Close this search box.

Effect of hypercholesterolemia on circulating and cardiomyocyte-derived extracellular vesicles – Scientific Reports

In vivo high cholesterol diet and blood collection

All procedures were approved by the National Scientific Ethical Committee on Animal Experimentation and the Semmelweis University’s Institutional Animal Care and Use Committee, and all experiments were performed in accordance with the ARRIVE guidelines. Male Wistar rats were fed either with standard (n = 11) or high cholesterol (n = 7) chow for 12 weeks. HC chow contained 2% cholesterol and 0.25% cholic acid. Animals were housed in a temperature (22 ± 2 °C)-, and humidity-controlled (50 ± 10%) room at a 12 h light/dark cycle and had free access to laboratory chow and drinking water ad libitum. After the 12 week period, the animals were anesthetized intraperitoneally with 60 mg/kg pentobarbital and their blood was collected from the abdominal aorta into Anticoagulant Citrate Dextrose-A vacuum tubes. Platelet-free plasma (PFP) was obtained by centrifugation twice with 2.500 × g at 4 °C, for 15 min as previously described73. PFP samples were stored at − 80 °C until processing (Fig. 1A). No animals were excluded from the study.

EV isolation from the plasma

Plasma EVs were isolated and purified as previously described by Onódi et al.25. In brief, extracellular vesicle isolation was performed by iodixanol (60 w/V% iodixanol in ultrapure water; Axis-Shield, Oslo, Norway) density gradient ultracentrifugation (24 h, 120.000 × g, 4 °C). The EV-rich DGUC fractions (Supplementary Figure S1) were loaded into a HiScreen Capto Core 700 column (GE Healthcare Life Sciences) and size exclusion chromatography-based purification was performed. Vezics system ( was used to implement the isolation.

In vivo metabolomics

Metabolomic analysis of 0.01 mL EV- or PFP sample was performed using a Biocrates MxP Quant 500 kit (Biocrates AG, Innsbruck, Austria) according to the manufacturer’s instructions, as detailed in the Online Supplementary Material. The analytical procedure was guided by the laboratory information management software Biocrates MetIDQ (version 7.13.11, DB109-Nitrogen-2850 Revision 31.995, Base Version 109, Runtime version 1.8.0_181, Runtime architecture amd64; Biocrates A.G., Innsbruck, Austria). Analytical system control and data acquisition were accomplished using Sciex Analyst version 1.5.3. The analytical results were evaluated by MetIDQ and results were analyzed in R74. Metabolites were excluded from the analysis if less than half of the measurements (< 7 for CTRL and < 4 for VEH) were below the limit of detection (LOD) in both groups independently for EV and plasma samples. For the analytes included, values below the LOD were imputed with a normal distribution around half of the LOD. For correlation analysis, metabolite types for which at least twelve plasma-EV measurement pairs were available were analyzed with a linear regression model. For both linear regression and statistical analysis, data were log10 transformed.

Culturing and HC treatment of AC16 cells

AC16 human cardiomyoblast cells from ATCC were cultured in Dulbecco’s Modified Eagle medium and Nutrient F12 1:1 mix (DMEM/F12) (Capricorn, Cat. no.: DMEM-12-A) supplemented with 12.5% heat-inactivated fetal bovine serum (FBS) (Corning, Cat. no.: 35-079-CV), 2 mM of L-glutamine (Corning, Cat. No.: 25-005.CI), 10 mM of HEPES (N-2-hydroxyethylpiperazine- N-2-ethane sulfonic acid) (Gibco, Cat. No: 15630-056) and 1% Antibiotic-Antimiotic Solution (Corning, Cat. No.: A5955-100ML). Cells were kept at 37 °C in a 5% CO2/95% air environment and subcultured by trypsinization (TrypLE) (Gibco, Cat. No.: 12604-021) when cells reached 90% of confluence. For HC treatment, cells were kept in medium supplemented with Refeed (Remembrane, Imola, Italy) hypercholesterolemic supplement or its vehicle (0.3% ethanol final concentration) for 48 h according to Makkos et al.75 (Fig. 2A). At the end of the treatment, cells were trypsinized and cell count was measured using a hemocytometer and viability was determined by Trypan blue (Cytiva, Cat. No.: SV3008401) staining. Cells were used only if viability reached 90%.

Oil Red O staining

AC16 cells were seeded on CELLview cell culture microscope slides (Greiner Bio-One, Austria) at a concentration of 20,000 cells/well. After 24 h, the cells were treated with HC treatment solution, then cultured for another 48 h at 37 °C in a CO2 incubator. The cells were fixed in 10% neutrally buffered formalin, washed with ultrapure water and then with 60% isopropanol. The cells were then stained with Oil Red O solution (3 mg/mL) (Sigma, USA, Cat. No.: O0625-25G) for 10 min. This was followed by another wash with isopropanol followed by ultrapure water, and the nuclei were labeled with 1:1.000 dilution of DAPI (Cell Signaling Technologies, Danvers, MA, USA, Cat. No.: 4083S) for 1 min, and the samples were covered with Prolong Gold (Thermo Scientific, Waltham, MA, USA) mounting medium and coverslips were applied. Samples were examined with a Leica SP8 confocal microscope. Oil-Red-O dye was excited at 552 nm and the emitted fluorescence was detected in the range of 645–767 nm.

Isolation of EV from AC16 cell culture supernatant

A total of 3.5 million AC16 cells were seeded in 175 cm2 culture dishes and 24 h later, the culture medium was changed to FBS-free medium. After 48 h of incubation, the culturing medium was collected in 50 mL centrifuge tubes and EVs were isolated using differential centrifugation. Cell supernatants were centrifuged at 300 × g at 4 °C for 10 min (Hettich Universal 320R; Rotor type: 1494 with Hettich 1427 adaptor). Then supernatants were centrifuged at 2,500 × g at 4 °C for 5 min (Hettich Universal 320R; Rotor type: 1494 with Hettich 1427 adaptor). The supernatants were transferred into 50 mL centrifuge tubes (Herolab Cat. no.: 253211) and centrifuged at 13,500 × g at 4 °C for 40 min (Hermle Z326K; Rotor type: 220.78 V20.). Next, they were transferred into ultracentrifuge tubes (Beckman Coulter, Cat. no.: 326823) and centrifuged at 174,900 × g at 4 °C for 3 h using an ultracentrifuge (Optima XPN-100; Rotor type: SW32Ti with adaptor 129.7). Pellets were resuspended in 120 µL phosphate-buffered saline (PBS) or 120 µL Tris-buffered saline (TBS) for atomic force microscopy (AFM). Protein concentration was measured with light absorbance at 280 nm by Implen N50 (Implen, München, Germany) nanophotometer.

Nanoparticle tracking analysis

EV samples were analyzed by Zeta View PMX 110 (Particle Metrix, Meerbusch, Germany) nanoparticle tracking analysis (NTA) machine calibrated with 100 nm polystyrene beads according to the manufacturer’s protocol. Samples were diluted in PBS to a concentration of 50–300 particles in a view of sight. At least 1 mL of sample was injected into the machine and automated measurements were performed at 11 positions throughout the measurement cell, with two readouts at each position. Positions for which the software recommended exclusion were excluded from the final evaluation.

Instrument parameters were set as: temperature 25 °C, sensitivity 85, frame rate 30 frames per second and shutter speed 100. Post-acquisition parameters were set as: minimum brightness 20, minimum size 5 pixels and maximum size 1000 pixels. Results were multiplied by the dilution factor.

For size distribution plot, the mean of all the measurements was calculated, data were smoothed with Loess regression and visualized with ± SEM.


For Western blotting, EV samples were lysed in radioimmunoprecipitation assay buffer (Cell Signaling Technologies, Danvers, MA, USA) supplemented with 1 mM of PMSF (Roche, Basel, Switzerland), and 0.1 mM of sodium fluoride, and complete protease inhibitor cocktail (Roche, Basel, Switzerland). Equal volumes of each sample were mixed with 1/4 volume of Laemmli buffer containing β-mercaptoethanol (Thermo Scientific, Waltham, MA, USA) and loaded on Tris–glycine sodium dodecyl sulfate–polyacrylamide gels (Bio-Rad, Hercules, CA, USA), and electrophoresed. Proteins were transferred onto a PVDF membrane (Bio-Rad, Hercules, CA, USA). Membranes were blocked in 5% bovine serum albumin (Bio-Rad, Hercules, CA, USA) in Tris-buffered saline containing 0.05% Tween-20 at room temperature for 2 h. Primary antibodies used were anti-TSG101 (ab83, Abcam, Cambridge, UK), anti-Alix (sc-53540, Santa Cruz Biotechnology, Dallas, Texas, USA), anti-CD81 (sc-166029, Santa Cruz Biotechnology, Dallas, Texas, USA), anti-albumin (sc-271605, Santa Cruz Biotechnology, Dallas, Texas, USA), Anti-ApoB (MABS2046, (Merck KGaA, Darmstadt, Germany), anti-fibrinogen (GTX54019, GeneTex, Irvine, Ca, USA) and anti-HSP70 (sc-66049, Santa Cruz Biotechnology, Dallas, Texas, USA). The presence of individual contaminating organelles was detected using the Organelle Detection Western Blot Cocktail (ab133989, Abcam, Cambridge, UK). The cocktail contained anti-sodium–potassium ATPase (plasma membrane), anti-ATP5A (mitochondria), anti-GAPDH (cytosol) and anti-Histone-H3 (nucleus) antibodies. Anti-mouse IgG, HRP-linked Antibody (7076s, Cell Signaling Technologies, Danvers, MA, USA) and Anti-rabbit IgG, HRP-linked Antibody (7074s, Cell Signaling Technologies, Danvers, MA, USA) secondary antibodies were used to detect the proteins. Signals were visualized using enhanced chemiluminescence kit (Bio-Rad, Hercules, CA, USA) using Chemidoc XRS + (Bio-Rad, Hercules, CA, USA) and analyzed with Image Lab software (Bio-Rad, Hercules, CA, USA).

Determination of lipid and phosphatidylcholine content of EVs

Lipid content was determined as described previously by Visnovitz et al.76. In brief, 50 mg of vanillin (Sigma, W310727) was dissolved in 50 mL of 17% phosphoric acid (Sigma, 79617) to create phosphor-vanillin reagent. 200 µL of 96% sulfuric acid was added either to 40 µL of 1,2-Dioleoyl-sn-glycero-3-phosphocoline (DOPC) (Sigma, P6354) liposome standards, or to 40 µL of EVs suspended in sterile filtered PBS. After being vortexed, samples and standards were incubated at 90 °C in a fume hood for 20 min. Tubes were cooled down and 120 µL of phospho-vanillin reagent was added to each tube and 280 µL of each sample was transferred into a 96-well plate and was incubated at 37 °C for 1 h. Absorbance was determined at 540 nm using a plate reader (Multiskan Go, Thermo Scientific). Total PC content was measured using a colorimetric assay (CS0001, Merck KGaA, Darmstadt, Germany).

AFM imaging and force spectroscopy of EVs

EV samples were diluted 100-fold with TBS buffer. One hundred µL of sample was deposited on freshly cleaved mica surface and incubated at 25 ± 1 °C for 30 min. Excess vesicles that did not adsorb on the substrate were removed by gently rinsing with ultrapure water, then 100 µL of TBS was added. These samples were examined using a Cypher ES atomic force microscope (Asylum Research, Santa Barbara, CA) using Olympus BL AC 40 TS cantilevers (nominal stiffness: 90 pN/nm, resonance frequency: 110 kHz, tip radius: 8 nm; Olympus, Japan) at 25 °C. Prior to measurements, cantilevers were calibrated by the thermal method in air77. Then, images were taken in non-contact mode, at 0.5–1 Hz line scanning frequency in buffer, oscillating the cantilever at its resonance frequency. Contact mode measurements were then performed for in situ force spectroscopy on selected spherical vesicles with topographical height exceeding 15 nm. During force spectroscopy, the cantilever was moved at speed of 1 µm/s from a pre-set height towards the vesicle until a load threshold of 100 pN was reached. It was then immediately retracted at the same speed. Deflection of the cantilever and thus force as a function of cantilever position (force-indentation curve or force curve) were recorded during the process.

Image and force curve analysis was implemented using the built-in algorithms of AFM driving software (IgorPro, WaveMetrics Inc., Lake Oswego, OR). Data were collected from a minimum of 5 individual experiments for each treatment group. Image analysis included only the particles with at least 10 nm height. Smaller particles were classified as non-vesicular structures and excluded from the analysis. EV diameter was determined as the diameter of a circle with an area equal to the surface area of the vesicle projected to the substrate. Vesicle height is defined as the difference between the highest point of the vesicle and the substrate. Young modulus of vesicles was obtained by fitting force-vesicle indentation curves from 0 (contact point) to 100 pN loading force with the modified Hertz model78.

Mass spectrometry of EV proteins

AC16 EV isolates were vacuum-dried and transferred on dry ice to UCD Conway Institute Mass Spectrometry Resource. Samples were then resuspended in 50 µL of 50 mM Tris HCl (Thermo Scientific, Waltham, MA, USA), sonicated, and protein concentration was measured using a bicinchoninic acid (BCA) assay kit (Thermo Scientific, Waltham, MA, USA). Samples were normalized to protein concentration and dissolved in 6M Urea (Thermo Scientific, Waltham, MA, USA). The samples were then reduced by adding 8 mM of dithiothreitol (Merck KGaA, Darmstadt, Germany) and stirred at 30 °C at 1000 rpm for 30 min. Samples were then carboxylated by adding 20 mM of iodoacetamide (Merck KGaA, Darmstadt, Germany) and stirred at 30 °C at 1000 rpm for 30 min. Next, samples were diluted with 50 mM of Tris HCL to reduce urea concentration below 2 M. Samples were digested by Trypsin (Promega (Corporation, Madison, WI, USA) at 37 °C with 1000 rpm in 1:30 enzyme:sample ratio overnight. Digestion was terminated by adding formic acid (Thermo Scientific, Waltham, MA, USA) to 1% final concentration. Samples were purified on Empore Solid Phase Extraction membranes (Merck KGaA, Darmstadt, Germany), eluted in 60% acetonitrile and 0.1% Trifluoroacetic acid, vacuum dried and in 0.1% formic acid and each sample was loaded onto an Evosep tip (Evosep Biosystems, Buchwaldsgade, Denmark). The Evosep tips were placed in position on the Evosep One, in a 96-tip box. The autosampler is configured to pick up each tip, elute and separate the peptides using a set chromatography method (30 samples a day)79. The mass spectrometer was operated in positive ion mode with a capillary voltage of 1.700 V, dry gas flow of 3 l/min and a dry temperature of 180 °C. All data were acquired with the instrument operating in trapped ion mobility spectrometry (TIMS) mode. Trapped ions were selected for ms/ms using parallel accumulation serial fragmentation (PASEF). A scan range of (100–1700 m/z) was performed at a rate of 5 PASEF MS/MS frames to 1 MS scan with a cycle time of 1.03s80. The following chromatography buffers were used: Buffer B: 99.9% acetonitrile, 0.1% formic acid. Buffer A: 99.9% water, 0.1% formic acid. Raw mass spectrometric data have been deposited to the ProteomeXchange Consortium via the PRIDE81 partner repository with the dataset identifier PXD044594.

Analysis of proteomics data

The raw data were searched against the Homo sapiens subset of the UniProt Swissprot database (reviewed, 12.11.2021) using the search engine Maxquant (release with specific parameters for trapped ion mobility spectra data dependent acquisition (TIMS DDA). Each peptide used for protein identification met specific Maxquant parameters, i.e., only peptide scores that corresponded to a false discovery rate (FDR) of 0.01 were accepted from the Maxquant database search. The normalized protein intensity of each identified protein was used for label-free quantitation (LFQ)83. LFQ intensities were analyzed according to the protocol of Tyanova & Cox37, using R74. Proteins that were labelled as reversed or potential contaminants or only identified by site were excluded (see supplementary material). Proteins that were detected in less than half of the samples in all groups were also excluded. Every other protein was defined as identified and processed for quantiative analysis. Data were log2 transformed and the missing values were imputed with normal distribution around the detection limit. For statistical analysis, log2 transformed data were used. To calculate fold changes, non-transformed data was used. STRING software84,85 was used to visualize protein interaction networks and for gene ontology (GO) enrichment analysis.

Monocyte activation assay

THP1 human monocytes expressing apoptosis-associated speck-like protein containing a CARD domain fused by green fluorescent protein (ASC-GFP, InvivoGen, Toulouse, France) were maintained in THP1 medium consisting of RPMI 1640 medium (Gibco, 21875-034), 10 V/V% heat inactivated FBS (Corning, 35-079-CV), 1% L-glutamine (Corning, 30-004-CI), 1% antibiotics-antimycotics (Corning, 25-005-CI), and 1% HEPES (Gibco, 15,-630-080) at a maximum of 6 × 105 cells/mL in a T175 flask. THP1 medium was also supplemented with 100 μg/mL Zeocin (InvivoGen, ant-zn-0.5) for transgene selection at every second passage. Cells were grown at 5% CO2 95% air, at 37 °C. All experiments were performed within 10 passages and repeated at least four times.

1 × 106 THP1-ASC-GFP cells in 24-well plates were treated with either 100 ng/mL LPS, 107–109 particles/mL HC-EV, CTRL-EV or VEH-EV, corresponding EV supernatants, or whole medium from AC16 cells overnight, and then cells were collected onto ice for flow cytometry analyses.

Cells were resuspended in PBS and fixed with 1% PFA at 4 °C for 10 min and washed twice. Flow cytometry was performed using BD FACSCalibur (BD Biosciences, San Jose, CA, USA) and evaluated using Flowing software (Turku Bioscience, Turku, Finland). THP1-ASC-GFP cells were gated first on live cells based on SSC and FSC followed by GFP+ population analysis. For each measurement, 10,000 events were counted.

Quantitative PCR

After treatment of THP-1-ASC-GFP cells as described above, cells were lyzed in quiazol lysis reagent (Quiagen, Germantown, MD, USA) and stored at − 80 °C. Total RNA isolation, cDNA synthesis and qPCR measurements were performed as described earlier86. Measurements were performed on a LightCycler 480 Real-Time PCR System (Roche Diagnostics, Basel, Switzerland) using LightCycler® RNA Master SYBR Green I reagent (Roche Diagnostics, Basel, Switzerland) with primers presented in Supplementary Table S2, with the following protocol: Initiation: 95 °C 2 min; Amplification: (45x) 95 °C 5 s, 57 °C 10 s, 72 °C 20 s. Data were analyzed by ΔΔCt calculation method according to Schmittgen and Livak 200887, using Hypoxanthine Phosphoribosyltransferase 1 (HPRT) as a housekeeping gene.


For statistical analysis and data visualization, the programming language R with ggplot274,88 was used. To create figures, Adobe Illustrator (Adobe, San Jose, CA, USA) was used. For statistical analysis, corresponding parametric statistical probes were applied with a significance level of 0.05. In more detail, multiple t-tests with Benjamini–Hochberg false discovery rate (FDR) adjustment were used for metabolomics data. To analyze AC16 EVs, including NTA, AFM and protein concentration data, and for the analysis of flow cytometry data on THP1-ASC-GFP cells, analysis of variance (ANOVA) with Tukey’s post-hoc test was applied. ANOVA with FDR adjustment followed by Tukey’s post-hoc test was used to analyze proteomics data. For GO enrichment analysis with FDR correction, STRING software was used84,85. Data are presented as mean ± standard error of the mean.

Ethics approval

All procedures were approved by the National Scientific Ethical Committee on Animal Experimentation and the Semmelweis University’s Institutional Animal Care and Use Committee (H-1089 Budapest, Hungary) in accordance with NIH guidelines (National Research Council (2011), Guide for the Care and Use of Laboratory Animals: Eighth Edition) and permitted by the government of Food Chain Safety and Animal Health Directorate of the Government Office for Pest County (project identification code: PE/EA/1912-7/2017; date of approval: November 2017).