMs also secrete EVs that are similar to lung macrophages, indicating growth or recruitment of cells with particular EV profile in lung cancer patients.¹⁰⁶ The quantification of EV diversity is also useful for profiling single EVs to uncover novel macrophage biology. Super-resolution methods provide resolution of nanoscale vesicle membrane and surface protein organization by overcoming the diffraction limit. These methods give details of molecular clustering and membrane heterogeneity, but do not directly measure mechanical or viscoelastic characteristics.

5.4. Atomic force microscopy (AFM)

AFM can be used to scan the vesicle surface topography on a nanoscale and at the same time measure mechanical properties using force-distance measurements. The method is able to deduce the stiffness, elasticity, and viscoelastic behavior of membranes, which is why it is especially useful in the study of EV nanomechanics. AFM utilizes a probe that is usually made of silicon or silicon nitride to scan the specimen's surface.¹⁰⁷ Upon contact of the specimen with the probe, the probe changes its position that can be measured through a laser beam. During scanning, position of the probe is recorded that allows generation of images through AFM. With resolution limit of around 1 nm, AFM enables imaging and quantification of EVs. For air mode of AFM, preparation of samples for EV imaging only requires immobilization of EVs on freshly cleaved mica, followed by scanning through the probe. In liquid-mode that can detect large sized vesicles as compared to air mode, EVs can be measured directly as they maintain their morphology and remains hydrated. The mica can be labeled with antibodies to capture EVs with specific antigen.¹⁰⁸ Chelnokova et al. have for the first time used the nanomechanical mapping mode of AFM to study the mechanical and structural properties of dispersed system of blood NPs after irradiations with X-rays. The authors analyzed the mechanical and structural properties (nonspecific adhesion force and elastic modulus) of lipoprotein isolates and exosomes from blood of Wistar rats that feed with high-fat diet after X-ray irradiation.¹⁰⁹ Sandira et al. studied the nanotopology of EVs derived from HEK293T and discussed their origin in the cells at the individual level with the help of high-speed AFM and exosomal markers (IgGCD81 and IgGCD63). Significant fluctuations in height were observed in larger EVs with a diameter greater than 100 nm. Also, IgG CD63 from mice and IgG control and IgGCD81 from rabbits displayed a “Y” shape structure. Moreover, the colocalization of exosomal marker antibodies was demonstrated with EVs, which have a diameter of up to 100 nm, not exceeding 100 nm, indicating CD63-CD81-enriched and depleted EVs.¹⁰⁸

5.5. Magnetic resonance imaging (MRI)

MRI has emerged as a crucial non-invasive technique for in vivo tracking of EVs with the help of magnetic nanoparticle-based contrast agents. A study describes that SPIOs or USPIOs (ultrasmall superparamagnetic iron oxide nanoparticles) may be loaded into EVs either through passive incubation or electroporation, with the latter facilitating much higher labeling efficiency of around 96% compared to 19% for incubation. These EVs are successfully traced in rodent models, especially piling up in damaged tissues. Nevertheless, they state that even though MRI possesses excellent spatial resolution and safety due to the non-ionizing nature of application, its usefulness is limited by relatively poor sensitivity, high cost, and incompatibility with metal implants.¹¹⁰ To overcome the MRI sensitivity limitations, Sancho-Albero et al. introduced a novel labeling method by creating superfluorinated EVs containing a perfluorocarbon molecule called PERFECTA that consisted of 36 equivalent 19F atoms. These vesicles, which were produced by incubating MSCs with a PERFECTA emulsion, retained their native morphology and markers and encapsulated an average of 1.95 × 10⁸ 19F atoms per EV.¹¹¹ In addition, Kim et al. evaluated the ability of mesenchymal stem cell-derived EVs for neural plasticity and restoration of motor function in marmosets stroke model. The results showed improved anatomical connectivity and decreased infarct volume by MCS-derived EVs, exhibiting their potential in clinical application in stroke therapy.¹¹²

5.6. Computed tomography (CT)/X-ray imaging

Visualization and distribution of EVs can be tracked using CT imaging. This technique involves labeling EVs with contrast agents and then using X-ray imaging, primarily with a CT scanner, to track their movement. However, because EVs are small, this method often requires high-contrast labeling and is typically used in conjunction with other imaging modalities to enhance sensitivity and facilitate detailed analysis. X-ray imaging and CT are extremely useful techniques for the identification of microstructural changes in tissue, but their use in direct visualization of EVs was not fully understood.¹¹³ Recent work in grating-based dark-field imaging showed enhanced sensitivity for contrast to discern minor tissue changes. In future, such advancements may allow indirect measurements of EV associated pathological alterations, especially where EVs participate in tissue remodeling or the onset stages of disease development. Photon-counting CT offers excellent dose efficiency and spectral sensitivity, enhancing visualization of tiny EVs with low noise.¹¹⁴ Lara et al. developed EVs carrying plasmonic gold-nanoparticles for CT-targeted delivery to the metastatic lung nodules of mice. Imaging with CT showed localization of EVs in tumor cells, and microscopic examination of tissue also confirmed localization of EVs in alveoli and tumor vasculature.¹¹⁵ AI-based functionalities in cardiac CT enhance value to functional and structural assessments, which can be adapted for EV subtype characterization and tissue remodeling analysis.¹¹⁶ Moreover, improvements in CT detector technology and image reconstruction provide increased spatial resolution and fewer artifacts, allowing for extensive EV morphological analysis.¹¹⁷ Together, these innovations provide accurate and non-invasive EV imaging, expanding diagnostic and therapeutic uses.

Overall, EV imaging techniques vary in terms of probing single vesicles and populations (Fig. 3). Morphology, membrane integrity, heterogeneity, and other properties can be used with methods like AFM, TEM, cryo-EM, and super-resolution microscopy, which have a single-vesicle resolution. By comparison, confocal fluorescence microscopy, BLI, X-rays and MRI typically measures the overall behavior of the vesicles, giving averaged data of vesicle distribution, localization, and uptake in a sample, but with an achievable low level of individual vesicle characteristics. It is worth mentioning that most methods of imaging use surface fixation of EVs, which may create vesicle-surface interactions as a significant physical parameter. The effect of adsorption to solid substrates can include the deformation or rupture or fusion of the vesicles, which may change the apparent morphology and mechanical properties and affect the interpretation of the data.

Fig. 3 Methods Used for Imaging of EVs. Different imaging techniques have different resolution limits. Based on the EV population, different approaches can be applied for imaging of EVs.

6. Techniques for characterization of EVs

6.1. Nanoparticle tracking analysis (NTA)

EVs can be characterized through NTA for quantification of their concentration, size and distribution (Fig. 4). NTA is a quick and sensitive technique that illuminates EVs in a solution with a laser beam and then tracks the movement of the dispersed light.^118,119 To evaluate the different isolation methods, secretory and storage conditions of EVs, the light-scattering method of NTA provides accurate measurement of particle size and distribution for identification of diseases in biological fluids.¹²⁰ Mladenović et al. compared nFCM (NanoAnalyzer) and NTA (ZetaView) for identifying SOPs and standard reference for fluorescent characterization of EVs. Quantitative fluorescence analysis was conducted by specifying ERF (equivalent reference fluorophore) through yellow-green FluoSpheres (FS). The standardizing procedure enables expression of fluorescent signals in EV to be displayed in terms of ERF units, allowing target protein measurement on single EV and whole population. The study provides absolute quantification of markers and detailed analysis of EV diversity and marker ratio. The standardized protocol uncovers the evaluation potential of nFCM and NTA and allows inter-platform data comparison.¹²¹ Interferometric nanoparticle tracking analysis (iNTA) is another development for particle analysis method that allows measurement of size and refractive index of EVs with enhanced sensitivity and accuracy. Kashkanova et al. used iNTA to distinguish between EVs and lipoproteins, and benchmark its performance in relation to conventional NTA. The results showed accurate quantification of EVs derived from plasma and in artificial-lipoprotein samples with different complexities. The conventional NTA was unable to provide EV numbers and cannot differentiate between lipoproteins and EVs, therefore iNTA can be a possible label-free characterization method of EVs.¹²² NTA quantifies the Brownian motion of discrete EVs, which gives the hydrodynamic size and the concentration of the particles. In addition to size distribution, diffusion can also provide indirect measurements of vesicle deformability and aggregation, but not membrane mechanics and surface interactions.

Fig. 4 Methods for characterization of EVs. These methods are based on morphology and biochemical composition of EVs. The common biochemical markers for EVs include surface biomarkers.

6.2. Dynamic light scattering (DLS)

DLS measures size of particles based on their Brownian motion in solution; the basis of Brownian motion is that lighter particles diffuse faster, and speed is relative to particle size. DLS measures the changes in scattered light, which are used to measure average hydrodynamic diameter and polydispersity. This method is used specifically for measurement of size distribution of EVs and their zeta potential as well.¹²³ Barranco et al. established the validity of DLS as a powerful tool for measurement of EV diversity and size distribution. They isolated EVs from porcine semen plasma through serial ultracentrifugation and characterized them by DLS. The DLS analysis found that the EVs derived from seminal plasma showed an asymmetric size distribution. The diameter of EVs varied from 13.55 to 824.99 nm, 77% of the vesicles had diameters within this range, and their size distribution pattern remained consistent among all ejaculates.¹²⁴ Similarly, Dave et al. precisely measured the size and stability through DLS by facilitating their role in increasing ATP level in brain cells. They characterized the naïve and engineered EVs which carry plasmid DNA coding BDNF (brain-derived neurotrophic factor) in different cells.¹²⁵ In another study by Afrisham et al., DSL was found to be important for comparison of EVs isolated from four different individuals through EXOCIB kit method and UC. The EVs were pooled from different samples, followed by measuring their size and zeta potential through DLS. The results showed that UC produced EVs with improved purity and larger hydrodynamic size (135.7 nm) in comparison to the kit which yield EVs with average size of 102.8 nm.⁴⁷ All these findings suggest the potential role of DLS in basic and applied research. Although DLS information mostly gives size data, vesicle heterogeneity and possible aggregation, it cannot determine shape, membrane stiffness, or fine-scale interfacial properties.

6.3. Förster resonance energy transfer (FRET)

FRET is a fast and sensitive approach that has been extensively employed in the field of biomedical and biological sciences. The mechanism of FRET involves energy transfer between two fluorescently labeled molecules (donor and acceptor) that are in close proximity. For successful applications of FRET, two important conditions must be fulfilled. Firstly, the emission spectra of donor molecule must overlap with absorption spectra of receptor molecule, and secondly the distance between recipient and donor molecule should be within the range of 1–10 nm. So far, most of the FRET-based studies are directed on detection of miRNAs in EVs. During analysis of EVs proteins, prolonged isolation and washing steps are required, which significantly increases the duration of diagnostic procedures. The FRET analysis based on co-localized protein signals and EV membrane does not require an isolation step.¹²⁶ Ishikawa et al. engineered hybrid EVs by fusing liposomes with EVs derived from insect cells expressing baculoviral fusogenic glycoprotein gp64 and programmed cell death 1 (PD-1) protein. The hybrid EVs were evaluated for the delivery of model molecules, Texas Red-labeled dextran (TR-Dex) into the cytoplasm. The incubation of hybrid EVs with HeLa cells increased the intracellular influx of hybrid EVs with PD-1 protein in comparison to the EVs lacking PD-1. The PD-1 in the hybrid EVs exhibited co-localization into late endosome after cellular uptake. The FRET analysis tracked the fusion of membrane of organelles with hybrid EVs.¹²⁷ In addition, Fisher et al. used minimal perturbative approach to increase the fusion between synthetic neutral liposomes (NLs) and EVs under acidic conditions and compared it with passive incubation under physiological conditions. The results showed increased fusion of synthetic liposomes with EVs in comparison to non-lipid components.¹²⁸

6.4. Surface plasmon resonance (SPR)

SPR is an optical method that uses SPR biosensor for analysis of EV subpopulation. The method involves immobilization of specific protein receptors on the active surfaces that are either coated with silver or gold NPs allowing label free, instantaneous detection of surface biomarkers present on EVs.¹²⁹ For quantification of molecular interaction on a surface, the detection set up uses quantification mechanism through resonant oscillation of electrons that are produced through incident light at the interface of positive and negative dielectric constant material. By targeting specific surface biomarkers on EVs, oligonucleotide aptamers (20–100 nucleotides) or antibodies are anchored to the gold NPs and employed for quantification of EVs derived from tumors. For the investigation of EVs, several different functional surfaces have been evaluated using this method, including gold nano-islands, gold nanopillars and gold nanoholes.¹³⁰ SPR is an important method for biochemical analysis of biomarkers with low abundance, and development over the last ten years has resulted in surface enhanced Raman spectroscopy (SERS). In this method, the signal is intensified, allowing improved analysis of single molecule that is attached with silver or gold metal nanostructures including cobalt, nickel, iron with antibody modifications. For multiplexing analysis of EVs with decrease concentration, SERS immunolabeled nanoprobes have also been synthesized recently.¹³¹ For detection and characterization of analytes, the SPR sensors use molecular interactions, like antibody-antigen, protein-small molecules interaction. Target molecules that are attached to the detecting surface bring local changes in optical resonance shift and refractive index of these sensor molecules. Additionally, SPR is a non-destructive method for identification of target molecules that are bound to the ligands on sensor surface. SPR sensors have a limited detection range (10–300 nm). The dimensions of EVs can be effectively surrounded by evanescent field of surface plasmon that helps in synergistic integration of plasmonic sensing in EV study. SPR sensing has multiple advantages compared to radioactive and fluorescent labeling techniques, including label-free and real-time analysis, evaluation of kinetics and affinity, and reduced costs and little reagent consumption.¹³² Jeong et al. presented a novel nanoplasmonic sensing technology for the sensitive detection of individual EVs. The nPLEX-FL (nano-plasmonic EV analysis with enhanced fluorescence detection) technology enhanced fluorescence signals from EVs using periodic gold nanohole structures, facilitating the sensitive and multiplexed investigation of individual EVs.¹³³

6.5. Raman spectroscopy (RS)

RS relies on the basic properties of interaction between electromagnetic radiation and matter. The spectral information from RS provides detailed information about the physical status and chemical composition of the analyte. The method has potential to discriminate between EV subpopulations and EV secreted from different cell types.¹³⁴ O’Toole et al. used RS for rapid and specific detection of EVs derived from plasma samples from burn-associated septic and non-septic patients. The specific glycoconjugates associated with bacterial sepsis can be tracked through differences in spectra of septic and non-septic patients.¹³⁵ Qin and his colleagues integrated RS and attentional neural network for improved classification of EVs from different pathogenic strains of bacteria by achieving 96% accuracy at Gram-staining level and >85% accuracy at the physiological level. The specific spectral peak at 1450 cm⁻¹ linked with lipid asymmetry was found to be main spectral marker for differentiating between EVs derived from antibiotic resistant strains versus sensitive strains.¹³⁶ In another study, RS was used to analyze the EVs isolated from breast cancer patients showing elevated level of nucleic acid and increased lipids profile in comparison to the healthy controls. The RS identified the difference between different types of EVs (large and small), and distinct spectral markers were found to be linked with tumorigenesis.¹³⁷ In addition, Buccini et al. used special types of Raman microscopy (confocal Raman and tip-enhanced Raman spectroscopy, or TERS) to study tiny chemical differences on the surface of individual EVs from cow's milk. TERS allowed them to examine different points on a single EV.¹³⁸ All these studies show the evolution of RS from a specialized analytical method to multidimensional platform for rapid and label-free analysis of EV composition and their functions.

6.6. Quartz crystal microbalance (QCM)

Quartz crystal microbalance (QCM) is a surface sensitive and label-free technology applied to real-time monitor adsorption and interaction dynamics of EVs. QCM can be used to measure resonance frequency and energy dissipation changes and therefore provides information about the mass, viscoelastic properties, and hydration layer of surface-bound vesicles. Notably, the vesicle-surface interactions impact QCM measurements significantly since adsorption, deformation or partial rupture of vesicles can interfere with frequency and dissipation measurements. Although QCM does not provide a resolution of single vesicles, it provides useful ensemble-level information about interfacial coupling and membrane softness, and is therefore a useful method of studying exosome–surface interactions when used with an appropriate physical model.¹³⁹ In a study, Kowalczyk et al. developed quartz crystal microbalance with dissipation (QCM-D)-based immunosensors for measuring the EVs concentration from human lung cancer cells using tetraspanins as biomarkers. The results showed strong correlation with NTA with lowest detection limit of 0.6–1.8 × 10⁴ particles per mL.¹⁴⁰

6.7. Flow cytometry (FCM)

FCM is the most appropriate technique for clinical sample analysis in comparison to EM and NTA. FCM is a powerful approach that allows multidimensional assessment of thousands of particles within seconds. Therefore, it is an effective technique for measurement, categorization and purification of particles from suspension. However, due to baseline noise and overlapping of particle light scattering, the standard FCM cannot classify a large number of particles, especially exosomes. To overcome these problems, top notch flow cytometers have been developed recently with improved sensitivity, high resolution images, fluorescence amplification and forward scatter detection. The two most commonly used approaches for FCM are Nanoflow cytometry and imaging flow cytometry.¹⁴¹

6.7.1. Nanoflow cytometry. Nanoflow cytometry, also known as nano-FCM or nFCM, is a specialized version of flow cytometry particularly designed to study biological nanoparticles that include but not limited to EVs, exosomes and viruses, etc.¹⁴² Lau et al. performed the proteomic and particle analysis of vesicles derived from HEK293 cells and from several batches of cell-derived vesicles (CDVs) with nanoflow cytometry. The analysis validated the presence of nicastrin (NCSTN), lysosome-associated membrane glycoprotein 1 (LAMP1) and CD63 in the CDVs, while in EVs along with CD9 and CD81, the prostaglandin F2 receptor negative regulator (PTGFRN) was dominating.¹⁴³ Similarly, Halvaei et al. identified a compound that affects release of EV. A high-throughput nanoscale flow cytometry was used to test 156 kinase inhibitors. Fluorescently labeled EVs with ZsGreen were examined for their size, and expression markers (CD63) were examined.¹⁴⁴ Huynh et al. reported the expression of CD9/CD63/CD81 in hypoxia-induced adipose-derived stem cell (ADSC) exosomes through flow cytometry, and the identity and purity of the exosomes.¹⁴⁵ In addition, Kobayashi et al. found that the detection of EVs markers was affected by the methods used for the isolation of EVs. They observed that ultracentrifugation and density gradients caused false positives in CD9 labeling using CD9 knockout EVs.¹⁴⁶

6.7.2. Imaging flow cytometry (IFCM). An integration of high-throughput flow cytometry with imaging capabilities of microscopy is IFCM. IFCM delivers a promising solution for high-throughput cell analysis with high precision in different fields, including environmental monitoring, green energy and biomedicines. But due to limitations in real-time data processing and imaging framerate, the existing IFC systems with real-time throughput have a restriction of 100–10 000 events per second, which is considered insufficient for cell analysis at a large scale.¹⁴⁷ Woud et al. provided an IFCM-based strategy to deal with artifacts produced during isolation of EVs though different methods. The method is also useful for determination of EVs numbers in a sample having a diameter of 400 nm and also for their phenotyping. The authors found single EVs events with >90% of the double positive fluorescence by using different combinations of markers (CD31, CD9, CFSE and Tetraspanins).¹⁴⁸ Tertel et al. analyzed the surface markers (CD81, CD63 and CD9) in EVs derived from mesenchymal stromal cells (MSC-EVs) by IFCM. Such an approach allowed for monitoring in real-time half-populations of EV in conditioned media without purifying them further to facilitate batch-to-batch therapeutic production consistency of EV. Based on IFCM, EV-specific subpopulations linked to the quality of human platelet lysate were found, which represented a paradigm towards standardized MSC-EV characterization.¹⁴⁹ The studies entail basic framework that allows application of IFCM for EV research in clinical diagnosis.

6.8. Mass spectrometry (MS)

MS is the most versatile and popular analytical approach to studying molecular contents of EVs. It is usually coupled with liquid chromatography (ultra-high performance or nanoscale)- electrospray ionization tandem mass spectrometry (LC/ESI-MS/MS) and help researchers with the analysis of EVs. Particularly, the nano-ESI-MS/MS offers high resolution and sensitivity, enabling the detection, quantification, characterization and identification of thousands of protein molecules even in a single EV sample. To elucidate the functional and structural architecture of EVs, MS-based technological platforms are gaining attention as fundamental tools. The fragmentation ions from negative and positive ESI tandem MS analysis in several biological samples give information about composition and unambiguous structural characteristics of proteins, helping with their identification. The understanding of molecular cargo especially proteins of EV has increased recently, due to increased sensitivity and requirement of small volume for MS, MS-based proteomic analysis. Several researchers have used ESI tandem MS analysis in conjunction with chromatographic methods (nano-LC, UPLC, UHPLC and HPLC) for profiling and structural characterization of proteins in various biological samples, biofluids, tissues and cancer cells.¹⁵⁰

6.9. Omics analysis of EVs

The term “omics analysis of EVs” describes in-depth analysis of the complete molecular makeup of EVs using high-throughput techniques to explore all molecular components that exist in them, such as proteins, lipids, metabolites, DNA and RNA, to offer a complete understanding of their biological functions and potential as disease diagnosis biomarkers.

6.9.1. Genomic and transcriptomic profiling. Genomic and transcriptomic profiling of EVs involves studying the nucleic acid content within EVs, respectively, to decode the genetic information which can provide insights into the cell of origin, disease condition, and potential biomarkers through examining the unique molecular profiles of these vesicles in body fluids, like urine or blood, essentially acting as a “liquid biopsy”, to study a patient's disease state without requiring tissue sample.¹⁵¹ Lázaro-Ibáñez et al. performed whole-genome sequencing (WGS) to identify DNA fragments present across all chromosomes and mitochondrial DNA in sEV subpopulations of different cells. Paired analysis revealed a higher abundance of RNA molecules compared to DNA, with DNA predominantly found in the high-density fractions, present on the outer surface of the vesicles, and both genomic and mitochondrial DNA are present in sEVs, exhibiting distinct DNA cargo.¹⁵² Analogously, the copy number variation produced through low coverage WGS of DNA associated with EVs was compared with DNA originated from patients with metastatic cancer.¹⁵³

6.9.2. RNA sequencing. RNA sequencing, also known as RNA-seq or cell sequencing, is a method for characterization and analysis of the RNA. The method provides detailed analysis of the EVs and their origin inside the cells, and also help researchers analyze their putative functions based on RNA molecules.^154,155 Giannopoulos-Dimitriou et al. performed high-throughput miRNA and proteomic analysis integrated with bioinformatic network analysis to examine molecular cargo of MRC-5 cells-derived exosomes. A total of 68 proteins were identified that are primarily involved in metabolic processes and organization of extracellular matrix. The miRNA sequencing analysis showed 72 types of different miRNAs, significantly involved in regulation of gene expression, metabolic processes and tumorigenic pathways.¹⁵⁶ Rozke et al. performed a high throughput sequencing analysis to compare the differential expression of snoRNAs of exosomes derived from non-infected and infected MDCK cells with influenza virus. The findings showed that the snoRNAs had a role in potentiating the viral activity. A total of 133 molecules of snoRNAs were found to be differentially regulated (with 131 overexpressed and 2 reduced expression), 2 were SCARNA, 38 were SNORA, and 93 were SNORD. Multiple snoRNA molecules previously associated with viral infection showed increased expression, including SNOD44, SNORD58, SNORD29, SNORD28 and SNORD27. Overall, 533 interactors associated with dysregulated snoRNAs were identified through the RNAinter resource with an allocated confidence value of ≥0.25. The primary clusters of predicted interactors were RNA-binding proteins and transcription factors.¹⁵⁷ In addition, Supradit et al. demonstrated the use of NGS technology in the discovery of exosomal miRNAs, particularly miR-192-5p and miR-194-5p, both of which showed increased expression in O. viverrini-infected group, suggesting their potential role as biomarkers for screening.¹⁵⁸

6.9.3. Lipidomic analysis. Lipidomic analyses of EVs include their complete lipid profile. Lipids from EVs also serve as disease biomarkers and are integral part of EVs. They are also responsible for structural and functional properties of EVs. EV synthesis and release into the extracellular environment depend on lipids. Faria et al. studied lipid composition of Giardia lamblia trophozoites, exosomes and microvesicles through LC-MS/MS. A total of 162 lipid species were characterized and classified into 8 lipid classes. The major classified lipids classes include phospholipids, such as phosphatidylinositol, phosphatidylglycerol, phosphatidylethanolamine, phosphatidylcholine and cardiolipins, sphingolipid classes ceramides, sphingomyelin and cholesterol, and 3 lipids subclasses include lyso PG, lyso PE and lyso PC.¹⁵⁹ Similarly, Zhu et al. investigated the NAFLD (Non-alcoholic fatty liver disease) biomarkers using information from urinary EVs through LC-MS/MS. They comprehensively compared them with the lipidomic profiles of EVs derived from NASH and NAFLD, which hold potential for advancing existing knowledge of disease progression and investigating their impact as a non-invasive approach for NASH diagnosis.¹⁶⁰ In addition, Krokidis and his colleagues performed targeted analysis of lipids in EVs derived from plasma of patients with Alzheimer disease and compared them with healthy controls. The authors compared selected lipids species, including sphingolipids, lysophospholipids, glycerophospholipids and glycerolipids in EVs derived from different sources. The study provides novel insights into lipid profile of EVs in the patients and explains the role of lipid biomarkers as early diagnosis and therapeutic targets.¹⁶¹

6.9.4. Proteomic analysis of EVs. The proteomic profile of EVs can be characterized efficiently and comprehensively through mass spectrometry-based protein analysis. The analysis of proteins within EVs is an innovative approach, showing great interest in EV research. Extensive investigations on proteins in EVs have elucidated their diverse functions over the past three decades. Mass spectrometric data helped researchers create online databases, such as Vesiclepedia (https://www.microvesicles.org) and EXoCarta (https://www.exocarta.org) for a list of proteins that exist in EVs. A variety of biochemical techniques have been employed to analyze low-level membrane proteins in EVs using MS-based analysis. Numerous protein interaction networks and protein datasets have elucidated a significant connection between EV proteins, thereby enhancing our understanding of vesicle biogenesis and pathological functions. For analysis of surface proteins of EVs in HMC-1 mast cell line and pancreatic cancer cells, non-membrane-permeable reagents, including SS-biotin and sulfo-NHS, have been used for chemical derivatization.¹⁶² Different studies published on EVs proteins from different sources infer the presence of a regulated protein-sorting mechanism that can be categorized through encapsulation of EV proteins in a random way, sharing common vesicular proteins.¹⁶³

The ESI fragment from positive and negative mode of tandem mass spectrometry experiments provides important information from different EV samples, which can be used to elucidate the chemical composition and structural characterization of proteins. The technique is extremely sensitive and only a small amount of sample is required. The proteomic profiling based on MS facilitated the better understanding of protein cargo in EVs. Nowadays, chromatographic techniques integrated with ESI tandem MS, such as HPLC-MS, UHPLC-MS and nano-LC-MS/MS are routine method of protein characterization for EVs derived from samples of disease and healthy patients.¹⁶⁴ Similarly, another method was developed for surface protein identification and validation from prostate-derived exosomes. The method was developed by combining proximity ligation assay with high-resolution mass spectrometry. The author designed a workflow for investigation of surface proteins of SF-sEV (small EVs derived from seminal fluid) and identified 1014 surface proteins and validated their presence on surface of SF-sEVs.¹⁶² Likewise, Mizgier et al. compared the proteomic profile of EVs derived from different samples. The EVs samples were isolated from plasma samples and gingival crevicular fluid of healthy pregnant females as well as pregnant females with gestational diabetes mellitus. Characterization of the EVs was performed through MS, electron microscopy, immunoassay and nanoparticle tracking. The findings showed that periodontitis and GDM pregnancies, along with attachment loss, probing depth, and severe bleeding, had an increased number of total, small and large GCF-EVs and reduced expression of CD81, CD9, and CD81/CD63 ratio. No significant variation was observed in protein expression and plasma EV concentration. Proteomic profile of GCF-EVs revealed the presence of both bacterial and host peptides. Gene ontology analysis revealed that proteins from GCF-EVs influence insulin response mechanism, glucose metabolism and immune-inflammatory responses.¹⁶⁵ Aparicio et al. analyzed the proteomic profile of EVs in sarcopenia and robust controls to identify differential protein expression that can help with disease diagnosis. EVs from plasma were isolated, and their proteomic cargo was analyzed. Proteins with different concentrations in sEVs between robust controls and sarcopenic patients were further confirmed through ELISA. Proteomic analysis revealed 157 proteins present in robust control and sarcopenic patients, of which 48 were new entries in the Vesiclepedia and ExoCarta databases. The findings support the identification of complement C1r subcomponent and platelet factor 4 as plasma diagnostic biomarkers in sarcopenic patients, paving the way for exploring their roles in disease pathogenesis.¹⁶⁶

6.9.5. Isobaric tags for relative and absolute quantification (iTRAQ). The iTRAQ is another method that facilitates multiplexed quantification of proteins in EVs through labelling with isobaric tags, allowing comparison of relative quantification in one experiment. Wu et al. used iTRAQ technology for quantitative proteomic analysis to examine exosomes derived from the synovial fluid of arthritic patients and compared them with those from healthy controls. The findings revealed 439 proteins, of which twenty were overexpressed in arthritis patients and five showed decreased expression. The bioinformatic analysis showed implications of these proteins in several different processes, such as immunological processes, activation of compliment proteins and antigen binding. The authors also compared their dataset with PXD023708, a publicly available data set on synovial fluid and found overlapping of 5 differentially expressed proteins (DEPs) in both data sets, and protein–protein interaction showed APOM, C4B and C3 as main components of tightly interactive network. The study identified 5 DEPs, including DPYSL2, MMP3, APOM, C4B and C3, as potential biomarkers of of OA. Another approach for analysis of EV proteins is tandem mass tag (TMT) that also provides quantitative expression of different proteins in samples.^167,168

6.9.6. Shotgun proteomics. Shotgun proteomics is an inductive approach for detailed analysis of proteins present in a sample. The procedure includes breakdown of proteins into peptides. These peptides are then separated through chromatography and then analyzed with mass spectrometry. The LC-MS/MS analysis of EVs proteins gives detailed information about their protein composition.¹⁶⁹ Surman et al. profiled proteome of exosomes derived from thyroid cancer cells and normal cells through LC-MS. The authors identified 1769 different proteins in the cancer cells and most of them were unique in given cells. The proteins related to the cancer progressions were abundantly found in cancer cells-derived exosomes. The label-free quantification approach showed several different proteins in the exosomes of thyroid cancer with increased expression.¹⁷⁰ The label-free quantification provides cost effective approach and is suitable for analysis of a large number of samples. Without any harsh chemical labeling, the technique provides information from signal intensities and spectral count. The method is suitable for protein analysis of EVs under different conditions.¹⁷¹ Together, the above-mentioned techniques offer unique advantages regarding accuracy and sensitivity for detailed analysis of EVs.

6.9.7. Western blotting or immunoblotting. Western blotting, also known as immunoblotting is also a common approach for proteins analysis in EVs research. The analysis is based on surface marker proteins in EVs (CD63, CD40, CD9, CD151 and CD81). These proteins are expressed on the membrane of EVs and are involved in protein internalization, such as Alix, Hsp90, Hsp70 and Tsg101. The technique is extensively used for identification, characterization and also for detection of post-translational modifications.¹⁷² The analysis of biological samples with western blot provides semi-quantitative data about proteins in both simple and complex samples. Firstly, the proteins are separated on the gel through electrophoresis and then transferred to a nitrocellulose membrane. Certain antibodies are added in the samples to bind with proteins. The proteins bands separated on the gel are visualized and categorized as antigen-negative and antigen positive samples in comparison with a standard ladder. Silva et al. engineered Exipi293F cells for expression of EVs, sorting proteins that are tagged with GFP. Increased payload of GFP into EVs was confirmed through western blotting.¹⁷³ Chen et al. studied EVs proteins isolated from embryonic stem cells (ESC-sEVs) and induced pluripotent stem cells (iPSC-sEVs) with LC-MS/MS and validated the results through western blot. However, no standard biomarkers specific to EVs proteins have been established yet. Collectively, different techniques are combined to help with detection and characterization of EVs, including their size, quantification, purity and morphology.¹⁷⁴

6.10. Digital droplet PCR (ddPCR)

Ko et al. used immune-droplet digital PCR (iddPCR) for single EV protein analysis. EVs were labeled with antibody-DNA conjugation and single EVs were incorporated into droplets thorough stochastic microfluidics. The study used in situ PCR, with fluorescently labeled reporter probes that convert and amplify the barcode signals into droplet imaging that can be read directly. The study showed multiplex protein analysis at single EV level as proof of principle study and opened new avenues in EV analysis.¹⁷⁵ Likewise, Liu et al. performed surface protein analysis of individual EVs through triplex droplet digital immune-PCR (ddiPCR). The method is quite versatile and uses multi-subpopulation of EVs. The ddiPCR successfully integrated two technologies with highly specific immune-PCR with droplet digital PCR of superior sensitivity. The method also combined ddiPCR with machine learning for improvement of the method in disease diagnosis.¹⁷⁶

Overall, the majority of common EV characterization techniques are ensemble-average measures of the properties of populations (Fig. 4). For example, NTA and DLS methods can measure the average size, concentration, polydispersity and surface charge of EVs, but do not identify individual vesicles. Other common characterization approaches of EVs use mostly give ensemble-level molecular or biophysical data. FCM has the capability of identifying surface markers of a population of vesicles, though high-resolution instruments can occasionally resolve individual-vesicle events. Protein, nucleic acid or lipid content in all vesicle populations are analyzed by MS, western blotting and PCR giving molecular composition but no physical/mechanical data. SPR and Raman spectroscopy provide ensemble-averaged information about surface interactions, binding kinetics, and molecular organization, and Raman spectroscopy can give limited information concerning membrane composition and hydration.

7. Conclusions

EVs play an important role in intercellular communication and have great potential as biomarkers and as a vehicle of therapeutic delivery. Accurate functional interpretation and clinical translation cannot be performed without reliable isolation, purification, labeling, and characterization as mentioned in this review. Each of the current methodologies has its own strengths and weaknesses as far as purity, yield, scalability, and reproducibility are concerned and integrated approach to methodology, namely application specific methodology is required instead of depending on a single approach. Novel platforms, such as microfluidic systems, provide a better standardization and analytical accuracy, and compliance with minimal information for studies of extracellular vesicles (MISEVs) criteria is required so that the data can be compared. In the future, the combination of artificial intelligence and machine learning in EV analysis will likely allow improving data interpretation, automating the analysis of images and particles, performing single-EV analysis, and optimizing and standardizing methods. Together, the integration of state-of-the-art experimental methods and AI-based analytics will increase the translational potential of EV-based diagnostics and therapeutics.

Conflicts of interest

There are no conflicts to declare.

Data availability

This is a review article. No primary research results, software or code have been included and no new data were generated or analyzed as part of this review.

Acknowledgements

This work was supported by the Natural Science Foundation of Jiangxi (20252BAC240711 and 20242BAB21031) and the Technological Project of Jiangxi Province (20232BCD44003). The illustrations were partially generated using BioRender.

References

1. X. Nie, Q. Li, H. Fatima, Y. Wang, F. Meng and J. Xie, et al., Advances in mammalian, plant, and bacterial-derived extracellular vesicles for cancer therapy, Chem. Eng. J., 2025, 523, 168637.
2. Z. Wang, X. Zhou, Q. Kong, H. He, J. Sun and W. Qiu, et al., Extracellular Vesicle Preparation and Analysis: A State-of-the-Art Review, Ad. Sci., 2024, 2401069.
3. X. Nie, Q. Li, Y. He, Y. Xu, S. Qiao and X. Wang, et al., Engineered bacterial extracellular vesicles for gastrointestinal diseases, J. Controlled Release, 2025, 385, 113972.
4. P. Ahmad, N. Estrin, N. Farshidfar, Y. Zhang and R. J. Miron, Isolation methods of exosomes derived from dental stem cells, Int. J. Oral Sci., 2025, 17(1), 50.
5. J. Xie, L. Van Hoecke, E. Van Wonterghem, G. Van Imschoot, V. Andries and L. Vereecke, et al., The gut-brain axis in Alzheimer's disease is shaped by commensal gut microbiota derived extracellular vesicles, Gut Microbes, 2025, 17(1), 2501193.
6. P. Ahmad, N. Estrin, N. Farshidfar, Y. Zhang and R. J. Miron, Mechanistic insights into periodontal ligament stem cell-derived exosomes in tissue regeneration, Clin. Oral Investig., 2025, 29(7), 357.
7. M. Wen, J. Wang, Z. Ou, G. Nie, Y. Chen and M. Li, et al., Bacterial extracellular vesicles: A position paper by the Microbial Vesicles Task Force of the Chinese Society for Extracellular Vesicles (3/2023), Interdiscip. Med., 2023, 1(3), e12046.
8. Q. Li, X. Chen, J. Xie and S. Nie, Engineered Bacterial Extracellular Besicles: Developments, Challenges, and Opportunities, Engineering, 2025, 291–307.
9. X. Liang, D. Gupta, J. Xie, E. Van Wonterghem, L. Van Hoecke and J. Hean, et al., Engineering of extracellular vesicles for efficient intracellular delivery of multimodal therapeutics including genome editors, Nat. Commun., 2025, 16(1), 4028.
10. X. Li, M. Zhu, L. Xiao, J. Li, K. Yu and Q. Zhang, et al., Extracellular vesicles derived from Schwann cells to enhance bone and dental tissue regeneration: a literature review, J. Nanobiotechnol., 2025, 23(1), 502.
11. Y. Liu, J. Hu and B. Yan, Enrichment of ER tubule-derived microsomes by differential centrifugation and immunoprecipitation, Biophys. Rep., 2024, 10(2), 61.
12. C. Coughlan, K. D. Bruce, O. Burgy, T. D. Boyd, C. R. Michel and J. E. Garcia-Perez, et al., Exosome isolation by ultracentrifugation and precipitation and techniques for downstream analyses, Curr. Protoc. Cell Biol., 2020, 88(1), e110.
13. X. Nie, Q. Li, H. Ji, S. Zhang, Y. Wang and J. Xie, et al., Bifidobacterium longum NSP001-derived extracellular vesicles ameliorate ulcerative colitis by modulating T cell responses in gut microbiota-(in) dependent manners, npj Biofilms Microbiomes, 2025, 11(1), 27.
14. S. Wang, Z. Ding, S. Wang, B. Hu, E. Wu and C. Xia, et al., Approaching effective differential centrifugal fractionation by combining image analysis with analytical ultracentrifugation, Langmuir, 2024, 40(4), 2191–2197.
15. Y. Tian, M. Gong, Y. Hu, H. Liu, W. Zhang and M. Zhang, et al., Quality and efficiency assessment of six extracellular vesicle isolation methods by nano-flow cytometry, J. Extracell. Vesicles, 2020, 9(1), 1697028.
16. A. Teixeira-Marques, S. Monteiro-Reis, D. Montezuma, C. Lourenço, M. C. Oliveira and V. Constâncio, et al., Improved recovery of urinary small extracellular vesicles by differential ultracentrifugation, Sci. Rep., 2024, 14(1), 12267.
17. S. Thangariyal, S. Bhat, A. Mittal, P. Debishree Subudhi, P. Negi and C. Bihari, et al., Comprehensive method for isolation and functional characterization of bacterial vesicles from human biological samples, Methods, 2025, 1095–9130.
18. P. Jiang, X. Ma, X. Wang, J. Huang, Y. Wang and J. Ai, et al., Isolation and Comprehensive Analysis of Cochlear Tissue-Derived Small Extracellular Vesicles, Adv, Sci., 2024, 2198–3844.
19. K. Tiszbein, I. Koss-Mikołajczyk and D. Martysiak-Żurowska, Unlocking the Secrets of Human Milk: Isolation and Characterization of Extracellular Vesicles, Adv. Nutr., 2025, 2156–5376.
20. L. Tengler, M. Tiedtke, J. Schütz, K. Bieback, S. Uhlig and M.-N. Theodoraki, et al., Optimization of extracellular vesicles preparation from saliva of head and neck cancer patients, Sci. Rep., 2024, 14(1), 946.
21. M. A. Livshits, E. Khomyakova, E. G. Evtushenko, V. N. Lazarev, N. A. Kulemin and S. E. Semina, et al., Isolation of exosomes by differential centrifugation: Theoretical analysis of a commonly used protocol, Sci. Rep., 2015, 5(1), 17319.
22. L. G. Rikkert, M. Engelaer, C. M. Hau, L. W. Terstappen, R. Nieuwland and F. A. Coumans, Rate zonal centrifugation can partially separate platelets from platelet-derived vesicles, Res. Pract. Thromb. Haemostasis, 2020, 4(6), 1053–1059.
23. E. I. Yakubovich, A. G. Polischouk and V. I. Evtushenko, Principles and Problems of Exosome Isolation from Biological Fluids, Biochem. (Mosc.) Suppl. Ser. A Membr. Cell Biol., 2022, 16(2), 115–126.
24. P. Li, M. Kaslan, S. H. Lee, J. Yao and Z. Gao, Progress in Exosome Isolation Techniques, Theranostics, 2017, 7(3), 789–804.
25. P. Duong, A. Chung, L. Bouchareychas and R. L. Raffai, Cushioned-Density Gradient Ultracentrifugation (C-DGUC) improves the isolation efficiency of extracellular vesicles, PLoS One, 2019, 14(4), e0215324.
26. N. Dilsiz, A comprehensive review on recent advances in exosome isolation and characterization: Toward clinical applications, Transl. Oncol., 2024, 50, 102121.
27. S. Busatto, G. Vilanilam, T. Ticer, W.-L. Lin, D. W. Dickson and S. Shapiro, et al., Tangential flow filtration for highly efficient concentration of extracellular vesicles from large volumes of fluid, Cells., 2018, 7(12), 273.
28. K. S. Visan, R. J. Lobb, S. Ham, L. G. Lima, C. Palma and C. P. Z. Edna, et al., Comparative analysis of tangential flow filtration and ultracentrifugation, both combined with subsequent size exclusion chromatography, for the isolation of small extracellular vesicles, J. Extracell. Vesicles, 2022, 11(9), e12266.
29. X. Hua, Q. Zhu, Y. Liu, S. Zhou, P. Huang and Q. Li, et al., A double tangential flow filtration-based microfluidic device for highly efficient separation and enrichment of exosomes, Anal. Chim. Acta, 2023, 1258, 341160.
30. D. Lou, Y. Wang, Q. Yang, L. Hu and Q. Zhu, Ultrafiltration combing with phospholipid affinity-based isolation for metabolomic profiling of urinary extracellular vesicles, J. Chromatogr. A, 2021, 1640, 461942.
31. J. Yu, D. Huang, H. Liu and H. Cai, Optimizing Conditions of Polyethylene Glycol Precipitation for Exosomes Isolation From MSCs Culture Media for Regenerative Treatment, Biotechnol. J., 2024, 19(9), e202400374.
32. Q. Huang, J. Wang, H. Ning, W. Liu and X. Han, Exosome isolation based on polyethylene glycol (PEG): a review, Mol. Cell. Biochem., 2025, 480(5), 2847–2861.
33. S. Tiwari, V. Kumar, S. Randhawa and S. K. Verma, Preparation and characterization of extracellular vesicles, Am. J. Reprod. Immunol., 2021, 85(2), e13367.
34. T. Shtam, V. Evtushenko, R. Samsonov, Y. Zabrodskaya, R. Kamyshinsky and L. Zabegina, et al., Evaluation of immune and chemical precipitation methods for plasma exosome isolation, PLoS One, 2020, 15(11), e0242732.
35. H. Odaka, K. Hiemori, A. Shimoda, K. Akiyoshi and H. Tateno, Platelet-derived extracellular vesicles are increased in sera of Alzheimer's disease patients, as revealed by Tim4-based assays, FEBS open Bio, 2021, 11(3), 741–752.
36. J. Guo, C. Wu, X. Lin, J. Zhou, J. Zhang and W. Zheng, et al., Establishment of a simplified dichotomic size-exclusion chromatography for isolating extracellular vesicles toward clinical applications, J. Extracell. Vesicles, 2021, 10(11), e12145.
37. D. Vogele, S. Wöhrle, B. S. Saller, K. Fröhlich, B. A. Barta and M. Cosenza-Contreras, et al., Size exclusion chromatography based proteomic and degradomic profiling of inflammasome-activated, murine bone marrow-derived dendritic cells highlights complex retention and release of cleavage products, Mol. Omics, 2024, 20(9), 595–610.
38. K. Takov, I. J. Teng and M. Mayr, Isolation of Circulating Extracellular Vesicles by High-Performance Size-Exclusion Chromatography, Methods Mol. Biol., 2022, 2504, 31–40.
39. J. Guo, C. Wu, X. Lin, J. Zhou, J. Zhang and W. Zheng, et al., Establishment of a simplified dichotomic size-exclusion chromatography for isolating extracellular vesicles toward clinical applications, J. Extracell. Vesicles, 2021, 10(11), e12145.
40. R. P. Fernandes, A. B. Ruiz, S. Bezemer, F. Detmers, P. Hermans and C. Peixoto, Targeted isolation of extracellular vesicles from cell culture supernatant using immuno-affinity chromatography, Sep. Purif. Technol., 2025, 358, 130312.
41. L. Zhu, H. T. Sun, S. Wang, S. L. Huang, Y. Zheng and C. Q. Wang, et al., Isolation and characterization of exosomes for cancer research, J. Hematol. Oncol., 2020, 13(1), 152.
42. Y. T. Kang, Y. J. Kim, J. Bu, Y. H. Cho, S. W. Han and B. I. Moon, High-purity capture and release of circulating exosomes using an exosome-specific dual-patterned immunofiltration (ExoDIF) device, Nanoscale, 2017, 9(36), 13495.
43. J. A. Welsh, D. C. I. Goberdhan, L. O'Driscoll, E. I. Buzas, C. Blenkiron and B. Bussolati, et al., Minimal information for studies of extracellular vesicles (MISEV2023): From basic to advanced approaches, J. Extracell. Vesicles, 2024, 13(2), e12404.
44. B. Barnes, T. Caws, S. Thomas, A. P. Shephard, R. Corteling and P. Hole, et al., Investigating heparin affinity chromatography for extracellular vesicle purification and fractionation, J. Chromatogr. A, 2022, 1670, 462987.
45. A. J. Zimmerman, G. P. de Oliveira, X. Su, J. Wood, Z. Fu and B. Pinckney, et al., Multimode chromatography-based techniques for high purity isolation of extracellular vesicles from human blood plasma, J. Extracell. Biol., 2024, 3, 3.
46. F. J. Ansari, H. A. Tafti, A. Amanzadeh, S. Rabbani, M. A. Shokrgozar and R. Heidari, et al., Comparison of the efficiency of ultrafiltration, precipitation, and ultracentrifugation methods for exosome isolation, Biochem. Biophys. Rep., 2024, 38, 101668.
47. R. Afrisham, V. Farrokhi, R. Moradi and S. Alizadeh, Comparison of the Characteristics of Circulating Small Extracellular Vesicles Isolated by Ultracentrifugation and a Commercial Kit, Recent Pat. Biotechnol., 2025, 19(4), 346–353.
48. Z. Chen, D. Wang, S. Gu, N. Wu, K. Wang and Y. Zhang, Size exclusion chromatography and asymmetrical flow field-flow fractionation for structural characterization of polysaccharides: A comparative review, Int. J. Biol. Macromol., 2024, 277(Pt 2), 134236.
49. C.-Y. Chou, P.-C. Chiang, C.-C. Li, J.-W. Chang, P.-H. Lu and W.-F. Hsu, et al., Improving the Purity of Extracellular Vesicles by Removal of Lipoproteins from Size Exclusion Chromatography-and Ultracentrifugation-Processed Samples Using Glycosaminoglycan-Functionalized Magnetic Beads, ACS Appl. Mater. Interfaces, 2024, 16(34), 44386.
50. X. Z. Cong, J. Feng, H. J. Zhang, L. Z. Zhang, T. Y. Lin and G. Chen, et al., Microfluidic Device-Based In Vivo Detection of PD-L1-Positive Small Extracellular Vesicles and Its Application for Tumor Monitoring, Anal. Chem., 2024, 96(6), 2658–2665.
51. A. Meggiolaro, V. Moccia, P. Brun, M. Pierno, G. Mistura and V. Zappulli, et al., Microfluidic strategies for Extracellular Vesicle isolation: towards clinical applications, Biosensors, 2022, 13(1), 50.
52. C. Zhang, X. Huo, Y. Zhu, J. N. Higginbotham, Z. Cao and X. Lu, et al., Electrodeposited magnetic nanoporous membrane for high-yield and high-throughput immunocapture of extracellular vesicles and lipoproteins, Commun. Biol., 2022, 5(1), 1358.
53. N. Valle-Tamayo, R. Pérez-González, G. Chiva-Blanch, O. Belbin, S. Serrano-Requena and S. Sirisi, et al., Enrichment of Astrocyte-Derived Extracellular Vesicles from Human Plasma, J. Vis. Exp., 2022, 186.
54. D. Brambilla, L. Sola, E. Chiodi, N. Zarovni, D. Fortunato and M. Criscuoli, et al., Separation of extracellular vesicles by DNA-directed immunocapturing followed by enzymatic release, Anal. Chem., 2020.
55. Y. Lu, B. Lin, W. Liu, J. Zhang, L. Zhu and C. Yang, et al., Isolation of PD-L1 Extracellular Vesicle Subpopulations Using DNA Computation Mediated Microfluidic Tandem Separation, Small methods, 2023, 7(9), 2300516.
56. C. V. Pham, R. Chowdhury, S. Patel, S. K. Jaysawal, Y. Hou and H. Xu, et al., An aptamer-guided fluorescence polarisation platform for extracellular vesicle liquid biopsy, J. Extracell. Vesicles, 2024, 13(9), e12502.
57. X. Guo, F. Hu, Z. Yong, S. Zhao, Y. Wan and B. Wang, et al., Magnetic Nanoparticle-Based Microfluidic Platform for Automated Enrichment of High-Purity Extracellular Vesicles, Anal. Chem., 2024, 96(18), 7212–7219.
58. M. Dehghani and T. R. Gaborski, Fluorescent labeling of extracellular vesicles, Methods Enzymol., 2020, 645, 15–42.
59. M. Boudna, A. D. Campos, P. Vychytilova-Faltejskova, T. Machackova, O. Slaby and K. Souckova, Strategies for labelling of exogenous and endogenous extracellular vesicles and their application for in vitro and in vivo functional studies, Cell Commun. Signaling, 2024, 22(1), 171.
60. C. Chen, N. Cai, Q. Niu, Y. Tian, Y. Hu and X. Yan, Quantitative assessment of lipophilic membrane dye-based labelling of extracellular vesicles by nano-flow cytometry, J. Extracell. Vesicles, 2023, 12(8), e12351.
61. M. Cha, S. H. Jeong, S. Bae, J. H. Park, Y. Baeg and D. W. Han, et al., Efficient Labeling of Vesicles with Lipophilic Fluorescent Dyes via the Salt-Change Method, Anal. Chem., 2023, 95(14), 5843–5849.
62. T. C. Pham, M. K. Jayasinghe, T. T. Pham, Y. Yang, L. Wei and W. M. Usman, et al., Covalent conjugation of extracellular vesicles with peptides and nanobodies for targeted therapeutic delivery, J. Extracell. Vesicles, 2021, 10(4), e12057.
63. Y. Jiang, L. Wang, P. Zhang, X. Liu, H. Di and J. Yang, et al., Chemoenzymatic Labeling of Extracellular Vesicles for Visualizing Their Cellular Internalization in Real Time, Anal. Chem., 2020, 92(2), 2103–2111.
64. R. Bhatta, J. Han, Y. Liu, Y. Bo, D. Lee and J. Zhou, et al., Metabolic tagging of extracellular vesicles and development of enhanced extracellular vesicle based cancer vaccines, Nat. Commun., 2023, 14(1), 8047.
65. N. Jiang, A. Saftics, E. Romano, I. Ghaeli, C. Resto and V. Robles, et al., Multiparametric profiling of HER2-enriched extracellular vesicles in breast cancer using Single Extracellular VEsicle Nanoscopy, J. Nanobiotechnol., 2024, 22(1), 589.
66. H. Zhang, Y. Mao, Z. Nie, Q. Li, M. Wang and C. Cai, et al., Iron Oxide Nanoparticles Engineered Macrophage-Derived Exosomes for Targeted Pathological Angiogenesis Therapy, ACS Nano, 2024, 18(10), 7644–7655.
67. K. Buttiens, C. Maksoudian, I. Perez Gilabert, C. Rios Luci, B. B. Manshian and S. J. Soenen, Inorganic Nanoparticles Change Cancer-Cell-Derived Extracellular Vesicle Secretion Levels and Cargo Composition, Resulting in Secondary Biological Effects, ACS Appl. Mater. Interfaces, 2024, 16(1), 66–83.
68. V. A. Toomajian, A. Tundo, E. E. Ural, E. M. Greeson, C. H. Contag and A. V. Makela, Magnetic Particle Imaging Reveals that Iron-Labeled Extracellular Vesicles Accumulate in Brains of Mice with Metastases, ACS Appl. Mater. Interfaces, 2024, 16(24), 30860–30873.
69. X.-C. Liu, H.-J. Zhang, R. Xu, J. Liu, H.-F. Xia and Q.-H. Xie, et al., A NIR-II quantum dot-assisted dual-color imaging strategy enables simultaneous tracking of two subtypes of extracellular vesicles in vivo, Chem. Eng. J., 2024, 492, 152242.
70. K. Zhang, J. Hu and Y. Hu, Visualization strategies of extracellular vesicles: Illuminating the invisible ‘dust’ in theranostics, J. Extracell. Vesicles, 2024, 4, 100061.
71. M. Zhang, L. Vojtech, Z. Ye, F. Hladik and E. Nance, Quantum Dot Labeling and Visualization of Extracellular Vesicles, ACS Appl. Nano Mater., 2020, 3(7), 7211–7222.
72. V. Vinduska, C. E. Gallops, R. O'Connor, Y. Wang and X. Huang, Exosomal Surface Protein Detection with Quantum Dots and Immunomagnetic Capture for Cancer Detection, Nanomaterials, 2021, 11, 7.
73. D. Levy, M. A. Do, J. Zhang, A. Brown and B. Lu, Orchestrating extracellular vesicle with dual reporters for imaging and capturing in mammalian cell culture, Front. Mol. Biosci., 2021, 8, 680580.
74. C. Olson, P. Zhang, J. Ku, R. Flojo, D. Boyes and B. Lu, A Novel Dual-Reporter System Reveals Distinct Characteristics of Exosome-Mediated Protein Secretion in Human Cells, Biol. Proced. Online, 2023, 25(1), 25.
75. M. José Sánchez, P. Leivar, S. Borrós, C. Fornaguera and M. Lecina, Enhanced quantification and cell tracking of dual fluorescent labeled extracellular vesicles, Int. J. Pharm., 2024, 667, 124921.
76. K. Wardhani, A. Levina, G. E. Grau and P. A. Fluorescent Lay, phosphorescent, magnetic resonance contrast and radioactive tracer labelling of extracellular vesicles, Chem. Soc. Rev., 2024, 53, 6779–6829.
77. S. Almeida, L. Santos, A. Falcão, C. Gomes and A. Abrunhosa, In Vivo Tracking of Extracellular Vesicles by Nuclear Imaging: Advances in Radiolabeling Strategies, Int. J. Mol. Sci., 2020, 21, 24.
78. F. N. Faruqu, J. T. Wang, L. Xu, L. McNickle, E. M. Chong and A. Walters, et al., Membrane Radiolabelling of Exosomes for Comparative Biodistribution Analysis in Immunocompetent and Immunodeficient Mice - A Novel and Universal Approach, Theranostics, 2019, 9(6), 1666–1682.
79. S. Patel, K. F. Schmidt, M. Farhoud, T. Zi, S. C. Jang and K. Dooley, et al., In vivo tracking of [(89)Zr]Zr-labeled engineered extracellular vesicles by PET reveals organ-specific biodistribution based upon the route of administration, Nucl. Med. Biol., 2022, 112–113, 20–30.
80. R. Szatanek, M. Baj-Krzyworzeka, J. Zimoch, M. Lekka, M. Siedlar and J. Baran, The Methods of Choice for Extracellular Vesicles (EVs) Characterization, Int. J. Mol. Sci., 2017, 18, 6.
81. M. L. Corona, I. Hurbain, G. Raposo and G. van Niel, Characterization of extracellular vesicles by transmission electron microscopy and immunolabeling electron microscopy, Cell-secreted vesicles: methods and protocols, Springer, 2023, p. 33–43.
82. F. Royo, C. Théry, J. M. Falcón-Pérez, R. Nieuwland and K. W. Witwer, Methods for Separation and Characterization of Extracellular Vesicles: Results of a Worldwide Survey Performed by the ISEV Rigor and Standardization Subcommittee, Cells., 2020, 9, 9.
83. E. Priglinger, J. Strasser, B. Buchroithner, F. Weber, S. Wolbank and D. Auer, et al., Label-free characterization of an extracellular vesicle-based therapeutic, J. Extracell. Vesicles, 2021, 10(12), e12156.
84. D. Leng, C. Chimedtseren, T. Wang, R. Su, S. Bao and X. Mo, et al., Proteomics analysis of body fluid exosomes of rheumatoid arthritis patients underwent oxhorn cupping therapy, PLoS One, 2024, 19(12), e0311526.
85. C. Aydoğan, Z. Günyel, S. Alharthi, H. Aslan, İ. Y. Erdoğan and Z. E. Rassi, Rapid Separation and Analysis of Exosomes in Milk Sample by on-Line Nano-Liquid Chromatography, Electrophoresis, 2025, 1418–1426.
86. C. Elsner, S. Ergün and N. Wagner, Biogenesis and release of endothelial extracellular vesicles: Morphological aspects, Ann. Anat., 2023, 245, 152006.
87. D. Buschmann, V. Mussack and J. B. Byrd, Separation, characterization, and standardization of extracellular vesicles for drug delivery applications, Adv. Drug Delivery Rev., 2021, 174, 348–368.
88. R. Rahmani, J. Kiani, W. Y. Tong, M. Soleimani, N. H. Voelcker and E. Arefian, Engineered anti-EGFRvIII targeted exosomes induce apoptosis in glioblastoma multiforme, J. Drug Target, 2023, 31(3), 310–319.
89. E. Cansever Mutlu, M. Kaya, I. Küçük, B. Ben-Nissan and A. Stamboulis, Exosome Structures Supported by Machine Learning Can Be Used as a Promising Diagnostic Tool, Materials, 2022, 15, 22.
90. A. Emelyanov, T. Shtam, R. Kamyshinsky, L. Garaeva, N. Verlov and I. Miliukhina, et al., Cryo-electron microscopy of extracellular vesicles from cerebrospinal fluid, PLoS One, 2020, 15(1), e0227949.
91. M. Kurtjak, S. Kereïche, D. Klepac, H. Križan, M. Perčić and V. Krušić Alić, et al., Unveiling the Native Morphology of Extracellular Vesicles from Human Cerebrospinal Fluid by Atomic Force and Cryogenic Electron Microscopy, Biomedicines., 2022, 10, 6.
92. M. I. Morandi, P. Busko, E. Ozer-Partuk, S. Khan, G. Zarfati and Y. Elbaz-Alon, et al., Extracellular vesicle fusion visualized by cryo-electron microscopy, PNAS Nexus, 2022, 1(4), pgac156.
93. A. Parra, I. Barranco, P. Martínez-Díaz, E. González, O. E. Albóniga and D. Cabrera, et al., Author Correction: Cryogenic electron microscopy reveals morphologically distinct subtypes of extracellular vesicles among porcine ejaculate fractions, Sci. Rep., 2024, 14(1), 18512.
94. Y. He, Y. Xing, T. Jiang, J. Wang, S. Sang and H. Rong, et al., Fluorescence labeling of extracellular vesicles for diverse bio-applications in vitro and in vivo, Chem. Commun., 2023, 59(44), 6609–6626.
95. J. Bos, Y. Abou Haydar, O. Mayboroda, P. H. Commere and D. Mazel, Extracellular vesicle production and membrane uptake promote repair and antibiotic tolerance in E. coli., bioRxiv, preprint, 2025.02.03.636191, 2025 DOI:10.1101/2025.02.03.636191.
96. A. Wallucks, P. DeCorwin-Martin, M. L. Shen, A. Ng and D. Juncker, Size photometry and fluorescence imaging of immobilized immersed extracellular vesicles, J. Extracell. Vesicles, 2024, 13(10), e12512.
97. J. E. Sorrells, E. M. Martin, E. Aksamitiene, P. Mukherjee, A. Alex and E. J. Chaney, et al., Label-free characterization of single extracellular vesicles using two-photon fluorescence lifetime imaging microscopy of NAD(P)H, Sci. Rep., 2021, 11(1), 3308.
98. E. Lazaro-Ibanez, F. N. Faruqu, A. F. Saleh, A. M. Silva, J. Tzu-Wen Wang and J. Rak, et al., Selection of fluorescent, bioluminescent, and radioactive tracers to accurately reflect extracellular vesicle biodistribution in vivo, ACS nano., 2021, 15(2), 3212–3227.
99. D. Gupta, X. Liang, S. Pavlova, O. P. B. Wiklander, G. Corso and Y. Zhao, et al., Quantification of extracellular vesicles in vitro and in vivo using sensitive bioluminescence imaging, J. Extracell. Vesicles, 2020, 9(1), 1800222.
100. H. Liu, S. Liu, Y. Xiao, W. Song, H. Li and L. W. C. Ho, et al., A pH-Reversible Fluorescent Probe for in Situ Imaging of Extracellular Vesicles and Their Secretion from Living Cells, Nano Lett., 2021, 21(21), 9224–9232.
101. G. I. Perez, M. H. Bachmann and M. Kanada, In Vivo Analysis of Heterogeneous Extracellular Vesicles Using a Red-Shifted Bioluminescence Resonance Energy Transfer Reporter Protein, Methods Mol. Biol., 2023, 2668, 23–32.
102. A. D. Elliott, Confocal Microscopy: Principles and Modern Practices, Curr. Protoc. Cytom., 2020, 92(1), e68.
103. S. Liu, B. Chen, W. Zou, H. Sha, X. Feng and S. Han, et al., Compressive confocal microscopy imaging at the single-photon level with ultra-low sampling ratios, Commun. Eng., 2024, 3(1), 88.
104. C. J. Kim, M. D. Kuczler, L. Dong, J. Kim, S. R. Amend and Y. K. Cho, et al., Extracellular Vesicle Uptake Assay via Confocal Microscope Imaging Analysis, J. Vis. Exp., 2022, 180.
105. R. P. McNamara, Y. Zhou, A. B. Eason, J. T. Landis, M. G. Chambers and S. Willcox, et al., Imaging of surface microdomains on individual extracellular vesicles in 3-D, J. Extracell. Vesicles, 2022, 11(3), e12191.
106. S. Dechantsreiter, A. R. Ambrose, J. D. Worboys, J. M. E. Lim, S. Liu and R. Shah, et al., Heterogeneity in extracellular vesicle secretion by single human macrophages revealed by super-resolution microscopy, J. Extracell. Vesicles, 2022, 11(4), e12215.
107. M. Skliar and V. S. Chernyshev, Imaging of Extracellular Vesicles by Atomic Force Microscopy, J. Vis. Exp., 2019, 151.
108. M. I. Sandira, K. Lim, T. Yoshida, E. S. Sajidah, S. Narimatsu and R. Imakawa, et al., Nanoscopic Profiling of Small Extracellular Vesicles via High-Speed Atomic Force Microscopy (HS-AFM) Videography, J. Extracell. Vesicles, 2025, 14(4), e270050.
109. I. A. Chelnokova, I. A. Nikitina and M. N. Starodubtseva, Mechanical properties of blood exosomes and lipoproteins after the rat whole blood irradiation with X-rays in vitro explored by atomic force microscopy, Micron, 2024, 184, 103662.
110. Z. Han, S. Liu, Y. Pei, Z. Ding, Y. Li and X. Wang, et al., Highly efficient magnetic labelling allows MRI tracking of the homing of stem cell-derived extracellular vesicles following systemic delivery, J. Extracell. Vesicles, 2021, 10(3), e12054.
111. M. Sancho-Albero, N. Ayaz, V. Sebastian, C. Chirizzi, M. Encinas-Gimenez and G. Neri, et al., Superfluorinated Extracellular Vesicles for In Vivo Imaging by (19)F-MRI, ACS Appl. Mater. Interfaces, 2023, 15(7), 8974–8985.
112. E. H. Kim, J. P. Son, G. S. Oh, S. Park, E. Hong and K. S. Lee, et al., Clinical Scale MSC-Derived Extracellular Vesicles Enhance Poststroke Neuroplasticity in Rodents and Non-Human Primates, J. Extracell. Vesicles, 2025, 14(6), e70110.
113. Z. Wang, Recent advances in X-ray grating-based dark-field imaging, TrAC, Trends Anal. Chem., 2024, 181, 118052.
114. J. Greffier, A. Viry, A. Robert, M. Khorsi and S. Si-Mohamed, Photon-counting CT systems: A technical review of current clinical possibilities, Diagn. Interv. Imaging, 2025, 106(2), 53–59.
115. P. Lara, S. Palma-Florez, E. Salas-Huenuleo, I. Polakovicova, S. Guerrero and L. Lobos-Gonzalez, et al., Gold nanoparticle based double-labeling of melanoma extracellular vesicles to determine the specificity of uptake by cells and preferential accumulation in small metastatic lung tumors, J. Nanobiotechnol., 2020, 18(1), 20.
116. P. M. Rubio, H. M. Garcia-Garcia, A. A. Sanchez, A. Alfaddagh, D. Andreini and A. Baggiano, et al., Key advances in cardiac computed tomography: A review of the most relevant studies published in 2024 on coronary and structural heart disease, J. Cardiovasc. Comput. Tomogr., 2025, 19(3), 291–298.
117. J. Gaztanaga and J. Lopez-Mattei, Modern CT detector technology and innovations in image reconstruction enhance cardiovascular CT, J. Cardiovasc. Comput. Tomogr., 2025, 19(1), 72–73.
118. N. Comfort, K. Cai, T. R. Bloomquist, M. D. Strait, A. W. Ferrante Jr and A. A. Baccarelli, Nanoparticle tracking analysis for the quantification and size determination of extracellular vesicles, J. Visualized Exp., 2021, 10(169), 1–22.
119. X. Chen, Q. Li, W. Zhang, Y. Xu, X. Nie and X. Wang, et al., Akkermansia muciniphila-derived extracellular vesicles alleviate colitis-related cognitive impairment via tryptophan metabolic reprogramming of the gut-brain axis, Gut Microbes, 2026, 18(1), 2611546.
120. G. Kowkabany and Y. Bao, Nanoparticle Tracking Analysis: An Effective Tool to Characterize Extracellular Vesicles, Molecules, 2024, 29, 19.
121. D. Mladenović, J. Brealey, B. Peacock, K. Koort and N. Zarovni, Quantitative fluorescent nanoparticle tracking analysis and nano-flow cytometry enable advanced characterization of single extracellular vesicles, J. Extracell. Biol., 2025, 4(1), e70031.
122. A. D. Kashkanova, M. Blessing, M. Reischke, J. O. Baur, A. S. Baur and V. Sandoghdar, et al., Label-free discrimination of extracellular vesicles from large lipoproteins, J. Extracell. Vesicles, 2023, 12(8), e12348.
123. J. M. Carnino, H. Lee and Y. Jin, Isolation and characterization of extracellular vesicles from Broncho-alveolar lavage fluid: a review and comparison of different methods, Respir. Res., 2019, 20, 1–11.
124. I. Barranco, L. Padilla, I. Parrilla, A. Álvarez-Barrientos, C. Pérez-Patiño and F. J. Peña, et al., Extracellular vesicles isolated from porcine seminal plasma exhibit different tetraspanin expression profiles, Sci. Rep., 2019, 9(1), 11584.
125. K. M. Dave, W. Zhao, C. Hoover and A. D'Souza, D SM. Extracellular Vesicles Derived from a Human Brain Endothelial Cell Line Increase Cellular ATP Levels, AAPS PharmSciTech, 2021, 22(1), 18.
126. S. Xiao, Y. Yao, S. Liao, B. Xu, X. Li and Y. Zhang, et al., Accurate and Convenient Lung Cancer Diagnosis through Detection of Extracellular Vesicle Membrane Proteins via Förster Resonance Energy Transfer, Nano Lett., 2023, 23(17), 8115–8125.
127. R. Ishikawa, S. Yoshida, S. I. Sawada, Y. Sasaki and K. Akiyoshi, Fusogenic Hybrid Extracellular Vesicles with PD-1 Membrane Proteins for the Cytosolic Delivery of Cargos, Cancers, 2022,(11), 14.
128. W. S. Fisher, J. Douglas, S. Roshan, R. Perez, S. Wei and L. Roberts, et al., Acidic Conditions Promote Clustering of Cancer Cell Derived Extracellular Vesicles and Enhance their Fusion with Synthetic Liposomes, Langmuir, 2025, 41(7), 4533–4544.
129. Y. Yang, C. Zhai, Q. Zeng, A. L. Khan and H. Yu, Multifunctional detection of extracellular vesicles with surface plasmon resonance microscopy, Anal. Chem., 2020, 92(7), 4884–4890.
130. J. M. Lopez Baltazar, W. Gu, M. Bocková and Q. Yu, Immunoassays for Extracellular Vesicle Detection via Transmembrane Proteins Using Surface Plasmon Resonance Biosensors, ACS Sens., 2024, 9(7), 3594–3603.
131. T. Shibuta, Y. Takada, S. Nishinosono, S. Yasuda, Y. Ono and Y. Hirooka, et al., Disease-specific signatures of circulating extracellular vesicles detected by the surface plasmon resonance imaging: a pilot study, Extracell. Vesicles Circ. Nucleic Acids, 2025, 6(1), 36–53.
132. J. Min, T. Son, J. S. Hong, P. S. Cheah, A. Wegemann and K. Murlidharan, et al., Plasmon-Enhanced Biosensing for Multiplexed Profiling of Extracellular Vesicles, Adv. Biosyst., 2020, 4(12), e2000003.
133. M. H. Jeong, T. Son and H. Im, Plasmon-Enhanced Characterization of Single Extracellular Vesicles, Methods Mol. Biol., 2023, 2668, 3–13.
134. J. Zini, H. Saari, P. Ciana, T. Viitala, A. Lohmus and J. Saarinen, et al., Infrared and Raman spectroscopy for purity assessment of extracellular vesicles, Eur. J. Pharm. Sci., 2022, 172, 106135.
135. H. J. O'Toole, N. M. Lowe, V. Arun, A. V. Kolesov, T. L. Palmieri and N. K. Tran, et al., Plasma-derived extracellular vesicles (EVs) as biomarkers of sepsis in burn patients via label-free Raman spectroscopy, J. Extracell. Vesicles, 2024, 13(9), e12506.
136. Y. F. Qin, X. Y. Lu, Z. Shi, Q. S. Huang, X. Wang and B. Ren, et al., Deep Learning-Enabled Raman Spectroscopic Identification of Pathogen-Derived Extracellular Vesicles and the Biogenesis Process, Anal. Chem., 2022, 94(36), 12416–12426.
137. A. Bonizzi, L. Signati, M. Grimaldi, M. Truffi, F. Piccotti and S. Gagliardi, et al., Exploring breast cancer-related biochemical changes in circulating extracellular vesicles using Raman spectroscopy, Biosens. Bioelectron., 2025, 278, 117287.
138. L. Buccini, A. Proietti, G. La Penna, C. Mancini, F. Mura and S. Tacconi, et al., Toward the nanoscale chemical and physical probing of milk-derived extracellular vesicles using Raman and tip-enhanced Raman spectroscopy, Nanoscale, 2024, 16(16), 8132–8142.
139. L. Bar, M. E. Villanueva, S. Neupane, G. Cordoyiannis and P. Losada-Pérez, QCM-D Study of the Formation of Solid-Supported Artificial Lipid Membranes: State-of-the-Art, Recent Advances, and Perspectives, Phys. Status Solidi A, 2023, 220(22), 2200625.
140. A. Kowalczyk, A. Gajda-Walczak, M. Ruzycka-Ayoush, A. Targonska, G. Mosieniak and M. Glogowski, et al., Parallel SPR and QCM-D Quantitative Analysis of CD9, CD63, and CD81 Tetraspanins: A Simple and Sensitive Way to Determine the Concentration of Extracellular Vesicles Isolated from Human Lung Cancer Cells, Anal. Chem., 2023, 95(25), 9520–9530.
141. J. A. Welsh, G. J. A. Arkesteijn, M. Bremer, M. Cimorelli, F. Dignat-George and B. Giebel, et al., A compendium of single extracellular vesicle flow cytometry, J. Extracell. Vesicles, 2023, 12(2), e12299.
142. H. Liu, Y. Tian, C. Xue, Q. Niu, C. Chen and X. Yan, Analysis of extracellular vesicle DNA at the single-vesicle level by nano-flow cytometry, J. Extracell. Vesicles, 2022, 11(4), e12206.
143. H.-C. Lau, I. Passalacqua, J.-H. Jung, Y. Kwon, D. Zocco and S.-S. Park, et al., Unraveling the surface marker signature of cell-derived vesicles via proteome analysis and nanoparticle flow cytometry, Sci. Rep., 2024, 14(1), 121.
144. S. Halvaei, N. Salmond and K. C. Williams, Identification of DYRK1b as a novel regulator of small extracellular vesicle release using a high throughput nanoscale flow cytometry screening platform, Nanoscale, 2025, 17(13), 8206–8218.
145. C. B. Huynh, N. B. Vu, T. T. Van and P. V. Pham, Effects of Exosomes From Hypoxia-Induced Adipose-Derived Stem Cells on Ameliorating Photoaging, Clin., Cosmet. Invest. Dermatol., 2025, 18, 1683–1702.
146. H. Kobayashi, T. Shiba, T. Yoshida, D. Bolidong, K. Kato and Y. Sato, et al., Precise analysis of single small extracellular vesicles using flow cytometry, Sci. Rep., 2024, 14(1), 7465.
147. S. M. Johnson, A. Banyard, C. Smith, A. Mironov and M. G. McCabe, Large extracellular vesicles can be characterised by multiplex labelling using imaging flow cytometry, Int. J. Mol. Sci., 2020, 21(22), 8723.
148. W. W. Woud, E. van der Pol, E. Mul, M. J. Hoogduijn, C. C. Baan and K. Boer, et al., An imaging flow cytometry-based methodology for the analysis of single extracellular vesicles in unprocessed human plasma, Commun. Biol., 2022, 5(1), 633.
149. T. Tertel, R. Dittrich, T. J. Kutzner and B. Giebel, Advanced quality control through single ev analysis: fingerprinting of msc-ev populations with unique capabilities for end-product and in-process control, Cytotherapy, 2024, 26(6), S88.
150. I. Jalaludin, D. M. Lubman and J. Kim, A guide to mass spectrometric analysis of extracellular vesicle proteins for biomarker discovery, Mass Spectrom. Rev., 2023, 42(2), 844–872.
151. I. Casanova-Salas, D. Aguilar, S. Cordoba-Terreros, L. Agundez, J. Brandariz and N. Herranz, et al., Circulating tumor extracellular vesicles to monitor metastatic prostate cancer genomics and transcriptomic evolution, Cancer Cell, 2024, 42(7), 1301–1312.e7.
152. E. Lázaro-Ibáñez, C. Lässer, G. V. Shelke, R. Crescitelli, S. C. Jang and A. Cvjetkovic, et al., DNA analysis of low- and high-density fractions defines heterogeneous subpopulations of small extracellular vesicles based on their DNA cargo and topology, J. Extracell. Vesicles, 2019, 8(1), 1656993.
153. B. Nguyen, N. C. Wong, T. Semple, M. Clark, S. Q. Wong and C. Leslie, et al., Low-coverage whole-genome sequencing of extracellular vesicle-associated DNA in patients with metastatic cancer, Sci. Rep., 2021, 11(1), 4016.
154. R. T. Miceli, T. Y. Chen, Y. Nose, S. Tichkule, B. Brown and J. F. Fullard, et al., Extracellular vesicles, RNA sequencing, and bioinformatic analyses: Challenges, solutions, and recommendations, J. Extracell. Vesicles, 2024, 13(12), e70005.
155. E. Schoger, F. Bleckwedel, G. Germena, C. Rocha, P. Tucholla and I. Sobitov, et al., Single-cell transcriptomics reveal extracellular vesicles secretion with a cardiomyocyte proteostasis signature during pathological remodeling, Commun. Biol., 2023, 6(1), 79.
156. A. Giannopoulos-Dimitriou, A. Saiti, A. Malousi, A. K. Anagnostopoulos, G. Vatsellas and P. M. Al-Maghrabi, et al., Molecular Profiling of A549 Cell-Derived Exosomes: Proteomic, miRNA, and Interactome Analysis for Identifying Potential Key Regulators in Lung Cancer, Cancers, 2024,(24), 16.
157. W. Rozek, M. Kwasnik, W. Socha, B. Czech and J. Rola, Profiling of snoRNAs in Exosomes Secreted from Cells Infected with Influenza A Virus, Int. J. Mol. Sci., 2024, 26, 1.
158. K. Supradit, K. Wongprasert, S. Tangphatsornruang, T. Yoocha, C. Sonthirod and W. Pootakham, et al., microRNA profiling of exosomes derived from plasma and their potential as biomarkers for Opisthorchis viverrini-associated cholangiocarcinoma, Acta Trop., 2024, 258, 107362.
159. C. P. Faria, B. A.-O. Ferreira, Á. A.-O. Lourenço, I. Guerra, T. Melo and P. Domingues, et al., Lipidome of extracellular vesicles from Giardia lamblia, PLoS One, 2023, 1932–6203.
160. Q. Zhu, H. Li, Z. Ao, H. Xu, J. Luo and C. Kaurich, et al., Lipidomic identification of urinary extracellular vesicles for non-alcoholic steatohepatitis diagnosis, J. Nanobiotechnol., 2022, 20(1), 349.
161. M. G. Krokidis, K. A. Pucha, M. Mustapic, T. P. Exarchos, P. Vlamos and D. Kapogiannis, Lipidomic Analysis of Plasma Extracellular Vesicles Derived from Alzheimer's Disease Patients, Cells., 2024, 13, 8.
162. E. A.-O. Manouchehri Doulabi, C. Fredolini, R. A.-O. Gallini, L. Löf, Q. Shen and R. Ikebuchi, et al., Surface protein profiling of prostate-derived extracellular vesicles by mass spectrometry and proximity assays, Commun. Biol, 2022, 2399–3642.
163. G. A.-O. Xu, R. A.-O. Huang, R. A.-O. Wumaier, J. A.-O. Lyu, M. A.-O. Huang and Y. A.-O. Zhang, et al., Proteomic Profiling of Serum Extracellular Vesicles Identifies Diagnostic Signatures and Therapeutic Targets in Breast Cancer, Cancer Res., 2024, 1538–7445.
164. R. Bandu, J. W. Oh and K. A.-O. Kim, Extracellular vesicle proteins as breast cancer biomarkers: Mass spectrometry-based analysis, Proteomics, 2024, 1615–9861.
165. M. L. Mizgier, G. Nardocci, V. Ramírez, M. J. Bendek, M. Hernández and C. Rojas, et al., Proteomic Insights Into Gingival Crevicular Extracellular Vesicles in Periodontitis and Gestational Diabetes: An Exploratory Study, J. Clin. Periodontol., 2025, 1600–051X.
166. P. A.-O. Aparicio, D. Navarrete-Villanueva, A. Gómez-Cabello, T. López-Royo, E. Santamaría and J. Fernández-Irigoyen, et al., Proteomic profiling of human plasma extracellular vesicles identifies PF4 and C1R as novel biomarker in sarcopenia, J. Cachexia Sarcopenia Muscle, 2024, 2190–6009.
167. J. Zhang, Z. Gao, W. Xiao, N. Jin, J. Zeng and F. Wang, et al., A simplified and efficient extracellular vesicle-based proteomics strategy for early diagnosis of colorectal cancer, Chem. Sci., 2024, 15(44), 18419–18430.
168. X. Wu, H. Li, F. Meng, T. H. Lui and X. Pan, iTRAQ proteomic analysis of exosomes derived from synovial fluid reveals disease patterns and potential biomarkers of Osteoarthritis, J. Orthop. Surg. Res., 2024, 1749–799X.
169. J. E. Nielsen, B. Honoré, K. Vestergård, R. G. Maltesen, G. Christiansen and A. U. Bøge, et al., Shotgun-based proteomics of extracellular vesicles in Alzheimer's disease reveals biomarkers involved in immunological and coagulation pathways, Sci. Rep., 2021, 11(1), 18518.
170. M. Surman, M. Wilczak, U. Jankowska, B. Skupień-Rabian and M. Przybyło, Shotgun proteomics of thyroid carcinoma exosomes - Insight into the role of exosomal proteins in carcinogenesis and thyroid homeostasis, Biochim. Biophys. Acta, Gen. Subj., 2024, 1872–8006.
171. C. Qin, D. Xu, H. Han, J. Fang, H. Wang and Y. Liu, et al., Dynamic and Label-Free Sensing of Cardiomyocyte Responses to Nanosized Vesicles for Cardiac Oxidative Stress Injury Therapy, Nano Lett., 2023, 23(24), 11850–11859.
172. K. Ono, Y. Okusha, M. T. Tran, K. Umemori and T. Eguchi, Western blot protocols for analysis of CCN proteins and fragments in exosomes, vesicle-free fractions, and cells, CCN Proteins: Methods and Protocols, Springer, 2022, p. 39–57.
173. A. M. Silva, E. Lázaro-Ibáñez, A. Gunnarsson, A. Dhande, G. Daaboul and B. Peacock, et al., Quantification of protein cargo loading into engineered extracellular vesicles at single-vesicle and single-molecule resolution, J. Extracell. Vesicles, 2021, 10(10), e12130.
174. Z. Chen, L. Luo, T. Ye, J. Zhou, X. Niu and J. Yuan, et al., Identification of specific markers for human pluripotent stem cell-derived small extracellular vesicles, J. Extracell. Vesicles, 2024, 13(2), e12409.
175. J. Ko, Y. Wang, J. C. T. Carlson, A. Marquard, J. Gungabeesoon and A. Charest, et al., Single Extracellular Vesicle Protein Analysis Using Immuno-Droplet Digital Polymerase Chain Reaction Amplification, Adv. Biosyst., 2020, 4(12), e1900307.
176. C. Liu, H. Lin, J. Guo, C. Yang, J. Chen and W. Pan, et al., Profiling of single-vesicle surface proteins via droplet digital immuno-PCR for multi-subpopulation extracellular vesicles counting towards cancer diagnostics, Chem. Eng. J., 2023, 471, 144364.

---