Article

ISSN 0006-2979, Biochemistry (Moscow), 2025, Vol. 90, No. 10, pp. 1376-1387 © Pleiades Publishing, Ltd., 2025.

Russian Text © The Author(s), 2025, published in Biokhimiya, 2025, Vol. 90, No. 10, pp. 1474-1486.

1376

Comparison of Methods for Concentration

Assessment of Extracellular Vesicles Isolated

from Different Biological Fluids

Gleb O. Skryabin

, Adel D. Enikeev

, Anastasiia A. Beliaeva

, Kirill I. Zhordania

Sergey A. Galetsky

, Dmitry V. Bagrov

1,2

, Oyatiddin T. Imaraliev

Ivan A. Karasev

, and Elena M. Tchevkina

1,a

Blokhin National Medical Research Center of Oncology, 115522 Moscow, Russia

Faculty of Biology, Lomonosov Moscow State University, 119234 Moscow, Russia

e-mail: tchevkina@mail.ru

Received July 17, 2025

Revised September 19, 2025

Accepted September 23, 2025

Abstract— Accurate quantification of extracellular vesicles (EVs) remains a significant challenge in biomedical

research. Although various analytical methods have been developed, their reliability is often limited by the

presence of non-vesicular nanoparticles and biological contaminants, particularly in biological fluids. More-

over, for some sources of EVs, such as uterine aspirates and gastric juice, quantitative evaluation of EVs has

not been investigated. The aim of the study is to perform comparative analysis of three EV quantification

methods: total protein content measurement, nanoparticle tracking analysis (NTA), and esterase activity as-

sessment using commercial FluoroCet exosome quantitation kit in EVs isolated from various biological fluids:

blood plasma, ascitic fluid, uterine aspirates, gastric juice, and medium conditioned by ovarian and non-small

cell lung cancer cells. All three methods demonstrated strong correlation for the EV samples derived from

the conditioned medium, supporting their validity for in vitro EV quantification in highly purified samples. In

contrast, blood plasma, ascitic fluid, and uterine aspirates exhibited discrepancies between the methods, likely

attributable to the presence of non-vesicular nanoparticles. Notably, the EVs from gastric juice demonstrated

strong correlation between the protein content and esterase activity, indicating prevalence of the vesicle-as-

sociated proteins and, potentially, unique EV composition in this fluid. The findings underscore the necessity

for multifactorial approach to EV quantification, taking into account factors such as sample origin and lim-

itations inherent to the specific method employed. These results may serve as a basis for the development

of standardized protocols for EV quantification, which is particularly relevant for clinical sample analysis.

DOI: 10.1134/S0006297925602217

Keywords: exosomes, extracellular vesicles, vesicle concentration, NTA, FluoroCet

* To whom correspondence should be addressed.

INTRODUCTION

Extracellular vesicles (EVs) comprise a hetero-

geneous group of nanoparticles enclosed into a lip-

id-bilayer membrane, which are released by cells

into environment. Exosomes (with sizes 30-150  nm)

and microvesicles (sizes up to 1  µm) are the most in-

vestigated types of EVs, which differ in mechanisms

of formation and molecular composition. EVs play a

key role in cell–cell communications transporting pro-

teins, lipids, RNAs and DNAs between the nearest and

distant cells and tissues, thus affecting physiological

and pathological processes in an organism. Tumor

cells excrete larger numbers of EVs in comparison

with the normal cells thus facilitating intercellular

exchange of tumor-associated molecules; that is why

their function have been investigated most thoroughly

in the context of carcinogenesis, where EVs mediate

malignization of normal cells, remodeling of stromal

microenvironment, evasion of immune surveillance,

EFFICIENCY OF METHODS FOR EVALUATION OF VESICLES CONCENTRATION 1377

BIOCHEMISTRY (Moscow) Vol. 90 No. 10 2025

acquiring drug resistance, formation of pre-meta-

static niche, and other processes [1]. Mechanism of

EV formation and selection of their molecular cargo

is strictly controlled; hence, their composition could

reflect molecular profile of the cells-producers; that

is why EVs are often considered as promising bio-

markers for non-invasive fluid-based diagnostics in

oncology and other areas of medicine [2]. Moreover,

due to the protective function of the membrane and

unique composition of the receptors on its surface,

EVs are subjects of intensive investigation in the area

of targeted drug delivery [3].

Considering growing interest in EVs for diagnos-

tics and therapy, the International Society for Ex-

tracellular Vesicles (ISEV) developed standards and

recommendations for working with these structures.

According to the MISEV recommendations (Minimal

Information for Studies of Extracellular Vesicles),

it is necessary prior to start of investigation to char-

acterize the preparation using several independent

methods [4]. These methods include morphologi-

cal analysis, identification of molecular markers,

as well as quantitative assessment including mea-

suring concentration of EVs and their size distrib-

ution.

Assessment of EV concentration is an essential

step of their characterization; however, at present

there is no universal technique known that has suf-

ficient accuracy, sensitivity, and specificity. The most

often used approaches are the following: measuring

concentration (total content) of protein in EV prepara-

tions, determination of concentration using nanoparti-

cle tracking analysis (NTA), enzymatic methods (such

as FluoroCet and ExoCet), as well as less often used

but promising technologies such as tunable resistive

pulse sensing (tRPS), flow cytometry, and fluorescent

modification of NTA [5, 6].

Determination of total protein content in the

samples with extracellular vesicles is performed us-

ing either classic colorimetric techniques (Bradford

protein assay and bicinchoninic acid assay), or using

highly sensitive fluorescent reagent kits such as Qubit

and NanoOrange. The latter are especially useful for

working with low-concentration samples, obtained,

for example, from the conditioned cell culture medi-

um (CM). The main advantages of the method are its

simplicity and reproducibility, however, the key draw-

back of the method is lack of specificity: the obtained

values include both proteins associated with vesicles,

as well as non-vesicular proteins including from vi-

ruses, supramolecular attack particles (SMAPs), chy-

lomicrons, exomeres, supermeres, as well as protein

complexes and aggregates, lipoproteins of different

density, and ribonucleic complexes such as vault-

type ribonucleoprotein complexes, etc. [7,  8]. The EV

preparations most contaminated with non-vesicular

structures are preparations isolated from clinical sam-

ples, particularly from blood [9, 10].

The NTA method allows quantitative evaluation

of size distribution and concentration of the particles

based on analysis of their Brownian motion in a solu-

tion. Unlike in the method of dynamic light scattering

(DLS), this method evaluates each particle in the solu-

tion separately, while DLS operates with ensemble of

particles, it is demanding to the sample concentration,

and is rarely used for assessment of concentrations.

The NTA method exhibits high sensitivity and it pro-

vides information on heterogeneity of particles in the

sample, however, similar to the DLS-based methods,

it does not allow distinguishing vesicles from non-ve-

sicular particles of the similar size such as lipopro-

teins, virus-like particles, and other structures men-

tioned above [11].

The enzymatic methods, such as the one based on

using the FluoroCet reagent kit, are based on deter-

mination of activity of acetylcholine esterase (AChE)

in the vesicles, membranes of which are, presumably,

enriched with this enzyme. This approach requires

minimal sample volume and exhibits high sensitivity,

but it depends on the level of expression of AChE,

which could vary depending on the cell origin and

type of biological fluid. Other methods, in particu-

lar the tRPS method based on recording changes of

electrical resistance, when individual particles pass

through a nanopore, or flow cytometry, which uses

antibodies against specific markers of EVs (CD9, CD63,

CD81), as well as mass-spectrometry allow more ac-

curate assessment of composition and origin of the

vesicles, but so far they remain more labor-intensive

and expensive [12,13]. Many of the alternative meth-

ods, such as, for example, methods based on immu-

noprecipitation, allow assessment of only particular

populations of EVs, but not the entire pool.

Hence, the issue with standardization of the meth-

ods for quantification of EVs and correct interpreta-

tion of the results, especially in the cases of clinical

preparations of EVs with high levels of contamination

(blood plasma, urine, etc.), has been unresolved yet.

At the same time, selection of the most adequate and

accessible method for quantification of EVs is one of

the most important task of the research in this area.

That is why, the goal of this study was comparison

of three methods of quantification of EVs – evalua-

tion of protein concentration in the EV preparations,

analysis of trajectories of nanoparticles, and analysis

of esterase activity with the help of commercial kit

for quantitative analysis of exosomes FluoroCet – in

the samples of EVs isolated from different biological

fluids. The following biological fluids were selected

as sources of EVs commonly used in the context of

studying EVs (blood plasma, ascitic fluid), as well as

less studied in the context of EVs biological fluids,

SKRYABIN et al.1378

BIOCHEMISTRY (Moscow) Vol. 90 No. 10 2025

but which attract significant interest with regard to

screening of oncomarkers and as drug targets, such

as uterine aspirates and gastric juice.

MATERIALS AND METHODS

Clinical samples. Clinical samples were ob-

tained from Blokhin National Medical Research Cen-

ter of Oncology, Ministry of Health of the Russian

Federation from the patients prior to surgery or any

other treatment. All donors of biomaterials signed

voluntary informed consent form to participate in

the study. Study protocol was approved by the local

ethics committee (Protocol no.  5, June 10, 2022, proj-

ect no. 22-15-00375; and Protocol no.  1, January 25,

2024, project 24-25-00052). Samples of aspirate from

uterus cavity (uterine aspirates, n  =  14) were collect-

ed with the help of a Pipelle type  C device from the

donors without oncological anamnesis. Immediately

after collection, the sample (volume varied from 0.2

to 1  ml) was diluted in 0.5  ml of cold phosphate buf-

fer (PB; Gibco, USA). Samples of ascitic fluid (n = 10)

were collected under sterile conditions in the process

of laparocentesis from the female patients with his-

tologically verified diagnosis of ovarian serous car-

cinoma. Sample volumes varied from 5 to 500  ml

(more often 10-25  ml). Samples of uterine aspirates

and ascites were provided by the department of on-

cogynecology; samples of gastric juice (GJ) – by the

Endoscopic department. Gastric juice samples (n = 18)

were collected during endoscopic examination of the

individuals without oncological diseases using a vid-

eo gastroscope GIF H-185 (OLYMPUS, Japan). Prior to

procedure, the patients were fasting (12  h without

food and 6 h without water). Volume of collected GJ

was from 2 to 10  ml; after collection it was diluted

in 5 ml of PB. GJ samples were collected from differ-

ent regions of the stomach. Peripheral blood samples

(n  =  12) were provided by the department of thoracic

oncology; they were collected from the patients with

verified diagnosis of non-small cell lung cancer into

vacuum tubes with EDTA. Blood plasma was produced

with a standard centrifugation technique.

Cell cultures. Ovarian cancer cell lines (OV-

CAR-3, OVCAR-4, OVCAR-8, SKOV3) and non-small cell

lung cancer cell lines (H460, H1299, A549) were cul-

tivated at 37°C in atmosphere of 5% CO

in a RPMI-

1640 and DMEM medium (PanEko, Russia) supple-

mented with 10% fetal bovine serum (FBS; HyClone,

Austria), 100 U/ml penicillin and 100 mg/ml strepto-

mycin (PanEko). To prepare exosome-free medium

FBS was used, which was first purified from native

vesicles with the help ultracentrifugation at 110,000g

for 16 h. To collect conditioned medium, cells were

seeded into 6 cell culture flasks with surface area

of175cm

, next day the medium was exchanged with

the exosome-free medium. When the cells reached

90% confluency the medium was collected, combined,

and used for isolation of small EVs.

Sample processing and isolation of EVs. Sam-

ples of conditioned medium after removal of cell de-

bris (centrifugation at 2000g for 15  min at 4°C) was

stored without freezing at 4°C up to 7days. Combined

supernatants were used for the following isolation of

EVs according to the protocol described previously

[14]. All clinical samples were processes no later than

1  h after collection and stored on ice for the duration

of the entire process. Tubes with whole blood were

centrifuged at 2000g for 15  min at 4°C to produce

blood plasma, which next was centrifuged again at

10,000g (30 min) and stored at −80°C until isolation of

EVs. Protocols for processing samples of uterine aspi-

rates and ascites [15], as well as of gastric juice [16]

were described in our previous studies. Isolation of

EVs was carried out using the method of differential

centrifugation according to the standard protocol [17];

all modifications of the technique for isolation of EVs

for each biological fluid have been described in the

respective abovementioned papers.

Nanoparticle tracking analysis. Size composition

and concentration of extracellular vesicles were deter-

mined using NTA method using a NanoSight LM10HS

device equipped with a LM14 temperature-control

system (Malvern Panalytical Ltd., United Kingdom), a

laser module LM 14C (405  nm, 65  mW), and high-res-

olution CMOS camera with (C11440-50B; Hamamatsu

Photonics, Japan). All measurements were carried out

in accordance with the ASTM E2834-12(2018) stan-

dard. Prior to examination samples were diluted in

a particle-free PB to final concentration ~1.5  ×  10

particles/ml. For each sample 12 60-s videos were re-

corded. Processing and combining of the data were

performed using the NTA 2.3 build 33 software (Mal-

vern Panalytical Ltd.). Primary NTA data are present-

ed in the Online Resource 1.

Transmission electron microscopy. Carbon coat-

ed grids (Ted Pella, USA) were first treated for 45  s in

an Emitech K100X device (Quorum Technologies Ltd.,

United Kingdom) to increase hydrophilicity of the sur-

face. Vesicle samples were diluted 5-40-fold in PB de-

pending on concentration determined with NTA; next

samples were applied onto grids and incubated for

30-60  s. Next, grids were stained twice for 45  s with

a 1% solution of uranyl acetate and dried at room

temperature. Images (no less than 10 for each sam-

ple) were acquired with a JEM-1400 electron micro-

scope (JEOL, Japan) operating at acceleration voltage

of 120 kV.

Analysis of protein and exosome concentra-

tion. Total protein content in the preparations of ex-

tracellular vesicles was determined using two methods

EFFICIENCY OF METHODS FOR EVALUATION OF VESICLES CONCENTRATION 1379

BIOCHEMISTRY (Moscow) Vol. 90 No. 10 2025

Fig. 1. Morphological and molecular characterization of extracellular vesicles (EVs) isolated from different biological fluids.

a) Examples of microphotographs of EVs from the investigated biological fluids obtained with the help of transmission

electron microscopy. Scale bar: 500 nm. b) Examples of Western blot analysis of exosomal proteins markers in the same

preparations of EVs. Lanes: 1) control, cell lysate of OVCAR-8 cell line; 2-6) vesicles isolated from five different biological

sources in the following order: uterine aspirates(2), ascitic fluid (3), gastric juice(4), blood plasma(5), conditioned medium

from the culture of OVCAR-8 cells (6).

depending of protein concentration in the sample. The

samples with low protein concentration (<1 µg/ml)

were analyzed using a NanoOrange® Protein Quan-

titation Kit (Thermo Fisher Scientific, USA) according

to the manufacturer’s recommendations. Fluorescence

was recorded with a SpectraMax M5e microplate

reader (Molecular Devices, USA). The samples with

high protein concentrations (such as ascites and plas-

ma) were examined using the Bradford protein assay

with reagents from Bio-Rad (#500-0006; Germany) ac-

cording to the manufacturer’s instructions. Measure-

ments were carried out with a Benchmark Plus micro-

plate spectrophotometer (Bio-Rad Laboratories, USA).

A FluoroCet™ Exosome Quantitation Kit was used en-

zymatic activity of acetylcholine esterase (System Bio-

sciences, USA). Analysis was carried out according to

the manufacturer’s protocol. Samples were incubated

with a substrate at room temperature; next measure-

ments were performed. Each sample was analyzed in

at least two technical replicates.

Immunoblotting and antibodies. Protocol used

for immunoblotting of the vesicle samples from dif-

ferent sources was described in our previous stud-

ies [14-16]. The following primary antibodies were

used: anti-Flotillin-2 (1  :  1000; #3436S; CST, USA);

anti- CD9 (1  :  2000; #13174; CST); anti-Stomatin

(1  :  500; #sc-134554; Santa Cruz, USA). Goat-anti-

rabbit antibodies were used as secondary antibodies

(1  :  80  000; #29902; CST).

Statistical data processing. Spearman’s Rank

Correlation was used to calculate correlation co-

efficients between the methods for quantification

of extracellular vesicles. Calculations were carried

our using the GraphPad Prism 9 software (GraphPad

Software, USA). p-values < 0.05 were considered sta-

tistically significant.

RESULTS

Characterization of vesicle of different ori-

gins. Morphology and size of EVs isolated from dif-

ferent biological fluids were evaluated with the help

of transmission electron microscopy and nanoparti-

cle tracking analysis. As can be seen in the present-

ed microphotographs (Fig. 1a), particles of spherical

SKRYABIN et al.1380

BIOCHEMISTRY (Moscow) Vol. 90 No. 10 2025

Table 1. Average values of size characteristics of extracellular vesicles isolated from different biological sources

(based on NTA data)

Source of EVs

Number

of samples, n

Average size,

nm (SEM)

Modal size,

nm (SEM)

Median size,

nm (SEM)

10th

percentile,

nm (SEM)

90th

percentile

nm (SEM)

Uterine

aspirate

14 139.0 (2.3) 102.9 (3.6) 122.8 (2.4) 59.4 (1.0) 235.9 (4.0)

Ascitic fluid 10 143.6 (4.7) 91.3 (3.9) 122.9 (4.5) 58.4 (1.3) 252.1 (8.4)

Gastric juice 18 145.1 (5.6) 88.4 (5.6) 130.1 (6.2) 54.3 (2.3) 261.1 (8.8)

Blood plasma 12 132.5 (4.3) 89.2 (4.1) 125.7 (4.4) 57.8 (2.4) 249.3 (6.9)

Conditioned

medium

13 127.5 (2.5) 95.0 (4.9) 112.2 (2.8) 56.4 (1.0) 211.6 (3.8)

Note. Values are presented in nm with standard error of the mean (SEM) in brackets.

or cap-like shape typical of vesicles have been ob-

served in all preparations.

In the next stage molecular verification of the ve-

sicular nature of the isolated particles was conducted.

In accordance with the ISEV recommendations the fol-

lowing proteins localized in different compartments

of EVs were used as positive markers of exosomes:

luminal component of the ESCRT complex TSG101;

component of lipid rafts, stomatin, suggested in our

previous study as an exosome marker [18]; and CD9–

component of tetraspanin-enriched microdomains. All

investigated samples were enriched with these pro-

teins in comparison with the cell lysates, which sup-

ports their vesicular nature (Fig. 1b).

Analysis of concentration and size-distribution of

the particles conducted using NTA demonstrated that

the average size of EVs varied from 127.5 nm (CM)

to 145.1 nm (GJ), while the modal values were in the

range 88.4-102.9nm. Complete results on size charac-

teristics are presented in Table 1.

To evaluate reliability of different approaches

for quantification of extracellular vesicles the sam-

ples obtained from five different sources (conditioned

medium of tumor cells (n  =  13), blood plasma (n  =  10),

ascitic fluid (n  =  12), uterine aspirates (n  =  14), and

gastric juice (n  =  18)) were analyzed. In all cases

vesicles were isolated using differential ultracentrif-

ugation. Three independent methods were used for

quantification: measurements of total protein content

(commercial reagent kit NanoOrange or Bradford

method depending on the sample concentration), NTA,

and fluorescent analysis of acetylcholine esterase ac-

tivity (further mentioned as FluoroCet).

Vesicles from the conditioned medium. The

highest convergence of the results was obtained

during analysis of EVs isolated from the conditioned

media of different tumor cells lines. Results obtained

for all samples demonstrated significant and pro-

nounced correlation between the three used methods

(Fig. 2).

Fig. 2. Correlation analysis of the results of quantification of extracellular vesicles isolated from the media conditioned

by ovarian cancer tumor cells (OVCAR-3, OVCAR-4, OVCAR-8, SKOV3) and by the non-small cell lung cancer cells (H460,

H1299, A549) using three methods. a)Comparison of FluoroCet with NanoOrange; b)comparison of NTA with NanoOrange;

c)comparison of NTA with FluoroCet. Each dot on the graph corresponds to the individual sample of extracellular vesicles

isolated from the indicated biological source.

EFFICIENCY OF METHODS FOR EVALUATION OF VESICLES CONCENTRATION 1381

BIOCHEMISTRY (Moscow) Vol. 90 No. 10 2025

Fig. 3. Correlation analysis of the results of quantification of extracellular vesicles isolated from blood plasma (upper panel)

and ascitic fluid (lower panel) using three methods. a)Comparison of FluoroCet with Bradford assay; b)comparison of NTA

with Bradford assay; c) comparison of NTA with FluoroCet. Each dot on the graph corresponds to the individual sample of

extracellular vesicles isolated from the indicated biological source.

Spearman’s Rank Correlation coefficients(r) were

0.95 for the pairs protein/NTA and NTA/FluoroCet,

and for the pair protein/FluoroCet – 0.97 (p  <  0.01).

With high purity of the isolated vesicles and ab-

sence of significant contribution of external proteins

or non-vesicular nanoparticles, the results of these

methods show similar patterns and allow comparing

concentrations of EVs in different samples. Hence, in

the case of using standardized protocol of isolation,

all three methods can be used for quantification of

relative concentration of EVs.

Vesicles isolated from blood plasma and ascitic

fluid. Unlike in the case of CM, in the samples iso-

lated from blood plasma significant differences were

revealed between the results of different methods.

The weakest correlation was observed between NTA

and FluoroCet (r  =  0.63), as well as between NTA and

protein content measured with the Bradford method

(r  =  0.63), which indicates presence in the plasma of

a large number of non-vesicular particles with siz-

es comparable with EVs (Fig. 3, upper panel). At the

same time, the correlation dependence in the pair

protein/FluoroCet was shown to be non-linear, which

was most pronounced in the samples with high pro-

tein concentration (>8µg/µl). This could be explained

by the possibility of presence of AChE-positive lipo-

protein complexes in the high-concentration samples.

Although AChE is not typical for the main lipopro-

teins, this enzyme has been found in the specialized

complexes, for example, being associated with the lip-

id membranes of erythrocytes via the GPI-anchors sta-

bilized by phospholipids (including cardiolipin) [19].

The enzyme could be released during hemolysis from

the membranes in composition of complexes with lip-

ids, which could ensure non-proportional contribution

to the signal at high protein concentrations. These

results highlight limitations for the use of both NTA

and FluoroCet in the native biological fluids with high

concentrations of proteins and lipoproteins. At the

same time measurements of total protein content also

do not ensure sufficient specificity for the reliable

quantification of vesicles, because significant fraction

of protein could be from the soluble components of

plasma not associated with EVs.

Patterns of the results obtained in analysis of ves-

icles from ascitic fluid were similar to those observed

in the case of blood plasma, but exhibited even more

pronounced divergence between the methods. The

lowest correlation was observed between NTA and

FluoroCet (r  =  0.42) and between NTA and Bradford

assay (r  =  0.33), which indicates higher level of for-

eign particles and protein complexes in the samples

(Fig.3, lower panel). Same as in the case of blood plas-

ma, quantitative dependence between the total protein

content and activity of acetylcholine esterase (Fluoro-

Cet) in the samples of ascitic fluid was non-linear.

SKRYABIN et al.1382

BIOCHEMISTRY (Moscow) Vol. 90 No. 10 2025

Fig. 4. Correlation analysis of the results of quantification of extracellular vesicles isolated from uterine aspirates using

three methods. a) Comparison of FluoroCet with NanoOrange; b) comparison of NTA with NanoOrange; c) comparison of

NTA with FluoroCet. Each dot on the graph corresponds to the individual sample of extracellular vesicles isolated from

the indicated biological source.

Fig. 5. Correlation analysis of the result of quantification of extracellular vesicles isolated from gastric juice using three

methods. a) Comparison of FluoroCet with NanoOrange; b) comparison of NTA with NanoOrange; c) comparison of NTA

with FluoroCet. Each dot on the graph corresponds to the individual sample of extracellular vesicles isolated from the

indicated biological source.

Considering that ascites often contain admixtures of

blood, this effect could be explained by the presence

of specific AChE-positive lipoprotein complexes sim-

ilar to those observed in plasma. At the same time,

contribution of variable expression of AChE on its

content in the extracellular vesicles obtained from

different types of cells cannot be ruled out. Hence,

similar to the case of blood plasma, EVs from ascitic

fluid require more stringent purification method, use

of several quantification techniques, and accurate in-

terpretation of the results.

Vesicles from uterine aspirates. The samples

obtained from uterine aspirates typically have low

total protein content (not exceeding 2  µg/µl), which

allowed avoiding saturation of the signal in enzymatic

reactions. Nevertheless, the degree of correlation be-

tween the methods was moderate (r  =  0.62 – for the

protein/FluoroCet pair; r  =  0.52 – for the protein/NTA

pair), while no significant correlation between the

results of FluoroCet and NTA was observed (Fig. 4).

Overall, the data indicate that quantification of

EVs from uterine aspirates requires using of several

approaches and, if possible, following normalization

based on marker proteins.

Vesicles from gastric juice. Unexpectedly, the

results obtained during analysis of EVs isolated from

gastric juice differed significantly from the results ob-

served during investigation of EVs from other biologi-

cal fluids. Unlike in other cases, there was very high

correlation between the protein content and activity

of acetylcholine esterase in the EVs samples isolated

from GJ (r  =  0.97; p  <  0.0001), which implied presence

of high proportion of vesicular proteins in the total

protein composition (Fig. 5).

Moreover, significant, although moderate, cor-

relation between the results of NTA and two other

methods was observed: r  =  0.63 – for the NTA/Fluoro-

Cet pair and r  =  0.67 – for the NTA/NanoOrange pair;

which indicated that contribution of non-vesicular

particles to the NTA signal was still present. It is im-

portant to note that total protein concentrations in

the samples of GJ were significantly lower (<0.5  µg/µl)

in comparison with other fluids, which could facilitate

more accurate evaluation of vesicles and decrease

EFFICIENCY OF METHODS FOR EVALUATION OF VESICLES CONCENTRATION 1383

BIOCHEMISTRY (Moscow) Vol. 90 No. 10 2025

Table 2. Spearman’s correlation coefficients between three methods of quantification of extracellular vesicles

derived from different types of biological fluids

Pair of compared methods

Spearman’s correlation coefficient

CM Plasma Ascites Aspirate GJ

Protein/FluoroCet 0.97 n/d* n/d* 0.62 0.97

NTA/Protein 0.95 0.63 0.33 0.52 0.63

NTA/FluoroCet 0.95 0.63 0.42 0.07 0.67

Note. Statistically significant values are shown in bold. * Correlation dependence is non-linear.

of the background signal. The obtained data allow

suggesting that the EV samples obtained from GJ

are characterized with high purity and absence of

contaminants, majority of which have protein com-

ponents. Hence, one of the significant results of this

study is characterization of gastric juice as a prom-

ising, previously not investigated in detail biological

source of EVs, which could potentially be important

for the search of markers and development of non-in-

vasive diagnostic approaches based on molecular

composition of EVs.

Summarizing the obtained results of comparative

analysis, it could be concluded that consistency of

the results between the used methods of quantifica-

tion of extracellular vesicles depends significantly on

the type of biological fluid from which the EVs were

produced (Table 2).

DISCUSSION

Heterogeneity of vesicles in size, composition,

and origin in combination with difficulties of distin-

guishing them from other interfering particles, such

as lipoproteins, protein complexes, viruses, and a

whole number of other particles, complicates stan-

dardization of approaches for biochemical and phys-

ical characterization of EVs. Despite the existence of

a wide variety of approaches developed and used in

investigation of EVs, they all have their peculiarities

and limitations in the detection range, accuracy, pro-

cessivity, and applicability for analysis of particular

parameters of EVs. At present, there are no methods,

which could be considered universal for the reliable

quantification of EVs.

The results obtained in our study confirm that

selection of the method for assessment of concen-

tration of EVs is critically dependent on the type of

analyzed biological fluid and degree of purity of the

sample. In the case of cell culture CM, where content

of ‘foreign proteins’ and nanoparticles is minimal, all

three methods (NTA, measurement of total protein,

and FluoroCet), demonstrated high degree of consis-

tency. This allows considering any of them as suitable

for relative quantitative assessment of vesicles in the

highly purified samples, which is achieved by the sec-

ond round of ultracentrifugation. The data obtained

in this study are in agreement with the previously

published results demonstrating significant effect of

the sample purity on the results of EVs quantification;

in particular, it was shown in the study by Escude-

ro-Cernudaetal.[20] that variations between the rep-

licates decreased from 43 to 15% with the decrease of

the level of admixtures present in the CM.

NTA is one of the most popular methods for

quantification of EVs, which uses the principle of

Brownian motion for determination of both size and

concentration of particles. However, its application for

analysis of complex biological fluids, such as blood

plasma and ascitic fluid, was shown to be severely

limited due to impossibility to distinguish true EVs

from lipoproteins and protein aggregates. These data

are in full agreement with the results of previous

studies showing that up to 70% of the particles de-

tected by the NTA method in blood plasma could be

not EVs [9].

Determination of total protein concentration re-

mains the most affordable and widely used method

for quantification of EVs. However, as was shown in

this study, high concentration of soluble proteins typ-

ical for blood plasma and ascitic fluid, decreases sig-

nificantly reliability of this approach. Similar limita-

tion has been reported previously: albumin and other

serum proteins could be responsible for up to 60% of

the total protein signal in the EV preparations from

blood plasma [21].

The enzymatic FluoroCet method based on mea-

suring activity of AChE demonstrated high sensitivi-

ty and linearity in the samples with low degree of

contaminations, such as conditioned cell culture me-

dium and gastric juice. However, in the EV samples

from blood plasma and ascites non-linear dependence

between the esterase activity and protein concentra-

tion was observed, which could be associated with

SKRYABIN et al.1384

BIOCHEMISTRY (Moscow) Vol. 90 No. 10 2025

both presence in the protein-enriched samples of

AChE-positive lipoprotein complexes, as well as with

uneven distribution of AChE between the populations

of vesicles [22]. In this context, the results reported

by Grigor’eva et al. seem to be of great interest [23].

It was shown that the preparations of exosomes iso-

lated from different biological fluids including blood

plasma inevitably contain a large number of lipo-

proteins (‘non-vesicles’) comparable in size with exo-

somes. Fraction of these structures could reach 40%

in plasma, which significantly distorts the results of

molecular analysis of exosomes. These data confirm

that the observed in our study non-linear dependence

between the AChE activity and protein concentration

in the EVs from protein-rich biological fluids could

be explained primarily by contamination with lipo-

proteins emphasizing the necessity of a complex ap-

proach for quantification of EVs, which should include

control of purity of the vesicle preparations.

The most interesting results were obtained during

analysis of EVs from gastric juice, which revealed ex-

tremely high correlation between the protein content

and activity of acetylcholine esterase (r  =  0.97). This

phenomenon has not been observed for other biologi-

cal fluids, and it allows suggesting that the aggressive

medium in stomach facilitates selective preservation

of vesicular proteins and degradation of proteins not

associated with vesicles. In comparison with blood

plasma and ascites, GJ contains lower amounts of pro-

teins and have relative low protein load, which facil-

itates reduction of artefacts during measurements. In

our previous study we for the first time isolated and

characterized EVs from gastric juice in accordance

with the ISEV recommendation including description

of their morphology, size characteristics, and com-

position of marker proteins [16]. Furthermore, we

demonstrated that the vesicles from GJ collected from

the patients with stomach cancer differ in size, dis-

tribution, and level of expression of tetraspanin CD9

from the vesicles collected from the GJ of the healthy

donors [24]. These data confirm potential of EVs in GJ

as a promising object for search of diagnostically sig-

nificant molecular signs of stomach cancer. This study

supplements these observations: high degree of cor-

relation between the independent methods of quan-

tification demonstrates applicability of GJ as a stable

and representative source of EVs for further molecule

studies devoted to search of potential biomarkers.

Use of different techniques at the same time

seems as the most effective strategy, which helps to

overcome limitations of individual approaches, when

used simultaneously. In particular, combination of en-

zymatic assay with the methods assessing size and

concentration (NTA or tRPS) seems reasonable to use

for examination of complex fluids (plasma, ascites)

supplementing this with determination of protein

content and calculation of the degree of purity (par-

ticles/protein ratio). This approach allows simulta-

neous evaluation of the yield of EVs and quality of

the preparation [25]. Lack of standardized methods

for quantification of EVs is a serious problem in the

studying of vesicles. It was shown both in our study

and in the studies published worldwide that the use

of different approaches often produces significantly

different results for the identical samples, which com-

plicates comparison of the results of different studies

and their clinical interpretation [26, 27].

In the context of standardization, it is necessary

to mention that at present there is no commonly ac-

cepted ‘gold standard’ for the method for isolation of

EVs from biological fluids, and, obviously, selection of

the method affects purity of preparations and amount

of contaminating particles of one or another origin.

The available techniques such as different variants of

ultracentrifugation including the ones using density

gradients, methods based of capture of nanoparticles

by biopolymers, chromatographic methods, methods

of microfiltration and ultrafiltration, methods based

on isolation of EVs based on binding of one or an-

other vesicular molecules have their advantages and

drawbacks with regards the balance between the

yield of EVs and purity of the preparations (absence

of non-vesicular contaminations). In this study the

method based on differential centrifugation/ultracen-

trifugation was used for isolation of EVs from dif-

ferent biological sources, which is used most often

and is recommended by ISEV. The revealed in this

study correlations between the results of different

methods for quantification of EVs indirectly support

the notion that this technique is ideally suitable for

quantification of EVs from the samples of cell culture

medium, which are characterized with minimal con-

taminations by non-vesicular particles with sizes and

density similar to EVs, as well as for quantification of

EVs from GJ, which, presumably, also have low level

of contaminations, at least of those of protein nature.

At the same time, as has been mentioned above, ab-

sence of correlation between the results of analysis of

EVs from other biological fluids indicates presence of

a large number of contaminating particles in the EV

preparations, which, undoubtedly, is associated with

the selection of the isolation technique. This problem

could be resolved to a large extent by using other

techniques (such as, for example, exclusion chroma-

tography, micro- and ultrafiltration, immunoaffinity

methods, and others); however, for many of these

methods loss of the total amount of isolated EVs or

of individual populations of vesicles is typical. Never-

theless, it could be assumed that in the case of using

these techniques, selection of the method for quan-

tification of EVs would affect the results of analysis

to a lesser degree.

EFFICIENCY OF METHODS FOR EVALUATION OF VESICLES CONCENTRATION 1385

BIOCHEMISTRY (Moscow) Vol. 90 No. 10 2025

Quantification of EVs is essential for comparison

of their concentrations in the clinical samples, espe-

cially considering the actively discussed at present

hypothesis on the increase of concentration of EVs

in the biological fluids (primarily in circulating flu-

ids) of oncology patients, which has been suggested

for diagnostic purposes and for monitoring relapses

[28-30]. Moreover, majority of the studies investigat-

ing functional significance of EVs and their in  vitro

and in vivo molecular composition in carcinogenesis

and tumor progression (as well as in pathogenesis of

other diseases) require very accurate enumeration of

EVs. Hence, the results obtained in our study are im-

portant for further progress in this area of research.

From the practical point of view, the obtained

data could be used for optimization of protocols for

quantification of EVs in clinical samples, as well as

the basis for selection of the method depending on

the type of biological fluid. In addition, our results

allow recommending gastric juice as a promising ob-

ject for the search of vesicular biomarkers with high

specificity and low background levels in the case of

using protein and enzymatic assays for quantification.

CONCLUSIONS

The conducted study clearly demonstrates that

none of the investigated methods of quantification

of extracellular vesicles is universally applicable for

all types of biological samples. The obtained results

confirm the necessity of differential approach for the

selection of the methods depending on the origin and

degree of purity of the analyzed preparations of EVs.

All three considered methods (NTA, total protein con-

tent, and analysis of esterase activity) could be suc-

cessfully used for the EV preparations isolated from

the standardized cell-conditioned medium, while ex-

amination of biological fluids derived from an organ-

ism requires combination of several approaches that

also should take into consideration degree of puri-

ty of the samples. Unique feature of the EVs from

gastric juice, which demonstrate exceptionally high

correlation between the protein content and activi-

ty of acetylcholine esterase revealed in this study, is

especially interesting. This indicates minimal level of

contamination of the samples with non-vesicular pro-

tein particles. The obtained data could be used by the

researchers for selection of the technique for quan-

tification of EVs of different origin, which especially

important in examination of clinical samples.

Abbreviations

AChE acetylcholine esterase

CM culture medium

EV extracellular vesicles

GJ gastric juice

NTA nanoparticle tracking analysis

PB phosphate buffer

Supplementary information

The online version contains supplementary material

available at https://doi.org/10.1134/S0006297925602217.

Acknowledgments

Experiments with transmission electron microscopy

were carried out at the Center for Collective Use “Elec-

tron microscopy in life sciences” of the Department

ofBiology, Lomonosov Moscow State University.

Contributions

G. O. Skryabin, E. M. Tchevkina – concept and su-

pervision of the study; G. O. Skryabin, A. D. Enikeev,

A. A. Beliaeva, S. A. Galetsky, D. V. Bagrov– conducting

experiments; K. I. Zhordania, O. T. Imaraliev, I. A. Kara-

sev– preparation of clinical materials; G. O. Skryabin,

A. D. Enikeev, D. V. Bagrov, E. M. Tchevkina – discus-

sion of the study results; G. O. Skryabin – writing text

of the paper; G. O. Skryabin, E. M. Tchevkina – editing

text of the paper.

Funding

This work was financially supported by the Russian

Science Foundation (grant no.24-25-00052, 2024-2025,

https://rscf.ru/project/24-25-00052/ [in Russian]).

Ethics approval and consent to participate

All procedures involving samples derived from hu-

mans were carried out in accordance with institution-

al, national and international standards. Voluntary

informed consent form was signed by all participants

of the study.

Conflict of interest

The authors of this work declare that they have

noconflicts of interest.

REFERENCES

1. Kalluri,R., and LeBleu, V.S. (2020) Thebiology, func-

tion, and biomedical applications of exosomes, Science,

367, eaau6977, https://doi.org/10.1126/science.aau6977.

2. Hu, C., Jiang, W., Lv, M., Fan, S., Lu, Y., Wu, Q., and

Pi,J. (2022) Potentiality of exosomal proteins as novel

cancer biomarkers for liquid biopsy, Front. Immunol.,

13, 792046, https://doi.org/10.3389/fimmu.2022.792046.

3. Han, Q.-F., Li, W.-J., Hu, K.-S., Gao,J., Zhai, W.-L., Yang,

J.-H., and Zhang, S.-J. (2022) Exosome biogenesis: ma-

chinery, regulation, and therapeutic implications in

cancer, Mol. Cancer, 21, 207, https://doi.org/10.1186/

s12943-022-01671-0.

SKRYABIN et al.1386

BIOCHEMISTRY (Moscow) Vol. 90 No. 10 2025

4. Welsh, J. A., Goberdhan, D. C. I., O’Driscoll, L., Buzas,

E. I., Blenkiron, C., Bussolati, B., Cai, H., Di Vizio, D.,

Driedonks, T. A. P., and Erdbrügger, U. (2024) Mini-

mal information for studies of extracellular vesicles

(MISEV2023): from basic to advanced approaches,

J.Extracell. Vesicle, 13, e12404, https://doi.org/10.1002/

jev2.12404.

5. Vogel, R., Coumans, F. A. W., Maltesen, R. G., Böing,

A.N., Bonnington, K.E., Broekman, M.L., Broom, M.F.,

Buzás, E.I., Christiansen,G., Hajji,N., Kristensen, S.R.,

Kuehn, M. J., Lund, S. M., Maas, S. L., Nieuwland, R.,

Osteikoetxea, X., Schnoor, R., Scicluna, B. J., Sham-

brook, M., de Vrij, J., and Pedersen, S. (2016) A stan-

dardized method to determine the Concentration

of extracellular vesicles using tunable resistive

pulse sensing, J. Extracell. Vesicles, 5, 31242, https://

doi.org/10.3402/jev.v5.31242.

6. Hartjes, T.A., Mytnyk,S., Jenster, G.W., van Steijn,V.,

and van Royen, M. E. (2019) Extracellular vesicle

quantification and characterization: common meth-

ods and emerging approaches, Bioengineering, 6, 7,

https://doi.org/10.3390/bioengineering6010007.

7. Jeppesen, D. K., Zhang,Q., Franklin, J. L., and Coffey,

R. J. (2023) Extracellular vesicles and nanoparticles:

emerging complexities, Trends Cell Biol., 33, 667-681,

https://doi.org/10.1016/j.tcb.2023.01.002.

8. Willms, E., Cabañas, C., Mäger, I., Wood, M. J., and

Vader, P. (2018) Extracellular vesicle heterogeneity:

subpopulations, isolation techniques, and diverse

functions in cancer progression, Front. Immunol.,

9, 738, https://doi.org/10.3389/fimmu.2018.00738.

9. Brennan,K., Martin,K., FitzGerald, S.P., O’Sullivan,J.,

Wu,Y., Blanco, A., Richardson,C., and McGee, M.M.

(2020) A comparison of methods for the isola-

tion and separation of extracellular vesicles from

protein and lipid particles in human serum, Sci.

Rep., 10, 1039, https://doi.org/10.1038/s41598-020-

57497-7.

10. Xu, R., Rai, A., Chen, M., Suwakulsiri, W., Greening,

D.W., and Simpson, R.J. (2018) Extracellular vesicles

in cancer – implications for future improvements in

cancer care, Nat. Rev. Clin. Oncol., 15, 617-638, https://

doi.org/10.1038/s41571-018-0036-9.

11. Comfort, N., Cai, K., Bloomquist, T. R., Strait, M. D.,

Ferrante, A.W.J., and Baccarelli, A.A. (2021) Nanopar-

ticle tracking analysis for the quantification and size

determination of extracellular vesicles, J. Vis. Exp.,

169, e62447, https://doi.org/10.3791/62447.

12. Maas, S. L. N., Breakefield, X. O., and Weaver, A. M.

(2017) Extracellular vesicles: unique intercellular de-

livery vehicles, Trends Cell Biol., 3, 172-188, https://

doi.org/10.1016/j.tcb.2016.11.003.

13. Tian, Y., Ma, L., Gong, M., Su, G., Zhu, S., Zhang, W.,

Wang, S., Li, Z., Chen, C., Li, L., Wu, L., and Yan, X.

(2018) Protein profiling and sizing of extracellular

vesicles from colorectal cancer patients via flow cy-

tometry, ACS Nano, 12, 671-680, https://doi.org/10.1021/

acsnano.7b07782.

14. Skryabin, G. O., Beliaeva, A. A., Enikeev, A. D.,

Bagrov, D. V., Keremet, A. M., Komelkov, A. V., Elkin,

D. S., Sylantieva, D. M., and Tchevkina, E. M. (2024)

Analysis of miRNAs miR-125a-5p, -27a-5p, -193a-5p,

-135b-5p, -451a, -495-3p and -136-5p in parental ovar-

ian cancer cells and secreted extracellular vesicles,

Adv. Mol. Oncol., 11, 113-123, https://doi.org/10.17650/

2313-805X-2024-11-1-113-123.

15. Skryabin, G. O., Komelkov, A. V., Zhordania, K. I.,

Bagrov, D.V., Enikeev, A. D., Galetsky, S. A., Beliaeva,

A. A., Kopnin, P. B., Moiseenko, A. V., Senkovenko,

A.M., and Tchevkina, E. M. (2024) Integrated miRNA

profiling of extracellular vesicles from uterine aspi-

rates, malignant ascites and primary-cultured ascites

cells for ovarian cancer screening, Pharmaceutics, 16,

902, https://doi.org/10.3390/pharmaceutics16070902.

16. Skryabin, G.O., Vinokurova, S.V., Galetsky, S.A., Elkin,

D. S., Senkovenko, A. M., Denisova, D. A., Komelkov,

A.V., Stilidi, I.S., Peregorodiev, I.N., Malikhova, O.A.,

Imaraliev, O. T., Enikeev, A. D., and Tchevkina, E. M.

(2022) Isolation and characterization of extracellular

vesicles from gastric juice, Cancers (Basel), 14, 3314,

https://doi.org/10.3390/cancers14143314.

17. Théry, C., Amigorena, S., Raposo, G., and Clayton, A.

(2006) Isolation and characterization of exosomes

from cell culture supernatants and biological flu-

ids, Curr. Protoc. Cell Biol., 3, 3.22.1-3.22.29, https://

doi.org/10.1002/0471143030.cb0322s30.

18. Skryabin, G.O., Komelkov, A.V., Galetsky, S.A., Bagrov,

D.V., Evtushenko, E.G., Nikishin, I.I., Zhordaniia, K.I.,

Savelyeva, E. E., Akselrod, M. E., Paianidi, I. G., and

Tchevkina, E. M. (2021) Stomatin is highly expressed

in exosomes of different origin and is a promising

candidate as an exosomal marker, J. Cell Biochem.,

122, 100-115, https://doi.org/10.1002/jcb.29834.

19. Beauregard, G., and Roufogalis, B. D. (1979) Charac-

terization of lipid-protein interactions in acetylcho-

linesterase lipoprotein extracted from bovine eryth-

rocytes, Biochem. J., 179, 109-117, https://doi.org/

10.1042/bj1790109.

20. Escudero-Cernuda,S., Eiro,N., Fraile,M., Vizoso, F.J.,

Fernández-Colomer, B., and Fernández-Sánchez,

M. L. (2025) Limitations and challenges in the char-

acterization of extracellular vesicles from stem

cells and serum, Mikrochim. Acta, 192, 311, https://

doi.org/10.1007/s00604-025-07147-4.

21. Tkach,M., and Théry,C. (2016) Communication by ex-

tracellular vesicles: where we are and where we need

to go, Cell, 164, 1226-1232, https://doi.org/10.1016/

j.cell.2016.01.043.

22. Liao,Z., Jaular, L.M., Soueidi,E., Jouve,M., Muth, D.C.,

Schøyen, T.H., Seale,T., Haughey, N.J., Ostrowski,M.,

Théry,C., and Witwer, K.W. (2019) Acetylcholinester-

ase is not a generic marker of extracellular vesicles,

EFFICIENCY OF METHODS FOR EVALUATION OF VESICLES CONCENTRATION 1387

BIOCHEMISTRY (Moscow) Vol. 90 No. 10 2025

J.Extracell. Vesicles, 8, 1628592, https://doi.org/10.108

0/20013078.2019.1628592.

23. Grigor’eva, A.E., Dyrkheeva, N.S., Bryzgunova, O.E.,

Tamkovich, S.N., Chelobanov, B.P., and Ryabchikova,

E. I. (2017) Contamination of exosome preparations,

isolated from biological fluids, Biomed. Khim., 63,

91-96, https://doi.org/10.18097/PBMC20176301091.

24. Skryabin, G., Enikeev, A., Beliaeva, A., Galetsky, S.,

Bagrov,D., Moiseenko, A., Vnukova,A., Imaraliev,O.,

Karasev, I., and Tchevkina, E. (2025) Distinctive fea-

tures of extracellular vesicles present in the gastric

juice of patients with gastric cancer and healthy sub-

jects, Int.J. Mol. Sci., 26, 5857, https://doi.org/10.3390/

ijms26125857.

25. Webber,J., and Clayton,A. (2013) How pure are your

vesicles? J.Extracell. Vesicles, 2, 19861, https://doi.org/

10.3402/jev.v2i0.19861.

26. Nelson, B. C., Maragh, S., Ghiran, I. C., Jones, J. C.,

DeRose, P. C., Elsheikh, E., Vreeland, W. N., and

Wang, L. (2020) Measurement and standardization

challenges for extracellular vesicle therapeutic de-

livery vectors, Nanomedicine (Lond), 15, 2149-2170,

https://doi.org/10.2217/nnm-2020-0206.

27. Erdbrügger, U., and Lannigan, J. (2016) Analytical

challenges of extracellular vesicle detection: a com-

parison of different techniques, Cytometry A, 89,

123-134, https://doi.org/10.1002/cyto.a.22795.

28. Zhou,E., Li,Y., Wu,F., Guo,M., Xu,J., Wang,S., Tan,Q.,

Ma, P., Song, S., and Jin, Y. (2021) Circulating extra-

cellular vesicles are effective biomarkers for pre-

dicting response to cancer therapy, eBioMedicine, 67,

103365, https://doi.org/10.1016/j.ebiom.2021.103365.

29. Pink, R. C., Beaman, E.-M., Samuel, P., Brooks, S. A.,

and Carter, D.R.F. (2022) Utilising extracellular vesi-

cles for early cancer diagnostics: benefits, challenges

and recommendations for the future, Br.J. Cancer, 126,

323-330, https://doi.org/10.1038/s41416-021-01668-4.

30. Stevic, I., Buescher,G., and Ricklefs, F.L. (2020) Mon-

itoring therapy efficiency in cancer through extra-

cellular vesicles, Cells, 9, 130, https://doi.org/10.3390/

cells9010130.

Publisher’s Note. Pleiades Publishing remains

neutral with regard to jurisdictional claims in published

maps and institutional affiliations. AI tools may have

been used in the translation or editing of this article.